|
|
Blood, Vol. 113, Issue 9, 1909-1918, February 26, 2009
Human erythrocytes bind and inactivate type 5 adenovirus by presenting Coxsackie virus-adenovirus receptor and complement receptor 1
Blood Carlisle et al.
113: 1909
Supplemental materials for: Carlisle et al
Studies of internalisation of human erythrocyte bound Ad5
Ad5 (109) was mixed with washed human erythrocytes (106) in 100 µl PBS at 37°C for 10 min to ensure binding, before sample was moved to 4°C or kept at 37°C for a further 30 min. All subsequent steps were then performed at 4°C. Erythrocytes were centrifuged at 700g for 3 min and after supernatant removal and washing in 1 ml PBS they were resuspended in 400 µl of 1% BSA/PBS and 2 µl FITC–anti-Ad5 antibody (AB1056F, Millipore) was added. After incubation for a further 60 min, erythrocytes were spun down, supernatant removed and post washing with 1 ml of 1% BSA/PBS, were resuspended in 500 µl PBS/BSA and analyzed by flow cytometry (see methods). Positive staining with 37°C erythrocytes + Ad5 + FITC–anti-Ad5 (red line) indicates Ad5 was not internalized and could still be detected on erythrocyte surface. Anti-C3b pull down, C3a release and C3 adduct formation assays
Plates (96 well, Immunosorb, Nunc) were coated with anti-C3c antibodies A0062 (Dako, Denmark) at a 1 in 100 dilution in PBS, 100 µl per well. After 24 hours at 4°C washing was performed (4 × 200 µl per well with PBS) and then blocking for 2 hours in 100 µl of PBS/1% BSA at 20°C performed. Blocker was removed, and 100 µl of plasma samples diluted 1 in 4 with PBS/1% BSA were added. After 1 hours at 20°C plasma was removed, washing was performed (4 × 200 µl PBS) and then lysis solution added and DNA recovered and extracted using the manufacturer’s protocol (G1N-350, Sigma) and Ad5 genome copies were quantified. C3a release assays were performed by adding Ad5 (10 9 ) to 10 µl of CPD plasma, CVF treated plasma, 56°C/30 min heated plasma or EDTA plasma. After 1 hour at 37°C samples were tested for C3a content according to the manufacturer’s protocol 550499 (BD Biosciences, USA) and results compared to plasmas that had been identically treated but not exposed to Ad5. N = 3, standard deviation shown. ** p < 0.005. Cryo-electron microscopy
Images of plunge-frozen microscope grids bearing erythrocyte ghosts and Ad5 were captured using an FEI F30 cryo-electron microscope operating at 300 kV under low dose conditions. The micrographs were scanned using a Zeiss scanner on a 7 µm raster and a total of 27 images showing virus-membrane interactions were excised using Boxer.3 Image reconstruction was performed in Spider,4 using the published Ad5 reconstruction (Macromolecular Structure Database, www.ebi.ac.uk/msd, entry EMD-1113),5 reoriented and shifted so that the fivefold vertex was coaxial with the y axis, to align the images. After the first round of reconstruction using this model, density was visible for the membrane, which had not been present in the alignment template, and this became stronger through repeated iterations, thus validating our approach to the reconstruction of this sample. Also appearing were tubular densities linking the capsid surface and the membrane; the reconstruction could be refined to a resolution of 40Å. The final map and supporting individual images of the additional peripheral fibre-mediated interaction of Ad5 with erythrocytes have been deposited in the Macromolecular Structure Database, entry EMD-1576.
Files in this Data Supplement:
- Figure S1. Binding of Ad5 to human erythrocytes does not result in internalisation (JPG, 16.4 KB)
-
Ad5 (109) was mixed with washed human erythrocytes (106) in 100 µl PBS at 37°C for 10 min to ensure binding, before sample was moved to 4°C or kept at 37°C for a further 30 min. All subsequent steps were then performed at 4°C. Erythrocytes were centrifuged at 700g for 3 min and after supernatant removal and washing in 1 ml PBS they were resuspended in 400 µl of 1% BSA/PBS and 2 µl FITC–anti-Ad5 antibody (AB1056F, Millipore) was added. After incubation for a further 60 min, erythrocytes were spun down, supernatant removed and post washing with 1 ml of 1% BSA/PBS, were resuspended in 500 µl PBS/BSA and analyzed by flow cytometry (see methods). Black = erythrocytes + FITC–anti-Ad5, blue = 4°C erythrocytes + Ad5 + FITC–anti-Ad5, red = 37°C erythrocytes + Ad5 + FITC–anti-Ad5, green = erythrocytes + 5 µl anti-CAR (RmcB) + Ad5 + FITC–anti-Ad5. Experiment was performed in triplicate, a representative histogram for each sample is shown.
- Figure S2. Ad5 does not bind murine erythrocytes (a) or macaque erythrocytes (b) in PBS (JPG, 45.7 KB)
-
Washed erythrocytes from mice or rhesus macaques, two of the most commonly used pre-clinical models for Ad5 pharmacokinetics,1 were incubated for 30 min at 37°C with Ad5. Erythrocyte binding of Ad5 was assessed by separating the liquid (yellow bar) and erythrocyte (red bar) fractions by centrifugation and performing QPCR specific for the Ad5 genome on each fraction. Data is represented as the % of the total input dose recovered, n = 4, error bars represent SEM, ** p < 0.005.
- Figure S3. Heparin addition can decrease Ad5 binding to human erythrocytes (JPG, 43.6 KB)
-
Ad5 in PBS or PBS/heparin (30µg/ml) was added to human erythrocytes that had been pre-incubated (15 min, 37°C) with anti-CAR antibody or not. After 30 min at 37°C, anti-CAR antibody was added or not (15 min, 37°C). Samples were then fractionated and assayed for Ad5 genome content as described in methods. N = 4, se shown, *p<0.005. With Ad5 in PBS significantly more erythrocyte binding (10-fold) was observed if anti-CAR antibody was added post incubation with Ad5 rather than pre-incubation with Ad5 (indicative of a secondary binding event taking place). However, if Ad5 was first incubated with heparin, binding in presence of post-addition anti-CAR antibody was halved, indicative of a role for HSPG in the secondary binding event.
- Figure S4. In plasmas which elicit no erythrocyte binding, Ad5 is not fixed by complement protein C3b (JPG, 43 KB)
-
Ad5 was incubated with a variety of plasmas before exposure to an immunosorb plate coated with anti-C3 antibody (see methods below), after washing the number of Ad5 genomes attached to the plate was quantified by QPCR. N = 4, SEM shown, p < 0.005.
- Figure S5. In plasmas which elicit no erythrocyte binding, Ad5 does not cause the release of complement protein C3a (JPG, 43.1 KB)
-
An ELISA was performed (see methods below) to detect the release of the anaphalatoxin C3a, upon addition of Ad5 to a variety of plasmas. N = 3, standard deviation shown, **p < 0.005.
- Figure S6 (JPG, 30.6 KB)
-
(A) Ad5 capsid proteins are covalently modified by complement. Ad5 was incubated with plasma and run on SDSPAGE as described for figure 2D. Western blotting was performed using a 1:1000 dilution of goat anti-Ad primary antibody (0151–9104, AbD Serotec) and a 1 in 2000 dilution of donkey anti–goat-HRP (SC2020, Santa Cruz). Lane 1 = Ad5, 2 = Ad5 + heparin plasma, 3 = Ad5 + EDTA plasma, 4 = heparin plasma, 5 = EDTA plasma. High molecular weight bands (see arrows) are stained in lane 2, but not in lanes 1 or 3. Reduction in the intensity of penton and fibre bands is evident in lane 2 compared to 1 and 3. (B) CAR ablated Ad5 (vKH3) does not bind erythrocytes in PBS but does bind in plasma. Washed human erythrocytes (5 × 109 cells/ml, suspended in PBS) were incubated for 30 min at 37°C with CAR binding ablated Ad5 variant vKH3 2 (2 × 108 vp/ml). Erythrocyte binding of vKH3 was assessed by separating the liquid (yellow bar) and erythrocyte (red bar) fractions by centrifugation and performing QPCR specific for the Ad5 genome on each fraction. Data is represented as the % of the total input dose recovered, n = 4, error bars represent SEM, ** p < 0.005.
|
|
