Virology 422 (2012) 37–45

Contents lists available at SciVerse ScienceDirect

Virology

j ourna l homepage: www.e lsev ie r .com/ locate /yv i ro

Role of arginine-56 within the structural protein VP3 of foot-and-mouth disease virus(FMDV) O1 Campos in virus virulence

Manuel V. Borca a, Juan M. Pacheco a, Lauren G. Holinka a, Consuelo Carrillo b, Ethan Hartwig a,Damià Garriga c, Edward Kramer a, Luis Rodriguez a, Maria E. Piccone a,d,⁎a Agricultural Research Service, U.S. Department of Agriculture, Plum Island Animal Disease Center, Greenport, New York, USAb Animal and Plant Health Inspection Service, U.S. Department of Agriculture, Plum Island Animal Disease Center, Greenport, New York, USAc Department of Molecular and Cell Biology, Centro Nacional de Biotecnología, CSIC, Darwin 3, 28049-Madrid, Spaind Department of Pathobiology and Veterinary Sciences, Univ. of Connecticut, Storrs, CT, USA

⁎ Corresponding author at: Plum Island Animal DiseaBox 848, Greenport, NY 11944-0848, USA. Fax: +1 631

E-mail address: mariapiccone@hotmail.com (M.E. Pi

0042-6822/$ – see front matter © 2011 Published by Eldoi:10.1016/j.virol.2011.09.031

a b s t r a c t

a r t i c l e i n f o

Article history:Received 22 June 2011Returned to author for revision 12 July 2011Accepted 13 September 2011Available online 27 October 2011

Keywords:Foot-and-mouth disease virusVirus attenuationStructural protein VP3

FMDV O1 subtype undergoes antigenic variation under diverse growth conditions. Of particular interest is theamino acid variation observed at position 56 within the structural protein VP3. Selective pressures influencewhether histidine (H) or arginine (R) is present at this position, ultimately influencing in vitro plaquemorphology and in vivo pathogenesis in cattle. Using reverse genetics techniques, we have constructed FMDVtype O1 Campos variants differing only at VP3 position 56, possessing either an H or R (O1Ca-VP3-56H andO1Ca-VP3-56R, respectively), and characterized their in vitro phenotype and virulence in the natural host. Bothviruses showed similar growth kinetics in vitro. Conversely, they had distinct temperature-sensitivity (ts) anddisplayed significantly different pathogenic profiles in cattle and swine. O1Ca-VP3-56H was thermo stable andinduced typical clinical signs of FMD, whereas O1Ca-VP3-56R presented a ts phenotype and was nonpathogenicunless VP3 position 56 reverted in vivo to either H or cysteine (C).

se Center, USDA/ARS/NAA, P.O.323 3006.ccone).

sevier Inc.

© 2011 Published by Elsevier Inc.

Introduction

Foot-and-mouth disease (FMD) is an infectious viral disease thataffects cloven-hoofed animals, including cattle, sheep, pigs, goats,camelids and deer. Its wide host range and rapid spread make FMD aninternational animal health concern, since all countries are vulnerableto accidental or intentional trans-boundary introduction (Grubman andBaxt, 2004; Rowlands, 2003). The disease is caused by foot-and-mouthdisease virus (FMDV), an Aphthoviruswithin the viral family Picornaviri-dae that exists as seven immunologically distinct serotypes: O, A, C, Asia1, and South African Territories (SAT) type 1, SAT2, and SAT3. The viralgenome consists of a single-stranded, positive sense RNA of about8200 nucleotides (nt). The open reading frame (ORF) encodes along polyprotein that is processed by virus-encoded proteases intofour structural proteins (VP1 through 4) and eight non-structuralproteins (L, 2A, 2B, 2C, 3A, 3B, 3C and 3D) (Belsham, 1993; Ryan et al.,1989). Due to high error rates inherent in RNA replication, viralRNA genomes display a high level of nucleotide variation (Drakeand Holland, 1999; Figlerowicz et al., 2003) that can affect replicationspeed (Jarousse et al., 1994; Nunez et al., 2001; Pelletier et al., 1998;Zhang et al., 1995), virulence (Conenello et al., 2007; Zurbriggen et al.,1989), antigenicity or biochemical properties (Li et al., 2010; Martin-

Acebes et al., 2010;Mateu et al., 1990; Szepanski et al., 1992). The avail-ability of infectious cDNA picornavirus clones provides an opportunityto manipulate specific genomic mutations and examine their effectson viral infection.

The genetic and antigenic variation has been extensively studiedfor FMDV both in field virus samples and laboratory cell-passedisolates (Carrillo et al., 2007; Cottam et al., 2009; Haydon et al.,2001; Meyer et al., 1994; Piatti et al., 1995; Sevilla et al., 1996).For type O viruses selective pressures influence whether a histidine(H) or arginine (R) residue is present at position 56 of VP3. It hasbeen reported that an R amino acid (aa) in position 56 of VP3 (R56) isacquired after serial passage in BHK-21 cells (Kitson et al., 1990;Marquardt et al., 1991) and that this mutation correlates with asmall plaque phenotype, a marker indicative of virus attenuation inmice (Jensen andMoore, 1993). Indeed, Sa-Carvalho et al. (Sa-Carvalhoet al., 1997) constructed a type A chimera virus expressing the O1Campos structural proteins carrying, among others aa substitutions,R56 in VP3 and demonstrated that this virus exhibited a small plaquephenotype in cells andwas highly attenuated in bovines. These authorssuggested that adaptation of typeO1 viruses to cell culture results in theability of the virus to utilize heparin sulfate (Jackson et al., 1996) as areceptor with a concomitant loss in virulence. In view of this data,Rodriguez-Pulido et al. (Rodriguez Pulido et al., 2009) introducedan R56H substitution in VP3 of an O1 Kaufbueren infectious cDNAclone (pO1K) to induce reversion of an attenuated phenotype. However,RNA transcripts derived from pO1K containing H56 did not exhibit

38 M.V. Borca et al. / Virology 422 (2012) 37–45

differences in virulence in suckling mice compared to RNA transcriptscarrying R56, indicating that the substitution itself was not sufficient torevert to the attenuated phenotype (Baranowski et al., 2003). It shouldbe noted that no complete genomic sequence was determined forthe viruses constructed in these previous works, so it is possiblethat additional residue substitutions may have affected these results.

Among the FMDV subtype O1 genomic sequences reported toGenBank, only the amino acids R or H are found in position 56 of VP3;however, the passage history of these viruses is unclear. Furthermore,there are no pathogenicity studies using viruses containing thesesequences. The O/JPN/2000 strain isolated in the 2000 Japan out-break exhibited a mild pathogenicity in bovines, and two differentviral plaque phenotypes of this strain appeared after two passages inprimary bovine cells. Genomic sequencing of the structural proteinsrevealed two aa substitutions (VP2 D133N and VP3 H56R) in themildlypathogenic small plaque viruswhen compared to the large plaque virusin sucklingmice (Kanno et al., 2002;Morioka et al., 2008). Nevertheless,these two mutations did not account for the characteristics of anotherPanAsia strain isolated from China (Bai et al., 2010). This study demon-strated that the BHK-21 cell-adapted O/Fujian/CHA/5/99 strain hadhigh virulence in suckling mice and formed larger plaques than thebovine-isolated O/Tibet/CHA/1/99, while sequence analysis revealedthat both strains carried a D residue in position 133 of VP2 and an Hin position 56 of VP3.

To clarify the role of these amino acids in FMDV virulence, wemutated the aa at position 56 of VP3 in a full-length infectiouscDNA clone derived from the FMDV O1 Campos/Bra/58 strain and

Fig. 1. Construction of the full-length cDNA clone of FMDV O1Campos/Bra/58 strain and in vdiagram of the FMDV genome (8168 nucleotides, (Pereda et al., 2002)) showing the polyC tabelow the genome represent the RT-PCR fragments used for the generation of the full-lengthA T7 RNA polymerase promoter was added at the 5′ end of the viral cDNA as well as a uniqueis also indicated. The mutations at position 2713 CAT→CGC resulting in the substitution of aof pO1Ca digested with selected enzymeswere translated in a rabbit reticulocyte system and thprime.Molecular weightmarkers and cleavage products are indicated on the right and left,previously described (Piccone et al., 2010) was used as marker (lane 1).

evaluated the biological properties and virulence of these virusesin swine and cattle. Our results demonstrate that identical virusesdiffering only in aa residue 56 of VP3, H56 (O1Ca-VP3-56H) or R56(O1Ca-VP3-56R), had similar growth kinetics in BHK-21 and LF-BKcells and slight differences in plaque size in BHK-21, LF-BK, IBRS-2and LK cells. In contrast, the viruses displayed very different patho-genicities in two natural host species. Virus O1Ca-VP3-56H inducedtypical clinical signs of FMD in swine and cattle, whereas virus O1Ca-VP3-56R was innocuous unless it reverted to either H or cysteine(C) at VP3-56 during the animal infection. These results clearlydemonstrate that the presence of R56 in VP3 is associated with virusattenuation during infections in natural host animals.

Results

Generation and characterization of pO1Ca

The full-length cDNA clone pO1Ca was constructed by overlappingPCR fragments flanked by unique restriction enzymes as described inMaterials and methods and shown schematically in Fig. 1A. The full-length cDNA was preceded by the T7 polymerase promoter and twoadditional G residues to improve transcription efficiency (Riederet al., 1993) and followed by a polyA tail of 15 residues and a uniqueEcoRV site. Plasmid containing the substitution H56R in VP3 derivedfrom the infectious clone pO1Ca was constructed as described inMaterials and methods.

itro translation of derivative RNA transcripts. (A) At the top of the figure is a schematicil (Cn) and the polyA (An) tail at the end of the 3′ untranslated region (UTR). The boxescDNA clone (named pO1Ca) including the restriction sites used in the cloning strategy.EcoRV site at the 3′ end to allow the synthesis of RNA transcripts. Restriction site EcoRIa H by R at position 56 of VP3 are indicated. (B) RNAs derived from in vitro transcriptione productswere then analyzed by PAGE on a 15% gel. Truncated proteins are indicated by arespectively. O1 Campos infected BHK-21 cell extract labeled with (35S) methionine as

39M.V. Borca et al. / Virology 422 (2012) 37–45

To check the ability of pO1Ca to function as a template for proteinsynthesis, RNA transcripts derived from EcoRV-linearized plasmidwere translated in vitro using reticulocyte lysates as described byPiccone et al. (1995). To facilitate the identification of the viral pro-teins, pO1Ca was digested with selected restriction enzymes thatremove the downstream portion of the viral genome from the plasmidfacilitating the interpretation of the viral protein pattern using SDS-PAGE. Fig. 1B shows that the polyprotein derived from transcripts ofpO1Ca was processed into the virus precursors and mature proteins,indicating its ability to undergo viral proteolytic processing.

In vitro characterization of the infectious clone-derived viruses

RNA transcript derived from pO1Ca-VP3-56H or pO1Ca-VP3-56Rwas used to transfect BHK-21 cells by electroporation, and thensupernatant from transfected cells was subjected to serial passagein LF-BK αVβ6 cells to amplify the clone-derived viruses. Sequencingof the entire genome of recovered viruses (O1Ca-VP3-56H and O1Ca-VP3-56R) verified that they were identical to the parental DNAs,confirming that only the predicted mutations were incorporatedinto the viruses. Sequencing results also revealed that nt changesin the third position of codons 55 and 57 (flanking position 56) ofVP3 introduced during the construction of pO1Ca-VP3-56R wereretained (data not shown), excluding the possibility of contaminationby the parental virus.

The in vitro growth characteristics of viruses O1Ca-VP3-56H andO1Ca-VP3-56R were comparatively evaluated in multi-step growthcurves in BHK-21 and LF-BK cells. Cells were infected at a low (0.01)or high (10) MOI and samples were collected at times post-infectionthrough 24 h. The results shown in Fig. 2A demonstrate that bothviruses exhibited similar growth characteristics and reached comparabletiters in these cell lines. Additionally, the plaque size and morphology ofthese viruses was analyzed in a standard plaque assay in cells fromdifferent origins (Fig. 2B). Both viruses produced a similar plaquesize in BHK-21 cells, while virus O1Ca-VP3-56R produced smallerplaques compared to O1Ca-VP3-56H in IBRS-2 and LF-BK cells, butclearly larger plaques compared to O1Ca-VP3-56H in LK cells.

Fig. 2. In vitro characterization of the FMDV O1Ca-VP3-56H and O1Ca-VP3-56R recovered virMOI 0.01 in BHK-21 and LF-BK (upper panel), or MOI 10 at 37 °C and 41 °C in LF-BK (bottomtemperature. Virus yield was determined by TCID50/ml at the time points indicated. The fi

O1Ca-VP3-56H and O1Ca-VP3-56R viruses in BHK-21, LF-BK, LK and IBRS-2 cells. Infected cat 48 h post infection. Plaques from appropriate dilutions are shown.

Virulence assessment of FMDV O1Ca-VP3-56H and O1Ca-VP3-56Rin cattle

Virulence of both viruses in cattle was assessed utilizing a wellestablished intradermolingual inoculation method (Henderson, 1949).Two cows were inoculated with 104 PFU/animal of either O1Ca-VP3-56H or O1Ca-VP3-56R, and infection and clinical signs were monitoreddaily as described by Pacheco andMason (2010). Results demonstratedthat the two animals inoculated with the O1Ca-VP3-56H strain (#42and #43) developed clinical FMD and reached the maximum clinicalscore (4 feet affected with vesicles) by 2 or 3 days post-inoculation(dpi). Both animals displayed fever (N40 °C) from 2 to 4 or to 6 dpi.Viremia lasted 4 to 5 days, starting at 1 dpi. Virus shedding wasdetected in nasal swabs from both animals beginning at 1 dpi tothe end of the collection period (7 dpi) (Fig. 3). Sequence analysisof the P1 polyprotein confirmed that viruses collected from secondaryreplication sites exhibited no differences from the inoculated virus(data not shown). Interestingly, cows inoculated with O1Ca-VP3-56R(#44 and #45) showed no clinical signs, no fever, no viremia and novirus shedding (Fig. 3). Additionally, no serum neutralizing antibodieswere detected 2 months after inoculation, no virus was isolated fromprobangs and no transmission occurred to two cows that were in directcontact since 1 dpi (results not shown). These results suggest thatO1Ca-VP3-56R does not undergo any detectable replicationwhen inoc-ulated into animals.

Virulence assessment of FMDV O1Ca-VP3-56H and O1Ca-VP3-56Rin swine

To examine the effect of H and R amino acids at position 56 of VP3on virus virulence in pigs, six animals were intradermally inoculatedwith approximately 104 PFU/animal of either O1Ca-VP3-56H orO1Ca-VP3-56R (Figs. 4 and 5, respectively) and monitored for clinicaldisease. All animals infected with O1Ca-VP3-56H (#01 through #06)developed classic FMD disease. Animals became febrile, depressed,lame and developed lesions in the non-inoculated feet and mouth.By the second week post-inoculation, the disease had subsided and

uses. (A) In vitro growth curves at 37 °C of O1Ca-VP3-56H and O1Ca-VP3-56R viruses atpanel). Cells were inoculated with the indicated MOI, and incubated at the described

gure is representative of two independent experiments. (B) Plaques corresponding toells were incubated at 37 °C under a tragacanth overlay and stained with crystal violet

Fig. 3. Assessment of O1Ca-VP3-56H and O1Ca-VP3-56R virus virulence in cattle. Steers were intradermolingual (IDL) inoculated with 104 PFU of either O1Ca-VP3-56H (animal #42and #43) or O1Ca-VP3-56R SK6 (animal #44 and #45). Presence of clinical signs (clinical scores), rise in body temperature over 40 °C (+), and virus yield (quantified as log10 ofviral RNA copies/ml of sample) in blood and nasal swabs were daily determined during a seven day observational period.

40 M.V. Borca et al. / Virology 422 (2012) 37–45

animals were returning to their normal physiological state. Presenceof viremia, detected by real-time RT-PCR, demonstrated the presenceof virus in blood starting on the first day of disease peaking by day 2to 3. Virus was also detected by rRT-PCR in nasal swabs of all 6animals starting as early as 2 dpi and peaking at 2 to 4 dpi. Conversely,the six animals inoculated with O1Ca-VP3-56R (#07 through #12)presented a heterogeneous phenotype. Two of the animals (#11and #12) did not display any FMD-related symptoms, presenting asclinically normal throughout the entire experimental period. According-ly, no virus could be detected in the bloodor nasal swabs of these animalsat any test time points. Conversely, the other four animals inoculatedwith O1Ca-VP3-56R presented a disease that was nearly indistinguish-able in intensity compared to that observed in animals infected withO1Ca-VP3-56H, differing only by a slight delay in the course of the dis-ease (approximately by 1–3 days). Consequently, in these four animalsviremia kinetics was also delayed by a similar lag of time in relation tothe appearance of clinical symptoms (Fig. 5).

Partial RNA sequencing of the viruses isolated from animals wasperformed to evaluate the genetic stability of the inoculated virusduring the infection in vivo. All virus isolates from local lesions inanimals inoculated with O1Ca-VP3-56H presented an H at position56 of VP3. As mentioned earlier, no virus was isolated from animalsinfected with O1Ca-VP3-56R that did not develop disease. Interestingly,all viruses isolated from the four symptomatic animals infected withO1Ca-VP3-56R possessed a mutation at position 56 of VP3; in three ofthem R was substituted by H and in the fourth case by C (data notshown). The above results indicate that animals inoculated withO1Ca-VP3-56R remain asymptomatic unless R at position 56 mutatesto either an H or a C.

Effect of R at position 56 of VP3 on thermal stability

Since temperature sensitivity has been linked with attenuationphenotype for FMDV (Butrapet et al., 2000; Omata et al., 1986), virusesO1Ca-VP3-56H and O1Ca-VP3-56R were analyzed for their ability toproduce plaques at 41 °C, a non-permissive temperature. Usingten-fold dilutions of the virus stocks, plaque assays were performedat low MOI and, after adsorption at 37 °C, cells were incubated at either37 °C or 41 °C for 48 h. While virus O1Ca-VP3-56H displayed a slightreduction in the number of plaques at 41 °C regarding those detectedat 37 °C, virus O1Ca-VP3-56R failed to produce plaques in LF-BK cellsat all dilutions tested (between 10−2 and 10−6) (data not shown).To further characterize the effect of temperature on the viruses'growth, LF-BK cells were infected at a high MOI (MOI=10) andincubated at 41 °C. Both viruses achieved lower titers at all testedtime points at 41 °C compared to 37 °C (Fig. 2A). In addition, virusO1Ca-VP3-56R grew slower than virus O1Ca-VP3-56H but reachedsimilar final yield, indicating that the substitution H56R had no signif-icant effect on virus production.

To further test the temperature sensitivity of these viruses, sampleswere heated up to 24 h at either 37 °C or 41 °C and aliquots were takenat different time points during the heat treatment and titrated on LF-BKcells. Infectivity titers of FMDV O1Ca-VP3-56R virus were stronglyreduced at both temperatures when compared with O1Ca-VP3-56H. An approximately 4 logs reduction titer was observed afterheating for 8 h at 37 °C or after 4 h at 41 °C (Fig. 6). Moreover, to deter-mine if H56 confers a better fitness to the viral population at a highertemperature, a serial passage experiment was performed in LF-BK cellsat 41 °C with these viruses. Sequencing analysis revealed that after

Fig. 4. Assessment of O1Ca-VP3-56H virus virulence in swine. Six pigs (animal #01 through #06) were intradermically inoculated into the foot bulb with 104 PFU of O1Ca-VP3-56H.Presence of clinical signs (clinical scores), rise in body temperature over 40 °C (+), and virus yield (quantified as log10 of viral RNA copies/ml of sample) in blood and nasal swabswere daily determined during a seven day observational period.

41M.V. Borca et al. / Virology 422 (2012) 37–45

three passages virus O1Ca-VP3-56R acquired an H codon at position 56,whereas virus O1Ca-VP3-56H conserved the original sequence (Table 1).The presented results indicate that the H56R substitution in VP3 drasti-cally reduces the thermal stability of the virus.

Discussion

Previous works have suggested that the H56R substitution isacquired during the process of virus adaptation to cell cultures,conferring an attenuated phenotype in mice and bovines with anincreased affinity for heparin (Baranowski et al., 2003; Jackson etal., 1996; Jensen and Moore, 1993; Rodriguez Pulido et al., 2009;Sa-Carvalho et al., 1997). In fact, R56 of VP3 is the critical viral residueused for recognition of the cellular heparin sulfate (HS) co-receptorby different strains of FMDV, as seen in the structure of the serotypeO1BFS in complex with various glycosaminoglycans (Fry et al., 1999).Nevertheless, none of these studies have assayed the infectivity of typeO transcripts or viruses exclusively differing in the aa residue atposition 56 of VP3 (either H or R) in the natural host. The purposeof this study was to experimentally demonstrate the role of each ofthese amino acids in this particular position on viral virulence inthe natural host. To this end, recombinant viruses were created bymodifying a full-length infectious cDNA clone of the O1 Campos/Bra/58 strain to introduce either H or R at VP3-56. Viruses O1Ca-VP3-56H and O1Ca-VP3-56R recovered after electroporation of BHK-21 cells showed similar growth kinetics and small differences inplaque sizes in cell culture. In contrast, these viruses displayed amarked difference in thermal inactivation at 41 °C. In vitro heatingof the viruses caused a pronounced loss of infectivity for virusO1Ca-VP3-56R compared to O1Ca-VP3-56H, suggesting a role forthe aa at VP3-56 in the thermal stability of the virion. At this tem-perature, plaque assays also showed significant differences between

both viruses at higher dilutions, although at MOI 10 they replicateto similar titers. These results suggest that the substitution H56Rmay affect the capsid assembly or stability of the virion. Aminoacid 56 of VP3 is a surface partially exposed residue, which also par-ticipates in interactions with VP1 and VP2 residues that stabilize theFMDV protomer (Fig. 7A). Hence, having an H instead of an R at thisposition may affect such interactions (Fig. 7B, C). Further experimentsare necessary to validate this assumption. In any case, H56 likely in-creased the thermal stability of the virus, since after two passages ofO1Ca-VP3-56R in cell culture at 41 °C VP3-R56 changed to VP3-H56.On the other hand, the identification of aa that increase the thermalstability of the virion without disrupting its infectivity could supportthe rational engineering of FMD vaccines (Mateo et al., 2003, 2008).

Another major difference between O1Ca-VP3-56H and O1Ca-VP3-56R relates to their ability to cause disease in the natural host. Princi-pally, cattle inoculated with O1Ca-VP3-56H showed typical clinicalsigns of FMD, while cattle inoculated with O1Ca-VP3-56R did notdisplay any clinical signs of disease. Furthermore, no neutralizingantibodies or virus in blood or saliva was detected in animals inoculatedwith O1Ca-VP3-56R. Since both viruses differ only in the aa at position56 of VP3, this data clearly indicates that the presence of R56 is respon-sible for the attenuation of O1Ca-VP3-56R in cattle. Additionally, the vir-ulence of both viruses was tested in swine. All animals inoculated withO1Ca-VP3-56H exhibited clinical signs of FMD while those inoculatedwith O1Ca-VP3-56R remained clinically normal with the exception ofsome animals that developed delayed FMD. Interestingly, the virusrecovered from lesions in these animals had a R56C or R56H mutationand no other mutations in their protein coding region, confirming theinvolvement of R56 in viral attenuation. Similar VP3 R to H or C substi-tutions were previously reported by Sa-Carvalho et al. (1997) in A24/O1Campos chimeric viruses recovered from infected cattle, although addi-tional substitutions accompanied that at position 56 of VP3.

Fig. 5. Assessment of O1Ca-VP3-56R virus virulence in swine. Six pigs (animal #07 through #12) were intradermically inoculated into the foot bulb with 104 PFU of O1Ca-VP3-56R.Presence of clinical signs (clinical scores), rise in body temperature over 40 °C (+), and virus yield (quantified as log10 of viral RNA copies/ml of sample) in blood and nasal swabswere daily determined during a seven day observational period.

42 M.V. Borca et al. / Virology 422 (2012) 37–45

As seen in Fig. 7, residue 56 is located at the interface between VP2and VP3 proteins, in close contact with the C-terminus of VP1, nearthe icosahedral three-fold axis. Structural comparisons betweenO1BFS and O1K, having R and H in position 56 of VP3, respectively,demonstrated important changes at the capsid surface between thetwo strains (Lea et al., 1995). These changes may explain the differentproperties exhibited by viruses carrying R or H in VP3-56 such asreactivity with MAbs (Kitson et al., 1990; Sa-Carvalho et al., 1997)or thermal stability properties. Additional studies with subtypeO1 viruses have shown that R56 of VP3 increases the positive chargeof the viral surface at this site and strongly enhances binding to nega-tively charged HS (Fry et al., 1999; Jackson et al., 1996). Fry et al.(1999) hypothesized that the high affinity to HS presumably hampersthe spread of infection by sequestering viruses from the target

Fig. 6. Temperature sensitivity of O1Ca-VP3-56H and O1Ca-VP3-56R viruses. Virusstocks were heated up to 24 h at either 37 °C or 41 °C and aliquots were taken at differenttimes during the heat treatment and titrated on LF-BK cells. Infectivity titers areexpressed in TCID50/ml. Threshold of detection is log101.80 TCID50/ml.

cells/tissues. Consistent with this hypothesis, a virus carrying R56cannot develop virulent infections in vivo unless it undergoes areversion to H or C, reducing HS affinity.

Materials and methods

Cell lines and virus

Baby hamster kidney (BHK-21) cells, secondary lamb kidney (LK)cells, bovine kidney (LF-BK) cells (Swaney, 1988) and a derivativeLF-BK cell line expressing the bovine αVβ6 integrin (LF-BK αVβ6)(LaRocco et al., unpublished data) were maintained as previouslydescribed (Piccone et al., 2010). Growth kinetics and plaque assays

Table 1Sequence comparison around the position VP3-56 between O1Ca-VP3-56H and O1Ca-VP3-viruses amplified at different temperatures.

O1Ca-VP3-56H O1Ca-VP3-56R

Passage 0 ctA cAT ttT ctg cgc ttcL H F L R F

Passage 3 at 37 °C ctA cAT ttT ctg cgc ttcL H F L R F

Passage 3 at 41 °C ctA cAT ttT ctg cAc ttcL H F L H F

Virus stocks were passaged three times in LF-BK cells at MOI 10 and the temperatureindicated. Changes in nucleotides and amino acid sequence are compared withoriginal virus (Passage 0).

Fig. 7. (A) Ribbon diagram showing the structure of a FMDV pentamer with the position of the five-fold, three-fold and two-fold icoshedral axes explicitly indicated. The biologicalprotomer is shown with standard coloring for the proteins VP1 (blue), VP2 (green), VP3 (red) and VP4 (yellow). The rest of the pentamer is shown in gray. Residue 56 of VP3 isshown as a space filling surface in cyan. (B, C, D) Close up of the interactions involving the position 56 of VP3, when the residue is R (B), H (C) and C (D). Hydrogen bonds aredepicted as discontinuous lines and the residues involved are explicitly labeled. Coordinates are taken from O1BFS (PDB id 1FOD). H and C were modeled using the programcoot (Emsley and Cowtan, 2004) with the preferred side chain rotamer conformation.

43M.V. Borca et al. / Virology 422 (2012) 37–45

were performed also as described previously (Piccone et al., 1995).The FMDV strain O1Campos/Bra/58 was obtained from the Institute ofVirology at the National Agricultural Technology Institute, Argentina.

Construction of the FMDV O1 Ca full-length cDNA clone and derivative

Total RNA from FMDV-infected BHK-21 cells was isolated with theTRIzol® system (Life-Technologies), and reverse transcribed bystandard methods using random hexamers (PROMEGA) and Super-Script II reverse transcriptase (Life-Technologies). cDNA fragmentswere amplified with the Expand High Fidelity PCR system (RocheApplied Science, Indianapolis, IN) using primers on unique restrictionsites naturally found in the FMDV O1Campos/Bra/58 sequence. Allnucleotide numbering included in this study refers to the sequencedeposited in GenBank under accession number AJ320488.

A total of 4 fragments covering the complete viral genome were am-plified (Fig. 1A). The 3′ and 5′ ends of the viral genome as well as the in-troduction of the polyC tail were obtained by a RACE-PCR technique. Forthe 3′ end, a 40mer primer (5′GAATTCCATATGgatatcT15GATTAAG3′)complementary to the 3′ end of the viral genome, including 15adenine residues, followed by an adaptor sequence containingan EcoRV site (lower case) was used for priming the cDNA synthe-sis. Thereafter, PCR was performed using sense primer nt 5416-5′GGCAGCAATTgaattcTTT3′ on a unique EcoRI site (lower case) andantisense primer 5′GAATTCCATATGgatatc3′ complementary to theadaptor sequence.

For the 5′ end, PCR was performed using sense primer 5′GGAATTCcatagTAATACGACTCACTATAGGTTGAAAGGGGGCGCTAGGC3′,which contained the sequences of a NdeI site (lower case), T7 promoter(underlined), two additional G nucleotides (bold) and 19 nucleotides

of the 5′ end of the viral genome; and antisense primer nt 350-5′TTggatccTGAAAGGCGGGCGTCGGG3′ which incorporated a StyIsite (lower case). To add the polyC tail, the PCR reaction wascarried out using sense primer 5′ TtcctaggC17TAGGTTTTACCGTCATTCCCwhich contained a StyI site (lower case) and antisense primer nt 3019-5′CAGTGCgcggccgcCTCAGGTGTC3′ on a unique NotI site (lower case).

All purified PCR products were cloned into the pGEM-T easyvector (Promega) and sequenced using selected primers. The as-sembly of the full-length cDNA clone was done using a multi-stepstrategy that implemented the unique restriction sites NdeI, StyI,NotI, EcoRI and EcoRV on pGEM vectors. Each step of the constructionwas monitored by restriction enzyme analysis to ensure the genera-tion of fragments of predicted sizes. The final plasmid containing thegenome-length cDNA of FMDV O1Campos/Bra/58 was designated aspO1Ca and completely sequenced using an Applied Biosystems 3730xl automated DNA sequencer (Applied Biosystems, Foster City, CA).

Substitution of aa H56 in the structural protein VP3 was carriedout by site-directed mutagenesis in plasmid pO1Ca using theQuickChange system (Stratagene) and forward primer nt 2695-5′GGCATGCCCGACGTTTCTGCGCTTCGAGGGT GGCGTACCGTAC3′ con-taining the desired change (in bold), following the manufacturer'sprotocol. The full-length mutant clone was completely sequencedto confirm the presence of expected modifications and absence ofunwanted substitutions.

Transfection and recovery of the infectious clone-derived viruses

Plasmid pO1Ca or its mutant version was linearized at the EcoRVsite following the poly(A) tract and used as a template for RNAsynthesis using the MegaScript T7 kit (Ambion, Austin, TX), according

44 M.V. Borca et al. / Virology 422 (2012) 37–45

to the manufacturer's protocols. BHK-21 cells were transfected withthese synthetic RNAs by electroporation (Electrocell Manipulator 600;BTX, SanDiego, CA) as previously described (Mason et al., 1994). Briefly,0.5 ml of BHK-21 cells at a concentration of 1.5×107 cells/ml in phos-phate-buffered saline (PBS) were mixed with 10 μg of RNA in a 4-mm-gap BTX cuvette. The cells were then pulsed once at 330 V, infiniteresistance, and a capacitance of 1000 μF; diluted in cell growthmediumand allowed to attach to a T-25 flask. After 4 h, the medium wasremoved, fresh medium was added and the cultures were incubatedat 37 °C for up to 24 h.

The supernatants from transfected cells were passed in LF-BKαVβ6 cells until a cytopathic effect (CPE) appeared. After successivepassages in these cells, virus stocks were prepared and the viralgenome completely sequenced using the Prism 3730xl automatedDNA sequencer (Applied Biosystems), as previously described (Picconeet al., 2010).

Viruses derived from plasmid pO1Ca and the derived mutantplasmid encoding the aa substitution H56R in VP3 are referred toin this work as O1Ca-VP3-56H and O1Ca-VP3-56R, respectively.

In vitro processing

Plasmid pO1Ca containing the complete genome sequence waslinearized with EcoRV and in vitro transcribed using the T7-basedMegaScript kit (Ambion, Inc., Austin, Tex.). RNA transcripts weretranslated in vitro by using rabbit reticulocyte lysates (Promega,Fitchburg, WI) in the presence of [35S] methionine and translationreaction mixtures were resolved by electrophoresis on 15% polyacryl-amide gels containing sodium dodecyl sulfate (SDS-PAGE) and fluoro-graphed, following previously described procedures (Piccone et al.,1995). Additionally, pO1Ca was digested with selected restrictionenzymes that eliminate part of the viral open reading frame, andused in in vitro transcription/translation assays

Virulence of O1Ca-VP3-56H or O1Ca-VP3-56R viruses in cattle and swine

Animal experiments were performed under biosafety level 3 condi-tions in the animal facilities at PIADC following a protocol approved bythe Institutional Animal Care and Use Committee. Two steers (300–400 kg) were infected per the intradermolingual inoculation method(Henderson, 1949) with approximately 104 PFU/cattle of FMDV O1Ca-VP3-56H or O1Ca-VP3-56R, diluted in MEM. Six pigs (20 kg) wereinfected intradermally (Pacheco and Mason, 2010) with 104 PFU/pigof either FMDV O1Ca-VP3-56H or O1Ca-VP3-56R, diluted in MEM. For7 days after inoculation, animals were clinically examined, includingrectal temperature recordings and serum and nasal swabs collection.After collection, clinical samples were aliquotted and frozen at−70 °C. One serum and one swab aliquot were used to perform viral ti-tration by rRT-PCR as described previously (Pacheco et al., 2010). Forcattle, clinical scoring was made based on lesions present in each foot,with amaximumof 4. For swine, the clinical scorewasbased on severityof affected digits and/or snout, as previously described (Pacheco andMason, 2010), with a maximum of 20.

Acknowledgments

We thank Nuria Verdarguer for valuable conversations on themodeling of FMDV type O, the Plum Island Animal Resource Unitpersonnel for their excellent assistance with the animal experiments,the USDA-APHIS Foreign Animal Disease Diagnostic Laboratory forproviding the LK cells and Kathy Apicelli for excellent assistancewith the figures. We also thanks Melanie Prarat for the editing themanuscript.

References

Bai, X., Bao, H., Li, P., Sun, P., Kuang, W., Cao, Y., Lu, Z., Liu, Z., Liu, X., 2010. Geneticcharacterization of the cell-adapted PanAsia strain of foot-and-mouth diseasevirus O/Fujian/CHA/5/99 isolated from swine. Virol. J. 7, 208.

Baranowski, E., Molina, N., Nunez, J.I., Sobrino, F., Saiz, M., 2003. Recovery of infectiousfoot-and-mouth disease virus from suckling mice after direct inoculation with invitro-transcribed RNA. J. Virol. 77 (20), 11290–11295.

Belsham, G.J., 1993. Distinctive features of foot-and-mouth disease virus, a member ofthe picornavirus family; aspects of virus protein synthesis, protein processing andstructure. Prog. Biophys. Mol. Biol. 60 (3), 241–260.

Butrapet, S., Huang, C.Y., Pierro, D.J., Bhamarapravati, N., Gubler, D.J., Kinney, R.M.,2000. Attenuation markers of a candidate dengue type 2 vaccine virus, strain16681 (PDK-53), are defined by mutations in the 5′ noncoding region and non-structural proteins 1 and 3. J. Virol. 74 (7), 3011–3019.

Carrillo, C., Lu, Z., Borca, M.V., Vagnozzi, A., Kutish, G.F., Rock, D.L., 2007. Genetic andphenotypic variation of foot-and-mouth disease virus during serial passages in anatural host. J. Virol. 81 (20), 11341–11351.

Conenello, G.M., Zamarin, D., Perrone, L.A., Tumpey, T., Palese, P., 2007. A single mutationin the PB1-F2 of H5N1 (HK/97) and 1918 influenza A viruses contributes to increasedvirulence. PLoS Pathog. 3 (10), 1414–1421.

Cottam, E.M., King, D.P., Wilson, A., Paton, D.J., Haydon, D.T., 2009. Analysis of Foot-and-mouth disease virus nucleotide sequence variation within naturally infectedepithelium. Virus Res. 140 (1–2), 199–204.

Drake, J.W., Holland, J.J., 1999. Mutation rates among RNA viruses. Proc. Natl. Acad. Sci.U. S. A. 96 (24), 13910–13913.

Emsley, P., Cowtan, K., 2004. Coot: model-building tools for molecular graphics. Acta.Crystallogr. D. Biol. Crystallogr. 60 (12), 2126–2132.

Figlerowicz, M., Alejska, M., Kurzynska-Kokorniak, A., 2003. Genetic variability: the keyproblem in the prevention and therapy of RNA-based virus infections. Med. Res.Rev. 23 (4), 488–518.

Fry, E.E., Lea, S.M., Jackson, T., Newman, J.W., Ellard, F.M., Blakemore, W.E., Abu-Ghazaleh,R., Samuel, A., King, A.M., Stuart, D.I., 1999. The structure and function of a foot-and-mouth disease virus-oligosaccharide receptor complex. EMBO J. 18 (3),543–554.

Grubman, M.J., Baxt, B., 2004. Foot-and-mouth disease. Clin. Microbiol. Rev. 17 (2),465–493.

Haydon, D.T., Samuel, A.R., Knowles, N.J., 2001. The generation and persistence of geneticvariation in foot-and-mouth disease virus. Prev. Vet. Med. 51 (1–2), 111–124.

Henderson, W.M., 1949. The quantitative study of foot-and-mouth disease virus.Report Series Agricultural Research Council, London, no. 8, Her Majesty's StationeryOffice.

Jackson, T., Ellard, F.M., Ghazaleh, R.A., Brookes, S.M., Blakemore, W.E., Corteyn, A.H.,Stuart, D.I., Newman, J.W., King, A.M., 1996. Efficient infection of cells in cultureby type O foot-and-mouth disease virus requires binding to cell surface heparansulfate. J. Virol. 70 (8), 5282–5287.

Jarousse, N., Grant, R.A., Hogle, J.M., Zhang, L., Senkowski, A., Roos, R.P., Michiels, T.,Brahic, M., McAllister, A., 1994. A single amino acid change determines persistenceof a chimeric Theiler's virus. J. Virol. 68 (5), 3364–3368.

Jensen, M.J., Moore, D.M., 1993. Phenotypic and functional characterization of mouseattenuated and virulent variants of foot-and-mouth disease virus type O1 Campos.Virology 193 (2), 604–613.

Kanno, T., Yamakawa, M., Yoshida, K., Sakamoto, K., 2002. The complete nucleotide se-quence of the PanAsia strain of foot-and-mouth disease virus isolated in Japan.Virus Genes 25 (2), 119–125.

Kitson, J.D., McCahon, D., Belsham, G.J., 1990. Sequence analysis of monoclonal anti-body resistant mutants of type O foot and mouth disease virus: evidence forthe involvement of the three surface exposed capsid proteins in four antigenicsites. Virology 179 (1), 26–34.

Lea, S., Abu-Ghazaleh, R., Blakemore, W., Curry, S., Fry, E., Jackson, T., King, A., Logan, D.,Newman, J., Stuart, D., 1995. Structural comparison of two strains of foot-and-mouth disease virus subtype O1 and a laboratory antigenic variant, G67. Structure3 (6), 571–580.

Li, J., Das, P., Zhou, R., 2010. Single mutation effects on conformational change andmembrane deformation of influenza hemagglutinin fusion peptides. J. Phys.Chem. B 114 (26), 8799–8806.

Marquardt, O., Adam, K.H., Straub, O.C., 1991. Detection and localization of single-basesequence differences in foot-and-mouth disease virus genomes by the RNasemismatch cleavage method. J. Virol. Methods 33 (3), 267–282.

Martin-Acebes, M.A., Vazquez-Calvo, A., Rincon, V., Mateu, M.G., Sobrino, F., 2010. Asingle amino acid substitution in the capsid of foot-and-mouth disease virus canincrease acid resistance. J. Virol. 85 (6), 2733–2740.

Mason, P.W., Rieder, E., Baxt, B., 1994. RGD sequence of foot-and-mouth disease virus is es-sential for infecting cells via the natural receptor but can be bypassed by an antibody-dependent enhancement pathway. Proc. Natl. Acad. Sci. U. S. A. 91 (5), 1932–1936.

Mateo, R., Díaz, A., Baranowski, E., Mateu, M.G., 2003. Complete alanine scanning ofintersubunit interfaces in a foot-and-mouth disease virus capsid reveals criticalcontributions of many side chains to particle stability and viral function. J. Biol.Chem. 278 (42), 41019–41027.

Mateo, R., Luna, E., Rincón, V., Mateu, M.G., 2008. Engineering viable foot-and-mouthdisease viruses with increased thermostability as a step in the development ofimproved vaccines. J. Virol. 82 (24), 12232–12240.

Mateu, M.G., Martinez, M.A., Capucci, L., Andreu, D., Giralt, E., Sobrino, F., Brocchi, E., Domin-go, E., 1990. A single amino acid substitution affects multiple overlapping epitopes inthe major antigenic site of foot-and-mouth disease virus of serotype C. J. Gen. Virol.71 (Pt 3), 629–637.

45M.V. Borca et al. / Virology 422 (2012) 37–45

Meyer, R.F., Pacciarini, M., Hilyard, E.J., Ferrari, S., Vakharia, V.N., Donini, G., Brocchi, E.,Molitor, T.W., 1994. Genetic variation of foot-and-mouth disease virus from fieldoutbreaks to laboratory isolation. Virus Res. 32 (3), 299–312.

Morioka, K., Fukai, K., Ohashi, S., Sakamoto, K., Tsuda, T., Yoshida, K., 2008. Comparisonof the characters of the plaque-purified viruses from foot-and-mouth disease virusO/JPN/2000. J. Vet. Med. Sci. 70 (7), 653–658.

Nunez, J.I., Baranowski, E., Molina, N., Ruiz-Jarabo, C.M., Sanchez, C., Domingo, E.,Sobrino, F., 2001. A single amino acid substitution in nonstructural protein 3Acan mediate adaptation of foot-and-mouth disease virus to the guinea pig. J.Virol. 75 (8), 3977–3983.

Omata, T., Kohara, M., Kuge, S., Komatsu, T., Abe, S., Semler, B.L., Kameda, A., Itoh, H.,Arita, M., Wimmer, E., et al., 1986. Genetic analysis of the attenuation phenotypeof poliovirus type 1. J. Virol. 58 (2), 348–358.

Pacheco, J.M., Mason, P.W., 2010. Evaluation of infectivity and transmission of differentAsian foot-and-mouth disease viruses in swine. J. Vet. Sci. 11 (2), 133–142.

Pacheco, J.M., Arzt, J., Rodriguez, L.L., 2010. Early events in the pathogenesis of foot-and-mouth disease in cattle after controlled aerosol exposure. Vet. J. 183 (1),46–53.

Pelletier, I., Duncan, G., Colbere-Garapin, F., 1998. One amino acid change on the capsidsurface of poliovirus sabin 1 allows the establishment of persistent infections inHEp-2c cell cultures. Virology 241 (1), 1–13.

Pereda, A.J., Konig, G.A., Chimeno Zoth, S.A., Borca, M., Palma, E.L., Piccone, M.E., 2002.Full length nucleotide sequence of foot-and-mouth disease virus strain O1 Campos/Bra/58. Brief rep. Arch. Virol. 147 (11), 2225–2230.

Piatti, P., Hassard, S., Newman, J.F., Brown, F., 1995. Antigenic variants in a plaque-isolateof foot-and-mouth disease virus: implications for vaccine production. Vaccine 13(8), 781–784.

Piccone, M.E., Rieder, E., Mason, P.W., Grubman, M.J., 1995. The foot-and-mouth diseasevirus leader proteinase gene is not required for viral replication. J. Virol. 69 (9),5376–5382.

Piccone, M.E., Pacheco, J.M., Pauszek, S.J., Kramer, E., Rieder, E., Borca, M.V., Rodriguez,L.L., 2010. The region between the two polyprotein initiation codons of foot-and-mouth disease virus is critical for virulence in cattle. Virology 396 (1), 152–159.

Rieder, E., Bunch, T., Brown, F., Mason, P.W., 1993. Genetically engineered foot-and-mouth disease viruses with poly(C) tracts of two nucleotides are virulent inmice. J. Virol. 67 (9), 5139–5145.

Rodriguez Pulido, M., Sobrino, F., Borrego, B., Saiz, M., 2009. Attenuated foot-and-mouth disease virus RNA carrying a deletion in the 3′ noncoding region can elicitimmunity in swine. J. Virol. 83 (8), 3475–3485.

Rowlands, D.J., 2003. Special issue— foot-and-mouth disease— preface. Virus Res. 91 (1), 1.Ryan, M.D., Belsham, G.J., King, A.M., 1989. Specificity of enzyme-substrate interactions in

foot-and-mouth disease virus polyprotein processing. Virology 173 (1), 35–45.Sa-Carvalho, D., Rieder, E., Baxt, B., Rodarte, R., Tanuri, A., Mason, P.W., 1997. Tissue culture

adaptation of foot-and-mouth disease virus selects viruses that bind to heparin andare attenuated in cattle. J. Virol. 71 (7), 5115–5123.

Sevilla, N., Verdaguer, N., Domingo, E., 1996. Antigenically profound amino acid substitutionsoccur during large population passages of foot-and-mouth disease virus. Virology 225(2), 400–405.

Swaney, L.M., 1988. A continuous bovine kidney cell line for routine assays of foot-and-mouth disease virus. Vet. Microbiol. 18 (1), 1–14.

Szepanski, S., Gross, H.J., Brossmer, R., Klenk, H.D., Herrler, G., 1992. A single pointmutation of the influenza C virus glycoprotein (HEF) changes the viral receptor-binding activity. Virology 188 (1), 85–92.

Zhang, H., Blake, N.W., Ouyang, X., Pandolfino, Y.A., Morgan-Capner, P., Archard, L.C.,1995. A single amino acid substitution in the capsid protein VP1 of coxsackievirusB3 (CVB3) alters plaque phenotype in Vero cells but not cardiovirulence in amouse model. Arch. Virol. 140 (5), 959–966.

Zurbriggen, A., Hogle, J.M., Fujinami, R.S., 1989. Alteration of amino acid 101 withincapsid protein VP-1 changes the pathogenicity of Theiler's murine encephalomyelitisvirus. J. Exp. Med. 170 (6), 2037–2049.