Recognition of decay accelerating factor and alpha(v)beta(3) by inactivated hantaviruses: Toward the development of high-throughput screening flow cytometry assays

Abstract

Hantaviruses cause two severe diseases in humans: hemorrhagic fever with renal syndrome (HFRS) and hantavirus cardiopulmonary syndrome (HCPS). The lack of vaccines or specific drugs to prevent or treat HFRS and HCPS and the requirement for conducting experiments in a biosafety level 3 laboratory (BSL-3) limit the ability to probe the mechanism of infection and disease pathogenesis. In this study, we developed a generalizable spectroscopic assay to quantify saturable fluorophore sites solubilized in envelope membranes of Sin Nombre virus (SNV) particles. We then used flow cytometry and live cell confocal fluorescence microscopy imaging to show that ultraviolet (UV)-killed SNV particles bind to the cognate receptors of live virions, namely, decay accelerating factor (DAF/CD55) expressed on Tanoue B cells and alpha(v)beta(3) integrins expressed on Vero E6 cells. SNV binding to DAF is multivalent and of high affinity (K(d) approximately 26pM). Self-exchange competition binding assays between fluorescently labeled SNV and unlabeled SNV are used to evaluate an infectious unit-to-particle ratio of approximately 1:14,000. We configured the assay for measuring the binding of fluorescently labeled SNV to Tanoue B suspension cells using a high-throughput flow cytometer. In this way, we established a proof-of-principle high-throughput screening (HTS) assay for binding inhibition. This is a first step toward developing HTS format assays for small molecule inhibitors of viral-cell interactions as well as dissecting the mechanism of infection in a BSL-2 environment.

Figure 1

Structural morphology and composition of SNV virions A. Low magnification wide field view of electron micrographs of Cesium Chloride-stained UV-irradiated SNV. B. Higher magnification of area marked with rectangle. C. Histogram of average size distribution of SNV virus particles shown in A (sample size 163 virions) mean 193±40 nm. D. Schematic cross-section of Bunyaviridae virus. The three RNA genome segments (S, M, and L) are complexed with the nucleocapsid protein to form ribonucleocapsid structures. The nucleocapsids and RNA-dependent RNA polymerase are packaged within a lipid membrane envelope that contains viral glycoproteins Gn and Gc. E. Western blot representing the preservation of antigens consequent to 30-sec of UV irradiation. Purified SNV preparations were subject to irradiation (“UV”) before being treated with SDS Laemmli buffer in a boiling water bath and the proteins analyzed by SDS-PAGE. Antibodies used in the detection were anti-nucleocapsid (“N”), anti-Gc or anti-Gn.

Figure 2

Plot of dequenching of R18 as a function of its mole% fraction in lipid vesicles (insert shows full range of experiment) in DOPC and DSC membranes at 23°C and 41°C. Horizontal tie lines are measures of dequenching of R18 solubilized in the envelope membrane of SNV at 23°C and 41°C, top and bottom, respectively. Each tie line intersects the standard calibration curves at a single point which corresponds to the mole fraction of R18 in (the envelope membrane) along the x-axis. The ovoids indicate the limiting errors in the points of intersection. Errors are based on the standard deviation in the dequenching values from over 5 labeling experiments.

Figure 3

UV killed SNV^R18 binds cognate receptors of live SNV. A. Flow cytometry measurements of SNV^R18 binding to Tanoue B cells. Total binding is normalized to samples in the presence of 1mM Ca²⁺ and samples measured under the conditions of: 40μg/ml polyclonal H319 anti-DAF antibodies; 10× unlabeled SNV and Mn²⁺. The similarity between Ca²⁺ and Mn²⁺ conditions suggests the absence α_vβ₃ expression on Tanoue B cells. B. Models of α_vβ₃ in bent, low affinity (Ca²⁺) and extended confirmation (Mn²⁺). Model of the extended conformation was assembled using the program O Jones et al 1991 [56]. Figure was adapted from Xiong et al [57]and prepared with Pymol version 0.99 Delano Scientific LLC (San Francisco, CA). Micrographs show effect of cation environment on the ability of SNV^R18 to bind to α_vβ₃. Ca²⁺ environment maintains α_vβ₃ in a bent low affinity state that leaves the PSI domain accessible to incoming virions, and enables binding of SNV to integrins. 1 mM Mn²⁺ environment activates α_vβ₃ into an unbent conformation that renders the PSI domain inaccessible to SNV and diminished binding. Graph shows flow cytometry measurements of binding of SNV^R18 to Vero E6 cells. 10× SNV, 6μM PSI domain peptides, and Mn²⁺ are used to inhibit binding to Vero E6 cells.

Figure 4

Flow cytometry measurements of SNV^R18 binding to Tanoue B cells. A. Paired scattergram plots of cell granularity (Side Scatter, SSC-H) versus size (Forward Scatter, FSC-C) and SSC-H-versus cellular fluorescence associated with the indicated quantity of SNV^R18. The ellipses shown in the scattergrams are electronic gates used to select cells (92.8% of total population of scatter events), which are considered valid for analysis. B. Pseudo-3D plot of log fluorescence versus concentration of SNVR18 titers.

Figure 5

A.Equilibrium binding curve of SNV^R18 to 10,000 Tanoue B cells in 10μl. Insert shows a plot of bound SNV^R18 versus residual free SNV^R18 particles. B. Plot of SNV^R18/cell versus log free [SNV^R18]. The analysis of the Sigmoidal binding data yielded a K_d of ≈ 26 pM. C. Competition binding curve of a fixed quantity of SNV^R18 and various arbitrarilly defined titers of unlabeled SNV. The EC50 value determined from Fig. 4B is used to determine the actual concentrations of SNV titers (see text for details).

Figure 6

A. Dot plot of SNV^R18-bearing cells from an assay in a 384 well plate. Row 1. Data plugs bisected by the line represent duplicate titration of SNV^R18 show increasing binding with increasing SNV^R18 in each well. Data marked with asterisks represent samples preblocked with 10x unlabeled virions. Data points along the baseline represent autofluorescence. Row 2. Competition assay of SNV and SNV^R18, data marked with a c represents control marker samples not mixed with SNV. Row 3 is a repeat of Row 1 in a 1% DMSO environment. B. Plot of median fluorescence intensity (MFI) versus added SNV^R18 of Rows 1&3. Open squares represent 1% DMSO solvent, triangles neat HHB buffer. Z′ values for each concentration shown in red were determined for 4 values at each point using Eqn. 4. x's represent samples preblocked with 10x unlabeled SNV. C. Plot of bound SNV^R18/cell versus titers of unlabeled SNV nominally measured in terms of grams of N protein. The EC50 value determined from Fig. 5B can then used to convert the grams of N protein to concentration of SNV as described in Fig. 5C.