on may 7, 2020 by guest · 8 pirbright institute, ash road, pirbright, surr ey, uk b; comision...

1

Development of a competitive ELISA for the detection of antibodies against the 3B 1

protein of foot-and-mouth disease virus 2

3

Ming Yang,a Satya Parida,

b Tim Salo,

a Kate Hole,

a Lauro Velazquez-Salinas,

c, * Alfonso 4

Clavijoa,#,

† 5

6

National Centre for Foreign Animal Disease, Winnipeg, Manitoba, Canadaa ; The 7

Pirbright Institute, Ash Road, Pirbright, Surrey, UK b; Comision Mexico-Estados Unidos 8

para la Prevencion de la Fiebre aftosa y otras Enfermedades Exoticas de los Animales, 9

Cuajimalpa, Mexico c; 10

11

12

Running Head: Detection of antibody against 3B protein of foot-and-mouth disease virus 13

14

#Address correspondence to Alfonso Clavijo, [email protected] 15

16

* Present address: Foreign Animal Disease Research Unit, Agricultural Research Service, 17

U.S Department of Agriculture, Plum Island Animal Disease Center, Orient Point, NY, 18

USA. 19

†Present address: Institute for Infectious Animal Diseases, College Station, TX, USA 20

21

This is a contribution from the National Centre for Foreign Animal Disease, Canada 22

23

CVI Accepted Manuscript Posted Online 4 February 2015Clin. Vaccine Immunol. doi:10.1128/CVI.00594-14Copyright © 2015, American Society for Microbiology. All Rights Reserved.

on May 13, 2020 by guest

http://cvi.asm.org/

Dow

nloaded from

http://cvi.asm.org/

2

Abstract 24

Foot-and-mouth disease (FMD) is one of the most highly contagious and economically 25

devastating diseases, that severely constrains the international trade of animals. 26

Vaccination against FMD is a key element in the control of FMD. However, vaccination 27

of susceptible animals raises critical issues, such as the differentiation of infected animals 28

from vaccinated animals. The current study developed a reliable and rapid test to detect 29

antibodies against the conserved, non-structural proteins (NSP) of the FMD virus 30

(FMDV) to distinguish infected animals from vaccinated animals. A monoclonal 31

antibody (mAb) against the FMDV NSP 3B was produced. A competitive enzyme-linked 32

immunosorbent assay (cELISA) for FMDV/NSP antibody detection was developed using 33

a recombinant 3ABC protein as the antigen and the 3B-specific mAb. Sera collected from 34

naïve, FMDV experimentally infected, vaccinated carrier and non-carrier animals were 35

tested using the 3B cELISA. The diagnostic specificity was 99.4% for naïve animals 36

(cattle, porcine and sheep) and 99.7% for vaccinated non-carrier animals. The diagnostic 37

sensitivity was 100% for experimentally inoculated animals and 64% for vaccinated 38

carrier animals. The performance of this 3B cELISA was compared to four commercial 39

ELISA kits using a serum panel established by the World Reference Laboratory for FMD 40

at Pirbright. The diagnostic sensitivity of the 3B cELISA for the panel of FMDV/NSP 41

positive bovine serum was 94%, which was comparable or better than commercially 42

available NSP antibody detection kits. This 3B cELISA is a simple, reliable test to detect 43

antibodies against FMDV non-structural proteins. 44

45

46


http://cvi.asm.org/

Dow

nloaded from

http://cvi.asm.org/

3

Introduction 47

Foot-and-mouth disease (FMD) is one of the most highly contagious and economically 48

devastating diseases of cloven-hoofed animals and it severely constrains the international 49

trade of animals and animal products. Vaccination against FMD, in addition to the 50

slaughter and movement restrictions of infected animals, is a key element in the control 51

of FMD. However, countries that vaccinate in the event of an outbreak must re-establish 52

their FMD-free status to the satisfaction of their trading partners (1, 2). Vaccination of 53

susceptible animals raises critical issues, such as the differentiation of infected animals 54

from vaccinated animals and the development of carrier status because of subclinical 55

infection in vaccinated animals. 56

FMD is caused by the FMD virus (FMDV), which is a member of the genus 57

Aphthovirus and the family Picornaviridae (3) and it exhibits seven serotypes, O, A, Asia 58

1, C, SAT 1, SAT 2 and SAT 3. FMDV has a positive sense, single-stranded ribonucleic 59

acid (RNA) genome of 8,400 nucleotides that code for twelve proteins. Four structural 60

proteins (VP1, VP2, VP3 and VP4) compose the viral capsid, and eight proteins are non-61

structural proteins (L, 2A, 2B, 2C, 3A, 3B, 3C and 3D). All 12 proteins allow the virus to 62

replicate in infected cells.(4-6). Antibodies to the 3ABC non-structural proteins (NSPs) 63

are a reliable indicator of infection regardless of the FMDV serotype. ELISAs for the 64

detection of antibodies against NSPs are widely used to differentiate between vaccinated 65

and infected animals because purified vaccines are free of NSPs, thus elicit antibodies 66

against only structural proteins (7). However, not all manufactures produce purified FMD 67

vaccines, and the degree of purity among FMD vaccine manufacturers is not always 68

identical (8). 69


http://cvi.asm.org/

Dow

nloaded from

http://cvi.asm.org/

4

Several tests for the detection of antibodies against NSPs were reported, and some of 70

these tests were made into commercially available kits. The tests produced by Svanova, 71

Bommeli, and UBI and several other published tests (9-15) are not ideal, because these 72

tests require species-specific conjugated antibodies. Separate assays are required to test 73

samples from different species (cattle, deer, goats, and sheep) and no reagents are 74

available for wildlife (2, 16). A wide range of animal species are susceptible to FMDV. 75

Therefore, a competitive ELISA (cELISA) would be advantageous because serum 76

samples from different species could be tested without changing reagents (17). cELISAs 77

are simple, easy to perform and species independent. Numerous cELISAs for the 78

detection of antibodies against NSPs were used to differentiate vaccinated animals from 79

infected animals (1, 18). However, polyclonal antibodies were used as the competitor in 80

these tests. The use of polyclonal antibodies cannot ensure consistent quality compared 81

to monoclonal antibodies (mAb) because of batch-to-batch variations. Sørensen et al. (17) 82

produced a mAb against NSP-3B and developed a cELISA using the same mAb (L74D5) 83

as the capture and detector antibody in a blocking ELISA. The disadvantage of this 84

ELISA system is that when the antigen binds to the capture antibody, the same epitope 85

that is recognized by the polyclonal or competition antibodies might be hidden, which 86

reduces the test sensitivity. The PrioCHECK NS uses a specific mAb against NSP and a 87

recombinant NSP protein in a cELISA format. One study demonstrated that the 88

PrioCHECK NS test is sensitive and very specific in the buffalo populations of eastern 89

Africa (16). However, the use of a commercial kit for regular diagnosis and surveillance 90

can be costly. The development of an effective in-house test for the detection of 91


http://cvi.asm.org/

Dow

nloaded from

http://cvi.asm.org/

5

antibodies against FMDV/NSP regardless of the species is necessary to make daily tests 92

more affordable. 93

A mAb against a conserved epitope located on the 3B NSP was produced. A cELISA 94

for FMD 3B-NSP antibody detection was developed using this mAb and a recombinant 95

3ABC protein. The assay was validated using a serum panel and samples collected from 96

naïve, experimentally inoculated, vaccinated, and carrier and non-carrier animals. 97

Development of a reliable NSP antibody assay would enable successful disease diagnosis 98

and the establishment of disease-free zones, and acceleration of global FMD control. 99

100

Materials and methods 101

102

Non-structural protein production. Cloning and expression of His-tagged 3ABC 103

recombinant protein was performed as described previously (1, 19). Recombinant 104

bacterial antigen was produced in cultures that were incubated for 2-3 hours at 37oC with 105

shaking until an OD 600 of 0.6–0.7 was reached. Protein expression was induced by the 106

addition of isopropyl β-D-1-thiogalactopyranoside (IPTG) to a final concentration of 1 107

mM, and the solution was incubated for an additional 3 hours at 37oC with shaking. The 108

culture was pelleted using centrifugation at 8000 ×g for 15 minutes at 4oC. The cell pellet 109

was completely resuspended at room temperature in BugBuster reagent (25 U/mL, 110

Novagen, Darmstadt, Germany) to extract the recombinant 6xHis-3ABC. Benzonase 111

nuclease (Novagen) was added, and the suspension was incubated at room temperature 112

with gentle shaking for 10-20 minutes. Insoluble cell debris was collected using 113

centrifugation at 10,000 xg for 20 minutes at 4oC. The pellet was completely resuspended 114


http://cvi.asm.org/

Dow

nloaded from

http://cvi.asm.org/

6

in the same volume of BugBuster reagent and centrifuged at 10,000 xg for 20 minutes at 115

4oC to collect the inclusion bodies. The final pellet was resuspended in 50 ml of 8 M 116

UREA, 0.1 M sodium phosphate buffer, 0.01 M Tris-HCl pH 8.0 until the inclusion 117

bodies (3ABC recombinant protein) became soluble and the solution was transparent. 118

Any insoluble material was removed using centrifugation at 10,000 xg for 20 minutes at 119

4oC. The collected supernatant containing the recombinant 3ABC NSP was aliquoted and 120

stored at -70oC. 121

122

Peptides, mouse immunization and monoclonal antibody production. The peptide, 123

H2N-CGPYAGPLERQKPLK –OH (20) derived from 3B1, was synthesized for 124

immunization and its purity was assessed using HPLC (New England Peptide, MA, 125

USA). The peptide was conjugated to the carrier protein-Keyhole limpet hemocyanin 126

(KLH) by New England Peptide. 127

The Canadian Science Centre for Human and Animal Health Animal Care Committee 128

approved animal use and study procedures (Animal use document #C-02-002). The 129

procedures for mouse immunization and mAb production were performed as previously 130

described (21). Hybridoma supernatants were screened using the recombinant 3ABC NSP 131

as the antigen in an indirect ELISA. The positive clone (F8-3Bp) was subcloned and the 132

isotype was determined using a mouse monoclonal antibody isotyping kit (Roche, 133

Indianapolis, IN, USA). 134

135

Development of the 3B cELISA. The optimal concentrations for recombinant 3ABC 136

protein, mAb and conjugated antibody dilutions were determined using checkerboard 137

titrations of known strongly positive, weakly positive and negative sera. The 3B cELISA 138


http://cvi.asm.org/

Dow

nloaded from

http://cvi.asm.org/

7

was performed by coating microtiter plates (Nunc-Immunoplate, Roskilde, Denmark) 139

with 100 μl/well of recombinant 3ABC antigen that was diluted based on the titration of 140

each batch in 0.06 M carbonate/bicarbonate buffer, pH 9.6, and incubated overnight at 141

4°C. After five washes with the washing buffer (0.05% Tween20 in 0.01 M phosphate-142

buffered saline (PBS-T)), 100 µL of saturation buffer (50 mM Tris-Cl, 150 mM NaCl, 143

0.05% NaN3, 0.2% BSA, 6% D-Sorbitol) was added to each well and incubated 144

overnight at 4°C. The saturation buffer was discarded, and the plates were dried at room 145

temperature for 5 hours, heat-sealed under vacuum and stored up to 12 months at 4°C. 146

Plates were washed 3 times with PBS-T for testing, followed by the addition of 50 μl of 147

heat-inactivated test sera (final 1:5 dilution) in duplicate and an equal volume of 148

hybridoma culture supernatants (1:1000) in PBS-T. The plates were incubated at 37°C for 149

1 hour with agitation. After washing five times, 100 μl of horseradish peroxidase (HRP)-150

conjugated goat anti-mouse IgG (1:2000) (Jackson ImmunoResearch Laboratories Inc.) 151

in PBS-T was added and incubated for 1 hour at 37 °C with subsequent washing. The 152

3,3’,5;-tetramethyl benzidine dihydrochloride substrate (TMB, Sigma-Aldrich, St Lucia, 153

MO, USA) was added, and color development was stopped after 15 minutes with the 154

addition of 50 μl/well of 2 M sulfuric acid. The optical density (OD) was determined at 155

450 nm using an automated plate reader (Photometer Multiskan Reader, Labsystems, 156

Foster, VA, USA). 157

Results were calculated based on strong positive (Q1), weak positive (Q2) and negative 158

reference sera (Q3). Test results (mean of serum tested in duplicate wells) were derived 159

using the following formula: Percentage of inhibition (%) = [(Q3 OD - test sample OD) / 160

(Q3 OD – Q1 OD)] X 100%. 161


http://cvi.asm.org/

Dow

nloaded from

http://cvi.asm.org/

8

162

Negative sera from bovine, porcine and ovine. Negative serum samples were collected 163

from naïve cattle (n = 503), pigs (n=508) and sheep (n=507) in Canada. Sera from naïve 164

cattle (n=353) and sheep (n=436) were also included in the evaluation in Mexico. 165

166

Raising of sera against the seven serotypes of FMDV in cattle, sheep, pigs and deer. 167

The FMDV strains (O1/UKG11/2001, A22 Iraq, A24/Cruzeiro/Br/55, C1 Noville, Asia 168

1/Shamir, SAT 1/BOT 1/68, SAT 1 KEN/4/98, SAT 2 SAU 1/2000 and SAT 3 ZIM 169

4/81) are reference strains that were obtained from the World Reference Laboratory for 170

FMD (WRLFMD) at the Pirbright Institute, UK. Baby hamster kidney clone 21 cells 171

(BHK-21) or lamb kidney primary cells (LK) were used to prepare viruses for the seven 172

FMDV serotypes. The cells were infected with FMDV in Glasgow’s MEM supplemented 173

with 2 mM L-glutamine and 50 µg/ml gentamycin. Viruses were harvested upon 174

observation of a complete cytopathic effect (CPE) and clarified using centrifugation at 175

2000 xg for 20 minutes. 176

All procedures involving experimental animal inoculations and care were performed 177

according to the Canadian Council of Animal Care guidelines. The local animal care 178

committee approved all animal procedures prior to initiation of the study. Cattle, sheep, 179

pigs and deer (1, 22) were inoculated with one of the FMDV strains. Cattle and sheep 180

were infected via intradermal/subdermal injections on the dorsal aspect of the tongue and 181

pigs were inoculated in their heel bulb with a total viral dose of 106 50% tissue culture 182

infectious doses (TCID50) at five sites. Two white-tailed deer (D16, D20) were inoculated 183

intranasally with 104 TCID50 of FMDV OUKG 11/2001 diluted in 5 ml of culture 184


http://cvi.asm.org/

Dow

nloaded from

http://cvi.asm.org/

9

medium. Blood samples were collected before (0 days post-inoculation (dpi)) and after 185

virus inoculation until 28-31 dpi. 186

187

Positive bovine serum panel for evaluation of the 3B cELISA. The bovine serum 188

panel (23) was established by the WRLFMD at the Pirbright Institute. This panel 189

evaluates new tests and provides quality control for new batches of existing tests. Thirty-190

six bovine serum samples were selected from a series of vaccination-challenge 191

experiments. The details of the experiments used to generate these samples were reported 192

previously (23). 193

194

Sera from control, vaccinated and carrier cattle. These sera were raised by the 195

WRLFMD at the Pirbright Institute. Cattle were vaccinated with O1 Manisa and 196

challenged with O UKG 34/2001 in two experiments, UV and UY (24, 25). The 197

vaccination dose in the second experiment (UY), was increased 10-fold compared to the 198

first (UV). Cattle were grouped into control (non-vaccinated), vaccinated non-carriers 199

and vaccinated carriers. The carrier state was confirmed by either positive virus isolation 200

or quantitative RT-PCR in the probang samples taken from individual animals (24, 25). 201

Serum samples were collected from these animals seven days before vaccination and at 202

seven days intervals until 168 days post challenge (dpc) (26, 27). However, not every 203

animal was sampled at all-time points. 204

205


http://cvi.asm.org/

Dow

nloaded from

http://cvi.asm.org/

10

Statistical analysis. Correlation analyses for comparisons of the two test methods (the 206

3B cELISA versus PrioCHECK NS) were performed using Microsoft Excel. The 207

statistical significance was set at P<0.05. 208

209

Results 210

211

Production of a monoclonal antibody against the 3B peptide. The KLH-conjugated 212

FMDV 3B peptide (H2N-CGPYAGPLERQKPLK–OH) was used as the immunogen to 213

produce the peptide-specific mAb (F8-3Bp). This mAb is an IgG1 isotype with kappa 214

light chains. The mAb was examined for its reactivity and specificity against the 215

recombinant 3ABC NSP and the 3B peptide using an indirect ELISA. Results indicated 216

that the mAb reacted with the recombinant 3ABC protein and the 3B peptide in the 217

ELISA with an OD higher than 1.0 after background subtraction. The mAb F8-3Bp 218

competed with the serum from a FMDV-infected animal. 219

220

Development and evaluation of the 3B cELISA. A cELISA to detect antibodies against 221

FMDV NSP was developed using the newly generated mAb (F8-3Bp) and the 222

recombinant 3ABC protein as the antigen. The optimal concentrations for the 223

recombinant 3ABC protein, mAb and conjugated antibody dilutions were optimized 224

using checkerboard titrations of known strong positive, weak positive and negative sera. 225

A total of 1518 negative sera (bovine = 503, porcine=508 and ovine=507) were 226

collected from Canada and tested using the 3B cELISA. The frequencies of the 227

percentage inhibition (PI) generated from these sera were normal distributed (Fig 1a). 228


http://cvi.asm.org/

Dow

nloaded from

http://cvi.asm.org/

11

The mean PIs for the negative sera were 4.43% (standard deviation (SD) =14.03), - 229

8.63% (SD=14.0) and -9.6% (SD=9.59) for bovine, porcine and ovine, respectively. 230

Similar results were obtained using sera collected from Mexico (bovine =353, 231

ovine=436) (Fig 1b). The mean PIs were -6.4 (SD=17.35) and -15.76 (SD=18.92) for 232

bovine and ovine, respectively. The negative cut-off values, which were calculated by the 233

addition of 3SD to the mean PIs of the negative serum, equaled 19-46% of the inhibition 234

for all animal species. Therefore, the cut-off value was set at <50% of inhibition (28). 235

Less than 1% of the tested negative samples collected from bovine, porcine and ovine 236

exceeded this cut-off value, which produced a diagnostic specificity of 99.4%. 237

A bovine serum panel for the evaluation of antibodies against FMDV NSP was tested 238

using the 3B cELISA and compared with other commercially available tests. This serum 239

panel (n=36) was selected from a series of vaccination-challenge experiments (23, 29). 240

All samples were seropositive by the virus neutralization test (VNT) (23). Comparison of 241

the 3B cELISA with commercially available tests indicated that the 3B cELISA detected 242

the highest percentage of positive sera (94%) among the NSP antibody detection tests 243

(Table 1). PrioCHECK NS detected slightly lower positive percentage (92%) than the 3B 244

cELISA. Bommedi, UBI and Svanova demonstrated positive results of 61-81%. The 3B 245

cELISA detected positive NSP antibody in 34 of 36 serum samples, and it failed to detect 246

positive antibodies against NSP in two samples (UV10 and UV 19) that were also 247

negative in the Bommedi, UBI and Svanova tests. These two samples were from 248

vaccinated (O1/Manisa) and challenged (O1/UKG) carriers. Their VNT titers were 178 249

and 256, respectively. Three serum samples (2 from the vaccinated carrier group, 1 from 250


http://cvi.asm.org/

Dow

nloaded from

http://cvi.asm.org/

12

the vaccinated non-carrier) with VNT titers > 512 exhibit a negative NSP antibody 251

response in the PrioCHECK NS (29). 252

Serum samples from vaccinated non-carriers in vaccination-challenge experiment #1 253

were tested using both the 3B cELISA and PrioCHECK NS to evaluate the 3B cELISA as 254

a test for differentiating vaccinated from infected animals (DIVA). The PrioCHECK NS 255

is the most commonly used test for FMD/NSP-specific antibody detection. Therefore, the 256

3B cELISA results were compared with the PrioCHECK NS. A total of 114 and 111 of 257

115 samples resulted in percentage inhibitions below (<) 50% for 3B cELISA and 258

PrioCHECK NS. These results indicate that there were no or little antibodies against NSP 259

present in these vaccinated non-carrier animals. One exceptional sample (UV8), which 260

was collected at 42 dpc showed a positive antibody response with a PI of 60%, which 261

produced a DIVA test specificity of 99.1% for the 3B cELISA (Fig 2a). Four samples 262

demonstrated positive antibody responses in the PrioCHECK NS with a test 263

specificity of 96.6% (Fig2b). All cattle in the control group (non-vaccinated, 264

challenged) exhibited a PI > 50% from 14 dpc, which indicates the presence of 265

antibodies to 3B NSP. The 3B cELISA results demonstrate a good correlation with the 266

PrioCHECK NS results with a correlation coefficient of 0.87 and 0.98 for the two 267

experiments, respectively (Fig 3). Moreover, the sera taken from vaccinated-carrier cattle 268

were tested using 3B cELISA and compared with previous PrioCHECK NS results (Fig 269

4). Seven of 11 cattle showed positive seroconversion in both 3B cELISA and 270

PrioCHECK NSs after 14 dpc. However, two carrier cattle (UV2 and UV14), which 271

were confirmed positive using reverse transcription PCR were negative in these two 272

tests. The other two carrier cattle (UV5 and UY76) showed positive antibody 273


http://cvi.asm.org/

Dow

nloaded from

http://cvi.asm.org/

13

responses against NSP in the PrioCHECK NS, but negative antibody responses in the 274

3B cELISA. The correlation coefficient between the 3B cELISA and the PrioCHECK 275

NS for both experiments was 0.70 with p< 0.0001 (Fig4c). 276

277

Detection of seroconversion in FMDV inoculated animals using 3B cELISA. The 278

antibody responses to FMDV/NSP in experimentally inoculated animal species (15 cattle, 279

9 porcine, 18 sheep, and 2 deer) were examined using the 3B cELISA. Antibodies to NSP 280

were undetectable from 0-5 dpi for swine, cattle and deer (Table 2). One of 18 sheep 281

showed a positive antibody response against NSP on 5 dpi. A total of 89-100% of 282

samples from all tested species demonstrated positive antibody responses at 9 dpi. Two 283

samples, one pig and one sheep, demonstrated positive NSP antibody response after 9 284

dpi. Therefore seroconversions occurred mainly between 6-9 dpi. All animal samples, 285

except one pig, remained positive for antibody responses until the end of the experiment 286

(28 dpi). These results indicated that the 3B cELISA detected antibodies against NSP in 287

all tested animal species. 288

Serum samples from animals experimentally inoculated with all seven serotypes were 289

tested to determine whether the 3B cELISA detected NSP antibodies in sera of different 290

serotypes. The results demonstrated positive antibody response to NSP at 8-10 dpi, and 291

responses remained positive until the end of the experiment (Fig 5). Two sera from sheep 292

experimentally inoculated with SAT 3 showed a positive seroconversion after 10 dpi and 293

remained positive until 28 dpi. These results showed that 3B cELISA is serotype 294

independent for NSP-specific antibody detection. 295

296

Discussion 297


http://cvi.asm.org/

Dow

nloaded from

http://cvi.asm.org/

14

The use of vaccines to FMD is an economic strategy to prevent and control this disease. 298

The differentiation between infected and vaccinated animals is also important in 299

surveillance efforts. An accurate test for detecting infection in susceptible animals is 300

needed, especially for infections after vaccination (30). ELISA for the detection of 301

antibodies against FMDV/NSP can distinguish between vaccinated animals and infected 302

animals. A good mAb that recognizes strain-independent epitopes on FMDV/NSP is 303

required to develop tests for FMDV/NSP antibody detection. Linear B-cell epitopes 304

located in the FMDV NSP were identified (31, 32). Höhlich et al. (20) identified six 305

linear B-cell epitopes in the NSPs. Three of six peptides were recognized by the sera 306

collected from all infected animals regardless of the infecting strains. None of the sera 307

from vaccinated animals contained antibodies against these peptides in their experiments. 308

Therefore, our study, used one (CGPYAGPLERQKPLK) of the 3B peptides (20) as the 309

immunogen for mAb production and the mAb produced against this peptide was used as 310

the basis for the 3B cELISA to differentiate vaccinated from infected animals. 311

The mAbs that are raised by peptides may not recognize original or recombinant 312

protein (33, 34). However, the mAb F8-3Bp that was produced in the current study 313

reacted with the 3B peptide and recombinant 3ABC. Similar results were also obtained 314

by Yang et al (35), who successfully produced mAbs against FMDV/NSP 3D using 315

immunization with recombinant 3D and boosted with a 3D peptide. Different species may 316

recognize different epitopes. Sera from FMDV-infected pigs failed to react with any 317

linear epitope located on the 3D protein (35). The peptide used for immunization in this 318

study was recognized by sera from all infected animals independent of animal species 319

(20). 320


http://cvi.asm.org/

Dow

nloaded from

http://cvi.asm.org/

15

A 3B cELISA was developed using the recombinant 3ABC protein and the 3B-specific 321

mAb. A cut-off value of 50% inhibition was established based on 2307 negative sera, 322

which provided a clear distinction between positive and negative sera. Diagnostic 323

specificities using this cut-off value were 99.4% for all tested animal species (bovine, 324

porcine and ovine). The diagnostic specificity of a previously developed 3ABC indirect 325

ELISA for bovine was estimated at 96.4% (15). Some previously developed tests were 326

designed to detect NSP antibodies predominantly in cattle, but these tests are less useful 327

in sheep and pigs. Sheep, in particular, may fail to develop detectable levels of these 328

antibodies because of the frequently subclinical nature of FMD (30, 36). Immune 329

responses to infection may vary within individuals, virus serotypes and animal species, 330

but our 3B cELISA detected antibodies against FMDV/NSP in cattle, swine and sheep in 331

a serotype-independent manner. However, more deer and buffalo samples may need to be 332

tested. 333

Evaluations of the performance of a new test require comparisons of the performance to 334

standard samples. A serum panel (n=36) collected from FMD vaccine-challenged animals 335

(WRLFMD, Pirbright UK) was used to evaluate the 3B cELISA. The test results 336

indicated that the 3B cELISA detected the highest percentage of positive sera (94%) 337

among the NSP antibody detection tests (PrioCheck NS, Bommedi, UBI and Svanova). 338

The 3B cELISA failed to detect two positive samples (UV10 and UV 19) from this panel. 339

These samples consisted of sera collected from cattle that were vaccinated and challenged 340

by direct contact. The most reliable and consistent route of infection in experimental 341

studies on FMD is generally via intradermal/subdermal injection of virus (2). Therefore, 342

one possible explanation for this observation may be that vaccinated cattle challenged by 343


http://cvi.asm.org/

Dow

nloaded from

http://cvi.asm.org/

16

contact with infected donors rather than by needle inoculation were generally 344

subclinically infected and produced low levels of NSP antibody because of limited viral 345

replication in the presence of high levels of structural antibodies produced against 346

vaccine (23). 347

The aim of this study was to develop a DIVA test, and samples collected from 348

vaccinated cattle in two vaccine-challenge experiments were analyzed. Theoretically, 349

antibody responses against FMDV/NSP in the vaccinated population should be similar to 350

the naive population. Serum samples collected from 11 vaccinated and challenged non-351

carrier cattle demonstrated positive virus isolation results from probang samples that 352

cleared by 28 days post-initial contact (24, 25) were tested using the 3B cELISA and 353

PrioCHECK NS. A total of 114 of the 115 samples from these animals (-7 dpv to168dpc) 354

showed negative results in the 3B cELISA, which indicated that no antibodies against 355

NSP were present in those serum samples. These cattle were challenged after vaccination, 356

but there was little or no FMDV replication because of induction of the protective 357

immune response. Therefore no antibodies against NSP were detected. 358

Samples collected from control cattle (non-vaccinated, challenged) were also tested 359

using the 3B cELISA and the results were comparable with the PrioCHECK NS results 360

(23). The PrioCHECK NS demonstrated the highest diagnostic specificity among the 361

commercial ELISA kits in a comparative study (37, 38). The 3B cELISA exhibits a good 362

correlation with the PrioCHECK NS results with a correlation coefficient of 0.87 and 363

0.98 for experiment 1 and 2, respectively. Clearly, the 3B cELISA can be used to detect 364

antibodies against FMDV/NSP and differentiate vaccinated from infected animals (39). 365


http://cvi.asm.org/

Dow

nloaded from

http://cvi.asm.org/

17

Vaccinations in FMD endemic countries or during an outbreak situation result in a high 366

possibility that the clinical diseases will be masked in animals with partial immunity that 367

are exposed to live virus (36). The challenge for FMD diagnosis is to identify vaccinated 368

animals that have had contact with live virus and become carriers. The sera from 369

vaccinated-carrier cattle in two vaccine-challenge experiments were tested using the 3B 370

cELISA and compared with PrioCHECK NS test results. Seven of 11 cattle showed 371

seroconversion in both the 3B cELISA and PrioCHECK NS tests after 14 dpc. Two 372

vaccinated-carrier cattle (UV2 and UV14), which were confirmed positive results using 373

RT-PCR, showed negative results in both the 3B cELISA and PrioCHECK NS tests. The 374

other two vaccinated-carriers (UV5 and UY76) showed positive antibody responses 375

against FMDV/NSP in the PrioCHECK NS test, but were negative in the 3B cELISA. 376

The development of antibodies to NSP correlates with the extent of virus replication. 377

Therefore, the immune response induced by vaccination may reduce virus replication and 378

the amount of FMDV/NSP produced, which results in a low anti-NSP antibody response 379

(17, 26, 40). One outcome could be that, the antibody response may fall below the 380

detection limit of the 3B cELISA because of a decreased level of FMDV/NSPs. However, 381

some vaccinated carrier animals fail to develop antibodies against the NSPs (11). Present 382

tests for antibodies to NSPs cannot completely guarantee that a population of vaccinated 383

animals exposed to live virus contains no carriers (2). Caution must be taken when 384

analyzing data from vaccinated animals. The definitive identification of carrier status or 385

subclinically infected animals requires the recovery of live FMDV from these animals or 386

the potentially use of real time RT-PCR ((2, 36). 387


http://cvi.asm.org/

Dow

nloaded from

http://cvi.asm.org/

18

In conclusion, the 3B cELISA that was developed using a 3B-specific mAb and a 388

recombinant 3ABC antigen is a simple and reliable diagnostic test for the detection of 389

FMDV 3B-specific antibodies. The 3B cELISA exhibited comparable performance to the 390

PrioCHECK NS test, and it out-performed other FMDV/NSP antibody detection tests. 391

The binding epitope of the mAb used in this 3B cELISA is well characterized and 392

conserved among the FMD seven serotypes. The test is serotype and species independent. 393

The advantages of the 3B cELISA over the PrioCHECK NS include (1) a reduced assay 394

time using pre-coated plates of under four hours for the 3B cELISA compared to an assay 395

time of 20 hours for the PrioCHECK NS and (2) a much smaller cost for the 3B cELISA 396

(< $1 per sample in duplicate) than the PrioCHECK NS test ($5-6 per sample). The 397

reduced time and cost to detect antibodies against NSP while maintaining diagnostic 398

sensitivity and specificity is critical for FMD diagnosis. 399

400

Acknowledgements 401

The authors gratefully thank the animal care staff, for expert animal services, and 402

Professor Soren Alexandersen and Dr. Charles Nfon for critical review of the manuscript. 403

This work was supported by funds from the Canadian Food Inspection Agency and 404

theDepartment for Environment, Food and Rural Affairs, UK (DEFRA). 405

406

Legends 407

FIG 1 Frequency distribution of the negative sera tested using the 3B cELISA. (a) 408

Serum samples collected from disease-free bovines (n = 503), porcine (n=508) and ovine 409

(n=507) in Canada and (b) bovines (n=353) and ovine (n=436) in Mexico. Recombinant 410


http://cvi.asm.org/

Dow

nloaded from

http://cvi.asm.org/

19

3ABC protein was coated onto microtiter plates. Equal volumes of diluted sera and mAb 411

(F8-3Bp) were added to the plates and allowed to compete with antibodies in serum 412

samples. Results are expressed as a percentage of inhibition. 413

FIG 2 Evaluation of the 3B cELISA using sera from vaccinated, non-carrier cattle. 414

Serum samples from vaccine experiment #1(UV3,4,6,7,8,12,15,16,18,20,21) were 415

collected from vaccinated non-carrier cattle at 7 days intervals from -7 dpv to 168 dpc 416

during the course of experiments. Samples tested using (a) the 3B cELISA, and (b) the 417

PrioCHECK NS. 418

FIG 3 Correlations between the 3B cELISA and PrioCHECK NS results in sera 419

collected from the control group (un-vaccinated and challenged). Correlation 420

coefficients between the 3B cELISA and the PrioCHECK NS were determined for 421

vaccine experiment #1 (a) and vaccine experiment #2 (b). 422

FIG 4 Test results of the 3B cELISA and PrioCHECK NS using sera collected from 423

vaccinated-carrier cattle. Serum samples were collected from vaccinated-carrier cattle 424

at 7 days intervals from -7 dpv to 168 dpc during the course of the two vaccination-425

challenge experiments and assayed by (a) 3B cELISA and, (b) PrioCHECK NS. (c) 426

Correlations between the 3B cELISA and PrioCHECK NS test. 427

FIG 5 Detection of FMDV/NSP antibodies in animals inoculated with FMDV seven 428

serotypes using the 3B cELISA. Sera collected from animals experimentally inoculated 429

with seven FMDV serotypes (a) O, (b) A24, (c) A22, (d) C1, (e) Asia 1, (f) SAT 1, (g) 430


http://cvi.asm.org/

Dow

nloaded from

http://cvi.asm.org/

20

SAT 2, (h) SAT 3 and tested using the 3B cELISA. C: cattle; S: sheep; P: porcine, and D: 431

deer. 432

References 433

1. Clavijo, A., E. M. Zhou, K. Hole, B. Galic, and P. Kitching. 2004. Development and 434

use of a biotinylated 3ABC recombinant protein in a solid-phase competitive ELISA for 435

the detection of antibodies against foot-and-mouth disease virus. J. Virol. Methods. 436

120:217-227. doi: 10.1016/j.jviromet.2004.05.007. 437

2. Alexandersen, S., Z. Zhang, A. Donaldson, and A. Garland. 2003. The 438

pathogenesis and diagnosis of foot-and-mouth disease. J. Comp. Pathol. 129:1-36. 439

3. Belsham, G. J. 1993. Distinctive features of foot-and-mouth disease virus, a member 440

of the picornavirus family; aspects of virus protein synthesis, protein processing and 441

structure. Prog. Biophys. Mol. Biol. 60:241-260. 442

4. Domingo, E., C. Escarmis, E. Baranowski, C. M. Ruiz-Jarabo, E. Carrillo, J. I. 443

Nunez, and F. Sobrino. 2003. Evolution of foot-and-mouth disease virus. Virus Res. 444

91:47-63. 445

5. Mason, P. W., J. M. Pacheco, Q. Z. Zhao, and N. J. Knowles. 2003. Comparisons of 446

the complete genomes of Asian, African and European isolates of a recent foot-and-447

mouth disease virus type O pandemic strain (PanAsia). J. Gen. Virol. 84:1583-1593. 448

6. Fry, E., D. Stuart, and D. Rowlands. 2005. The structure of foot-and-mouth disease 449

virus, p. 71-101. In Anonymous Foot-and-Mouth Disease Virus. Springer. 450

7. Sørensen, K., K. Madsen, E. Madsen, J. Salt, J. Nqindi, and D. Mackay. 1998. 451

Differentiation of infection from vaccination in foot-and-mouth disease by the detection 452

of antibodies to the non-structural proteins 3D, 3AB and 3ABC in ELISA using antigens 453

expressed in baculovirus. Arch. Virol. 143:1461-1476. 454

8. Doel, T. 2001. Repeated administration of maximum payload emergency vaccines 455

made from inactivated purified antigen concentrates do not induce significant titles of 456

antibodies against non-structural proteins of foot-and-mouth disease virus. Report of a 457

Session of the Research Group of the Standing Technical Committee of the European 458

Commission for the Control of Foot-and-Mouth Disease–EUFMD. 88-92. 459

9. Bergmann, I. E., V. Malirat, E. Neitzert, E. Beck, N. Panizzutti, C. Sanchez, and 460

A. Falczuk. 2000. Improvement of a serodiagnostic strategy for foot-and-mouth disease 461

virus surveillance in cattle under systematic vaccination: a combined system of an 462


http://cvi.asm.org/

Dow

nloaded from

http://cvi.asm.org/

21

indirect ELISA-3ABC with an enzyme-linked immunoelectrotransfer blot assay. Arch. 463

Virol. 145:473-489. 464

10. De Diego, M., E. Brocchi, D. Mackay, and F. De Simone. 1997. The non-structural 465

polyprotein 3ABC of foot-and-mouth disease virus as a diagnostic antigen in ELISA to 466

differentiate infected from vaccinated cattle. Arch. Virol. 142:2021-2033. 467

11. Mackay, D., M. Forsyth, P. Davies, A. Berlinzani, G. Belsham, M. Flint, and M. 468

Ryan. 1998. Differentiating infection from vaccination in foot-and-mouth disease using a 469

panel of recombinant, non-structural proteins in ELISA. Vaccine. 16:446-459. 470

12. Shen, F., P. Chen, A. Walfield, J. Ye, J. House, F. Brown, and C. Wang. 1999. 471

Differentiation of convalescent animals from those vaccinated against foot-and-mouth 472

disease by a peptide ELISA. Vaccine. 17:3039-3049. 473

13. Kweon, C. H., Y. J. Ko, W. I. Kim, S. Y. Lee, J. J. Nah, K. N. Lee, H. J. Sohn, K. 474

S. Choi, B. H. Hyun, and S. W. Kang. 2003. Development of a foot-and-mouth disease 475

NSP ELISA and its comparison with differential diagnostic methods. Vaccine. 21:1409-476

1414. 477

14. Srisombundit, V., N. Tungthumniyom, W. Linchongsubongkoch, C. 478

Lekcharoensuk, L. Sariya, P. Ramasoota, and P. Lekcharoensuk. 2013. Development 479

of an inactivated 3C< sup> pro</sup>-3ABC (mu3ABC) ELISA to differentiate cattle 480

infected with foot and mouth disease virus from vaccinated cattle. J. Virol. Methods. 481

188:161-167. 482

15. Mohapatra, J. K., L. K. Pandey, A. Sanyal, and B. Pattnaik. 2011. Recombinant 483

non-structural polyprotein 3AB-based serodiagnostic strategy for FMD surveillance in 484

bovines irrespective of vaccination. J. Virol. Methods. 177:184-192. 485

16. Bronsvoort, B. M., S. Parida, I. Handel, S. McFarland, L. Fleming, P. Hamblin, 486

and R. Kock. 2008. Serological survey for foot-and-mouth disease virus in wildlife in 487

eastern Africa and estimation of test parameters of a nonstructural protein enzyme-linked 488

immunosorbent assay for buffalo. Clin. Vaccine Immunol. 15:1003-1011. doi: 489

10.1128/CVI.00409-07 [doi]. 490

17. Sørensen, K., K. De Stricker, K. Dyrting, S. Grazioli, and B. Haas. 2005. 491

Differentiation of foot-and-mouth disease virus infected animals from vaccinated animals 492

using a blocking ELISA based on baculovirus expressed FMDV 3ABC antigen and a 493

3ABC monoclonal antibody. Arch. Virol. 150:805-814. 494

18. Rodriguez, A., J. Dopazo, J. Saiz, and F. Sobrino. 1994. Immunogenicity of non-495

structural proteins of foot-and-mouth disease virus: differences between infected and 496

vaccinated swine. Arch. Virol. 136:123-131. 497


http://cvi.asm.org/

Dow

nloaded from

http://cvi.asm.org/

22

19. Clavijo, A., M. Lin, J. Riva, M. Mallory, F. Lin, and E. Zhou. 2001. Development 498

of a competitive ELISA using a truncated E2 recombinant protein as antigen for detection 499

of antibodies to classical swine fever virus. Res. Vet. Sci. 70:1-7. 500

20. Hohlich, B. J., K. H. Wiesmuller, T. Schlapp, B. Haas, E. Pfaff, and A. 501

Saalmuller. 2003. Identification of foot-and-mouth disease virus-specific linear B-cell 502

epitopes to differentiate between infected and vaccinated cattle. J. Virol. 77:8633-8639. 503

21. Yang, M., A. Clavijo, R. Suarez-Banmann, and R. Avalo. 2007. Production and 504

characterization of two serotype independent monoclonal antibodies against foot-and-505

mouth disease virus. Vet. Immunol. Immunopathol. 115:126-134. doi: 506

10.1016/j.vetimm.2006.10.002. 507

22. Moniwa, M., C. Embury-Hyatt, Z. Zhang, K. Hole, A. Clavijo, J. Copps, and S. 508

Alexandersen. 2012. Experimental foot-and-mouth disease virus infection in white tailed 509

deer. J. Comp. Pathol. 147:330-342. doi: 10.1016/j.jcpa.2012.01.010 [doi]. 510

23. Parida, S., L. Fleming, D. Gibson, P. A. Hamblin, S. Grazioli, E. Brocchi, and D. 511

J. Paton. 2007. Bovine serum panel for evaluating foot-and-mouth disease virus 512

nonstructural protein antibody tests. J. Vet. Diagn. Invest. 19:539-544. 513

24. Cox, S. J., C. Voyce, S. Parida, S. M. Reid, P. A. Hamblin, D. J. Paton, and P. V. 514

Barnett. 2005. Protection against direct-contact challenge following emergency FMD 515

vaccination of cattle and the effect on virus excretion from the oropharynx. Vaccine. 516

23:1106-1113. doi: 10.1016/j.vaccine.2004.08.034. 517

25. Cox, S. J., C. Voyce, S. Parida, S. M. Reid, P. A. Hamblin, G. Hutchings, D. J. 518

Paton, and P. V. Barnett. 2006. Effect of emergency FMD vaccine antigen payload on 519

protection, sub-clinical infection and persistence following direct contact challenge of 520

cattle. Vaccine. 24:3184-3190. doi: S0264-410X(06)00064-8 [pii]. 521

26. Parida, S., S. Cox, S. Reid, P. Hamblin, P. Barnett, T. Inoue, J. Anderson, and D. 522

Paton. 2005. The application of new techniques to the improved detection of persistently 523

infected cattle after vaccination and contact exposure to foot-and-mouth disease. Vaccine. 524

23:5186-5195. 525

27. Parida, S., J. Anderson, S. J. Cox, P. V. Barnett, and D. J. Paton. 2006. Secretory 526

IgA as an indicator of oro-pharyngeal foot-and-mouth disease virus replication and as a 527

tool for post vaccination surveillance. Vaccine. 24:1107-1116. doi: 528

10.1016/j.vaccine.2005.09.006. 529

28. Jacobson, R. H. 1998. Validation of serological assays for diagnosis of infectious 530

diseases. Rev. Sci. Tech. 17:469-526. 531

29. Perkins, J., S. Parida, and A. Clavijo. 2007. Use of a standardized bovine serum 532

panel to evaluate a multiplexed nonstructural protein antibody assay for serological 533


http://cvi.asm.org/

Dow

nloaded from

http://cvi.asm.org/

23

surveillance of foot-and-mouth disease. Clin. Vaccine Immunol. 14:1472-1482. doi: 534

CVI.00227-07 [pii]. 535

30. Alexandersen, S., Z. Zhang, and A. I. Donaldson. 2002. Aspects of the persistence 536

of foot-and-mouth disease virus in animals—the carrier problem. Microb. Infect. 4:1099-537

1110. 538

31. Sharma, G. K., J. K. Mohapatra, L. K. Pandey, S. Mahajan, B. S. Mathapati, A. 539

Sanyal, and B. Pattnaik. 2012. Immunodiagnosis of foot-and-mouth disease using 540

mutated recombinant 3ABC polyprotein in a competitive ELISA. J. Virol. Methods. 541

185:52-60. doi: http://dx.doi.org/10.1016/j.jviromet.2012.05.029. 542

32. Uttenthal, Å., S. Parida, T. B. Rasmussen, D. J. Paton, B. Haas, and W. G. 543

Dundon. 2010. Strategies for differentiating infection in vaccinated animals (DIVA) for 544

foot-and-mouth disease, classical swine fever and avian influenza. Expert Review of 545

Vaccines. 9:73-87. 546

33. McCullough, K. C., J. R. Crowther, and R. N. Butcher. 1985. A liquid-phase 547

ELISA and its use in the identification of epitopes on foot-and-mouth disease virus 548

antigens. J. Virol. Methods. 11:329-338. 549

34. Van Regenmortel, M. 2001. Antigenicity and immunogenicity of synthetic peptides. 550

Biologicals. 29:209-213. 551

35. Yang, M., A. Clavijo, M. Li, K. Hole, H. Holland, H. Wang, and M. Y. Deng. 552

2007. Identification of a major antibody binding epitope in the non-structural protein 3D 553

of foot-and-mouth disease virus in cattle and the development of a monoclonal antibody 554

with diagnostic applications. J. Immunol. Methods. 321:174-181. doi: 555

10.1016/j.jim.2007.01.016. 556

36. Kitching, R. P. 2002. Identification of foot and mouth disease virus carrier and 557

subclinically infected animals and differentiation from vaccinated animals. Revue 558

Scientifique Et Technique-Office International Des Épizooties. 21:531-535. 559

37. Brocchi, E., I. Bergmann, A. Dekker, D. Paton, D. Sammin, M. Greiner, S. 560

Grazioli, F. De Simone, H. Yadin, and B. Haas. 2006. Comparative evaluation of six 561

ELISAs for the detection of antibodies to the non-structural proteins of foot-and-mouth 562

disease virus. Vaccine. 24:6966-6979. 563

38. Kittelberger, R., G. Mackereth, M. Sewell, J. Keall, R. Clough, C. Pigott, and J. 564

O'Keefe. 2008. Specificity of non-structural protein enzyme-linked immunosorbent 565

assays for the detection of serum antibodies against foot-and-mouth disease virus in a 566

target population in New Zealand. N. Z. Vet. J. 56:227-232. 567

39. Paton, D. J., K. de Clercq, M. Greiner, A. Dekker, E. Brocchi, I. Bergmann, D. J. 568

Sammin, S. Gubbins, and S. Parida. 2006. Application of non-structural protein 569


http://cvi.asm.org/

Dow

nloaded from

http://cvi.asm.org/

24

antibody tests in substantiating freedom from foot-and-mouth disease virus infection after 570

emergency vaccination of cattle. Vaccine. 24:6503-6512. 571

40. Parida, S., L. Fleming, Y. Oh, M. Mahapatra, P. Hamblin, J. Gloster, and D. J. 572

Paton. 2008. Emergency vaccination of sheep against foot-and-mouth disease: 573

significance and detection of subsequent sub-clinical infection. Vaccine. 26:3469-3479. 574

doi: 10.1016/j.vaccine.2008.04.026; 10.1016/j.vaccine.2008.04.026. 575

576


http://cvi.asm.org/

Dow

nloaded from

http://cvi.asm.org/


http://cvi.asm.org/

Dow

nloaded from

http://cvi.asm.org/

1 TABLE 1 Comparison of test results of FMDV NSP antibody evaluation panel (n = 36). All animals 2

(vaccinated and control group) were challenged and shown to be seropositive by VNT. Test results 3

except 3B cELISA have been previously reported by Parida et al (2007) 4

___________________________________________________________________________________ 5

Animal group 3B cELISA PrioCHECKa Bommedi

b UBI

c Svanova

d 6

Pos/total Pos/total Pos/total Pos/total Pos/total 7

___________________________________________________________________________________8

Vaccinated carrier 18/20 18/20 14/20 11/20 10/20 9

Vaccinated non-carrier 4/4 3/4 3/4 2/4 2/4 10

Control carrier 7/7 7/7 7/7 6/7 6/7 11

Control non-carrier 5/5 5/5 5/5 5/5 4/5 12

13

Total seropositive # 34/36 33/36 29/36 24/36 22/36 14

Total seropositive % 94% 92% 81% 67% 61% 15

___________________________________________________________________________________ 16

17 aPrioCHECK NS: Cedi Diagnostics B.V. Lelystad, The Netherlands. 18

bBommedi:CHEKIT-FMD-3ABC bo-ov, Bommeli Diagnostics, Liebefeld-Bern, Switzerland. 19

cUBI: UBI FMDV NS ELISA (CATTLE), United Biomedical Inc, Hauppauge, NY. 20

dSvanova: Svanovir FMDV 3ABC-Ab ELISA, Svanova Biotech AB, Uppsala, Sweden. 21

22

23

24

25

TABLE 2 Summary of the FMDV NSP specific antibody response from different animal 26

species inoculated with different serotypes of FMDV 27

___________________________________________________________________________________ 28

5 dpi 7 dpi 9 dpi 28 dpi 29

Pos # Pos % Pos # Pos % Pos # Pos % Pos # Pos % 30

___________________________________________________________________________________ 31

Porcine (n=9) 0 0% 1 12.5% 8 89% 8 89% 32

Sheep (n=18) 1 6% 14 82% 17 94% 18 100% 33

Cattle (n=15) 0 0% 8 53% 15 100% 15 100% 34

Deer (n=2) 0 0% 2 100% 2 100% 2 100% 35

___________________________________________________________________________________ 36

37


http://cvi.asm.org/

Dow

nloaded from

http://cvi.asm.org/

on may 7, 2020 by guest · 8 pirbright institute, ash road, pirbright, surr ey, uk b; comision...

Documents