Fig 1.
CVB3 effectively infected virus receptor-negative cells and tissues.
(A and B) Caco-2 cells were infected with coxsackievirus B3 at 0.05 TCID50 for 24h. The purified pellet was isolated from culture supernatant by differential ultracentrifugation(A) and analyzed by western blot with CVB3 structural protein VP1 and VP2 specific-antibody. (C) The ultracentrifugal pellet containing CVB3 virion was analyzed by transmission electron microscope. Bar = 50nm. (D) Western blot analysis of the main receptor CAR and co-receptor DAF in various mouse organ tissues. (E) Real-time PCR data for the copy number of CVB3 RNA in different mouse organ tissues after infecting with CVB3 that isolation by differential ultracentrifugation. Six weeks-old of mice were orally inoculated after fasting at 5×106 TCID50 per mouse for 24h. n = 5. (F, G and H) The expression levels of CAR and DAF in different primary immune cells or cell lines were determined by western blot analysis(F), RT-PCR(G) and flow cytometry(H). (I) Real-time PCR data for the copy number of CVB3 RNA in different primary immune cells or cell lines after incubating with ultracentrifugal CVB3 pellet. Data are shown as mean ± SD of three independent.
Fig 2.
CVB3 entry into receptor-negative cells depended on exosomes.
(A) Caco-2 cells were infected with coxsackievirus B3 at 0.05 TCID50 for 24h. The purified pellet containing exosomes and virion were isolated from culture supernatant by differential ultracentrifugation. (B) Western blot analysis of the purified pellet and pellet-deleted supernatant by CVB3 VP1 and exosomal markers CD9 and Alix. (C, D and E) The purified pellets were isolated from supernatant of CVB3-infected wild-type cells(WT cells) or knockout cells(KO cells). Equal number of WT or KO Caco-2 cells were infection with 0.05 TCID50 of CVB3 for 24h and the supernatants were collected, followed by differential ultracentrifugation(C). The protein levels of CVB3 VP1, exosomal markers CD9 and CD63 in different pellet were measured by western blot analysis(D). And the comparison of CVB3 RNA copy number by real-time PCR analysis(E). Data are shown as mean ± SD of three independent experiments. (F) Western blot analysis of viral protein VP1 in different type of cells after incubating with purified pellets that isolated from CVB3-infected WT cells or Rab27a-KO cells. Purified pellet containing exosomes and virion was isolated from equal numbers of infected WT cells or Rab27a-KO cells and then incubated each cell type for 12h.
Fig 3.
Exosomes as carrier vehicles for CVB3 virions.
(A) Schematic presentation of immuno-magnetic isolation method to separate microvesicles(MVs) virion and exosomes from CVB3-infected Caco-2 cells. Caco-2 cells were infected with CVB3 at a 0.05 TCID50 for 1h, then the cells were washed and switched to medium supplemented with EV-depleted FBS for the production of MVs, exosomes and virus particles. (B and C) The purified exosomes and MVs were analyzed by Nanoparticle Tracking Analysis(NTA)(B) and transmission electron microscopy(C). (D) Purified exosomes, MVs and CVB3 virion derived from infected-cells above described were analysed by western blot with exosomal markers CD9, CD63 and Alix, non-exosomal markers GM130 (Golgi marker) and calnexin (ER marker), MVs markers Annexin A1 and A5. CVB3-infected cell lysates as positive control. (E) The CVB3 virion pellet isolated by immuno-magnetic method was were analyzed by NTA. The exosomes-CVB3 attachment section was indicated with the black arrow. (F, G and H) The iodixanol density gradient procedure to separate the pellet that from ultracentrifugal supernatants of CVB3-infected cells. A 10–50% Optiprep gradient was collected in 10 fractions(F1-F10) and analysed by western blot. CVB3 virion and exosomes was determined using specific anti-VP1 and anti-CD9, respectively(H). (I) Fractions 6–7 from Optiprep gradient purification were pooled, and transmission/scanning electron microscopy analysis of exosomes-CVB3 attachment. Viral particles were indicated with the white arrow. (J) Quantitation of CVB3 virion coupled with exosomes from images of electron microscopy in different cell lines. n = 25 exosomes from three independent experiments. Data are shown as mean ± SD. (K and L) The molecular weight cut off (MWCO) of ultrafiltration column(50kDa) was used to concentrate culture medium, followed by anti-VP1 beads or blank-beads immuno-selection. Then the CVB3 pellet and blank-beads pellet was determined by western blot analysis. (M and N) RT-PCR analysis of CVB3 genomic RNA in purified exosomes and MVs that isolated from CVB3-infected Caco-2 cells by differential ultracentrifugation according to description above. Total RNA extracted from CVB3-infected cells as positive control. (O) Western blot analysis of viral protein VP1 in recipient 293T cells. The 293T cells were incubated with CVB3 virion, purified exosomes or MVs that isolated from CVB3-infected Caco-2 or HL-1 cells for 24h.The whole cell lysates as determined by specific VP1 antibody. All data are presented as the mean± SD of three independent experiments.
Fig 4.
Characterization of exosome-CVB3 attachment.
(A and B) The proportion of exosomes-CVB3 complex in total exosomes was determined by NTA. Fractions 5–7 from Optiprep gradient purification were pooled, and subjected to immuno-magnetic selection using VP1 antibody. Different symbols correspond to independent experiments. (C) The numbers of exosomes from WT Caco-2 cells or TSG101-KO Caco-2 cells with or without CVB3 infection was determined by NTA. Culture supernatants was collected from equal number of WT or KO Caco-2 cells at 24h post-infection with a 0.05 TCID50 of CVB3. (D) Quantitation of CVB3 virion bind with exosomes from images of electron microscopy in WT or KO Caco-2 cells. n = 20 exosomes from three independent experiments. (E and F) CVB3 virion mixed with purified exosomes at 4°C or 37°C temperature to form exosomes-virion complex followed by immuno-selection with anti-CD9. The exosomes-CVB3 pellets from different temperature condition were determined using viral anti-VP1 and anti-CD9 through western blot. Data are shown as mean ± SD of three independent experiments. (**p<0.01, ***p<0.001).
Fig 5.
Exosome coupling with CVB3 virions depended on CAR but not DAF.
(A and B) The exosomes and MVs derived from CVB3-infected cells or mock cells were analyzed by western blot with antibodies of viral receptors(CAR and DAF), exosomal markers(CD9) and MVs markers(Annexin A1). EVs were collected from equivalent amounts of culture medium, conditioned by equal numbers of cells, for equal lengths of time. The whole cell lysates as positive control. (C) Western blot analysis of CAR and DAF in stable knockdown Caco-2 cells(CARKD cells and DAFKD cells). (D and E) Purified exosomes derived from WT cells, CARKD Caco-2 cells or DAFKD Caco-2 cells were incubated with free CVB3 virion, followed by immuno-selection with anti-CD9(D). Then each group of the precipitated pellets were analysed by western blot (E). (F) Western blot analysis of CAR and DAF in stable overexpression Caco-2 cells(CARhigh cells and DAFhigh cells). (G and H) Purified exosomes derived from WT cells, CARhigh Caco-2 cells or DAFhigh Caco-2 cells were incubated with free CVB3 virion, followed by immuno-selection with anti-CD9 (D). Then each group of the precipitated pellets were analyzed by western blot (H). (I and J) Purified exosomes derived from WT Caco-2 cells were incubated with specific anti-CAR and then treated with CVB3-GFP virion, followed by CD9 immuno-selection. The PBS or IgG incubation as the negative control treatment(I). The relative fluorescence units (RFU) of each group was determined by Microplate Reader at 488nm of excitation wavelength (J). Data are shown as mean ± SD of three independent experiments. (***p<0.001). (K and L) Purified exosomes from HT-29 or RD cells were incubated with free CVB3 virion, followed by immuno-selection with anti-CD9(K). Then each group of the precipitated pellets were analysed by western blot (L). (M) Confocal co-localization analysis of CVB3 virion(green) and exosomal markers (CD9 and CD63, red) in Caco-2 cells. Caco-2 cells were infected with CVB3-GFP at 0.05 TCID50 for 24h and then incubated with anti-CD9 or anti-CD63 followed by incubated with fluorescent dye-tagged secondary antibody. Nuclei were stained with DAPI (blue). Bar = 10μm. All data are presented as the mean± SD of three independent experiments.
Fig 6.
Exosomes carrying CVB3 exhibited highly efficient infection of receptor-negative cells through various endocytic pathways.
(A and B) Purified exosomes derived from WT cells or CARKD Caco-2 cells were incubated with CVB3 virion for 4h. Then each group of exosomes-CVB3 complex was treated CAR-knockdown 293T cells(293TCAR-KD cells) for different time-point(A). The copy number of viral RNA in 293TCAR-KD cells was determined at various time points by real-time PCR analysis(B). Data are shown as mean±SD of three independent experiments. (C, D and E) Purified exosomes derived from WT cells or CARKD Caco-2 cells were incubated with CVB3-GFP virion, then each group of exosomes-CVB3-GFP complex was treated NIH 3T3 cells for different time-point(C). The condition of CVB3-infected cells was determined by flow cytometry analysis(D) and western blot(E). (F) Determination of virus titers by plaque assays. Each group of exosomes-CVB3 mixture was incubated monolayers of NIH 3T3 cells for 72 h, and then plaque assays were carried out to determine virus titers. Representative plaque formation on 3T3 cell monolayers were showed. (G, H and I) Equal copy number of CVB3-GFP virion mixed with CARKD-exo or WT-exo and then injected into the tail veins of BALB/c mice(n = 5)(G). The tissue paraffin sections were examined at 24h post-injection for virion distribution(green) by immunofluorescence staining(H).The fluorescence intensity of GFP signal was examined through ImageJ software and a minimum of six different fields were measured(I). Data are shown as mean±SD of three independent experiments. (J and K) Confocal microscopy analysis of different routes for exosomes carrying CVB3 enter into recipient 293TCAR-KD cells. Different drug or blocking-antibody were applied to 293TCAR-KD cells followed by treating with free CVB3-GFP at 0.1 TCID50 or exosomes-CVB3-GFP complex(contain 0.1 TCID50 of CVB3-GFP) for 4h. DiI was used to label the cell membrane(Red)(J). The fluorescence intensity of accumulation GFP puncta was examined through ImageJ software and a minimum of six different fields cells were counted(K). Bar = 10μm. (L)Real-time PCR analysis of CVB3 RNA copy number in 293TCAR-KD cells. After pretreating with drug inhibitors or blocking-antibodies, the cells were infected with free CVB3-GFP at 0.1 TCID50 or exosomes-CVB3-GFP complex(contain 0.1 TCID50 of CVB3-GFP) for 4h. (M and N) Purified exosomes from HeLa cells after transfecting with CAR-orange fluorescent fusion protein vector(CAR-OFP vector) mixed with CVB3-GFP virion, followed by incubating 293TCAR-KD cells for 2h(M). Confocal co-localization analysis of CVB3 virion(green) and CAR receptor (red) in cells. Arrows point to CVB3-CAR colocalization(N). Bar = 10μm. All the data are shown as mean ± SD of three independent experiments. (*p<0.05, **p<0.01, ***p<0.001, ns: no significance).
Fig 7.
Inhibition of exosomes coupling with CVB3 attenuated the pathogenetic events in vivo.
(A and B) The exosomes isolated from wild-type (WT) or Rab27a-KO mice of serum was quantified using NTA(A) and western blot(B). Each group contained 5 mice. (C, D and E) Real-time PCR data for the comparison of CVB3 RNA expression level in splenic T(C), splenic B(D) and monocytes(E) cells in WT or Rab27a-KO mouse at various time points. Six weeks-old of Rab27a-KO mice or WT mice were orally inoculated after fasting at 5×106 TCID50 per mouse. n = 5 biologically independent animals per group. (F and G) The percentage of CVB3-infected splenic T(F) and splenic B(G) cells were determined by flow cytometric analysis at indicative time points. Cells were stained with anti-CVB3 and Alexa Fluor 488-conjugated anti-mouse IgG antibody. (H, I and J) Real-time PCR analysis of inflammatory factors expression in cardiac tissue of CVB3-infected mice or mock mice at various time points. (K and L) Left ventricular ejection fraction(LVEF) and left ventricular fractional shortening(LVFS) of CVB3-infected mice was measured though M-mode echocardiography. Rab27a-KO mice or WT mice(n = 5) were orally inoculated at 5×106 TCID50 per mouse and the echocardiographic data was recorded at 3 day post-infection. (M) Levels of serum neutralizing antibody(NAb) titer in CVB3-infected mice. Different symbols correspond to independent experiments. (N) Cellular immune responses of CVB3-infected WT or Rab27a-KO mice was determined by IFN-γ ELISpot assay at 2, 4 and 6 day post-infection. (O and P) The body weight change (O) and survival rate (P) were respectively monitored daily until 14 day post-infection. All the data are shown as mean ± SD of three independent experiments. (*p<0.05, **p<0.01,***p<0.001).
Fig 8.
Exosomes provide a permissive route for CVB3 infection result in expanding viral tropism.
CVB3 infects human through GI system and intestines are the entry portal of the viral infection. After enter in alimentary tract by the mouth, the virus initially replicates in intestine epithelial cells and subsequently spreads to the circulation and various organs. Meanwhile, exosomes also release from virus-infected epithelial cells into extracellular space. In this study, we find that exosomes are prevalent and robust vehicles for the delivery of CVB3 virions, resulting in entry of tissue organs such as spleen and myocardial vessel that lacked virus receptors with high efficiency. Persistent infection of lymphocytes provides CVB3 virus with an important way of disrupting the normal function of immune cells and the homeostasis-related immunological response to antigens, resulting in enteroviral myocarditis.