Evaluation of a type 2 modified live porcine reproductive and respiratory syndrome vaccine against heterologous challenge of a lineage 3 highly virulent isolate in pigs

Virus and cells

The PRRSV field strain (TSYM-142575; GenBank: KY769953) used in the present study was originally isolated from a farm in Taiwan with severe post-weaning respiratory distress and continuously high mortalities between 2014 and 2017 (Hou et al., 2019). The previous study had demonstrated that TYSM-strain was a highly virulent strain and could induce severe clinical sign and high mortality in the field and experimental studies. Briefly, tissue samples from TSYM-infected pigs were homogenized, centrifuged, filtered and inoculated. The inocula was prepared by four-passage inoculation on pulmonary alveolar macrophages (PAM) for use in pig challenges and 10 further propagations on a MARC-145 cell line (ATCC CRL-12231) for neutralizing antibody and ELISpot assays in this study. The viral stocks were tittered for calculating the 50% tissue culture infective dose (TCID₅₀) and identified by reverse transcription polymerase chain reaction (RT-PCR) and gene sequencing. The TSYM-isolate was classified into lineage 3 of type II PRRSV by phylogenetic analysis and shared 85.7% ORF5 identity at the nucleotide level with the P129 vaccine strain (GenBank: AF494042). Meanwhile, both stocks were confirmed negative for pseudorabies virus (PRV), type II porcine circovirus (PCV2) and classical swine fever virus (CSFV) by molecular assays.

Animal experiment

Eighteen male, Landrace-Yorkshire pigs were introduced from a PRRSV, PRV, and CSFV-free pig farm at the age of 18 days. All pigs were negative for PRRSV under ELISA and real-time RT-PCR detections. The pigs were randomly divided into three groups (six pigs for each group) by using the Random function (EXCEL, Microsoft) and raised in separate rooms. The MLV pigs were administered Fostera PRRS MLV vaccines (LOT: 251726A; Zoetis) at 3 weeks of age (−28 days post-challenge; −28 DPC). Pigs in the unvaccinated (UnV) and mock control groups received saline as a placebo on the same day. Four weeks after the vaccination (0 DPC), the MLV and UnV groups were intra-nasally and intra-muscularly inoculated with 10⁶ TCID₅₀ of TSYM-strain PRRSV inocula. The mock pigs were administered a PAM culture medium instead. Blood samples were taken at −28, −21, −14, −7, 0, 4, 7, 10, 14 and 21 DPC for serological and virological analysis. During the experimental period, all clinical and laboratory personnel were blind to swine groups and samples. All pigs were then humanely euthanized by electrical stunning and exsanguination for pathological examination at 21 DPC. The experiment was conducted within the experimental house of the Graduate Institute of Veterinary Pathobiology according to animal ethical principles and the protocol was approved by the Institutional Animal Care and Use Committee of National Chung Hsing University (IACUC number: 107-045).

Clinical observations

Clinical signs were evaluated at the same time every morning by the same investigator. The body temperatures were simultaneously detected by using BioThermo LifeChips implants (Destron Fearing) and classified as fever when temperatures were above 40 °C. Clinical scores were evaluated for activity, ranging from 0 (normal) to 3 (severe), and respiratory distress, ranging from 0 (normal) to 6 (severe), following the guidelines of previous studies (Halbur et al., 1995; Jolie, Mulks & Thacker, 1995). Individual body weights were measured at −28, 0, 10 and 21 DPC for average daily weight gain (ADWG) calculations. The amounts of daily food intake (FI) and the food conversion rate (FCR) were also recorded for measuring the appetite and growth performance of the pigs.

Viremia and serological measurement

Serum samples were submitted for real-time RT-PCR for PRRSV quantification as previously described (Hou et al., 2019). The PRRSV-specific antibody was measured by using IDEXX PRRSX3 Ab test kits and strictly following the manufacturer’s instructions. The serum neutralizing antibody assay was performed according to previous methods (Chia et al., 2010; Yoon et al., 1994).

Pathological evaluation

Briefly, the macroscopic lesion score of lung affected by pneumonia is estimated as previous descriptions (Halbur et al., 1995). This method is based on assigning a number to each lobe to reflect the approximate percentage of the entire lung represented by that lobe. For microscopic measurement, sections were taken from all lobes of lung and histopathological lesions were scored as the previous study (Halbur et al., 1995). The severity is estimated based on the extent and magnitude of interstitial pneumonia: 0, no lesion; 1, mild/focal; 2, moderate/multifocal; 3, moderate/diffuse (alveolar wall accounting for greater than 50% of the measuring section); 4, severe/diffuse (alveolar wall accounting for greater than 75% of the measuring section). The severities of both macro- and microscopic lung lesions were scored by three pathologists under blind tests.

PRRSV-specific IFN-γ ELISpot assay

PRRSV-specific interferon-gamma (IFN-γ) responses were measured by using pre-coated porcine IFN-γ ELISpot plates (Mabtech) according to the manufacturer’s directions and previous studies (Park et al., 2014).

Statistical analysis

All statistical analysis was performed using IBM SPSS statistical software (version 20). Continuous data, including fever day, ADWG, viremia and serology, were verified for the normality (Shapira-Wilk test) and homogeneity of variance (Levene test) and measured for statistical significance by using ANOVA and Student’s t test. Post Hoc analysis was then done by Tukey’s test. For not normally distributed and categorial data, including body temperature, clinical scores, lung lesion scores and ELISpot, Kruskal-Wallis and Mann–Whitney tests were rather applied. Statistical significances were defined as a P value less than 0.05.

Animal exclusion and mortality

During the experimental period, one pig from the MLV group was excluded due to PRRSV-unrelated death. The pig showed acute clinical signs (including open-mouth breathing, vomiting, and cyanosis) after 5 min of administration with viral inoculum and died immediately. Pathological examination showed massive edema and hemorrhage in multiple viscera. These findings indicated that a severe anaphylactic reaction was induced by viral inoculum and the death was not caused by PRRSV infection.

One pig from the UnV group was humanely euthanized at 14 DPC due to severe depression, anorexia and dyspnea. Proliferative necrotizing pneumonia and non-suppurative encephalitis were noted during histopathological examination, which were frequently observed after challenge with virulent TSYM-isolate in our previous study (Hou et al., 2019). The PRRSV viremia titer reached 7.14 copies/µL and showed no decline until death.

As stated above, the final PRRSV-associated mortality rates of the mock, MLV and UnV groups were 0, 0 and 17%, respectively, in the present study.

Changes in body temperature

Following TSYM-strain challenge, the UnV group showed obviously increased body temperature at 1 DPC and presented persistent fever from 5 to 13 DPC. The MLV group showed slightly raised body temperatures after challenge and then gradually recovered from 4 DPC (Fig. 1A). Briefly, the level of fever in the MLV group was milder than that in the UnV group, and the average days of fever in the MLV group (3.0 ± 0.5) was also significantly lower than the UnV group (6.2 ± 0.5) (P < 0.001; Fig. 1B). The temperature values of the mock group remained steady during the challenge period.

Clinical signs and growth performance

After challenge, the UnV group showed high morbidity in terms of significant depression during 5–13 DPC, and signs of respiratory distress gradually developed from 8 DPC, with notable distress persisting until the end of the trial. In contrast, most of the pigs in the MLV group showed only mild depression after challenge, and the magnitude of respiratory distress was consistently less than the UnV group throughout this experiment (Fig. 2).

Prior to challenge, the average body weights were 16.6 ± 0.3, 17.8 ± 0.4, 17.4 ± 0.1 kg for the mock, UnV and MLV groups, respectively, and no statistically significant differences were apparent among these three groups. However, the ADWG of the UnV group was significantly decreased compared to that of the MLV and mock groups in both the acute and whole infection periods, as shown in Fig. 3. The average productivity of the UnV group was only 13% (98 g/day) of the mock group (763 g/day) in the acute phase. In contrast, a lesser degree of growth retardation was observed in the MLV group, with 80% preservation of the ADWG compared to the mock pigs. Similarly, the amounts of food intake and the food conversion ratio also revealed the same trends among the different groups.

Macroscopic and microscopic lung lesion scores

As shown in Fig. 4A, the mean macroscopic lesion areas of the UnV and MLV groups were 37.8 ± 11.3 and 19.6 ± 6.1%. The UnV group showed significantly more severe interstitial pneumonia (P<0.01) than those in the mock group, and the MLV group showed milder interstitial pneumonia compared to the UnV group (P = 0.12). Histopathologically, lung lesions characterized by alveolar septa thickening, which was caused by type II pneumocyte hypertrophy and hyperplasia and mononuclear inflammatory cells infiltration, were observed in both the UnV and MLV groups (Figs. 4D and 4E). The mean microscopic scores were 1.97 ± 0.19 and 1.36 ± 0.09 for the UnV and MLV groups, respectively, which indicates that the degree of interstitial pneumonia in MLV group was obviously milder than that in the UnV group (Fig. 4B). The differences in severity of microscopic interstitial pneumonia were significant between the UnV and mock group (P < 0.01), but not reach a significant level between the MLV and mock groups (P = 0.16). Both the macro- and microscopic lesion scores of the mock pigs showed no evidence of PRRSV-associated lung lesions (Fig. 4C).

PRRSV viremia quantification

Following vaccination, low viremia titers in the MLV group were detected at −14 DPC (14 days post-vaccination; 14 DPV; viral titer 1.87 ± 0.27 log₁₀ copies/µL) and 0 DPC (28 DPV; viral titer 0.87 ± 0.26 log₁₀ copies/µL), and the UnV and mock groups remained undetectable before challenge (Fig. 5). After challenge with the TSYM-strain, viremia was detected as early as 4 DPC in all challenge groups. At 4 DPC, the UnV pigs displayed high serum viral titers (6.84 ± 0.23 log₁₀ copies/µL) and were significantly higher (P < 0.05) than that of the MLV group. Meanwhile, the serum viral titer of the MLV group was 10¹ to 10^1.5 less than that of the UnV group at 7, 10 and 14 DPC (P = 0.18, 0.08 and 0.22, respectively). PRRSV nucleic acid was not detected in the serum of the mock pigs throughout this study.

Serological response

As shown in Fig. 6A, all pigs were negative for PRRSV-specific IgG antibodies at the time of vaccination (0 DPV; −28 DPC) and seroconversion was first detected at 14 DPV in the MLV group. All vaccinated pigs were seropositive for PRRSV-specific antibodies at 0 DPC and the MLV group had significantly higher (P < 0.05) antibody titers than those of the UnV group at 7 DPC. The mock group showed no PRRSV-specific IgG antibodies throughout this trial.

In the neutralizing antibody assay, no groups showed neutralizing effects in response to the challenge strain (less than 2).

Responses of PRRSV-specific IFN-γ secreting cells

The cellular response was evaluated in frequencies of IFN-γ SC in PBMC. Prior to challenge (0 DPC), the frequencies of PRRSV-specific IFN-γ SC in all groups were less than an average of 10 cells per 10⁶ PBMC. Upon challenge with the TSYM-strain, the frequency of PRRSV-specific IFN-γ SC reached an average of 13.8 ± 8.9 and 26.0 ± 9.5 cells per 10⁶ PBMC at 21 DPC in the UnV and MLV groups, respectively (Fig. 6B). The result of Kruskal-Wallis test indicated that the P value was 0.08, and Post Hoc analysis revealed a moderate difference between the mock and MLV group at the time of 21 DPC (P = 0.07).