A novel composite immunotoxin that suppresses rabies virus production by the infected cells

Abstract

Using strepavidin as a scaffold, we have assembled a composite immunotoxin that consists of recombinant Pseudomonas exotoxin A subunit (PE38) and recombinant 25-D1.16 Fab fragment which recognizes the SIINFEKL (pOV8) peptide from ovalbumin in association with H-2K(b) MHC class I protein. The composite immunotoxin exercises cytotoxicity against H-2K(b+) cells sensitized with pOV8 peptide but not with irrelevant peptide. Specific binding of the immunotoxin to H-2K(b+) cells infected with recombinant rabies virus (RV) expressing pOV8 epitope (RV-pOV8) resulted in the suppression of the production of virus particles by the infected cells. This strategy allows readily produce different immunotoxins with desired specificity by combining various targeting and toxin molecules. The results provide a proof of concept that composite immunotoxins can be utilized as novel immunotherapeutics to stop virus spread in the acute phase of the infection allowing winning time for the development of protective immune response.

Figure 1. Assembly of ImTx

SDS PAGE at non-reducing condition shows that all biotinylated 25-D1.16 Fab and PE38 molecules taken at the equimolar (1:1) ratio were bound to accessible biotin binding sites of Streptavidin. Lane 5 shows that multiple Streptavidin/Fab/PE38 species are formed at (Fab/PE38):Streptavidin ratio 1:1. At the 3:1 ratio, mostly high molecular weight species were seen indicating that majority of Streptavidin binding sites were occupied (Lane 6). When the ratio was increased to 5:1, free Fab and PE38 species were present in the mixture inactive of complete saturation of the Streptavidin binding sites (Lane 7). Lane 1, 2,3 and 4 show Streptavidin, 25-D1.16 Fab, PE38 and a mixture of the Fab and PE38 proteins, respectively.

Figure 2. Binding specificity of composed immunotoxin

(A) RMA-S cells sensitized with peptide pOV8 but not with VSV peptide were specifically stained with ImTx containing Alexa Fluor 488 Streptavidin (Str-Alexa) as established by flow cytometry analysis on Beckman Coulter FACS analyser. (B). Fluorescent microscopy of the same cells showed that fluorescent-labeled ImTx was specifically bound to pOV8-sensitized RMA-S cells. Representative images from 2 independent experiments are shown; interference reflection microscopy images are on the top and fluorescent images on the bottom. The fluorescence intensity was adjusted by subtracting cell autofluorescence (green).

Figure 3. The cytotoxicity of composite immunotoxin

RMA-S cells sensitized with either pOV8 (▲) or VSV (●) peptide were incubated with ImTx containing active PE38 subunit at indicated concentrations for 4 h at 37° C in atmosphere of 5% CO₂; pOV8-sensitized cells were also incubated with ImTx containing inactive PE38^Asp-553 (▵). The cells viability was evaluated by the MTS assay and percent specific cytotoxicity was determined (see Material and Methods) for each concentration of ImTx.

Figure 4. Detection of pOV8-K^b proteins on the surface of RV-pOV8 infected EL4 cells

A. Staining with 25-D1.16 tetramer showed presence of pOV8-K^b molecules on the surface of EL4 cells infected with RV-pOV8 but not with control RV virus. B. RV nuclear protein is detected in the cytoplasm of EL4 cells infected with both RV-pOV8 and control RV virus. C. The double staining of RV-pOV8- and RV-infected cells demonstrated that practically all RV-pOV8-infected IL4 cells were pOV8-K^b positive.

Figure 5. ImTx suppresses RV-pOV8 production by EL4 infected cells

ImTx containing active PE38 subunit as opposed to that with inactive PE38^Asp-553 subunit significantly suppressed production of recombinant RV-pOV8 virus by the infected cells. The difference in RV-pOV8 virus particles measured in the supernatant was statistically significant as established by Student's t-tests for paired data. *: P<0.02. Presence of ImTx in the supernatant practically did not affect production of a control RV virus by the infected cells. The recombinant RV that has Gag gene cloned into the backbone similar to that of RV-pOV8 was used as a control virus. The control RV-Gag virus showed somewhat impaired ability to replicate in EL4 cells as compared to RV-pOV8 most likely due to a toxic effect of Gag protein on the infected cells.

Figure 6. Exposure of RV-pOV8-infected cells to ImTx results in substantial decrease of the cell surface level of pOV8-K^b proteins while the total level of MHC class I expression remains the same

A. Left panel. EL4 cells were infected with RV-pOV8 in the presence (green) or absence (red) of ImTx and stained with fluorescent-labeled 25-D1.16 Fab/tetramer (see Materialand Methods for details). The level of pOV8-Kb expression of the infected cells was determined by mean fluorescent intensity. Right panel. EL4 cells were sensitized with either pOV8 or VSV peptides at various concentrations and the cells were stained with fluorescent-labeled 25-D1.16 Fab/tetramer. The level of pOV8-Kb expression was evaluated by measuring MFI of the staining. The difference in peptide concentration required to achieve levels of pOV8-Kb expression similar to that found on RV-pOV8-infected cells in the presence or absence of ImTx (see left panel) amounted to approximately 100 fold (indicated by errors). B. Uninfected cells EL4 cells as well as those infected with either RV-pOV8 or control RV viruses were incubated in the presence (green) or absence (red) of ImTx or in the presence of ImTx containing inactive PE38Asp-553 subunit (blue) and were stained with H-2Kb-specific AF6-88.5.3 antibodies followed by FITC-labeled anti-mouse antibody. In control experiment the cells were incubated with FITC-labeled anti-mouse antibody only (grey). The level of H-2Kb expression was found to be similar in all cases.