A Vaccine Strain of the A/ASIA/Sea-97 Lineage of Foot-and-Mouth Disease Virus with a Single Amino Acid Substitution in the P1 Region That Is Adapted to Suspension Culture Provides High Immunogenicity

There are seven viral serotypes of foot-and-mouth disease virus (FMDV): A, O, C, Asia 1, and Southern African Territories 1, 2, and 3 (SAT 1–3). Unlike serotype O FMDV vaccine strains, vaccine strains of serotype A FMDV do not provide broad-range cross-reactivity in serological matching tests with field isolates. Therefore, the topotype/lineage vaccine strain circulating in many countries and a highly immunogenic strain might be advantageous to control serotype A FMDV. We developed a new vaccine strain, A/SKR/Yeoncheon/2017 (A-1), which belongs to the A/ASIA/Sea-97 lineage that frequently occurs in Asian countries. Using virus plaque purification, we selected a vaccine virus with high antigen productivity and the lowest numbers of P1 mutations among cell-adapted virus populations. The A/SKR/Yeoncheon/2017 (A-1) vaccine strain has a single amino acid mutation, VP2 E82K, in the P1 region, and it is perfectly adapted to suspension culture. The A/SKR/Yeoncheon/2017 (A-1) experimental vaccine conferred high immunogenicity in pigs. The vaccine strain was serologically matched with various field isolates in two-dimensional virus neutralization tests using bovine serum. Vaccinated mice were protected against an A/MAY/97 virus that was serologically mismatched with the vaccine strain. Thus, A/SKR/Yeoncheon/2017 (A-1) might be a promising vaccine candidate for protection against the emerging FMDV serotype A in Asia.


Introduction
Foot-and-mouth disease (FMD) is an acute contagious disease that affects clovenhoofed animals such as cows, pigs, sheep, goats, and deer. It induces fever; lameness; and vesicles on the mouth, tongue, snout, teats, and feet [1,2]. The FMD virus (FMDV) belongs to the genus Aphthovirus of the family Picornaviridae and is composed of a single-stranded, positive-sense RNA genome. The virus consists of seven serotypes: A, O, C, Asia1, and South African Territories 1, 2, and 3 (SAT 1-3).
An effective FMD vaccine or vaccine strain should be antigenically similar to causative viruses and evoke powerful immune responses [3]. The FMDV, O1 Manisa, and O PanAsia-2 vaccine strains of serotype O provide broad-spectrum protection. However, vaccine strains of serotype A FMDV do not provide broad-range cross-reactivity in serological matching tests with field isolates because unlike serotype O, serotype A FMDV is antigenically diverse [4][5][6]. Therefore, the topotype/lineage vaccine strain circulating in many countries might be advantageous to control serotype A FMDV. In addition, more immunogenic vaccines are necessary to improve the range of protection [7].

Serial Passaging of FMDV and Virus Plaque Purification
The A/SKR/Yeoncheon/2017 isolate was serially passaged 5 times in LFBK cells, 4 times in adherent BHK-21 cells, and 7 times in BHK-21 suspension cells for virus culture. Virus plaques were purified to obtain a single viral clone [19]. BHK-21 adherent cells in a 6-well plate were infected with the serially passaged A/SKR/Yeoncheon/2017 virus. After adsorbing the cell-adapted virus for 1 h, the cells were overlaid with melted SeaPlaque agarose (Lonza, Basel, Switzerland) mixed with neutral red (Sigma-Aldrich) and incubated at 37 • C in 5% CO 2 for 72 h. Three viral plaques were picked, and the purified viruses in the agar plugs were resuspended in DMEM by pipetting. After centrifugation, the supernatant was used to infect adherent BHK-21 cells, which were incubated at 37 • C in 5% CO 2 . The harvested viruses were serially passaged once in BHK-21 adherent cells and twice in BHK-21 suspension cells to increase the viral titer.

Infectivity Tests in CHO-K1 and BHK-21 Adherent Cells
Viral titers were measured and calculated using the Reed and Muench method at a 50% tissue culture infective dose (TCID 50 ) [20]. CHO-K1 cells were used to test the dependence of FMDV on heparan sulfate (HS) as a receptor [21,22]. To measure the titer in permissive cells expressing integrin, which is the receptor of wild-type FMDV, we performed virus titrations in BHK-21 adherent cells, as described previously herein.

Genome Amplification, Nucleotide Sequencing, and Amino Acid Sequence Alignment
Viral RNA was extracted from 100 µL of supernatant of plaque-purified A/SKR/ Yeoncheon/2017 (A-1, A-2, and A-3) virus and serially passaged once in BHK-21 adherent cells and twice in BHK-21 suspension cells, as mentioned in Section 2.2, using the MagNA Pure 96 System (Roche, Basel, Switzerland). The RNA was treated using a one-step PCR inhibitor removal kit (ZYMO Research, Irvine, CA, USA), and single-stranded cDNA was prepared by reverse transcription using an oligo-(dT) 18

Production of Inactivated Antigen
We prepared 3 shaking flasks containing BHK-21 suspension cells (3 × 10 6 cells/mL), and the medium was replaced with CD-BHK-21 production medium before virus infection. The cells in the flasks were infected with A/SKR/Yeoncheon/2017(A-1) that had been plaque-purified and serially passaged once in BHK-21 adherent cells and twice in BHK-21 suspension cells, as mentioned in Section 2.2, at a multiplicity of infection of 0.001, and cultured in a CO 2 shaking incubator. Inactivated FMDV antigen for the quantification of antigen and the experimental A/SKR/Yeoncheon/2017(A-1) vaccine were obtained as reported [24]. Supernatants were collected at 0, 8, 12, and 16 h post-infection, and the virus was inactivated by binary ethylenimine treatment (final concentration, 3 mM) in a shaking incubator at 26 • C for 24 h. The inactivation step was repeated at 26 • C for 24 h in a new flask. The supernatant was neutralized by treatment with sodium thiosulfate. For antigen concentration and purification, the inactivated viral supernatant was mixed with 3% Polyethylene glycol (PEG) 6000 at 4 • C overnight and centrifuged (10,000× g). The pellet was resuspended in TN (Tris, 50 mM; NaCl, 100 mM) buffer (pH 7.2). The inactivated 146S antigen (intact virion of FMDV) was quantified by sucrose density gradient centrifugation [25]. Briefly, the resuspended pellet was layered onto a 15-45% sucrose density gradient and ultracentrifuged at 110,000 × g at 4 • C in a SW41Ti rotor (Beckman Coulter, Brea, CA, USA) for 3 h. The 146S antigen was quantified by spectrophotometry at 259 nm. The FMD antigen, purified from a layer between 30% and 35% sucrose, was placed on formvar-coated grids and negatively stained with 1% uranyl acetate. The FMD antigen was imaged with a transmission electron microscope (Hitachi 7100, Tokyo, Japan).

Vaccination of Pigs and Cows for Serology Test
Nine-week-old pigs and 8-week-old cows verified as negative for FMDV-neutralizing antibody (neutralizing antibody titer < 1:4) were used with the approval of the Animal Care and Use Committee of the APQA (approval no. 2018-423), and the experiment was performed in an Animal Bio-Safety Level 3 facility at the APQA [26]. A total of 3 pigs and 5 cows were intramuscularly injected with the A/SKR/Yeoncheon/2017(A-1) experimental vaccine (2 mL/dose). The pigs were given a booster injection of the vaccine at 4 weeks postvaccination (WPV). The vaccine was formulated with the A/SKR/Yeoncheon/2017(A-1) vaccine antigen (15 µg/dose) and Montanide ISA 206 (Seppic, Paris, France) as a water-inoil-in-water (W/O/W) double emulsion. Blood samples of pigs were used for VNTs and those of cows were used for 2D VNTs.

VNTs for Homologous and Heterologous Viruses
Blood samples were collected from pigs at 1, 2, 3, 4, 5, and 6 WPV. Sera were separated and heat-inactivated at 56 • C for 30 min. VNT was carried out according to the Manual of Diagnostic Tests and Vaccines for Terrestrial Animals [27]. FMDV A/SKR/Yeoncheon/2017, passaged 4 times in LFBK cells, was used for VNT as homologous virus, and A/VIT/2013, A/SKR/Pocheon/2010, and A22/IRQ/24/64 passaged 4 times in LFBK cells were used for VNT as heterologous viruses. A neutralization reaction was performed using serially diluted serum and 100 TCID 50 of FMDV at 37 • C for 1 h. Moreover, the neutralized viruses were placed in the microplates, and then LFBK cells were added. The microplate was incubated at 37 • C for 48-72 h to assess the cytopathic effect. The neutralizing antibody titer was calculated as the reciprocal number of the maximum dilution of serum that neutralized the 100 TCID 50 of FMDV.  [27]. Serum samples were obtained from 5 cows vaccinated with the A/SKR/Yeoncheon/2017(A-1) experimental vaccine at 28 days post-vaccination (DPV). The same field viruses used for VNT and the A/SKR/Yeoncheon/2017(A-1) vaccine virus were used for 2D VNT. The neutralizing antibody titer of the vaccine serum for 100 TCID 50 of each virus was estimated by regression. The r1 value, which is the antigenic relationship between the vaccine strain and the field strain, was calculated as the neutralizing antibody titer against the field strain/neutralizing antibody titer against the vaccine virus. r1 values ≥ 0.3 were interpreted as cross-protective and r1 values < 0.3 as non-protective.   [28]. The pigs were intradermally challenged with 10 5 TCID 50 of FMDV A/SKR/Yeoncheon/2017 that had been passaged 2 times in pigs in the heel bulb of 1 foot at 7 DPV.

Blood Sampling and Analysis
Blood samples were collected once every 2 days, and oral swabs were collected daily from 0 to 8 DPC using the BD Universal Viral Transport Kit (BD Biosciences, Franklin Lakes, NJ, USA). FMD viral RNA was identified from viral RNA extracted from serum samples by reverse-transcription (RT) real-time PCR. The cador Pathogen 96 QIAcube HT Kit (Qiagen, Hilden, Germany) was used to extract viral RNA, and RT real-time PCR was conducted as previously reported [29]. Clinical signs were monitored daily after the challenge and were scored using the following criteria: (a) lameness (1 point); (b) vesicles in the hoof and foot (1 or 2 points for each affected hoof and foot, except the foot intradermally challenged); and (c) vesicles on the snout, lips, or tongue (1 point for each affected area) (maximum, 10 points) [30].

Statistical Analysis
Unpaired t-tests were conducted using GraphPad Prism Software (version 5.0, Graph-Pad Software, La Jolla, CA, USA). A p-value < 0.05 was considered significant.

Virus Adaptation to BHK-21 Suspension Cells over Serial Passages
After five serial passages in LFBK cells, four in BHK-21 adherent cells, and seven in BHK-21 suspension cells, the viral titer of A/SKR/Yeoncheon/2017 in BHK-21 adherent cells gradually increased and reached 10 7 TCID 50 /mL ( Figure 1). A/SKR/Yeoncheon/2017 had a titer of 10 2.5 TCID 50 /mL in CHO-K1 cells after being serially passaged five times in LFBK cells, four times in BHK-21 adherent cells, and six times in BHK-21 suspension cells (L5B4S6). These results showed that the virus became cell-adapted and employed HS as a receptor.

Virus Adaptation to BHK-21 Suspension Cells over Serial Passages
After five serial passages in LFBK cells, four in BHK-21 adherent cells, and seven in BHK-21 suspension cells, the viral titer of A/SKR/Yeoncheon/2017 in BHK-21 adherent cells gradually increased and reached 10 7 TCID50/mL (Figure 1). A/SKR/Yeoncheon/2017 had a titer of 10 2.5 TCID50/mL in CHO-K1 cells after being serially passaged five times in LFBK cells, four times in BHK-21 adherent cells, and six times in BHK-21 suspension cells (L5B4S6). These results showed that the virus became cell-adapted and employed HS as a receptor.

Comparison of Receptor Usage and Amino Acid Substitutions in the P1 Region in Plaque-Purified Viruses
Plaque-purified A/SKR/Yeoncheon/2017 viruses A-1, A-2, and A-3 were tested for viral titers in BHK-21 and CHO-K1 cells and for amino acid substitutions in the P1 region. The A-1, A-2, and A-3 viruses used HS as a receptor. Their titers were higher than 10 7 TCID50/mL, and the titer of A-1 was the highest in BHK-21 adherent cells (Table 1). A-1, A-2, and A-3 had different amino acids substitutions in the P1 region. A-1 had a single glutamic acid-to-lysine substitution in the VP2 region, VP2 E82K. A-2 additionally had two amino acid substitutions, including a glutamic acid-to-glycine transition in the VP2 region (VP2 E131G) and VP1 K42T. A-3 had two amino acid substitutions, including a glutamic acid-to-lysine transition in the VP2 region (VP2 E131K).

Comparison of Receptor Usage and Amino Acid Substitutions in the P1 Region in Plaque-Purified Viruses
Plaque-purified A/SKR/Yeoncheon/2017 viruses A-1, A-2, and A-3 were tested for viral titers in BHK-21 and CHO-K1 cells and for amino acid substitutions in the P1 region. The A-1, A-2, and A-3 viruses used HS as a receptor. Their titers were higher than 10 7 TCID 50 /mL, and the titer of A-1 was the highest in BHK-21 adherent cells (Table 1). A-1, A-2, and A-3 had different amino acids substitutions in the P1 region. A-1 had a single glutamic acid-to-lysine substitution in the VP2 region, VP2 E82K. A-2 additionally had two amino acid substitutions, including a glutamic acid-to-glycine transition in the VP2 region (VP2 E131G) and VP1 K42T. A-3 had two amino acid substitutions, including a glutamic acid-to-lysine transition in the VP2 region (VP2 E131K).

Neutralizing Antibody Response after Immunization with A/SKR/Yeoncheon/2017 Experimental Vaccine for Homologous and Heterologous Viruses in Pigs
To analyze the neutralizing antibody response after vaccinating pigs with the A/SKR/ Yeoncheon/2017 (A-1) experimental vaccine formulated with the ISA 206 adjuvant VNT was performed for the A/SKR/Yeoncheon/2017 field virus using pig sera collected from 0 to 6 WPV (Figure 3). A VN antibody titer of ≥ 1:45 was observed in all pig sera (3/3) at Vaccines 2021, 9, 308 8 of 14 2 WPV, and a VN antibody titer of ≥ 1:100 was observed in all pig sera (3/3) at 3 WPV for homologous virus. The VN titer increased until 3 WPV and remarkably increased after boosting (VN antibody titer ≥ 1:1000; p < 0.05, t-test). In addition, VNT for heterologous viruses included in the A/ASIA topotype was performed on the serum samples collected at 1, 3, and 5 WPV (Figure 4). The VN antibody titer for heterologous field viruses was not significantly different form the VN antibody titer for homologous field virus (p > 0.05, t-test) at 1 and 5 WPV. However, The VN antibody titers for the heterologous field viruses were different from that for homologous field virus (p < 0.05, t-test) at 3 WPV. VN antibody titers for A/VIT/2013 and A/SKR/Pocheon/2010 were > 1:1000 and that for A22/IRQ/24/64 was > 1:500 in all pig sera after boosting (5 WPV).

Neutralizing Antibody Response after Immunization with A/SKR/Yeoncheon/2017 Experimental Vaccine for Homologous and Heterologous Viruses in Pigs
To analyze the neutralizing antibody response after vaccinating pigs with the A/SKR/Yeoncheon/2017 (A-1) experimental vaccine formulated with the ISA 206 adjuvant VNT was performed for the A/SKR/Yeoncheon/2017 field virus using pig sera collected from 0 to 6 WPV (Figure 3). A VN antibody titer of ≥ 1:45 was observed in all pig sera (3/3) at 2 WPV, and a VN antibody titer of ≥ 1:100 was observed in all pig sera (3/3) at 3 WPV for homologous virus. The VN titer increased until 3 WPV and remarkably increased after boosting (VN antibody titer ≥ 1:1000; p < 0.05, t-test). In addition, VNT for heterologous viruses included in the A/ASIA topotype was performed on the serum samples collected at 1, 3, and 5 WPV (Figure 4). The VN antibody titer for heterologous field viruses was not significantly different form the VN antibody titer for homologous field virus (p > 0.05, ttest) at 1 and 5 WPV. However, The VN antibody titers for the heterologous field viruses were different from that for homologous field virus (p < 0.05, t-test) at 3 WPV. VN antibody titers for A/VIT/2013 and A/SKR/Pocheon/2010 were > 1:1000 and that for A22/IRQ/24/64 was > 1:500 in all pig sera after boosting (5 WPV).

Vaccine Matching Using Vaccinated Cow Sera
Cross-reactivity with heterologous field viruses included in the ASIA topotype was assessed via 2D VNT using sera from five cows inoculated with the A/SKR/Yeonch

Heterologous Virus Challenge in Vaccinated Mice
Mice vaccinated with the A/SKR/Yeoncheon/2017 (A-1) experimental vaccine formulated with the ISA 201 adjuvant were challenged with A/MAY/97 virus, which was not antigenically matched with A/SKR/Yeoncheon/2017 (A-1) ( Figure 6). In mice vaccinated with vaccine including 1 μg of antigen per head, which is the vaccine dose equivalent to 1/15 of a pig dose in the FMDV A/MAY/97 challenge at 10 DPV, the survival rate was 100% ( Figure 6A). The survival rate after challenge with FMDV at 21 DPV was 100% for mice injected with vaccine containing > 0.1 μg antigen per mouse ( Figure 6B).

Heterologous Virus Challenge in Vaccinated Mice
Mice vaccinated with the A/SKR/Yeoncheon/2017 (A-1) experimental vaccine formulated with the ISA 201 adjuvant were challenged with A/MAY/97 virus, which was not antigenically matched with A/SKR/Yeoncheon/2017 (A-1) ( Figure 6). In mice vaccinated with vaccine including 1 µg of antigen per head, which is the vaccine dose equivalent to 1/15 of a pig dose in the FMDV A/MAY/97 challenge at 10 DPV, the survival rate was 100% ( Figure 6A). The survival rate after challenge with FMDV at 21 DPV was 100% for mice injected with vaccine containing > 0.1 µg antigen per mouse ( Figure 6B). lated with the ISA 201 adjuvant were challenged with A/MAY/97 virus, which was not antigenically matched with A/SKR/Yeoncheon/2017 (A-1) (Figure 6). In mice vaccinated with vaccine including 1 μg of antigen per head, which is the vaccine dose equivalent to 1/15 of a pig dose in the FMDV A/MAY/97 challenge at 10 DPV, the survival rate was 100% ( Figure 6A). The survival rate after challenge with FMDV at 21 DPV was 100% for mice injected with vaccine containing > 0.1 μg antigen per mouse ( Figure 6B).  Table 2). Viral RNA was not detected in serum samples of the two pigs (no. 3 and no. 4) that were clinically protected, and a significantly low viral RNA level (about 10 copies) was detected in the serum of pig no. 2. The VN antibody titers in pigs no. 2, no. 3, and no. 4 were > 1:180 at 7 DPV. In pig no. 1, the serum viral RNA level was lower than that in the PBS control group. However, the viral RNA level in pig no. 1 was 100 times that in pig no. 2, and the VN antibody titer was only 1:16. The three pigs in the PBS control group showed severe clinical signs and high serum viral RNA levels.

Discussion
The most frequent lineage among serotype A FMDV in the pool 1 region is A/ASIA/Sea-97, accounting for about 25% of the total number of animals with FMD infected with all serotypes between 2019 and 2020 [31]. A/SKR/Yeoncheon/2017 belongs to the A/ASIA/Sea-97 lineage and is similar to strains occurring in various Asian countries, such as Vietnam, Myanmar, Thailand, and Russia [9]. However, several strains of the A/ASIA/Sea-97 lineage occurring in Southeast Asia do not match the internationally applied FMD vaccine strains such as A22 Iraq, A MAY 97, A TUR 06, and A IRN 05 in 2D-VNT by the FMD World Reference Laboratory [32]. Therefore, A/SKR/Yeoncheon/2017 might be useful as a vaccine strain for use in Asian countries.
Amino acid substitutions in the capsid of FMDV serotype A have been observed in various regions, such as the HS-binding pocket, fivefold symmetry axis, and G-H loop, which can harbor antigenically significant residues [11,13,33,34]. The most frequently reported amino acid mutation in serotype A FMDV is VP2 E131K [10]. Residues 130 and 131 are part of the EF loop and are important antigenic sites [14]. Furthermore, amino acid changes in the VP2 region are frequently reported to occur in combination with mutations in VP1 or VP3 [35][36][37]. In contrast, an amino acid substitution at position 82 of VP2 has been reported in only a few studies on serotype A FMDV [13,34].
We showed that a A/SKR/Yeoncheon/2017 (A-1) vaccine strain harboring only the VP2 E82K substitution in the P1 region uses HS as a receptor in BHK-21 suspension cells and that its 146S antigen production was suitable for large-scale production as the purified 146S antigen was > 2 µg/mL of culture supernatant. Therefore, we suggest that the amino acid substitution at position 82 of VP2 might be critical for adaptation in BHK-21 suspension cells, although it has not been frequently reported. The change of the negatively charged E to the positively charged K amino acid might be advantageous for binding to the HS receptor, as an increased number of surface positive charges has been correlated with increased interaction with the negatively charged HS receptor [14,38]. Residue 82 of VP2 is located in the surface-exposed VP2 βC strand and has been detected in serotype A virus serially passaged in BHK-21 cells [13]. In that study, VP2 E82K was accompanied by VP1 Q157R in cell-adapted serotype A FMDV. In cell-adapted A22 IRQ virus, it has been speculated that E82G on the surface of VP2 might change the structure of the VP1 and VP3 GH loops [34]. Therefore, a substitution at VP2 82 might affect receptor attachment and antigenicity. However, we only observed a substitution in VP2 E82K, but not any in VP1 or VP3, in the A/SKR/Yeoncheon/2017 (A-1) vaccine strain. Therefore, we suggest that the antigenic characteristics of this vaccine strain might be similar to those of wild-type virus because the critical antigenic sites, including the VP1 GH loop, are not altered. We also observed an amino acid substitution, A93T, in a non-structural region (2C) of the A/SKR/Yeoncheon/2017 (A-1) vaccine strain (data not shown). Amino acid substitutions in the 2C region have been reported in cell-adapted picornavirus [39][40][41]. The possibility that the substitution in the 2C region plays a role in optimization of the replication processes in different environments cannot be excluded, as amino acid substitutions in the 2C region have been observed to be critical for FMDV to adapt to a new host [42].
Cell adaptation of FMDV is achieved through selective pressure on the viral quasispecies and increased cell tropism due to genetic variation [43,44]. Therefore, we performed virus plaque purification after cell adaptation through serial passages, and we obtained three purified viruses with different characteristics. The three cell-adapted viruses (A-1, A-2, A-3) had different plaque phenotypes and 146S production levels (data not shown), as well as growth efficiencies. A-1 yielded the highest viral titer and 146S production. Furthermore, it had the lowest number of amino acid substitutions in the P1 region. A-2 and A-3 had a common amino acid substitution, VP2 E131G/K, and substitution at this position has been frequently reported [10,12,34,45]. We found no amino acid mutations except VP2 E82K in the P1 region after two additional passages of the A-1 vaccine seed virus in BHK-21 suspension cells using CD-BHK-21 medium (data not shown). However, the possibility that additional alterations occur when the virus is cultured in different cells or media cannot be excluded.
A highly immunogenic (high-potency) vaccine is important for protection against FMDV. FMD vaccine-induced immune system stimulation is influenced by the amount of antigen, antigenicity of the strain, adjuvant, and vaccine formulation [3]. We tested experimental vaccines, not a commercial vaccine, and various adjuvants, and the same A/SKR/Yeoncheon/2017(A-1) antigen was used in the experimental vaccine because the purpose of this study was to develop a novel vaccine strain with high immunogenicity (the antigenicity of the vaccine strain). Although we used various adjuvants, we observed high VN antibody titers and protection rates in all animals. Therefore, we will report the immunogenicity and potency of the final version of the A/SKR/Yeoncheon/2017 vaccine formulation in a further study.
For serotype A FMDV, the VN antibody titer at which animals are protected with 95% probability has been estimated to be approximately log 10 2.1 in cows [46]. We observed a maximum VN antibody titer of 1:1024 (log 10 3.3) at 21 DPV (1 µg antigen/dose) in mice (data not shown) and of 1:256 (log 10 2.4) at 21 DPV (15 µg antigen/dose) in pigs. In the early protection experiment in pigs, we observed no or low-level (score 2) clinical signs in pigs vaccinated and challenged with FMDV at 7 DPV. In addition, the VN antibody titer was > 1:100 in three of four pigs. Therefore, we suggest that vaccination with the A/SKR/Yeoncheon/2017 vaccine formulated with 15 µg antigen, ISA 206, 10% aluminum hydroxide gel, and saponin [28] might effectively provide early protection at 7 DPV, as well as protection after a second vaccination, in pigs. Because early protection is related not only to the vaccine strain, but also to the adjuvant used, the early protection rate might be enhanced through optimization of the vaccine formulation.
Cross protection has been recently reported in cattle vaccinated with an emergency vaccine against serotype A FMDV [47]. Cattle vaccinated with an A/MAY/97 vaccine poorly matching with an A/ASIA/Sea-97 lineage virus were protected against the A/ASIA/Sea-97 lineage virus. In line with this, we observed that mice vaccinated with the A/SKR/ Yeoncheon/2017 experimental vaccine were protected against an A/MAY/97 virus not matching the vaccine strain.

Conclusions
We used virus plaque purification to select a virus with high antigen productivity and the fewest amino acid mutations in the P1 region among cell-adapted virus populations. We thus selected a new vaccine strain, A/SKR/Yeoncheon/2017 (A-1), which belongs to the A/ASIA/Sea-97 lineage that frequently occurs in the pool 1 region. We showed that this vaccine strain harbored only the VP2 E82K substitution in the P1 region and provided high immunogenicity in pigs. The novel A/SKR/Yeoncheon/2017 (A-1) strain is a promising vaccine candidate for protection against the emerging FMDV serotype A in Asia.