Foot-and-mouth disease virus localisation on follicular dendritic cells and sustained induction of neutralising antibodies is dependent on binding to complement receptors (CR2/CR1)

Figures

• 

  Fig 1. Ex vivo FMDV immune complex deposition assays. Confocal microscopy images of cryosections from naïve mice spleens after addition of different forms of FMDV. (A) FMDV antibody (IB11), (B) FMDV antigen (O1/Manisa/TUR/69) or (C) FMDV ICs (O1/Manisa/TUR/69 antigen and IB11 antibody) were pre-incubated for 1 hour with 5% NMS prior to addition to the cryosections. FMDV (green) was labelled with polyclonal rabbit anti-FMDV-O1 and detected using anti-rabbit 488. (A) Absence of signals for FMDV in spleen cryosections when treated with antibody alone, (B) few isolated cells stained positive for FMDV upon addition of antigen alone and (C) large bright green clusters showing the binding of FMDV ICs in the B cell follicles in the spleen. Nuclei stained blue (DAPI). Scale bars = 100μm.

• 

  Fig 2. Flow cytometric analysis of splenocytes from mice treated with mAb 4B2 or control IgG, comparing cell subsets and availability of CR. Flow cytometry was used to identify the availability of complement receptors in mice after treatment with mAb 4B2 and the percentage of cell subsets, compared to control mice treated with IgG1. Spleen samples were taken at (A, D) early time points and (B, E) late time points from mice treated with 4B2 or IgG1 and naïve animals. At the early time points (A) there is a trend whereby mice treated with mAb 4B2 show a smaller number of positive cells to the CR antibodies, compared to the IgG1 or control groups, although this is not significant; 8C12 p = 0.312; 7G6 p = 0.061; 7E9 p = 0.194. By the late time points (B) mice treated with 4B2 had significantly reduced binding of the three anti-CR antibodies (p = 0.03) to their cells compared to the control mice. A representative histogram (C) of the number of cells positive for mAb 7G6 (CR2/CR1) from a mouse in the control group compared to a mouse from the mAb 4B2 treated group at the late time points. The percentage of the different splenic cell subsets CD8 and CD4 T cells, B cells (B220), macrophages…

• 

  Fig 3. Effect of pre-treatment with mAb 4B2 on the binding of PAP on the FDC networks in mouse spleen. BALB/c mice were treated with 500 μg of 4B2 (n = 3) or IgG1 (n = 4) control 24 hours before immunisation intravenously with peroxidase anti-peroxidase (PAP). Naïve mice were untreated. Spleen samples were collected in OCT compound from mice culled 1-day post inoculation with PAP. A) Cryosections were analysed via confocal microscopy for the presence of PAP associated to the FDC network. Confocal microscopy images are arranged in rows and columns according to the treatment and the staining. B) Mice from the 4B2 treatment group had significantly less PAP bound to FDCs compared the control group, p value of 0.05 using the non-parametric Mann-Whitney U test to compare the medians of the two treatment groups. PAP panels: show PAP labelled green, detected with anti-rabbit 488. PAP was detected in FDC networks in IgG1 control mice, but not in 4B2 treated mice. Absence of signal to PAP in naïve mouse spleen. FDC Network: light zone FDCs labelled red with Alexa Fluor 594-conjugated anti-mouse CD21/CD35 (CR2/CR1) antibody, clone 7E9; FDC clusters were detected in all groups. Macrophages:…

• 

  Fig 4. Viraemia in 4B2 treated and control-treated mice in response to FMDV infection. The presence of viraemia in serum samples from mice treated with 4B2 or IgG1 was investigated by (A) plaque assay and (B) qRT-PCR. Serum samples were collected from 4B2 and IgG1 treated mice at 2 and 7 dpi. The quantity of virus in the mouse serum at 2 dpi is expressed as (A) log 10 of the number of plaque forming units (PFU) per 1 ml serum and (B) log 10 of the genome copy number (GCN), with CT values ≥ 35 for 3D FMDV deemed negative and recorded as 0. Each data point represents an individual animal, and the line represents the median values. Naïve mice at 2 dpi and all serum samples harvested from 7 dpi were negative for viraemia. P values of <0.05 using the non-parametric Mann-Whitney U test to compare the medians of the two treatment groups.

• 

  Fig 5. Effect of pre-treatment with 4B2 preventing the binding of FMDV on FDCs in mouse spleens. Confocal microscopy images are arranged in rows and columns according to the treatment, day post infection and the staining. BALB/c mice treated with 500 μg of either mAb 4B2 or IgG1 control on day -1 before challenge with FMDV. FMDV-infected mouse spleen samples were collected at (A) 7, (B) 14, (C) 21 and (D) 28 dpi from IgG1 control mice and 4B2 mice (n = 8 per group per timepoint). Naïve mouse spleens were taken at 7 dpi (n = 4) and 28 dpi (n = 4). FMDV panels: show FMDV protein labelled green with biotinylated llama single domain anti-FMDV 12S antibody VHH-M3 and streptavidin Alexa-Fluor-488. FMDV was detected in the IgG1 control group at all timepoints. FMDV was not detected in the spleens of the 4B2 treated group at any of the time points, with the exception of four mice, three harvested at day 14 and one at day 21. There was an absence of signal for FMDV in naïve mouse spleen at all time points. FDC Network: show FDCs labelled red with Alexa Fluor 594-conjugated anti-mouse CD21/CD35 (CR2/CR1) antibody, clone 7E9; FDC clusters were detected at all time points in control, 4B2 treated…

• 

  Fig 6. Quantification of FMDV antigen and RNA in spleens from 4B2-treated and isotype control treated mice detected by confocal microscopy and RT-qPCR. Spleen samples were collected from BALB/c mice at 7, 14, 21 and 28 dpi (n = 8/group/timepoint) following treatment with either mAb 4B2 or IgG1 isotype control one day prior to IP challenge with 106.2 TCID50 of FMDV/O/UKG/34/2001.Sections were cut using a cryostat and a cross-section was taken of the spleens by consistently collecting 16 sections per animal with an approximate 70 μm gap between each section. A) FDC networks were visualised, imaged and counted using mAb 7E9, an anti-CR2/CR1 antibody. There were no significant differences in the total number of FDC networks in the group treated with mAb 4B2 compared to the IgG1 isotype control. B) FMDV was detected using a biotinylated llama single domain anti-FMDV 12S antibody VHH-M3, and the percentage of FDC networks which were positive for FMDV was calculated. Mice treated with mAb 4B2 had significantly less FMDV in their FDC networks compared to the isotype control mAb, with a P value of ≤0.001 from 7, 21 and 28 post infection. C) The samples were analysed by RT-qPCR for the presence…

• 

  Fig 7. Co-localisation of FMDV with FDCs. BALB/c mice were infected with FMDV/O/UKG/34/2001 and high-resolution images were taken of spleen samples using a STED confocal microscope. A) Spleen taken from an infected mouse 7 dpi, demonstrating the co-localisation of FMDV (green) with FDCs (red). B) Morphometric analysis using ImageJ confirmed that FMDV was preferentially associated with FDCs in spleen tissues (n = 7) and significantly greater than the null hypothesis that the pixels were randomly distributed, with a p value of 0.0023.

• 

  Fig 8. Effect of 4B2 treatment on titres of FMDV neutralising antibodies in mouse serum. FMDV neutralising antibodies were evaluated from serum samples taken from BALB/c mice at 7, 14, 21 and 28 dpi with FMDV. Mice had either been pre-treated with mAbs 4B2 or IgG1 1 day prior to FMDV infection. Naïve mice were used as controls. Each data point represents an individual animal, and the line represents the median antibody titre. Neutralising antibody titres are expressed as the serum dilution that neutralised 50% of 100 TCID50 of the virus. Using the non-parametric Mann-Whitney U test to compare the medians of the two groups, at 14 dpi p value of 0.001, 21 dpi p value of 0.004 and 28 dpi p value of 0.003.

• 

  Fig 9. Effect of 4B2 treatment on titres of FMDV specific antibodies. FMDV-specific antibodies in serum samples from mice treated with mAb 4B2 or an isotype control antibody (IgG1) were detected by ELISA. Serum samples were collected at 7, 14, 21 and 28 dpi and tested for (A, B) IgG and (C) IgM antibodies. Naïve sera were included on all ELISA plates and the cut-off was set at 1.5 times the OD of the naïve sera. Antibody titres were either expressed as (A, C) the reciprocal log10 of the last positive dilution or (B) using a known FMDV IgG standard to plot the concentration of IgG antibodies in mg/ml. Each data point represents an individual animal, and the bars represent median values. Using the non-parametric Mann-Whitney U test to compare the medians of the two groups, there were no statistically significant differences at any time points.

• 

  Fig 10. Effect of 4B2 treatment on titres of IgG antibodies specific to the FMDV G-H loop and the avidity of FMDV specific IgG antibodies in mouse serum. BALB/c mice treated with 500 μg of either mAb 4B2 or IgG1 control on day -1 before challenge with FMDV and sera was collected at 7, 14, 21 and 28 dpi. A) An indirect peptide ELISA showed that mice treated with 4B2 had significantly less antibodies to the FMDV G-H loop compared to the IgG1 control group, with a P value of ≤0.05 using the non-parametric Mann-Whitney U test. Naïve sera were included on all ELISA plates and the cut-off was set at 1.5 times the OD of the naïve sera. B) The avidity of antibodies was measured using biolayer interferometry and was performed using an Octet Red96e. FMD O1/Manisa/TUR/69 VLP were bound to streptavidin sensors and dipped into three dilutions of sera per mouse. Each data point represents the mean avidity of these measurements for each individual mouse, represented as -Log10 of the KD (M) value. Sera which produced a negative response rate, and therefore had too few antibodies bound to FMD VLPs to reach the limit of detection, are recorded as 0. The results demonstrate that mice treated with 4B2 had…