Isolation and characterisation of a human-like antibody fragment (scFv) that inactivates VEEV in vitro and in vivo

Abstract

Venezuelan equine encephalitis virus (VEEV) belongs to the Alphavirus genus and several species of this family are pathogenic to humans. The viruses are classified as potential agents of biological warfare and terrorism and sensitive detection as well as effective prophylaxis and antiviral therapies are required.In this work, we describe the isolation of the anti-VEEV single chain Fragment variable (scFv), ToR67-3B4, from a non-human primate (NHP) antibody gene library. We report its recloning into the bivalent scFv-Fc format and further immunological and biochemical characterisation.The scFv-Fc ToR67-3B4 recognised viable as well as formalin and ß-propionolactone (ß-Pl) inactivated virus particles and could be applied for immunoblot analysis of VEEV proteins and immuno-histochemistry of VEEV infected cells. It detected specifically the viral E1 envelope protein of VEEV but did not react with reduced viral glycoprotein preparations suggesting that recognition depends upon conformational epitopes. The recombinant antibody was able to detect multiple VEEV subtypes and displayed only marginal cross-reactivity to other Alphavirus species except for EEEV. In addition, the scFv-Fc fusion described here might be of therapeutic use since it successfully inactivated VEEV in a murine disease model. When the recombinant antibody was administered 6 hours post challenge, 80% to 100% of mice survived lethal VEEV IA/B or IE infection. Forty to sixty percent of mice survived when scFv-Fc ToR67-3B4 was applied 6 hours post challenge with VEEV subtypes II and former IIIA. In combination with E2-neutralising antibodies the NHP antibody isolated here could significantly improve passive protection as well as generic therapy of VEE.

Figure 1. Titration of monkey sera after the final boost.

Either VRS-purified culture supernatants from VEEV TC83 infected Vero cell or uninfected cells (negative control) were immobilised on 96well-microtiter plates. For the detection of a VEEV specific antibody titer, serial dilutions of pre-immune serum (PI 543) and immune serum (S543) after the sixth boost were applied. Bound antibodies were detected by rabbit anti-monkey IgG conjugated to horseradish peroxidase and staining with TMB.

Figure 2. Reactivity of scFv-Fc ToR 67- 3B4 to Alphavirus subspecies and species.

Panel 2A shows the antigen binding efficiency of ToR67-3B4 on different immobilised VEEV subtypes at a concentration of 1 µg/ml, n = 1 for all data points. Panel 2B illustrates the cross-reactivity of scFv-Fc ToR67-3B4 to a range of Alphavirus species analysed by a sandwich ELISA. Antigens were captured by an anti-Alphavirus mAb mix, specifically bound virus was detected by biotinylated scFv-Fc ToR67-3B4 (1∶5000). The insert in the upper right corner of the bar chart shows the positive control and demonstrates that all Alphaviruses except SINV are specifically captured by the antibody mixture. Detection of virus was performed with the same biotinylated anti- Alphavirus mAb mix. All viral antigens were applied with a TCID₅₀/ml of 5×10⁷ to 10⁸. Culture supernatant of non-infected Vero cells was used as negative control. The mean values of two ELISAs from two independent experiments are shown. In panel 2C scFv-Fc ToR67-3B4 was applied as capture antibody in combination with the cognate Alphavirus-specific antibody mixture for group-specific detection. Viral antigens were applied with a TCID₅₀/ml of 5×10⁷ to 10⁸. Culture supernatant of non-infected Vero was used as negative control. The mean values of two ELISAs from three independent experiments are shown. Panel 2D shows the detection limit for vaccine strain TC83 (subtype IA/B) using capture antibody mAb VEE WIS1 paired with biotinylated scFv-Fc ToR67-3B4 detector antibody (1∶5000). Virus was titrated in 2-fold dilutions. The data represents 2 separate experiments with 3 replicates of each concentration. Abbreviations used in this legend are BHK: Baby hamster kidney cells, VEEV: Venezuelean equine encephalitis virus, WEEV: Western equine encephalitis virus, EEEV: Eastern equine encephalitis virus, CHIKV: Chikungunya virus, SINV: Sindbis virus and SFV: Semliki Forest virus.

Figure 3. Immunological analysis of VEEV glycoproteins.

Panel A and B show the immunoblot analysis of VEEV glycoproteins with scFv-Fc ToR67-3B4 and monoclonal antibodies. Proteins from VEEV strain TC-83 were either resolved on an 8–15% gradient (panel A) or a linear 12% polyacrylamide gel (panel B) under denaturing and non-reducing conditions at 56°C (panel A) or at 56°C and 95°C (panel B) under denaturing conditions with or without 100 mM DTT. After Western blotting the membrane was cut in stripes and was probed with either scFv-Fc ToR67-3B4 (Panel A lane 2 and 6; panel B all lanes) or E1- and E2-specific mAbs (Panel A lane 1 and 5: SFV 8/6 (E1-specific); lane 3: WIS-VEE1 (E1-specific); lane 4: 1A3B7 (E2-specific)). Scale indicates protein size in kDa. Panel C depicts the immunohistochemistry of Vero cells infected with VEEV TrD, 1 day post infection. Photomicrographs were obtained from infected (I–IV) and non–infected (V) cells with a 10fold magnification lens. Cells in panel I and II were stained for VEEV antigen with biotinylated scFv-Fc ToR67-3B4, infected cells in panel III and IV were detected with biotinylated mAb SFV 8/6 as positive control. Non-infected Vero cells in panel V were stained with scFv-Fc ToR67-3B4 as negative control.

Figure 4. In vitro neutralisation activity of scFv-Fc fusion ToR67-3B4 to different VEEV strains and Everglades virus.

VEEV strains of subtype IA/B (Panel A and B) and Everglades virus (strain Fe37c, Panel C) with a TCID₅₀/ml of 5×10⁴ were incubated with serial dilutions of scFv-Fc ToR67-3B4 and other mAbs at 37°C for 2 hours. Starting concentrations were between 0.5 to 1.0 mg/ml. Subsequently the mixtures were subjected to cell culture infection. Residual infectious activity of virus was estimated by specific immunostaining one day post infection with VEEV-specific antibodies. Absorbance (OD450 nm) values obtained with virus samples not preincubated with any antibody served as positive control and were set as 100% infectivity. Non-infected Vero cells were used as negative control. The mean values of two independent experiments are shown.

Figure 5. scFv-Fc ToR67-3B4 protects mice against VEEV disease.

BALB/c mice were challenged by the aerosol route with approximately 100 LD₅₀ VEEV strain TrD (Panel A and B), Mena II, Fe37c and Mucambo virus (BeAn8, Panel C and D). Six, twenty-four, forty-eight and seventy-two hours later they were injected with 100 µg scFv-Fc ToR67-3B4 intraperitoneally (Panel A, B and C, n = 6 or 10). As negative control mice remained either untreated or were injected with a human IgG1 antibody (Panel B) or a nonspecific antibody (anti-WEEV, Panel D, n = 10 or 5). Animals were observed twice daily for clinical signs of infection and were culled when appropriate using humane endpoints.