Ebola virus mucin-like glycoprotein (Emuc) induces remarkable acute inflammation and tissue injury: evidence for Emuc pathogenicity in vivo
- 1.4k Downloads
Ebola virus (EBOV) is one of the most virulent pathogens to humans. Recently, the largest-ever outbreak of Ebola virus disease (EVD) in West Africa, 2013–2016, resulted in the unprecedented damage to human health and social economy. However, there is currently no licensed vaccine or antiviral available against EVD. Clinically, EBOV infection can result in exaggerated inflammatory responses and multiorgan damage, although the pathogenesis (including the inflammatory pathogenesis) and the viral virulence factor(s) involved in the excessive inflammation induction and viral pathogenicity are largely unclear. In vitro studies have suggested that the envelope glycoprotein (GP1,2) of EBOV mediating virus entry may also directly contribute to the viral pathogenesis, as cell surface expression of GP1,2 can induce adherent cell rounding and detachment and sterically mask some cell surface molecules, such as β1-integrin (reviewed in Ning et al., 2017). GP1,2 is composed of dimerized subunits GP1 and GP2, locating on the cell or virion membrane; an additional notable feature of GP1,2 is that it contains a heavily O-glycosylated mucin-like region within GP1. The functional mapping of GP1,2-induced cellular morphological changes showed that both the mucin-like region and cell surface expression by a transmembrane domain (TMD) are required (Ning et al., 2017; Francica et al., 2009; Takada et al., 2000; Yang et al., 2000; Chan et al., 2000), suggesting a role of the surface expressed EBOV mucin-like glycoprotein (Emuc) in the GP1,2 effect. However, the biological effects of the glycoproteins have not been investigated systematically in vitro and in vivo, and there is no direct evidence in vivo available for the potential pathogenicity of the glycoproteins.
Next, we conducted gene delivery of the adenoviral vectors to BALB/c mice for further exploring the effects of Emuc in vivo. The viral vectors (1 × 108 PFU) were respectively injected into the adductor magnus muscles of BALB/c mice with local hair depilated. Intramuscular route was chosen for real-time in situ observation; additionally, the infection through broken skins and muscles is likely one of the main EBOV transmission routes (Singh et al., 2015; Zaki et al., 1999). In the following days, the clinical feature was closely monitored. Intriguingly, as early as 24 h post infection (hpi), the mice inoculated with ADV-Emuc-TVA showed piloerection and nervousness and the injected hind leg muscles of most mice (7/9) showed mild redness and contracture (Fig. 1F and data not shown). In the preliminary experiments, we observed that the clinical manifestations could last for days, gradually reduced from 72–96 hpi and were invisible after ~1 week (data not shown). Considering that the viral vectors are replication-deficient, the quick recovery of the mice could be associated with the rapid clearance of the injected viral vectors as well as the degradation of expressed proteins and tissue repair. By contrast, no noticeable clinical signs of illness were observed throughout the infection course in the control groups (Fig. 1F). These observations suggest that Emuc expression specifically caused disease, though temporarily, in the context of local transduction by the replication-deficient adenovirus.
Following the euthanization of mice at 48 hpi (n = 3) or 72 hpi (n = 6), the injected muscles and controls were dissected for further analyses. Firstly, EGFP signals could be detected in the muscles inoculated with viruses under an animal imaging system (Fig. 1G), manifesting the successful transduction of the adenoviral vectors. Subsequently, the viral infection and transgene expression were further analyzed by immunofluorescence staining of paraffin sections using the antibodies against TVA or EGFP. The expression of TVA and EGFP could be detected in multiple cell types (such as myocytes, fat cells, fibroblasts, undifferentiated mesenchymal cells, etc.) of the muscular tissues and connective tissues (Fig. S2 and data not shown), consistent with the broad host cell range of adenoviral vectors. To analyze the histopathological changes, sections were further stained by hematoxylin and eosin (H&E). Intriguingly, the histopathologic examination revealed that Emuc expression by the ADV-Emuc-TVA transduction resulted in massive inflammatory cell infiltration and notable tissue damage in the muscular tissues and connective tissues in the samples at both 48 and 72 hpi, while the infections of the control viruses by contrast only caused slight confined inflammation (Figs. 1H and S3). On close examination, ADV-Emuc-TVA infection induced the infiltration of numerous inflammatory cells (including lymphocytes, monocytes, and neutrophils) into the muscular tissues and connective tissues, particularly the muscular fibers, endomysium, perimysium, connective tissue matrices, and adipose tissues (Figs. 1H, S3, and S4). Moreover, upon the ADV-Emuc-TVA infection, substantial disintegration and necrosis of muscle fibers were evident, congestion and expansion of blood capillaries in the diseased area could be observed, and hyperplasia of connective tissues accompanied with the inflammatory cell infiltration and small abscess formation were also distinct pathological changes (Fig. S4). Meanwhile, the IFA and H&E analyses of continuous slices showed that in the Emuc-expression positive areas, diffuse inflammation and tissue damage were always notable, and vice versa (Fig. 1I), indicating the close association of Emuc expression and the pathological changes. To further quantify the pathological changes, the degrees of pathological changes of muscular tissues and connective tissues in comparison to mock-infected samples were accordingly scored. The control viruses caused minimal to mild pathological changes which were comparable between the ADV and ADV-TVA groups, reflecting the inevitable host immune response against adenoviral vector transduction; however, ADV-Emuc-TVA induced distinct and much severer pathological changes compared with the control groups, further confirming the specific pathogenicity of Emuc in vivo (Figs. 1J and S5). Additionally, according to the severity scores, the pathological changes of the control groups, especially the mild inflammation in the connective tissues, seemed to be further alleviated at 72 hpi, compared to 48 hpi, whereas the severe histopathologic changes of Emuc expression group were more persistent (Figs. 1J and S5). Thus, the differences of the pathological severity degrees between ADV-Emuc-TVA and the control groups were more evident at 72 hpi although the specific pathogenic effects caused by Emuc could be obviously observed at both 48 and 72 hpi. Given this, 72 hpi can be a better time point for evaluating the specific pathogenicity of Emuc in this experimental model. Taken together, these data demonstrate that even in the context of local and transient expression by the replication-deficient adenoviral vector, Emuc can cause remarkable acute inflammation and tissue injury, suggesting that Emuc likely is a notable pathogenic factor of EBOV in vivo.
In vitro, Emuc expression caused rounding and detachment but not death of cultured adherent cells, and i.e., the morphologically changed cells remained viable (Simmons et al., 2002); however, in mouse muscles, Emuc induced not only cell pathologic changes but also cell death, revealing the distinct physiological virulence of Emuc and also highlighting the importance of in vivo investigation. Besides the cytopathic effects (CPE) and tissue damage, inflammation is also a significant pathologic effect mediated by Emuc, reminiscent of the excessive inflammation during EBOV infection in clinic (Feldmann and Geisbert, 2011; Baize et al., 2002; Cilloniz et al., 2011) and the capacity of GP1,2 and shed GP (both containing Emuc) to activate inflammatory reaction of immune cells in vitro (Ning et al., 2017). As shown in Figure 1I, many muscle fibers which seemed not to be infected by ADV-Emuc-TVA but were infiltrated by the inflammatory cells were damaged as well, suggesting that Emuc-mediated inflammation also plays important roles in the cell and tissue injury. Taken together, we consider that the tissue lesion could be mediated directly by the cytotoxicity of Emuc and indirectly by the host inflammatory responses triggered by Emuc although it remains undefined to what extents the two effects of Emuc, respectively, contribute to the tissue damage. Additionally, tissue injury and inflammation can trigger and promote each other under the physiological conditions. These effects likely combinedly contribute to the Emuc pathogenic process in vivo (Fig. S6).
As a highly pathogenic virus, EBOV is classified as a biosafety level-4 pathogen, and the manipulations of viral infections are strictly limited by the special containment facilities. As demonstrated here, in vivo gene transfer by adenoviral vectors may represent a safe and convenient experimental model which is especially valuable in the physiological function elucidation of the individual virulent proteins encoded by highly dangerous pathogens including EBOV and the development of antivirals specifically targeting the certain virulent proteins such as Emuc identified in this study. The full-length GP of EBOV has been generally used as the immunogen in vaccine researches (Ohimain, 2016) although Emuc is an unnecessary part for GP1,2-mediated entry and is highly variable in amino acid sequence across EBOV strains (Jeffers et al., 2002). Given the virulence of Emuc as shown here, it will be interesting to investigate whether Emuc-deleted GP is a safer and better immunogen. In summary, the present study demonstrated that Emuc can not only induce morphological change of adherent cells in vitro but also distinct cell and tissue damage and acute inflammation in mouse muscles, revealing and characterizing the Emuc pathogenicity both in vitro and in vivo. These findings provide direct evidence of the Emuc pathogenicity in vivo for the first time and also critical clues on EBOV pathogenesis and particularly the inflammatory pathogenesis.
We thank the Core Facility and Technical Support of Wuhan Institute of Virology for technical assistants. This work was supported by the National Natural Science Foundation of China (Grant Nos. 31621061 and 31600144), the National Basic Research Program (973 Program) (No. 2013CB911101), the Science and Technology Basic Work Program (2013FY113500), the National Key Research and Development Program of China (2016YFC1200400 and 2016YFE0113500), the strategic priority research program of the Chinese Academy of Sciences (XDPB0301), the Hubei Provincial Natural Science Foundation of China (2016CFB124), and the “One-Three-Five” Research Program of Wuhan Institute of Virology.
HW supervised the study; YJN conceived the study and designed the experiments; ZK, YJN, JX, YQM and KF performed experiments; ZK, YJN, JX, DL, MW, YZ, ZH and HW analyzed the data; FD, YQM, MW, YZ and ZH provided the reagents or materials and contributed to the completion of the study; YJN wrote the manuscript. All the authors reviewed the results and approved the final version of the manuscript.
Yun-Jia Ning, Zhenyu Kang, Jingjun Xing, Yuan-Qin Min, Dan Liu, Kuan Feng, Manli Wang, Fei Deng, Yiwu Zhou, Zhihong Hu, and Hualin Wang declare that they have no conflict of interest. All institutional and national guidelines for the care and use of laboratory animals were followed.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.