Detection of antibodies to structural proteins of foot-and-mouth disease virus in swine meat juice

Abstract

Established test methods for detecting foot-and-mouth disease virus (FMDV) rely on sample collection from live animals. However, circumstances exist in which it is not possible to collect the desired samples. Meat juice has been explored as an alternative for the detection of FMDV and has previously proven successful by real-time reverse transcription polymerase chain reaction and lateral flow strip test. Meat juice has not yet been assessed for the detection of antibodies to FMDV. This study, therefore, evaluated meat juice for the detection of antibodies to structural proteins by existing serotype-specific solid phase competitive enzyme-linked immunosorbent assays. Antibodies to FMDV structural proteins were detected in meat juice from experimentally infected pigs beginning 6- or 7-days post-infection (DPI) and continued until 21 to 28 DPI. Sera were tested in tandem and followed similar antibody detection patterns. The results show that meat juice can be used for detection of anti- FMDV structural protein antibodies.

Global consumption of meat and meat products increases every year as much of the world becomes more developed (1). Meat has become a prominent part of the average modern-day diet as poverty levels decrease (1). Animal trade between countries has become commonplace to satisfy the increased consumption of meat (2). To supply the rising demand, increased animal production and improved growth efficiency methods have been employed (3). However, surges in animal rearing pose a greater risk for infectious disease outbreaks. Among cloven-hoofed animals, foot-and-mouth disease (FMD) is an infectious disease that can cause reduced appetite, fever, vesicular lesions, and death in infected animals (4). The disease is caused by the FMD virus (FMDV) (4). Foot-and-mouth disease outbreaks can decimate agricultural industries and severely affect supply chains (5).

The methods for testing animals suspected of FMDV recommended by the World Organization for Animal Health include real-time reverse transcription polymerase chain reactions (rRT-PCRs), antigen enzyme-linked immunosorbent assays (Ag ELISAs), and virus isolation. These tests typically use blood, vesicular fluid, milk, epithelial tags, or swabs when they are available. In certain situations, desirable sample types may not be accessible.

The importation of animal products is one situation that presents a risk, as FMDV is known to remain viable for more than 120 d in tissues such as lymph nodes and bone marrow if they are kept at 4°C (6). The fast-paced nature of the meat industry can result in the detection of FMD outbreaks after animals have already been slaughtered. In these unique circumstances, meat juice as a sample type can minimize disruption at abattoirs while maintaining traceability. Meat juice has been proven to be an effective alternative for the detection of FMDV by rRT-PCR (7). The effectiveness of meat juice for detection of FMDV by other test methods such as ELISAs has not yet been explored.

In this study, we report the successful detection of antibodies to FMDV structural proteins in meat juice using an adapted diagnostic serotype specific solid-phase competitive ELISA (SPCE).

Swine meat juice and sera were obtained from pigs experimentally inoculated with FMDV serotype A (A22 IRQ 24/64), serotype Southern African Territories 2 (SAT2 ZIM 5/81), and serotype ASIA1 Shamir in 3 separate studies. The viruses were obtained from the World Reference Laboratory for FMD at The Pirbright Institute and propagated at the National Centre for Foreign Animal Disease (NCFAD), as previously described (7). The experimental procedures, sample collection, processing, and storage have been previously described (7). In addition, 48 meat juice samples from naïve pigs were tested to establish cut-off values for each SPCE.

Sera and meat juice samples were tested using an SPCE developed at the NCFAD as previously described (8), with minor modifications. Briefly, 96-well Nunc MaxiSorp micro-plates (439454; ThermoFisher Scientific, Waltham, Massachusetts, USA) were coated with 100 μL per well of an optimal dilution of rabbit anti-FMDV serum specific for each serotype A, SAT2, or ASIA1 in coating buffer (0.06 M carbonate buffer, pH 9.6) overnight at 4°C. The plates were then washed × 5 using 0.01 M phosphate buffered saline (PBS; pH 7.2) with 0.05% Tween detergent (PBST) followed by the addition of 100 μL per well of antigen for each FMDV serotype optimally diluted in blocking buffer (5% normal rabbit serum, 10% normal bovine serum in PBST) and incubated at 37°C on an orbital incubator shaker. After 1 h, the plates were again washed × 5 with PBST. Duplicate wells of 50 μL per well of each serum sample (diluted 1:5 in blocking buffer) or meat juice sample (diluted 1:2.5 in blocking buffer) or blocking buffer alone (diluent controls) were added to specific wells, immediately followed by the addition of 50 μL per well of guinea pig anti-FMDV serum diluted in blocking buffer to all wells. Plates were incubated for 1 h at 37°C on an orbital incubator shaker, washed × 5 with PBST, and followed with 100 μL per well of polyclonal donkey anti-guinea pig IgG (H&L) conjugated to horseradish peroxidase (706-035-148; Jackson ImmunoResearch Laboratories, West Grove, Pennsylvania, USA) diluted in blocking buffer. Plates were again incubated for 1 h at 37°C on an orbital incubator shaker. After washing × 5 with PBST, 100 μL of 2-component 3,3′,5,5′-tetramethylbenzidine (TMB) substrate (KP-50-76-03; Mandel Scientific, Guelph, Ontario) was dispensed into each well and incubated for 10 min at room temperature in the dark with shaking. The TMB color change was stopped with 100 μL of TMB stop solution (KP-50-85-06; Mandel Scientific) and the optical density (OD450) of each well determined on a SpectraMax Plus 384 Microplate Reader (Molecular Devices, San Jose, California, USA). The result for each sample was expressed as a percent inhibition:

$$\begin{array}{l}\text{Percent\hspace{0.17em}inhibition}=100-(\text{mean\hspace{0.17em}OD\hspace{0.17em}of\hspace{0.17em}}2\hspace{0.17em}\text{replicate\hspace{0.17em}test}\\ \text{samples}÷\text{mean\hspace{0.17em}OD\hspace{0.17em}of\hspace{0.17em}the\hspace{0.17em}}2\hspace{0.17em}\text{replicate\hspace{0.17em}diluent\hspace{0.17em}controls})\times 100\end{array}$$

Meat juice samples were also tested by the previously developed NCFAD 3ABC competitive ELISA (9). The assay failed to detect antibodies to FMDV non-structural proteins in meat juice.

Statistical analysis was performed using GraphPad Prism version 8.4.2 (San Diego, California, USA). Receiver-operating characteristic (ROC) curve analysis was used to estimate sensitivities and specificities at various cut-off percent inhibitions. Multiple comparisons of meat juice samples from various days post-infection (DPI) were also performed by 2-way analysis of variance (ANOVA) (Tukey’s multiple comparison test). Correlation analysis of antibody detection in serum and meat juice was performed to determine Pearson correlation coefficients.

Using both ROC curve analysis, mean + 3 standard deviation and the distribution of the percent inhibition for 40 negative meat juice samples (Figures 1 A, B, C), the cut-off percent inhibition value was 30 for SPCE A, SAT2, and ASIA1. Based on the cut-off for meat juice, diagnostic specificity was 100% for all 3 serotypes tested. The determined cut-off also minimized cross-reactivity between FMDV serotypes. The cut-off percent inhibition for serum SPCE has previously been established as 50 (8).

Figure 1.

Relative levels of antibodies to the structural proteins of foot-and-mouth disease virus in swine meat juice from the biceps femoris. Antibody levels are expressed as percent inhibition of negative and positive meat juice samples from (A) A22 IRQ 24/64, (B) SAT2 ZIM 5/81, and (C) ASIA1 Shamir experimentally infected pigs. Cut-off percent inhibition was set to 30 based on receiver operating characteristic curve analysis of meat juice samples.

Positive levels of anti-FMDV structural protein antibodies were detected in meat juice at DPI 12 and remained relatively high at DPI 21 for the A22 IRQ 24/64 experiment (Figure 2 A). The rise in antibody response at DPI 12 and 21 were significant compared to DPI 1 (P < 0.0001). There was a correlation between anti-FMDV structural proteins antibody detection in meat juice and sera from corresponding animals (r² = 0.77; P < 0.0001) (Figure 3 A).

Figure 2.

Percent Inhibition values showing the level of antibodies to structural proteins of foot-and-mouth disease virus in meat juice collected from the biceps femoris and serum from pigs experimentally infected with (A) A22 IRQ 24/64, (B) SAT2 ZIM 5/81, and (C) ASIA1 Shamir. The cut-off percent inhibition value for meat juice is 30. The cut-off value for serum is 50.

Figure 3.

Correlation in antibody response to foot-and-mouth disease virus (FMDV) structural proteins in meat juice and serum. Antibodies to structural proteins of FMDV in meat juice collected from the biceps femoris and serum from the same animals were determined by serotype specific competitive enzyme-linked immunosorbent assay. The Pearson correlation coefficients between antibody detection in meat juice and serum for pigs experimentally infected with (A) A22 IRQ 24/64, (B) SAT2 ZIM 5/81, and (C) ASIA1 Shamir were then determined with GraphPad Prism.

For the SAT2 experiment, positive levels of anti-FMDV structural proteins antibodies were detected in one of the meat juice samples at DPI 7, 100% at DPI 14 and 21, and 75% at DPI 28 (Figure 2 B). There was a significant difference in antibody response at DPI 14 and later time points compared to DPI 1 (P < 0.05). Similarly, there was a correlation between anti-FMDV SAT2 structural protein antibody detection in meat juice and sera from corresponding animals (r² = 0.84; P < 0.0001) (Figure 3 B).

For the ASIA1 experiment, anti-FMDV structural protein antibodies were detected in all samples at DPI 14 (Figure 2 C), with a statistically significant difference at DPI 14 relative to DPI 2 (P < 0.0014). On the contrary, 75% of sera from corresponding animals were positive at DPI 6 as opposed to 0% for meat juice at this time point. Nevertheless, there was a correlation between anti-FMDV ASIA1 structural protein antibody detection in meat juice and sera from corresponding animals (r² = 0.64; P < 0.0002) (Figure 3 C).

No cross-reactivity was observed between FMDV A antibody positive meat juice when tested on the FMDV SAT2 and ASIA1 SPCE. Similarly, FMDV SAT2 antibody-positive meat juice did not cross-react on the respective FMDV A and ASIA1 SPCE. However, 50% of FMDV ASIA1 antibody positive meat juice cross-reacted on both FMDV A and SAT2 SPCE. The same samples on both tests cross-reacted similarly (Table I). In both cases, the percent inhibitions for FMDV ASIA1 were higher than the cross-reacting percent inhibitions on FMDV A and SAT2. Due to the lack of meat juice samples, we were unable to assess the analytical specificity of the assays against other vesicular disease agents, such as SVDV, SVA, and VSV.

Table I.

Cross-reactivity between serotype specific competitive enzyme-linked immunosorbent assays (SPCE) for foot-and-mouth disease virus (FMDV) serotypes A, ASIA1, and SAT2.

	Percent inhibition (%)
	Pig + DPI	A22 SPCE	SAT2 SPCE	ASIA1 SPCE
A22 meat juice	169-12	71.56	26.43	25.85
	199-12	71.99	26.44	24.22
	174-21	65.21	25.36	11.86
	181-21	63.66	26.15	9.41
SAT2 meat juice	211-14	9.61	53.49	−3.50
	203-21	8.47	44.67	−2.14
	218-14	−5.26	28.04	−19.58
	236-14	−5.25	31.07	−19.37
ASIA1 meat juice	253-11	24.14	9.61	39.59
	246-14	42.52	33.38	58.97
	258-14	24.09	25.37	48.40
	259-14	46.72	40.81	53.35

Percent Inhibition values showing the level of antibodies to structural proteins of FMDV in meat juice collected from the biceps femoris of pigs experimentally infected with A22 IRQ 24/64, SAT2 ZIM 5/81, and ASIA1 Shamir. Each MJ sample was tested in the homologous and heterologous SPCE to determine cross-reactivity. The cut-off percent inhibition value for meat juice is 30. DPI — days post-infection.

We have previously shown meat juice to be a good sample matrix for detection of FMDV antigen and nucleic acids. This means meat juice can be used in parallel with preferred sample types but will be most useful in situations when the preferred samples are not available. We have now shown that antibodies to FMDV structural proteins can be detected in meat juice concurrent with detection in serum. Antibodies to FMDV structural proteins begin circulating in serum of most infected animals at 5 to 7 DPI (10). Since meat juice is a mixture of serum and other cellular exudates (11), presence of antibodies in serum should result in antibodies in meat juice as shown by our data for DPI 11 or later. These data are in agreement with published data for the detection of antibodies to hepatitis E virus (12), classical swine fever virus (13,14), African swine fever virus (15), Japanese encephalitis virus (16), porcine circovirus type 2 (17), and Aujeszky’s disease virus (18). The level of antibodies to FMDV structural proteins in meat juice was also lower than in serum. The probable reason is that any serum in meat juice is diluted by other liquid components contained in myocytes. This can also explain the detection of positive antibody levels in serum, but not in some meat juice samples at DPI 6 and 7. It has also been shown previously that antibody levels in meat juice are significantly lower than in serum (13,19). Sera from FMDV-infected animals usually permit for diagnosis of ongoing infection through viral nucleic acid and resolved infection through the detection of antibodies (10). As revealed by our previous (7) and current data, meat juice provides similar diagnostic opportunities as serum with the additional benefit that FMDV RNA is detected in MJ for a longer period than in serum. This allows for a longer overlapping period (DPI 6 or 7 to 21), during which both FMDV nucleic acids and antibodies to FMDV structural proteins can be detected in the same meat juice sample. Foot-and-mouth disease virus detection in serum is usually shortlived, with viremia disappearing before DPI 7 (7).

Furthermore, the detection of FMDV structural protein antibodies in meat juice implies meat juice cannot be used for differentiation of infected and vaccinated animals at this point since detection of antibodies to FMDV non-structural proteins in meat juice was unsuccessful. Further assay optimization may be required to determine the potential use of meat juice for the detection of antibodies to FMDV non-structural proteins. Complement proteins in meat juice are potentially interfering with non-structural proteins ELISAs. Serum is usually heat inactivated to prevent interference by complement proteins. However, meat juice cannot be heat-inactivated due to the high protein content that results in a coagulated matrix not suitable for an ELISA.

In conclusion, swine meat juice is a suitable sample for the detection of antibodies to FMDV structural proteins. Taken together with our previous detection of FMDV RNA in meat juice, this sample type is valuable for the diagnosis of FMD during the acute and convalescent phases of the disease. Therefore, in the absence of traditional samples, such as in a retroactive disease investigation or for imported meat, meat juice can be used for the diagnosis of FMD when necessary.