Macaque care.

The animal experiment was approved by the ethics committee for animal experimentation of IRBA-CRSSA (Institut de Recherche Biomédicale des Armées - Centre de Recherche du Service de Santé des Armées). The ethics committee number for the VEE macaque immunisation was 2008/03.1. The immunisation with inactivated virus particles was performed according to the governmental french ethical guidelines: “Partie reglementaire du livre II du code rural (Titre I, chapitre IV, section 5, sous section 3: expérimentation sur l'animal)”, “Décret 87–848 du 19-10/1987 relatif aux expériences pratiquées sur les animaux vertébrés, modifié par le décret 2001/464 du 29/05/2001”, “Arrêté du 29 octobre 1990 relatif aux conditions de l'expérimentation animale pour le ministère de la défense” and “Instruction 844/DEF/DCSSA/AST/VET du 9 avril 1991 relative aux conditions de réalisation de l'expérimentation animale”. Animal care procedures were in compliance with the regulations detailed under the Animal Welfare Act [33] and in the Guide for the Care and use of Laboratory Animals [34]. Animals were kept under stable temperature (22±2°C) and relative humidity (50%) conditions, with 12 hours of artificial lighting per day. Their cages were individual (6 per room) and each had a perch. Animals were fed twice a day, once with dried food, once with fresh fruits and vegetables, and water was provided concurrently. The food intake and general behaviour were observed during feeding by animal technicians, who were able to access a veterinarian if needed. Systematic veterinary visits to each NHP-room were conducted twice a week. General anesthesia was administered prior to any blood or bone marrow collection. Additionally, subsequent analgesics were administered in the following days by animal technicians if pain was suspected, according to the observations of animal behaviour.

Mice infection.

The use of mice for passive protection studies for research purposes complied with UK guidelines. Animal protocols adhered to the Animals (Scientific Procedures) Act 1986, and protocols were approved by the UK Home Office. The study was performed under project licence 30/2425 and was approved internally by a Defence Science and Technology Laboratory (Dstl) ethical review process.

Cell culture and virus production.

All viruses used in this study represent models for biowarfare agent relevant Alphavirus species and are either part of the strain collection of the Armed Forces Scientific Institute for Protection Technologies – NBC Protection (WIS) or are part of the strain collection of the Defence Science and Technology Laboratory, Porton Down, UK. Strains of VEEV from subtypes IA/B (Trinidad donkey; TrD), IC (P676), ID (3880), IE (Mena II), IF (78V), Everglades virus (formerly subtype II, strain Fe37c), Mucambo virus (formerly subtype IIIA, strain BeAn8), Pixuna virus (formerly subtype IV), Cabassou virus (formerly subtype V, strain CaAr508) and Rio Negro virus (formerly subtype VI, strain AG80) were kindly supplied by Dr. R.E. Shope (University of Texas Medical Branch, USA). Eastern equine encephalitis virus (EEEV) strain H178/99 and Western equine encephalitis virus (WEEV) strain H160/99 used in this study were received from the National Collection of Pathogenic Viruses (NCPV), UK.

Alphaviruses were either propagated in Baby Hamster Kidney (BHK), Vero-B4 (African green monkey kidney) or L929 (murine fibroblast) cells at 37°C and 5% CO₂ in a biosafety 3 facility according to standard procedures as described [18], [35]–[37]. The cell lines were obtained from the DSMZ-ACC 33 (Deutsche Sammlung von Mikroorganismen und Zellkulturen GmbH), Braunschweig, Germany or the European Collection of Animal Cell Cultures, Porton Down, Salisbury, UK.

All viruses were harvested from infected cells when 50–75% of the cell monolayer showed evidence of viral cytopathic effect (CPE). Virus titers were either determined by the 50% tissue culture infective dose (TCID₅₀/mL) method according to Spearman and Kaerber or plaque assay [38]–[40]. Titers typically ranged from 1.0×10⁷ to 1.2×10¹⁰ TCID₅₀/ml.

Purification of Alphaviruses.

Virus containing supernatants from infected Vero cells were either purified by affinity chromatography on Matrex Cellufine Sulfate Medium™ (Virus Recovery System, VRS, Chisso America Inc., NY, USA) or by isopycnic density gradient centrifugation as described below. Matrex Cellufine Sulfate Medium™ (VRS) is a cellulose bead medium functionalised with a low concentration of sulfate esters that operates similar to a cation-exchange resin and has a high affinity for enveloped viruses. It selectively adsorbs complete virus particles as well as viral coats according to their charge. Briefly, 50 mL resin was equilibrated with adsorption buffer (0.01 M phosphate buffer, pH 7.5). Up to 200 mL of virus containing prefiltered cell culture supernatant was loaded onto the column which then was washed twice with 0.01 M phosphate buffer, pH 7.5. Elution of virus particles was performed with 1 M NaCl.

Purification of virus particles was carried out in two steps. Prepurification was performed by the sucrose cushion ultracentrifugation method (20% sucrose cushion), which causes low mechanical stress and allows the concentration and collection of morphologically intact particles after centrifugation at 112,000× g for 2 to 3 hours. The pellet was resuspended in 0.5 to 1 mL phosphate buffered saline (PBS) and further purified by isopycnic density gradient centrifugation (20 to 60% sucrose) for 18 hours at 217,500× g. The virus containing fraction was removed and stored at −80°C until subjected to further analysis.

Inactivation of Alphaviruses.

Inactivation of all viral antigens was performed with a final concentration of 0.1% ß-propionolactone (ß-PL, Sigma-Aldrich, Taufkirchen, Germany). Immediately before use a 10% ß-PL solution was prepared and 0.1 ml of this dilution was incubated with 5 ml of virus containing cell culture supernatant (pH 8 to 8.5) for 1 h at 4°C and 4 hours at 37°C with constant stirring. After 2 and 4 hours the pH of the cell culture supernatant was controlled and adjusted if necessary. In order to achieve a complete hydrolysis of remaining ß-PL after the inactivation the supernatant was kept at 4°C for another 12 to 18 hours. After low speed centrifugation of the sample for 30 min at 1000× g the viral inactivation was verified by inoculation of Vero cells and monitoring for cytopathic effects (CPE) for 3 to 5 days.

Animal immunisation.

A male cynomolgus macaque (Macaca fascicularis) was immunised by intramuscular injection with 100 µl of the commercially available veterinary vaccine Fluvac Innovator Triple EFT plus EHV (Fort Dodge Animal Health, USA). This all-around vaccine is normally applied to horses in order to protect them from equine encephalitis caused by EEE, WEE and VEE viruses. In addition, the vaccine contains inactivated rhino- and influenza viruses as well as tetanus toxoid to protect horses from equine rhinopneumonitis, influenza and tetanus.

days after this first immunisation the macaque was immunised with 100 µl of VRS purified ß-PL inactivated VEEV TC83 (TCID₅₀/ml of 2×10¹⁰) in combination with complete Freund's adjuvant. Subsequently, 5 additional antigen injections of 100 µl were performed after 65, 90,150, 210 and 350 days with incomplete Freund's adjuvant. Ten days after the final antigen injection the immune library was generated.

After each boost immune sera were gathered, inactivated for 20 min at 56°C and tested for VEEV specific antibody titer by ELISA. For this purpose either VRS-purified culture supernatants from VEEV TC83 infected Vero cells or uninfected cells were immobilised on 96 well-microtiter plates (Maxisorp™, Nunc, Wiesbaden, Germany). For the determination of VEEV specific antibody titers serial dilutions of pre-immune-serum (PI 543) and immune sera (S543) were applied. Specifically bound antibodies were detected by incubation with rabbit anti-monkey IgG conjugated to horseradish peroxidase (1∶10.000, Sigma-Aldrich, Taufkirchen, Germany) for 30 minutes at room temperature. Staining was performed with 3-3′, 5-5′ -tetramethylbenzidine (TMB, Serva, Heidelberg, Germany) and stopped with 2 M sulphuric acid after 10 minutes. Absorbance was measured at 450 nm.

Library construction and screening of recombinant antibodies.

Lymphocytes of the macaque bone marrow were sampled after 6 boosts and subjected to RNA preparation by using Tri Reagent (Molecular Research Center Inc, Cincinnati, USA). The isolated RNA was reverse transcribed to cDNA and diverse combinations of forward and reverse primers were used to amplify DNA, coding for the variable regions of antibodies, VLκ and VH [41]. In order to obtain the cognate sublibraries, the PCR fragments were subcloned into the pGemT vector (Promega, Madison, Wisconsin) as described [41]–[45]. For the phage-displayed library the precloned VH and VL repertoire was reamplified and specific restriction sites were inserted by PCR [42]. For the construction of the library the reamplified VH and VL genes were subcloned into the phagemid pHAL14 in two successive ligation steps. The VH PCR products were subcloned into pHAL14 as HindIII-NcoI fragment, the VL PCR products were inserted as NotI-MluI fragment.

The phagemid pHAL14 was chosen as library vector because it allows phage display and the production of soluble antibody fragments without the necessity of recloning. pHAL14 is a high-copy plasmid and contains the scFv under the control of the lacI promoter/operator region which is inducible by IPTG. A His- and c-Myc-tag are encoded at the C-terminus in frame with the scFv and allow the straightforward purification of the recombinant antibody [45].

All cloning and ligation procedures necessary for the library construction were performed in E. coli XL1-Blue MRF' (Stratagene, Amsterdam, The Netherlands) according to Sambrook et al., 1989 [46]. For DNA plasmid and PCR product purification the commercial Plasmid Mini Kit and QIAquick PCR Purification Kit from QIAGEN, Hilden, Germany, were used.

The library was packaged by using Hyperphage [47], [48]. Therefore, 1 ml of the antibody gene library stock was grown to an OD600 nm of 0.4 to 0.5 and about 1.25×10¹⁰ E. coli XL1-Blue MRF' transformants were infected with 2.5×10¹¹ Hyperphage.

E. coli cells and Hyperphage were incubated for 30 min at 37°C without shaking, followed by a further incubation for 30 min with shaking at 250 rpm. The infected cells were harvested by centrifugation and transferred to glucose-free TY-medium containing 100 µg/ml ampicillin and 50 µg/ml kanamycin in order to repress the lac promoter of pHAL14 and allow the expression of the scFv::pIII fusions. After incubation overnight at 250 rpm and 30°C bacteria were removed by centrifugation and phage particles in the culture supernatant were precipitated as previously described [49]. The precipitated and packaged antibody phage library was then resuspended, titered and stored at 4°C until used for the panning and screening [49].

Surface presentation onto phage of the library was examined by SDS-PAGE, Western blotting and anti-pIII immuno-staining as described [50].

The process for the selection of specific scFv from a phage-displayed library is referred to as “panning,”and, in principle, involves the selection of Abs on the basis of their affinity.

The panning of VEEV specific scFv was performed on VRS purified VEEV presented by specifc mAbs coated onto 96 well microtitre plates (Maxisorb, Nunc, Wiesbaden, Germany) with minor modifications according to [12].

The panning and amplification of specifically bound phage was repeated three times before proceeding with postpanning analysis and each panning included a preselection and selection step. In order to remove unspecific binders, the library was preincubated in parallel with non-infected Vero cells and a nonspecific murine IgG. Therefore, 100 µl of the phage library was incubated for 1 h at room temperature (RT) onto a mixture of coated and concentrated VRS supernatant from non-infected Vero cells plus 1% BSA (bovine serum albumin) in PBST (PBS plus 0.01% Tween 20) containing 10 µg/ml of the murine IgG D1-4G2-4-15 [51].Washing and reamplification of specifically bound phage was performed as described [12].

Sequencing and sequence analysis.

Sequencing was performed by GATC Inc. (Konstanz, Germany) using the oligonucleotide primer MHLacZ-Pro_f (5′ ggctcgtatgttgtgtgg 3′). The antibody gene fragments were analysed by using VBASE2 (http://www.vbase2.org) and Kabat's database of sequences of immunological interest (www.bioinf.org.uk) as previously described [45], [52].

Construction of stable eukaryotic tranfectants, expression and production of scFv-Fc ToR67-3B4.

VEEV specific scFv gene fragments were subcloned from the immune library vector pHAL14 into the mammalian expression vector pCMV2.5-hIgG1-Fc by using the restriction sites NcoI and NotI [49]. The mammalian expression vector allows the expression of scFvs in frame with the human IgG1 gene. The fusion is under the control of the CMV promoter. Moreover, the vector contains besides the bla-gene a neomycin phosphotransferase (Neo) expression cassette for antibiotic selection with G418.

For the stable production of VEEV specific scFv-Fc fusion proteins, Chinese hamster ovary K1 cells (CHO-K1), from the American Type Culture Collection, (ATCC, Rockwell, MD, No. CCL61) were transfected by using 80 µl Polyfect® (Qiagen GmbH, Hilden, Germany) plus 4 to 5 µg of plasmid DNA. Stable clones were selected for resistance to the aminoglycoside antibiotic Geneticin (G418). For this purpose CHO-K1 cells were cultivated to 60% to 80% confluence overnight in 1000 mm² Petri dishes (Nunc, Vienna, Austria) in non-selective Dulbecco's Modified Eagle's Medium (DMEM/HAM's F-12), a nutrient mixture supplemented with L-glutamine and sodium bicarbonate, containing 10% (v/v) fetal calf serum (FCS) and 0.1% (w/v) penicillin and 0.1% (w/v) streptomycin (PAA, Parsing, Austria) at 37°C in 4% CO₂ atmosphere. In order to support the DNA-lipid complex formation, plasmid DNA and Polyfect® were preincubated in serum- and antibiotic-free medium for 10 minutes at room temperature prior to transfection. While the lipid-DNA complex formation took place, CHO-K1 cells were washed with PBS. Afterwards, 7 ml of medium with FCS and penicillin/streptomycin was added to the complex. The whole mixture was then immediately added to the washed cells by gently swirling. Cells were incubated for 3 h with the transfection complex. Subsequently the medium was removed and replaced with fresh non-selective DMEM/HAM's F-12 containing 10% (v/v) FCS and 0.1% (w/v) penicillin/streptomycin before the plates were incubated overnight at 37°C in 4% CO₂. A further medium change was performed one day after the transfection. The transfected cells were trypsinised and re-sown 1∶20 or 1∶50 in new Petri dishes on selective medium, containing additionaly 700 mg/ml G418. Medium change was performed every 3 to 4 days and first stable and G418-resistant clones were visible about 3 weeks after transfection. Four and five weeks after transfection single clones were isolated and cultivated onto 24 well plates (Nunc, Vienna, Austria). Since the productivity of individual clones varies, we tested the supernatant of 48 clones for VEEV-specific scFv-fusion production in ELISAs as described. Finally, three of the best expressing CHO-K1-clones were cultivated as suspension culture in a miniPerm bioreactor (Sarstedt, Nümbrecht, Germany) in DMEM/HAM's F-12 containing 10% (v/v) (FCS), 1% (w/v) penicillin/streptomycin and 700 mg/ml G418 at 37°C in 4% CO₂. The culture supernatant was harvested twice a week and replaced by fresh medium. Purification of recombinant fusion proteins was done by immunoaffinity chromatography on goat anti-human (GAH) sepharose.

Biotinylation of mAbs and scFv-Fc fusions.

One to two mg of mAb or scFv-Fc was dissolved in sodium bicarbonate buffer, pH 8.5, and incubated with an aliquot of the biotin N- hydroxysuccinimide ester (long arm, water soluble) from Vector laboratories, CA, USA, equal to 1/10 the weight of the protein to be labeled. The mixture was incubated for 2 hours at room temperature. In order to stop the reaction 10 mg of glycine was added and unreacted biotin was removed by gel filtration with PD-10 desalting columns containing Sephadex G25 (GE Healthcare, USA) according to the suppliers' protocol.

Indirect ELISA.

Sucrose density gradient-purified antigen from strains TrD, P676, 3880, Mena II, 78 V, Fe37c, BeAn8, Pixuna, CaAr508 and AG80 (subtypes IA/B, IC, ID, IE, IF, formerly subtypes II IIIA, IV, V and VI) were coated on 96-well Maxisorp™ microtiter plates purchased from Nunc, Wiesbaden, Germany. The viral protein concentration was determined by SDS-PAGE and scanning dosimetry, dilutions were performed in 50 mM bicarbonate buffer pH 8.5. After blocking with PBS containing 1% (w/v) skimmed milk powder (Sigma-Aldrich, UK) plus 0.1% (v/v) Tween 20 detection of VEEV strains was performed with different log₁₀ dilutions of scFv-Fc ToR67-3B4 for 2 hours at room temperature (RT). The starting concentration was 1 µg/ml. Subsequently the microwell plates were washed with PBS-T and the presence of bound antibody was detected by using peroxidase-conjugated goat-anti-human IgG (1∶4.000, AbD Serotec, UK) as secondary antibody. TMB-substrate (Sigma-Aldrich, UK) was used as chromogenic substrate, absorbance was measured at 450 nm.

Direct Sandwich ELISA.

All sandwich ELISAs were performed in 96 microwell plates (Maxisorp™, Nunc, Wiesbaden, Germany) that were either coated with 3 µg/ml antibody or scFv-Fc by incubation for 2 h at 37°C or overnight at 4°C. For the detection of VEEV we used either the monoclonal antibody (mAb) VEE-WIS1 or a mixture of mAbs VEE-WIS1, SFV3/4 and SFV 12/2 as capture antibodies [9], [16]–[19]. MAb VEE-WIS1 (E1-specific) regognises specifically VEEV of subtype IA/B, while the antibody mixture binds Old world as well as New world Alphaviruses. In order to capture WEEV specifically mAb SVF 12/2 was applied. The ELISA to detect EEE virus was done on plates coated with a mixture of mAbs SFV 8/6 (E1-specific), SFV3/4 and SFV 12/2 [9], [16]–[19]. Subsequently antibody coated microtiter plates were washed and blocked with 1% FCS in PBS-T during 1 h at room temperature. After further washes, wells were additionally overlayed with liquid plate sealer (Candor Bioscience GmbH, Wangen, Germany), stored at 4°C and used for ELISA studies within three to six weeks. Samples with virus were incubated for 2 h at room temperature on precoated microwell plates and afterwards extensively washed. Bound viruses were detected by using biotinylated scFv-Fcs or mAbs in appropriate dilutions (1∶500, 1∶2000 or 1∶5000) and incubation for 1 h at 37°C. After three further washes the conjugate Streptavidin-horseradish peroxidase (PSA, GE Healthcare, USA) was applied to the wells in a 1∶ 6000 dilution and plates were incubated for 30 min at RT with agitation. Bound biotinylated mAbs and fusions were detected by using TMB staining (commercial solution from Serva, Heidelberg, Germany) that was stopped with 2 M sulphuric acid after 10 minutes. Absorbance was measured at 450 nm.

Immunoblot analysis.

ScFv-Fc fusion proteins and several monoclonal antibodies as positive controls were used to detect VEEV specific envelope proteins.

Therefore, purified VEEV particles were disrupted either by incubation for 20 minutes at 56°C or 5 minutes at 95°C in Laemmli sample buffer [53] containing either no or 1 mM DTT. The viral proteins were separated by 15% SDS-PAGE and blotted onto a nitrocellulose membrane (Protran, BA85, Sigma-Aldrich, Taufkirchen, Germany). The membrane was blocked with PBS plus 1% Casein, pH7.4 (Thermo Fisher Scientific GmbH, Dreieich, Germany) for 1 h at RT. For the detection of virus-specific proteins, the membrane was incubated with either biotinylated scFv-Fc (Tor67-3B4: 1∶5000) fusions or mAbs (SFV 8/6: E1-specific mAb, 1∶10.000; WIS-VEE1: E1-specific mAb, 1∶1.000; 1A3B7: E2-specific mAb, 1∶1.000) for 1.5 h at RT in the dark. After 3 washes with PBS-T the membrane was incubated with the conjugate streptavidin-alkaline phosphatase (1∶1000, GE Healthcare, Munic, Germany) for 30 min. Staining and visualisation of specific proteins was performed with 5-Brom-4-Chlor-3-Indolyl-Phosphat/Nitro Tetrazolium Blue Chloride (NBT-BCIP, Thermo Fisher Scientific GmbH, Dreieich, Germany) according to standard procedures.

Immunohistochemistry.

Vero cells (100 µl of 2×10⁵ cells/ml) were grown in 96-well microtiter plates for 24 h and subsequently infected with serial log₁₀ dilutions of either VEEV TC83 or TrD. Non-infected Vero cells served as negative control. Specific staining of VEEV infected cells was performed with biotinylated scFv-Fc ToR67-3B4 and mAb SFV 8/6 as positive control. In detail, one day post infection the infected cells were fixed with 3% formalin in PBS for 3 h at 4°C. Afterwards the fixed samples were incubated with their cognate antibodies in either a 1∶5000 or 1∶10.000 dilution in PBSF-T (PBST plus 1% FCS) for 1 h in a humid chamber at 37°C. After 3 consecutive washes in PBS-T the fixed cells were incubated for 30 min with the conjugate streptavidin - horseradish peroxidase (PSA, GE Healthcare, USA) diluted 1∶6000 in PBS-FT. After further extensive washing and in order to visualise infected cells microscopically, the samples were incubated with the precipitating colorimetric peroxidase substrate TMB (KPL, Germany) for 10 minutes. Staining was stopped by rinsing the plates with Millipore-purified water.

Neutralising peroxidase-linked antibody (NPLA) assay.

The NPLA assay for scFv-Fc fusions and complete IgG antibodies was done according to the technique of Jensen [54] with several modifications. All assays were performed in 96 well microwell plates and Vero cells were used as host for infection. Fifty µl of two-fold dilutions of scFv-Fc fusions or mAbs (starting concentration between 0.5 to 1.0 mg/ml) were incubated with an equal volume of VEE virus strains TC-83, TrD or Fe3-7c with a TCID₅₀/ml of 5×10⁴ for 2 hours at 37°C in 96-well plates. Following this incubation, 100 µl of freshly trypsinised Vero cells at a concentration of 4×10⁵ cells/ml were transferred to the antibody virus mixture.

As positive control for infection, virus samples not preincubated with any antibody as well as virus samples preincubated with VEEV specific antibodies that do not display any neutralising activity like mAb WIS VEE1 were used. Moreover, virus samples preincubated with mAb SFV 8/6 and 1A3B-7, both antibodies with known neutralizing capacity ([55] and Hülseweh, personal communication) were applied as a positive control and non-infected Vero cells served as negative control. Twenty to twenty-four hours post infection all cell monolayers were fixed with 3% formalin in PBS for 3 h at 4°C and infection of cells was demonstrated by specific immunostaining of viral antigens. After washing with PBS-T, the monolayers were overlaid with 100 µl of a 1∶5000 dilution of the VEEV-specific biotinylated mAb SFV 8/6 and incubated for 2 h at RT. Bound biotinylated mAbs were detected by using the streptavidin HRP conjugate (GE Healthcare, USA, 1∶6.000) and visualisation was performed with precipitating TMB as substrate. The staining reaction was stopped by adding 100 µl of 1 M sulphuric acid and absorbance was measured at 450 nm in a microwell plate reader. The absorbance values were correlated to relative infectious activity. Thereby absorption at 450 nm (OD450 nm) values of the positive control, namely Vero B4 cells infected with VEEV, were set as 100% infectivity. A reduction equal or greater than 50% in A450_nm in the wells was considered as indicative for neutralisation.

Plaque reduction Assay.

The ability of scFv-Fc ToR67-3B4 (25 µg) to neutralise virus infectivity was determined by mixing the construct with VEEV strains TrD, Fe37c or BeAn8 (approximately 100 pfu) and incubating at 4°C overnight. Residual infectious virus was estimated by a standard plaque assay in L929 cells as described [37].

Determination of the in vivo half-life of scFv-Fc ToR67-3B4.

Sera were harvested from the marginal tail veins of three BALB/c mice (6–8 weeks old; Charles River Laboratories, UK). Five days later, 100 µg scFv-Fc ToR67-3B4 were injected by the intraperitoneal route. Serum samples were collected 8 h, 1 d, 3 d, 5 d, 10 d, 15 d and 25 d post-administration. The concentration of scFv-Fc ToR67-3B4 in sera was determined by ELISA using sucrose density gradient-purified β-propiolactone inactivated antigen of VEEV strain TC-83. Antibody concentrations were estimated by comparison of the absorbance values generated by diluted serum samples with a standard curve prepared with scFv-Fc ToR67-3B4.

Statistical methods.

Statistical analysis was performed using GraphPad Prism software (http://www.graphpad.com). To determine the half-life of scFv-Fc ToR-3B4 in vivo, a linear regression model was fitted to the log transformed data to estimate the rate of change over time. A change of 0.30 units on the log scale equates to a 50% reduction of the data on the original scale. This value is divided by the estimated model slope to obtain the half-life.

Passive protection studies.

BALB/c mice, 6–8 weeks old, were used for all passive antibody protection studies. Mice were challenged by the airborne route with 100LD50 strain TrD [56], [57]. The LD50s for strains Mena II, Fe37c and Mucambo virus have not been determined. However, titres were used in the collision nebuliser that have reliably achieved 100% mortality in previous experiments [36], [37]. Afterwards mice remained either untreated or were injected intraperitoneally with 100 µg scFv-Fc ToR67-3B4, human IgG (Sigma-Aldrich, Taufkirchen, Germany) or control construct (anti-WEEV). Mice received just one injection at a single time-point (6, 24, 48 or 72 h) post-challenge and were monitored for clinical signs of infection (piloerection, hunching, inactivity, excitability and paralysis) twice daily by an observer who was unaware of the treatment allocations. Animals were scored according to the severity of the clinical signs and were humanely culled as appropriate [58]. These experiments therefore record the occurrence of severe disease rather than mortality. Even though it is rare for animals infected with virulent VEEV and showing signs of severe illness to survive, our use of humane endpoints should be considered when interpreting any virus dose expressed here as 50% lethal doses (LD₅₀). In one experiment, sera were harvested by cardiac puncture from mice that were alive fourteen days post-challenge. Samples were assayed for VEEV-specific antibodies using sucrose density gradient-purified β-propiolactone-inactivated antigen from strain TC-83.