Remember me
Background:
Foot-and-mouth disease (FMD) is a contagious disease affecting cloven footed animals. Existing vaccines although effective, has limitations of short immunity, needs cold chain maintenance, high biosafety levels facilities for production and inability to differentiate infected from vaccinated animals. Virus-like particles (VLPs) provide an alternative strategy and the structural integrity of VLPs at moderate temperatures or acidic pH plays major role as they destabilize the capsid and exhibits poor immunogenicity when used as vaccine. Stabilized capsids are developed using bioinformatics approach to predict amino acids in the structural proteins of the FMD virus, which may provide thermostability and can withstand live virus challenge.
Methods:
Baculovirus expression system offer an attractive method for producing VLPs of foot-and-mouth disease virus, which mimics the native virus and elicits protective immune response. In-silico bioinformatics structural analysis was harnessed to predict thermostable amino acids on the VP2 and VP3 proteins of FMDV serotype A/IND/40/2000 by employing molecular modelling. Consequent mutations were introduced into the VP2 and VP3 proteins, F62Y and H142D, respectively, resulting in the formation of a double mutant (AM-3) and VLPs expressed in Tn5 cells. The AM-3 VLPs were tested in thermostability test in-vitro for different temperature and time points and also tested by in-vivo in guinea pigs for vaccine efficacy.
Results:
The demonstration of capsids of FMDV in the transmission electron micrograph confirmed the expression of AM-3 VLPs. Thermostability studies revealed AM-3 had significantly low degradation (62.5%) on 15 days post storage at 37°C amongst all VLPs and validated by in-vivo studies. Finally, the VLPs conferred 90% protection in guinea pigs and could serve as a thermostable candidate vaccine.
Conclusion:
The thermostable AM-3 VLPs produced based on bioinformatics without compromising the structural integrity, could confer protective immunity in preclinical studies and which can serve as a potential thermostable candidate vaccine for FMDV serotype A/IND/40/2000 for controlling FMD in ruminants.
1 IntroductionFoot-and-mouth disease virus (FMDV), the causative agent of FMD, belonging to the genus Aphthovirus of the Picornaviridae family, is a highly contagious disease of the cloven-hoofed animals. It causes huge economic loss in ruminants in terms of productivity and reproductive losses which needs to be tackled by control program. The diversity of FMDV exists in the form of seven serotypes (O, A, C, Asia1, SAT1, SAT2, and SAT3) and numerous subtypes (Bachrach, 1968). The FMD vaccine contains binary ethylene imine (BEI) inactivated virus (Bahnemann, 1990) of prevailing serotypes (O, A and Asia1 in India) with oil adjuvants and being used in the control program. The limitations of the inactivated trivalent FMD vaccine include short duration of immunity, stringent bio-containment facility for vaccine manufacturing, maintenance of cold chain, and vaccine related outbreaks. Further, the thermolabile nature of the whole virion disintegrates, leading to poor immunogenicity, warranting deployment of new strategies to improve vaccine stability by maintaining the structural integrity of 146S antigen (virus capsid with inactivated RNA genome is referred as 146S antigen, where S indicates Svedberg unit) (Scott et al., 2017) of the FMDV through bioinformatics approach.
The genome of FMDV consists of P1, P2 and P3 polyproteins. P1 encompasses structural proteins such as VP1, VP2, VP3 and VP4; whereas P2 and P3 encode the non-structural proteins which are responsible for virus replication and translation. P1 and 2A (cleavage protein) together form the empty capsid. The capsid of FMDV is composed of 60 copies each of the structural proteins VP1, VP2, VP3, and VP4. The four structural proteins (VP1-4) together form a single protomer (5S), five protomers join to form a pentamer (12S), and 12 such pentamers combine to form an empty capsid or virus-like particles (VLPs) of 75S which elicits antigenicity and immunogenicity as that of whole virus (Rincón et al., 2014; Cao et al., 2009). The interpentameric (IP) region consists of eight chains of peptides, two each of VP1, VP2, VP3, and VP4. Improving the stable interactions between the peptide chains at the IP interface could offer stability to capsid and prevent its dissociation into intermediates (Kotecha et al., 2015). Earlier VLPs with improved stability were produced by mutating residues at the IP region (Porta et al., 2013). Mutagenic studies revealed that H142 of VP3 is essential for genome uncoating and remains conserved among the serotypes of FMDV (Curry et al., 1995) as it is the major site of electrostatic repulsion for dissociation of capsid under low pH (van Vlijmen et al., 1998). A single mutation of A65H in VP2 in the infectious FMDV serotype C-S8c1 strain increased the thermostability by optimizing the electrostatic interactions (Rincón et al., 2014).
The development of thermostable VLPs is a potential method for optimizing the stability of the vaccine than the infectious virus, as suggested earlier by the introduction of H93C of VP2 mutation at the IP region of FMDV serotype A22 Iraq (Porta et al., 2013). Earlier attempts to predict the mutations by bioinformatics studies for FMDV VLPs of Asia1/IND/63/1972 successfully showed increase thermal stability amongst the mutated VLPs (Aparna et al., 2024). Although, A22 Iraq strain shares high homology with A/IND/40/2000 (88%), interpentameric region of both the strains are less similar to each other which need to be confirmed with thermostability studies. Therefore, we envisaged developing thermostable mutant VLPs of FMDV serotype A/IND/40/2000 using bioinformatics predictions to validate the stability of VLPs through in-vitro and in-vivo methods. The mutated stable VLPs produced in insect cells were confirmed in western blot and transmission electron microscopy imaging. The thermostability of VLPs was assessed based on shelf-life studies and differential scanning fluorescence assay. We have successfully demonstrated for the first time the protective antibody titer of the wild type and thermostable VLPs of FMDV serotype A in guinea pigs. The challenge studies further confirmed the protective efficacy of the thermostable VLPs as candidate vaccine and also favors the use of differentiation of infected from vaccinated animals (DIVA) as guinea pigs immunized with VLPs did not show 3AB antibodies in the serum samples by blocking ELISA.
Our results suggested that a double mutant, F62Y:H142D (F62Y mutation in VP2 and H142D in VP3) designated as AM-3 VLPs produced in insect cells had shown better thermostability and immunogenicity than the wild type and other mutant VLPs. The bioinformatics-based design of stable capsid has a potential for developing a safe and alternative FMD vaccine, which will be highly preferred in the final stages of the progressive control pathway (PCP) of FMD as per the World Organization for Animal Health (WOAH) guidelines. For prevention and eventual disease eradication, thermostable VLPs based screening ELISAs and vaccines will be highly preferable over the inactivated virus due to their inherent non-infectious nature. Here, we report the innovative bioinformatics approach used in predicting the mutations and eventually validating the in-silico methods by including appropriate positive and negative controls. The results of bioinformatic analysis were validated by producing VLPs in insect cells and tested both in-vitro and in-vivo and could serve as an alternative to current inactivated FMD vaccine for resource limited region for FMD control and eradication.
2 Material and methods2.1 Viruses and cellsFMDV serotype A/IND/40/2000 virus (P10) and Sf-21 (Spodoptera frugiperda) and Tn5 (Trichoplusia ni) insect cells available at ICAR - IVRI, Bengaluru, were used to produce VLPs. The Baby Hamster Kidney-21 cell line (BHK-21) procured from the American Type Culture Collection (ATCC, USA) was cultured in Glasgow’s modified Eagle’s medium (HiMedia, India) supplemented with 10% fetal bovine serum (HyClone™, USA) was used for virus production.
2.2 Construction of transfer plasmidsThe P1-2A-3C region of FMDV serotype A/IND/40/2000 was cloned in pFastbac1 vector as reported earlier (Bhat et al., 2013; Hassine et al., 2020; Gashti et al., 2024). Briefly, RNA was isolated from virus, cDNA was synthesized, P1-2A and 3C genes were amplified separately (Supplementary Table 1). The amplified genes were subsequently cloned into the pFastbac1 vector (Invitrogen, USA). The 3C gene of FMDV was mutated at positions 38 and 48 (G38SF48S) by site-directed mutagenesis, and a single cassette of P1-2A-3C was further cloned and inserted into a transfer pFastbac1™ plasmid (pFastbac1-A-P1-2A-3C).
2.3 In-silico prediction modeling using bioinfomatics to determine the amino acids contributing to the stability of capsidsHomology modelling (HM) of the IP region of the FMDV serotype A/IND/40/2000 (Accession no. HM854025) was constructed using A22 Iraq crystal structure (PDB code: 4GH4) as the template using PRIME module of the Schrodinger software suite interfaced with Maestro module (Schrödinger, 2019). Intra and inter-molecular interactions at the IP region were visualized and analyzed using the PyMol (www.pymol.org) software. In-silico point mutations were introduced using ‘coot’ and the rotamer orientation with minimal clashes were used for structure stability predictions.
Molecular dynamic (MD) simulations were performed using the Maestro tool of the Schrodinger suite (Schrödinger, 2019). Prepared mutant/controls structures were solvated using simple point charge, water molecule with 150 mM salt concentration ensuring the charge neutrality. Unrestrained MD simulations were carried out for 10 nanoseconds on each of the mutants on the residues lying within a 30 Å radius from the mutation site. Seven out of the 8 chains forming the IP region were represented as the receptor, while the chain containing the site of mutation is termed as the ligand and binding free energy (ΔG) value was calculated following a published protocol (Kotecha et al., 2015). The molecular mechanics/Poisson-Boltzmann surface area (MM-PBSA) module of the AMBER package was used for calculating the binding free energy (ΔG) value of all the mutations (Gohlke and Case, 2004; Genheden and Ryde, 2015) including the wild type and the negative control. The details of the mutations, their binding free energy values and S-ELISA reactivity were given in the Table 1.
Mutant#Description of mutations in VP2, VP3, and IP regionBinding free energy (ΔG) (kcal/mol)#S-ELISA reactivityDescription of mutants predicted by bioinformatics with binding free energy values (ΔG) of VLPs of FMDV serotype A/IND/40/2000 and their antigenic characterization using Sandwich ELISA.
#The molecular mechanics/Poisson-Boltzmann surface area (MM-PBSA) module of the AMBER package was used for calculating the ΔG values from the dynamic trajectories of every mutant and A-WT structure. AM refers to mutations in either the VP2 and/or the VP3 protein of FMDV.
$Optical density or absorbance indicates the mean ± standard error (n=4) of the absorbance value at a wavelength of 492 nm measured in an ELISA reader. The positive control with inactivated antigen had an A492 absorbance of 2.65 ± 0.07, whereas the negative control Tn5 cell lysate had an A492 absorbance of 0.103 ± 0.05.The bolded value indicates the best mutant among all and therefore used in the further experiments in the research work.
2.4 Generation of recombinant baculovirus with mutations in the structural genes of FMDVThe predicted mutations (single or double) were introduced in the IP region (VP2-H3 helices and the proximal VP3) by PCR-based site-directed mutagenesis using KOD mutagenesis kit (Toyobo©, Japan) with wild type as control. The double mutants (AM-3 and AM-4) were constructed after confirming the presence of the first mutation by sequencing, and the recombinant plasmid containing the first mutation was used as template for second mutation. The resultant plasmid was confirmed to contain two mutations and was termed as the double mutants AM-3 and AM-4 (Table 1). The mutations were confirmed by sequencing and the plasmid was transformed into DH10 Bac™ E. coli competent cells (Thermo Fisher Scientific, USA). The recombinant clones (white colony) were selected, grown in LB broth and used for bacmid DNA isolation. The bacmid DNA was transfected into Sf-21 cells using Cellfectin (Thermo Fisher Scientific Inc., USA) to generate recombinant baculovirus for each mutant separately. The recombinant baculovirus was plaque purified and passaged in Sf-21 cells upto P4 to serve as stock virus.
2.5 Characterization of VLPsThe Tn5 cells were used for expression of 9 recombinant baculovirus VLPs. The expressed VLPs were harvested as described earlier (Terhuja et al., 2015) and characterized by sandwich ELISA (S-ELISA) (Bhat et al., 2013), electro-immunoblot transfer assay, transmission electron microscopy (TEM) (Ruiz et al., 2014) and immunofluorescence assay (Basagoudanavar et al., 2015).
2.6 Bulk production and quantification of VLPsThe recombinant P4 stock virus for each VLP was used at five multiplicities of infection (m.o.i) for bulk production in Tn5 cells by monolayer method and VLPs were harvested on 3-day post infection (dpi) when 80% cells showed cytopathic effect. The standard 146S antigen of FMDV serotype A was prepared as earlier (Barteling and Meloen, 1974) and used for VLP quantification in S-ELISA. S-ELISA was performed (Bhat et al., 2013) by diluting the 146S standard antigen (8 µg to 4 ng) and the VLPs by two-fold serial dilution. The absorbance (A492 nm) values of the standards were used for construction of standard curve using Curve expert 1.4 and VLPs were quantified.
2.7 Evaluating the thermostability of VLPsVLPs were concentrated by ultracentrifugation in sucrose gradient (Ganji et al., 2018) and exposed to three different temperatures of 37, 45, and 56°C for different time points and the thermal stability was assessed by S-ELISA.
The initial concentration of 4 µg of each mutant and the wild type VLPs were stored at different temperatures of -20, 4, 24, and 37°C for 90 days. The stored VLPs were tested at 15 days intervals starting from day 0 to 90 post-storage (n=3/mutant/time point). The relative degradation of VLPs (%) tested in S-ELISA was calculated using the following formula:
2.8 Characterization of the unfolding temperature of VLPsDifferential scanning fluorescence (DSF) was performed to determine the unfolding temperature of VLPs using Protein binding fluorophore and SYPRO Orange dye in real-time PCR machine (Applied Biosystems real-time PCR machine 7500). Briefly, the temperature was ramped from 25 to 95°C with 1°C increments at an interval of 3 min in duplicate wells of a MicroAmp Optical 96-well plate. SYPRO orange fluorescence was produced with an excitation and emission wavelengths of 472 and 570 nm, respectively. The signal was detected using an ROX filter, and the minimum of the negative first derivative of the fluorescence curve was taken as the melting temperature (Tm). Each assay was performed pre and post exposing VLPs at 37, 45, and 56°C for 30 min and 60 min (n=3/mutant/time point).
2.9 Animal experimentationAnimal experiments were conducted as per the guidelines of Institutional Animal Ethics Committee (IAEC) of Indian Veterinary Research Institute (IVRI) campus, Hebbal, Bengaluru, India.
2.10 Determining the potency of stable VLPs by dose response study in guinea pigsDunkin Hartley guinea pigs were divided into five groups of six animals in each group and injected with VLP vaccine formulated with Montanide ISA 201VG adjuvant. Groups 1 to 5 were immunized (I/M) with 1, 2, 4, 8 and 12 µg/dose/animal, respectively. Animals were boosted on 30-day post vaccination (dpv) with respective antigen. On 58 dpv animals were challenged with 50 µL of serotype A 100 GPID50 virus (Median guinea pig infective dose50) by intradermally tracking in the left rear footpad (intra dermoplanter in the metatarsal pads). Further, animals were observed for ten days post challenge (dpc) for the appearance of lesions on the footpad. presence of lesions on any one of the non-inoculated foot pads indicates the animal is not protected, while absence of lesions in all the three non-inoculated footpads suggested that the animal is protected. To evaluate the antibody response against VLPs antigen, 2 mL of blood was collected intracardially on 0, 28 and 56 dpv and the sera were tested by Virus Neutralization Test (VNT) as per World Organisation for Animal Health (WOAH) (2022) protocols.
2.11 Comparison of potency of the AM-3 and wild-type VLPs in guinea pigsGuinea pigs (n=44) were divided into four groups with eleven animals per group. Groups 1 to 4 were immunized intramuscularly at the left hind limb quadriceps muscle with PBS, 12 µg of AM-3 VLPs, 12 µg of wild type A-WT VLPs and 4 µg of FMDV A inactivated 146S antigen, respectively. Animals were boosted on 30 dpv with the respective antigens and challenged on 58 dpv with 100 GPID50 virus intra-dermally by tracking in the left rear footpad. Animals were observed for ten dpc for lesion scoring to determine percentage protection and potency of the VLPs. The antibody titres of the sera on 0, 28 and 56 dpv were carried out by VNT.
2.12 Detection of FMDV non-structural protein antibodies by 3AB NSP blocking ELISAThe 3AB NSP blocking ELISA was performed as reported (Hosamani et al., 2022). Briefly, the recombinant 3AB antigen was coated in 96 well MaxiSorp (Nunc, USA) ELISA plate and incubated for 60 min at 37°C. The guinea pig sera were diluted (1:1) in blocking buffer and added to the antigen-coated plate. Following washing with Tris (25 mM) buffered saline containing Tween-20 (0.025%), 10H9D8 mAb HRPO conjugate was added to the plate and incubated for 30 min at room temperature. Post washing, 50 µL TMB substrate was added and incubated for 15 min in dark at room temperature and the reaction was stopped by addition of 50 μL of 1.5 M H2SO4. The absorbance was measured at 450 nm and percentage inhibition (PI) was calculated using the following formula:
Where ODTS is the mean OD of the test serum and ODNCS is the mean OD of the negative control serum.
2.13 Statistical analysisGraphPad Prism version 9.5.1 was used for statistical analysis. Data pertaining to degradation of VLPs and shelf-life at different temperature and time point, Tm values to find the thermal stability at high temperatures and neutralizing antibody titres of guinea pigs were analyzed by two-way ANOVA using Holm-Sidak post-hoc test. The dose to protection was modelled by simple binary logistic regression. Kaplan-Meier survival curve analysis was used for comparison of % inhibition of FMDV 3AB blocking ELISA with protection status among different vaccinated groups.
3 Results3.1 Homology modelling and molecular dynamics of FMDV structural proteinsBased on previous studies on other strains (Kotecha et al., 2015; Porta et al., 2013) and structure analysis at the IP region of the constructed model, F62, H93, S97, and V90 of VP2 and H142 of VP3 were selected for in-silico bioinformatic analysis (Figure 1). The differences in the local interactions due to mutations were analyzed by static and MD simulation trajectory models. In AM-3 double mutant (F62Y:H142D), two new interactions at the IP region were formed involving K88 of VP2 and E138 of VP3 at either end of the H3 helix (Figure 2; Supplementary Figure 1). Binding free energy (ΔG) of the serotype A wild type (A-WT) and mutants including controls were calculated (Table 1). The negative control was designed (F62N of VP2 and H142R of VP3) in such a way that it would disrupt the hydrophobic interactions of F62 and produce steric hindrance at the H142 site, thereby destabilizing the IP region. Of all the mutant structures analyzed by bioinformatics, AM-3 showed the lowest ΔG value (1091.9 kcal/mol), while A-WT (1201.1 kcal/mol) and the negative control mutant (F62N and H142R) showed higher ΔG values (1230.1 kcal/mol) as given in Table 1.
H3 helix region of VP2 forming a part of the inter-pentameric region: The crystal structure of the A22 Iraq strain (PDB code: 4GH4) was chosen as a template for constructing the homology model of FMDV serotype A/IND/40/2000. (A) Cartoon representation of the interpentameric region prepared using Schrodinger Maestro software. Residues interacting at the IP region are shown in the sticks. The IP region contains 8 peptide chains comprising 2 units each of VP1, VP2, VP3, and VP4. The amino acids V90, H93, S97 and F62 of VP2 and H142 of VP3 are targeted for mutation studies. Sequence and secondary structure features of VP2 (B) and VP3 (C). The sites of mutation F62 in VP2 and H142 in VP3 are marked with red stars below the sequence. Red arrows indicate the residues that interact with the WT or mutant residue at the IP region.
Schematic representation of the interactions in the double mutant, AM-3 of the FMDV serotype A determined using LigPlot: Panels (a) and (b) depict the changes in the intramolecular interaction due to mutation of F62Y in VP2. Similarly, panels (c) and (d) depict the same between wild-type and H142D in VP3. Arc with spikes indicate hydrophobic interactions. Substitution of F with Y in the VP2 region increased the hydrophobic interactions to five, of which two hydrophobic interactions involved E138 and D142 of the VP3 region of the neighboring pentamers. Substitution of H with D (c) resulted in reduced electrostatic repulsion in the VP3 region, which in turn provided the advantage of increasing the hydrophobicity by attracting Y62 and K88 of the VP2 region in the neighboring pentamer (d).
3.2 Characterization of recombinant baculovirus expressed VLPsRecombinant baculovirus containing the mutation in the VP2 and/or VP3 proteins were generated for eight predicted mutants and one wild type VLPs. They were characterized by S-ELISA and other techniques. The S-ELISA results showed that the optical density of the mutated VLPs was comparable with the wild type (Table 1). However, AM-3 showed the highest O.D of 2.52 than the other VLPs. The electro-immunoblot transfer assay demonstrated the expression of 25, 33 and 81 kDa corresponding to VP1/VP3, VP0, and P1-2A confirming the self-cleavage of polyprotein (P1) (Supplementary Figures 2.1, 2.2). The ultracentrifuged fractions of S-ELISA showed fractions no. 14, 15 and 16 had the highest O.D values of above 2 as compared with other fractions (<1 O.D) (Supplementary Figure 3). The pooled and concentrated VLPs demonstrate the presence of 25–30 nm particles in the TEM images confirming the formation of intact AM-3 VLPs (Figure 3). The AM-3 VLPs reacted with a neutralizing mAb, which was previously characterized with 146S whole virus antigen, which confirmed the presence of similar antigenic epitopes in VLPs as that of whole virus. Further, the appearance of similar fluorescence in immunofluorescence test between the AM-3 and A-WT corroborated the intactness of the antigenic site (Supplementary Figure 4).
Demonstration of the transmission electron microscopic image of stable AM-3 virus-like particle: The AM-3 VLP was purified using different sucrose density gradients by ultracentrifugation. The fraction with peak reactivity was taken for TEM imaging. Samples were stained with 1% uranyl acetate for 1 min and dried at 37°C for 1h. Grids were examined under the transmission electron microscope (Tecnai, T12, USA) operated at 80 kV and images were acquired. The blue arrow indicates the AM-3 VLP in the size of 25–30 nm corresponding to an empty capsid.
3.3 Quantification of expressed VLPs for stability and animal experiment studiesThe 146S antigen was purified from the supernatant of FMDV serotype A infected BHK-21 cells by ultracentrifugation that showed a clear single white band (Supplementary Figure 5) with a concentration of 188 µg/mL and ratio of 259 nm/239 nm was 1.4 which indicates the purity of antigen. The concentration of VLPs was quantified based on the standard antigen O.D. values obtained in the S-ELISA, the standard curve was plotted using R2 value. We obtained maximum concentration of 0.1 to 0.2 µg/mL of mutant and wildtype VLPs in one 150 cm2 harvested flask.
3.4 Stability of the VLPs at different temperatures and time pointsS-ELISA reactivity of mutants and wild type VLPs subjected to different temperatures and time intervals was shown in Figure 4. At 37°C (Figure 4A), AM-3 showed significantly lower degradation of 62.5% on 15 days post storage (dps) as compared to A-WT VLPs (>80% degradation) [95% CI: -27.65 to -10.65], AM-1 (P<0.05) and AM-8 (P<0.01) showed >70% degradation. On 30 dps, AM-3 showed a significantly lower degradation of <70% compared to A-WT (P<0.001) [95% CI: -29.85 to -12.85]. The stability was poor as evidenced by the antigen degradation of >90% after 60 dps for all the VLPs. At 24°C (Figure 4B), the degradation of the mutant VLP was in the range of 23 to 42.5%, and the difference was not statistically significant amongst all the VLPs, except A-WT VLPs which showed 47.5% antigen degradation on 15 dps (P>0.05) [95% CI: -38.90 to -10.10]. On 30 dps, AM-3 showed a significantly lower degradation of 42.5% in comparison with A-WT (72% degradation) (P<0.001) [95% CI: -43.92 to -15.12] and other VLPs (>90%). At 4°C (Figure 4C), the antigen degradation of mutant VLPs on 15, 30, 45, and 60 dps was comparable with that of A-WT (P>0.05); however, significantly lower degradation of 46% was recorded for AM-3 as compared to 62.5% for A-WT on 75 dps (P<0.001) [95% CI: -30.26 to -2.73]. At -20°C (Figure 4D), the antigenic degradation of mutant VLPs on 15, 30, 45, 60, and 75 dps was comparable with that of A-WT (P>0.05). However, AM-3 had a significantly lower degradation of 23.5% as compared to A-WT on 75 dps (P<0.001) [95% CI: -37.64 to -17.36].
Stability of the mutant and wild type virus-like particles (VLPs) at different temperatures stored for 90 days: (A-D) plots indicate the relative degradation of VLPs stored at 37, 24, 4 and -20ºC. Antigen content on day 0 post-storage was considered 100%. Degradation (%) of the VLPs was calculated by considering the difference between the initial (Day 0 post-storage) antigen content vis-à-vis the day of testing. The aliquots of VLPs stored at different temperatures were subjected to S-ELISA at a 15-day interval from day 0 to 90 to determine the antigen stability by relative degradation (%). The data were analyzed by two-way ANOVA. Orthogonal contrast was done by comparing the A-WT with the mutants (AM). Holm-Siddak post-hoc test was used for comparing each mutant with the wild type. Significance was set at a 5% level. Each point in the line chart indicates mean ± standard error (n=3/time point). The mutant AM-3 showed significantly less degradation than A-WT at 37ºC on day 15 post-storage (P<0.01). The P values for the effect of the mutant, time and interaction were <0.001.
The effect of exposure to high temperature on the stability of different VLPs is presented in Figure 5. At 37°C (Figure 5A), the reactivity of the mutant VLPs was comparable with that of A-WT for 0 to 120 min; however, at 240 min, it was significantly greater (0.46 OD) for AM-3 as compared to A-WT (P<0.05) [95% CI: -0.06 to -0.98]. At 45°C (Figure 5B), the reactivity of all the mutants was low and comparable with that of A-WT up to 240 min (P>0.05) except for AM-3, which had a significantly greater reactivity between 30 to 240 min (P<0.01) [95% CI: -0.04 to -1.05]. At 56°C (Figure 5C), no difference in the reactivity of mutant VLPs was observed as compared to A-WT at any time point (P>0.05) [95% CI: -0.31 to - 0.55].
Effect of exposure to different high temperatures on the stability of mutant and A-WT VLPs: (A–C) plots show optical density values of VLPs exposed to 37, 45, and 56°C for 4 h. The VLPs were incubated at 37, 45 and 56ºC for 30, 60,120, and 240 min and the stability of VLPs was assessed by Absorbance (A492) values using S-ELISA. The data were analyzed by two-way ANOVA. Orthogonal contrast was done by comparing the A-WT with each mutant (AM1-8). Holm-Siddak post-hoc test was used for pair-wise comparison of the mean difference. Significance was set at a 5% level. Each point in the line chart indicates mean ± standard error (n=3/time point). The experiment was repeated thrice and result is presented. The P-values for the effect of the mutant, time, and interaction were <0.05.
The results of exposing the mutant VLPs to three different temperatures were studied using differential scanning fluorescence assay (Figure 6; Supplementary Figure 6; Supplementary Table 2). Before exposing VLPs to high temperatures, there was no significant difference in their melting temperatures (Tm) and the negative control sample did not show any peak. At 37°C, there was no difference in the Tm of the mutant VLPs for 30 and 60 min (P>0.05) [95% CI: 2.58 to 3.65] incubation. However, when exposed to 45°C, the mean difference in the Tm (°C) of AM-3 and A-WT was 3.18 and 3.20 at 30 and 60 min, respectively (P<0.001) [95% CI: 3.01 to 4.22]. However, AM-3 demonstrated significantly high melting temperatures at 45°C and 56°C as compared to A-WT for 30 min (P<0.001) [95% CI: 0.47 to 1.9] incubation indicating a better thermostability. Based on the stability studies, it was observed that AM-3 VLPs are more stable compared to other VLPs.
Differential scanning fluorescence (DSF) assay showing the melting temperature (Tm) curve of the VLPs of the FMDV serotype A: Thermostability of wild type and mutant VLPs were analyzed by DSF assay. The Tn5 cells expressed VLP were pelleted over 20% sucrose by ultracentrifugation at 1, 31,101 g for 5 h at 10°C. The concentrated VLPs were quantified by S-ELISA with known FMDV 146S standard antigen and 4 μg of VLPs each were used for the DSF assay. SYPRO orange fluorophore served as a signal at a concentration of 5X. Part (a) shows the Melt curve analysis showing the Tm of the VLPs before exposure to high temperature, at 37°C for 30 min and at 37°C for 60 min, Part (b) shows the Tm of the VLPs at 42°C for 30 min and at 42°C for 60 min. Part (b) shows the Tm of the VLPs at 56°C for 30 min and at 56°C for 60 min and Red, blue and black arrows indicate the Tm of VLP mutants, AM-3, AM-1 and A-WT, respectively. Exposure of VLPs to three different high temperatures for 60 min (n=3/mutant/time point) was analyzed by two-way ANOVA with orthogonal contrast using the Bonferroni post-hoc test. The melting temperature of the VLP mutants was comparable (P>0.05) when exposed to 37°C for 60 min; however, AM-3 demonstrated significantly high melting temperature at 45°C (b) and 56°C (c) as compared to A-WT at 30 min (P<0.001) [95% CI: 0.47 to 1.9] indicating thermostability.
3.5 Determination of protective efficacy of VLPs by dose response study in guinea pigsOn the basis of the results of the stability experiments, stable AM-3 and wild-type VLPs were chosen for further animal experimentation. The serum neutralizing antibody titres of the immunized guinea pigs were presented in Figure 7a as a log10 value and Table 2 represents the rise in serum neutralizing antibodies post immunization. The mean log10 titres of group 1–5 values were calculated as 0.85, 1.13, 1.25, 1.40 and 2.48, respectively. The dose to protection was modelled by simple binary logistic regression and the results indicated that AM-3 VLP at 12 µg/dose conferred complete protection following challenge (Figure 7b; Supplementary Figure 7).
Dose response study of thermostable AM-3 VLPs in the guinea pig: A total of 36 seronegative guinea pigs were divided into six groups and six animals for each group. The PBS and thermostable AM-3 VLP at different doses such as (in µg), 1, 2, 4, 8 and 12 were used to immunize the 1, 2, 3, 4, 5 and 6 groups of guinea pigs, respectively. Serum was collected on days 28 and 56 post-vaccination to evaluate the vaccine-induced antibody titer. On day 58 post-vaccination all the animals were challenged with 50 µL of guinea pig adapted FMDV serotype A virus at 100GPID50 through intra-dermally by tracking in the left hind footpad. The absence of lesions on the non-inoculated foot pads on day 10 post-challenge was considered protected and the immunogenicity of AM-3 VLP was calculated. (a) Serum neutralization titer at each dpv was analyzed by two-way repeat measures ANOVA. (b) Simple binary logistic regression model was fit to find the log odds of protection (Y) using Graph Pad Prism 9.5.1. The results indicated that at thermostable AM-3 VLP at ≥ 8 µg conferred > 75% protection following challenge.
GroupGuin
Comments (0)