Canine Coronavirus Infection Modulates the Biogenesis and Composition of Cell-Derived Extracellular Vesicles

Abstract

Coronavirus (CoV) has persistently become a global health concern causing various diseases in a wide variety of hosts, including humans, birds, and companion animals. However, the virus-mediated responses in animal hosts have not been studied extensively due to pathogenesis complexity and disease developments. Extracellular vesicles (EVs) are widely explored in viral infections for their intercellular communication, nanocarrier, and immunomodulatory properties. We proposed that coronavirus hijacks the host exosomal pathway and modulates the EV biogenesis, composition, and protein trafficking in the host. In the present study, Crandell-Rees feline kidney (CRFK) cells were infected with canine coronavirus (CCoV) in an exosome-free medium at the multiplicity of infection (MOI) of 400 infectious units (IFU) at various time points. The cell viability was significantly decreased over time, as determined by the 3-(4, 5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay. Post-infection EVs were isolated, and transmission electron microscopy (TEM) showed the presence of small EVs (sEVs) after infection. NanoSight particle tracking analysis (NTA) revealed that EV sizes averaged between 100 and 200 nm at both incubation times; however, the mean size of infection-derived EVs was significantly decreased at 48 h when compared to uninfected control EVs. Quantitative analysis of protein levels performed by dot blot scanning showed that the expression levels of ACE-2, annexin-V, flotillin-1, TLR-7, LAMP, TNF-α, caspase-1, caspase-8, and others were altered in EVs after infection. Our findings suggested that coronavirus infection impacts cell viability, modulates EV biogenesis, and alters cargo composition and protein trafficking in the host, which could impact viral progression and disease development. Future experiments with different animal CoVs will provide a detailed understanding of host EV biology in infection pathogenesis and progression. Hence, EVs could offer a diagnostic and therapeutic tool to study virus-mediated host responses that could be extended to study the interspecies jump of animal CoVs to cause infection in humans.

Keywords: animal coronavirus; coronavirus; extracellular vesicles; immunomodulation.

Figure 1

Crandell–Rees feline kidney (CRFK) cell viability after canine coronavirus (CCoV) infection. CRFK cells were infected with CCoV at MOI of 400 infectious units (IFU) in an exosome-free medium and incubated for 48 h and 72 h; post-incubation cells were further incubated with 3-(4, 5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) dye solution at 37 °C for 3–4 h and absorbance was read at 570 nm. Statistical analysis of obtained data points was performed using one-way analysis of variance (ANOVA) with Tukey post hoc analysis. Statistical significance is indicated by the mean ± standard deviation (SD) and is defined as p ≤ 0.01 (**).

Figure 2

CRFK-derived extracellular vesicle (EV) characterization after CCoV infection. The morphology, size distribution (nm), and concentration (particles per mL) of CRFK-derived control EVs and infection EVs after CCoV infection were characterized. (A) transmission electron microscopy (TEM) images showing morphologies of CRFK-derived control sEVs at 48 h (small EVs indicated by red arrow); (B) nanosight particle tracking analysis (NTA) showing size distribution pattern of CRFK-derived EVs; (C) mean particle size; (D) particle concentration per mL, at 48 h and 72 h; (E) graphs showing densitometric analysis of dot blot (Supplementary Figure S2A) of classical EV biomarker, cluster of differentiation (CD)63, via Bio-Rad imaging program and GraphPad Version 5 software in isolated EVs at 48 h and 72 h; (F) total DNA, RNA, and protein content of CCoV-infected CRFK-derived control and infected EVs at 48 h and 72 h. Statistical analysis of obtained data points was performed using one-way ANOVA with Tukey post hoc analysis. Statistical significance is indicated by the mean ± SD and is defined as p ≤ 0.05 (*).

Figure 3

Host receptor and virus-specific protein expression levels after CCoV infection. Schematic (A) western blot and (B) dot blot showing expression of coronavirus host receptor, angiotensin-converting enzyme (ACE2), at 48 h and 72 h in the control and infection EVs; (C) graphs showing densitometric analysis of dot blot of ACE2 via Bio-Rad imaging program and GraphPad Version 5 software; (D) schematic dot blot images and densitometric graph for retroviral protein, syncytin-1, via Bio-Rad imaging program and GraphPad Version 5 software in isolated EVs at 48 h and 72 h. Statistical analysis of obtained data points was performed using one-way ANOVA with Tukey post hoc analysis. Statistical significance is indicated by the mean ± SD and is defined as p ≤ 0.05 (*).

Figure 4

The effect of CCoV infection on EV biogenesis and membrane proteins. Graphs showing densitometric analysis of dot blot of (A) flotillin-1 at 48 h and 72 h (Supplementary Figure S2B), (B) biogenesis protein annexin-V at 48 h and 72 h (Supplementary Figure S2C), (C) multifunctional transmembrane receptor CD44 at 48 h and 72 h (Supplementary Figure S2D), and (D) canine-specific adhesion molecule E-cadherin at 48 h and 72 h (Supplementary Figure S2E), in isolated control and infection EVs. Statistical analysis of obtained data points was performed using one-way ANOVA with Tukey post hoc analysis. Statistical significance is indicated by the mean ± SD and is defined as p ≤ 0.05 (*) and p ≤ 0.01 (**).

Figure 5

Activation of pathogen recognition molecules in response to CCoV infection. Graphs showing densitometric analysis of dot blot of (A) double-stranded RNA sensor TLR3 level at 48 h and 72 h (Supplementary Figure S2F), (B) lipopolysaccharide sensor TLR6 level at 48 h and 72 h (Supplementary Figure S2G), and (C) RNA sensor TLR7 level at 48 h and 72 h (Supplementary Figure S2H), via Bio-Rad imaging program and GraphPad Version 5 software in isolated control and infection-derived EVs. Statistical analysis of obtained data points was performed using one-way ANOVA with Tukey post hoc analysis. Statistical significance is indicated by the mean ± SD and is defined as p ≤ 0.05 (*).

Figure 6

The effect of CCoV infection on heat shock protein and caspases. Graphs showing densitometric analysis of dot blot of (A) heat shock protein HSP100 (Supplementary Figure S2I) at 48 h and 72 h, (B) caspase-1 (Supplementary Figure S2J) level at 48 h, and (C) caspase-8 level (Supplementary Figure S2K) at 48 h and 72 h, in isolated control and infection-derived EVs. Statistical analysis of obtained data points was performed using one-way ANOVA with Tukey post hoc analysis. Statistical significance is indicated by the mean ± SD and is defined as p ≤ 0.05 (*), p ≤ 0.01 (**) and p ≤ 0.001 (***).

Figure 7

Activation of immune response after CCoV Infection. Graphs showing densitometric analysis of dot blot of (A) TNF-α level at 48 h and 72 h (Supplementary Figure S2L), (B) TGIF-1 level at 48 h and 72 h (Supplementary Figure S2M), (C) TGIF-2 level at 48 h and 72 h (Supplementary Figure S2N), (D) MCCL22 level at 48 h and 72 h (Supplementary Figure S2O), and (E) iNOS level at 48 h and 72 h (Supplementary Figure S2P), in isolated control and infection-derived EVs. Statistical analysis of obtained data points was performed using one-way ANOVA with Tukey post hoc analysis. Statistical significance is indicated by the mean ± SD and is defined as p ≤ 0.05 (*).