Abstract
Endemic and epidemic outbreaks of porcine reproductive and respiratory syndrome virus (PRRSV) are causing large economic losses in commercial pig production worldwide. Given the complexity of controlling this disease with vaccines or other biosecurity measures, the selection of pigs with a natural resilience to this infection has been proposed as an alternative approach. In this context, we previously reported a vaccine-based protocol to classify 6-wk-old female piglets from one farm into resilient and susceptible phenotypes. Subsequent analysis showed that resilient sows had fewer lost piglets during a PRRSV epidemic. In the present study, we validated the results in four additional farms by showing a robust effect on the percentage of piglets lost (P < 0.05). We were able to associate the resilient phenotype with a 2% to 4% reduction in piglet losses on sow farms in both endemic and endemic/epidemic situations. Also consistent with previous results, susceptible sows delivered on average, almost 0.5 more piglets born per parity (P < 0.05). However, we show here that resilient sows have a longer stayability in the farm (+57 d; P < 0.05) and +0.3 more successful parities (P < 0.05), which balances the total number of piglets born and born alive in the full productive life of the sow between the two groups. Resilient sows thus contribute towards a more sustainable production system, reducing sow replacement and piglet mortality. The validation of this protocol on four independent production farms paves the way for the study of the genetic variation underlying the resilient/susceptible classification, with a view to incorporating this information into selection programs in the future.
Keywords: phenotyping, porcine reproductive and respiratory syndrome virus, production tradeoff, reproductive failure, resilience, sow
The response to an attenuated PRRSV vaccine can be used to phenotype female piglets for PRRSV resilience. This not only improves sow productivity during PRRSV outbreaks but also increases their farm stayability.
Introduction
Porcine reproductive and respiratory syndrome virus (PRRSV) can cause late-term abortions, prolonged anestrus, an increase in stillborn and mummified piglets, coughing, and respiratory problems in sows, while respiratory symptoms and reduced growth performance are often observed in young pigs (Lunney and Chen, 2010). Its high prevalence in intensive pig production areas and its impact on the reproductive and productive performance of the pig population (Fraile et al., 2010; Montaner-Tarbes et al., 2019) places this virus among the most economically important diseases affecting commercial pig production in North America, Europe, and Asia (Nathues et al., 2017; Zhang et al., 2022). In addition, this disease can also affect the ability of sows to remain on the farm with negative implications for welfare and sustainability due to an increase in the number of prematurely culled sows and, consequently, an increase in their replacement rate (Renken et al., 2021).
In sows, reproduction is influenced by both infectious and noninfectious factors (environment, nutrition, and management) and by factors intrinsic to the sow, including genetics. A resilient sow is one that is able to maintain reproductive performance in the face of various on-farm challenges. A common challenge in sow farms is PRRSV infection. Epidemiological models applied to viral infections have shown that selection for resilience should reduce both the likelihood and the impact of epidemics (Doeschl-Wilson and Kyriazakis, 2012). It has also been described that there is a large variation in the reproductive performance among sows in response to PRRSV infection, thereby suggesting that this trait could be responsive to selection (Rashidi et al., 2014; Serão et al., 2014; Harding et al., 2017). Therefore, a key step in the development of selection programs for disease resilience is the identification of resilient phenotypes.
In a previous publication (Abella et al., 2019), we reported that PRRSV-naïve female piglets vaccinated with a PRRSV-modified live vaccine could be classified as resilient to a future reproductive challenge with a field PRRSV strain if their serum was negative for PRRSV at 7- and 21-d post vaccination (DPV), and as susceptible if either sample was positive. Subsequently, we were able to demonstrate an association between the response to PRRSV exposure at a very early age in the rearing period and their future reproductive outcome in PRRSV-positive farms (Abella et al., 2019). In addition, we also tested how this classification might affect the overall productivity of the farm (Abella et al., 2021). So far, we have only tested this classification protocol on one large farm. The aim of this research work is to validate the phenotyping method in multiple farms and assess its potential to enhance the sustainability and profitability of pig production systems. Furthermore, the identification and selection of resilient sows may result in potential tradeoffs due to the potential genetic antagonism between production and health status (Prunier et al., 2010).
Materials and Methods
The experimental procedures were conducted in accordance with authorization 7,700 issued by the Catalan Department of Agriculture, Livestock, Fisheries and Food (Section of biodiversity and hunting) after being approved by the Ethics Committee for Animal Experimentation of the University of Lleida.
Farms
This study included two PRRSV-negative multiplication farms with 350 and 950 Landrace sows and four PRRSV-positive production farms with 1,120 to 1,700 Landrace × Large White sows. In both farms, the Large White Sire line was from Hendrix Genetics. All farms utilized a weekly farrowing batch system with all-in/all-out management system and were owned by two large Spanish integration companies (Pinsos del Segre S.A, Lleida, Spain and Vall Companys S.A, Lleida, Spain). One of the multiplication farms was a farrow-to-finish farm, and the rest of the farms were farrow-to-wean with the nursery phase located on-site. The piglets of these farms were never mixed with piglets from other pig production flows. The routine vaccination program included sow immunization against swine parvovirus, Aujeszky disease virus, swine influenza virus, Erysipelothrix rhusiopathiae, Escherichia coli, and Clostridium perfringens in all the included farms. Additionally, piglets were vaccinated against Mycoplasma hyopneumoniae and porcine circovirus type 2 (PCV2) at weaning (three weeks of age). PRRSV-modified live vaccines were not used either in the multiplication or in the production farms during the study period. An MLV PRRSV vaccine was only used for phenotyping the animals as thoroughly described later. No major signs of pig disease were present during the experiment, except for PRRSV during an outbreak period in two of the production farms. PRRSV diagnosis is explained in the next section.
Monitoring of PRRSV status on farms
PRRSV status for these farms was monitored throughout the trial following internationally accepted recommendations (Holtkamp et al., 2011) and previous publications from our research group (Abella et al., 2019). Herd classification for PRRSV was based on determining both the shedding and exposure status of the herd. Methods for testing shedding and exposure to the virus include direct detection by quantitative reverse transcriptase PCR (qRT-PCR) and antibody testing, respectively. The shedding status of the PRRS virus was classified as either negative or positive. A negative shedding status indicated that all the serum samples tested by qRT-PCR were negative, indicating an absence of viral shedding in the herd. Conversely, a positive shedding status indicated that at least one serum sample tested positive by qRT-PCR, providing evidence of viral shedding and transmission within the herd. The exposure status was also classified as either negative or positive. Negative exposure status indicated that there was absence of antibodies to the virus in the tested samples, while positive exposure indicated the presence of antibodies to the virus. Finally, the health status of the farm was classified as endemic or epidemic based on the absence or presence of overt reproductive problems in the sow farm, respectively. These problems (epidemic situation) were determined by a significant increase in abortions and/or lost piglets (stillborns and mummified) compared to the baseline situation (endemic situation) as previously published by our research group (Fraile et al., 2020). Mummified fetuses were established following the definition by Ladinig et al. (2014) for mummified fetuses. Other viral (PCV2, swine parvovirus, and swine influenza virus) and bacterial (swine leptospirosis and skin diamond disease) diseases were ruled out as causes of overt reproductive problems.
After detecting viral shedding in the production farm, the samples were sent to a laboratory specialized in PRRSV epidemiology (Grup de Sanejament Porci, Lleida, Spain). In this laboratory, the open reading frame 5 region (ORF5) of positive serum samples were sequenced using Sanger technology as previously described (Guzmán et al., 2021). All sequences were aligned using the Multiple Sequence Comparison by Log-Expectation (MUSCLE) algorithm in Geneious 10.0.7 (Biomatters, Ltd., Auckland, New Zealand) with default settings to compare and highlight variations across the sequences. A similarity matrix with all the sequences collected was built and compared with the database of sequences of the laboratory that included the vaccine used for the phenotyping method. Original ORF5 sequences from the farm specimens are available from the authors upon request.
Phenotyping for PRRSV resilience
At birth, 517 and 942 Landrace × Large White female piglets from each multiplication farm were ear tagged. Piglets were classified as resilient (R) or susceptible (S) to PRRSV according to a previously published method (Abella et al., 2019). Thus, piglets were vaccinated at 6 to 7 wk of age with a modified live PRRSV vaccine (MLV-PRRSV—Porcilis PRRS) according to manufacturer’s recommendations (2 mL by intramuscular dose; that is equivalent to approximately 105 TCID50 of PRRSV DV strain per animal). Blood samples were taken at 0, 7, and 21 d post-vaccination (DPV) and collected in tubes (Vacutainer, Betson Dickinson Ltd) to obtain serum. The vaccination procedure was carried out on another farm to avoid transmission of the PRRSV vaccine strain to the sow farm. PRRSV viremia was determined using a semi-quantitative TaqMan PCR as previously published (Abella et al., 2016). The assay results were reported as positive or negative depending on the cycle threshold (Ct) value (Ct < 40 is positive). If the piglet was reported as positive at 7 DPV, it was not tested again at 21 DPV. In line with the previous results in Abella et al. (Abella et al., 2016), a piglet was phenotyped as resilient (R) if its serum was negative for PRRSV according to the PCR outcome at 7 and 21 DPV. Conversely, a piglet was classified as susceptible (S) if one of the samples was positive at 7 or 21 DPV.
Production data recording
After being phenotyped as R or S, at 7 mo of age (130 kg BW, standard deviation (SD) 15 kg), 772 gilts were subsequently moved to the production farms following the standard operating procedures in place at both companies. These females represent sows retained for production after natural mortality during fattening and regular culling for lameness, leg conformation, number of functional teats, and other causes such as umbilical or inguinal hernias. On arrival, the gilts were first allocated to the quarantine unit of the production farms, where they were vaccinated against swine parvovirus, Aujeszky disease virus, swine influenza virus, Erysipelothrix rhusiopathiae, PCV2 and Mycoplasma hyopneumoniae. Once in the reproduction unit, they were artificially inseminated to meet weekly production goals. Sows were culled if they returned to estrus more than twice, suffered chronic lameness, rectal or vaginal prolapse, or showed a body condition of <2 (on a scale of 1 to 4). The farm staff were not aware of the phenotype (R or S) of the sows throughout the whole experiment. The farrowing date and the number of piglets born alive (NBA), stillborn (NSB), and mummified (NMU) per litter were then recorded for almost 3 yr in all the production farms. The total number of piglets lost per litter (NLP) was calculated as the sum of NSB and NMU. NSB and NMU were combined into a single trait to exclude misdiagnosis between them, which cannot be ruled out in non-experimental farm recording schemes. In addition, the total number of piglets born per litter (NTB) was calculated by summing NBA and NLP. All sows produced at least one litter. For each sow, the number of successful parities was recorded. Moreover, the length of productive life (taken as a proxy of the stayability of the sow in the farm) was calculated as the difference between the date of the last and the first parity, in days.
Statistical analysis
A descriptive statistic has been carried out for the number of piglets born alive (NBA), stillborn (NSB), mummified (NMU), lost (NLP = NSB + NMU), total born (NTB = NBA + NLP), total number of NBA and NTB during the productive life of the sow (NBA_life and NTB_life, respectively), the number of successful parities and the stayability of the sow at farm level and merging all the farms together. Data was also pooled according to the global PRRSV status of the farms (endemic vs endemic/epidemic). All the analyses were performed using JMP Pro, Version 17 (SAS Institute Inc., Cary, NC).
Univariate analysis
A univariate analysis was performed with all explanatory variables (farm, resilience criteria [R/S], parity, and global PRRSV status [endemic vs endemic/epidemic]) to identify the set of variables associated with %NLP/NTB, NTB, NBA, NBA_life, NTB_life, number of successful parities and sow stayability. On the other hand, sow was included as a random variable to account for repeated measures per animal. Explanatory variables with P-values ≤0.20 (calculated using an F-test for fixed effects and a Wald-test for random effects) were considered for inclusion in the multivariable model (Dohoo et al., 2012). Before including the variables in the multivariable model, explanatory variables were tested for correlation using Spearman rank correlation to avoid multicollinearity in the multivariable model for continuous variables. Collinearity was considered when the correlation coefficient was >0.50. In the case of nominal variables, a Chi-square test was also carried out to test for collinearity between them.
Multivariable model
A stepwise forward selection was performed as described by Dohoo et al. (2012) to build the multivariable model. In summary, all variables with a P-value ≤ 0.20 in the univariate analysis were included in the multivariable model. The variables with P-values >0.05 were excluded one at a time until obtaining the final model, with all variables having P-values ≤0.05. The order of inclusion in the model followed the value of the P-value from the lowest to the highest. In addition, Akaike’s Information Criterion (AIC) value was used to compare the goodness of fit of the model when excluding the variables by targeting the lowest value. Moreover, a scatterplot of residuals vs predicted values was carried out to check for homoscedasticity. Multicollinearity was tested on the final set of predictors included in the multivariable model by measuring the variance inflation factor (VIF ≤ 5). In addition, all predictors not included in the initial multivariable model because they had a P-value higher than the cutoff (P-value >0.20) in the univariate analyses were included again, one at a time, back into the final multivariable model, to test whether these variables remained nonsignificant in the presence of potential confounders (Dohoo et al., 2012). Potential interactions between variables retained in the final model were also included to test for them. Finally, the Tukey-Kramer test was used to perform a pairwise comparison between least squares means (LS-Means) of fixed effects.
Results
Results of the resilience criteria
PRRSV was never detected by qRT-PCR in the serum of 331 out of 1,459 female piglets (22.7%) from both multiplication farms after vaccination with an MLV-PRRSV vaccine. In contrast, the virus was detected in 77.3% of the piglets of both multiplication farms at 7 DPV (956 piglets) and 21 DPV (172 additional piglets, negative at 7 DPV but positive at 21 DPV). Gilts from multiplication farm 1 were allocated to production farm A and gilts from multiplication farm 2 were allocated to production farm B, C, and D (Table 1) according to the standard operating procedures for both pig production companies.
Table 1.
Distribution of the phenotyped sows (resilient and susceptible) on production farms and farm PRRSV status for exposure and shedding throughout the trial
| Production farm | Number of resilient sows | Number of susceptible sows | PRRSV status for shedding | PRRSV status for exposure | Global PRRSV status |
|---|---|---|---|---|---|
| A | 135 | 247 | Positive | Positive | Endemic/epidemic |
| B | 47 | 79 | Positive | Positive | Endemic/epidemic |
| C | 48 | 49 | Negative | Positive | Endemic |
| D | 9 | 158 | Negative | Positive | Endemic |
PRRSV, porcine reproductive and respiratory syndrome virus.
PRRSV status in the farms
The multiplication farms (1 and 2) stayed PRRSV negative throughout the trial. However, PRRS virus shedding status was classified as positive in two of the production farms (A and B) which remained PRRSV positive during 16 and 20 wk (duration of the outbreak period), respectively. As previously described, a positive shedding status meant that at least one serum sample tested positive by qRT-PCR (evidence of virus shedding and transmission in the herd) and, during this period, overt reproductive problems were observed (epidemic situation) based on a significant increase in abortions and/or lost piglets (stillborn and mummified piglets) compared to the baseline situation (endemic situation). Thus, farms A and B were classified as epidemic or endemic for PRRSV depending on the clinical signs compatible or not with PRRSV infection and on the laboratory results (positive or negative for shedding) throughout the trial (Table 1). On the other hand, production farms C and D were classified as negative in terms of virus shedding (no viremia detected by qRT-PCR) but positive in exposure status to PRRSV (presence of antibodies to the virus even in the absence of overt clinical problems). Thus, farms C and D were classified as endemic to PRRSV throughout the study (Table 1). Finally, the PRRSV strain used in the phenotyping method (Porcilis PRRS) had 85 and 88% homology, based on ORF5 sequence, with the field PRRSV strains detected in the production farms A and B, respectively, during the PRRSV outbreak. Figure 1 shows the genetic distance, based on ORF5 sequence, between the field PRRSV-1 strains, PRRSV-1 modified live vaccines, and a PRRSV-2 reference strain. Moreover, farms A and B were infected with a different field PRRSV strain as shown in Figure 1.
Figure 1.
Phylogenetic tree of the porcine reproductive and respiratory syndrome virus (PRRSV) open reading frame 5 regions (ORF5) of the field strains obtained from farm A and B during the PRRSV outbreak, five modified live PRRSV-1 vaccines (Porcilis PRRS—NCBI: KJ127878.1, Unistrain PRRS—NCBI: GU067771.1, Ingelvac PRRS—NCBI: KT988004.1, Pyrsvac-183—NCBI: DQ345726.1, Suvaxyn PRRS—NCBI: MK876228.1) and one PRRSV-1 (Lelystad—NCBI: NC_043487.1) and PRRSV-2 (VR2332 - NCBI: AY150564.1) reference strains available in GenBank (in brackets its reference for each strain above). The phylogenetic tree was calculated using the Geneious 10.0.7 program (Biomatters, Ltd., Auckland, New Zealand) by Multiple Sequence Comparison by Log-Expectation (MUSCLE) algorithm. The bootstrap analysis was set to 1,000 replicates. The phylogenetic tree is rooted in the PRRSV-2 reference strain. The scale bar indicates the nucleotide substitutions expressed on a unit basis.
Descriptive statistics for production data
Table 2 shows the mean and SD for the number of piglets born alive (NBA), stillborn (NSB), mummified (NMU), lost (NLP = NSB + NMU), total born (NTB = NBA + NLP), %NLP/NTB, the number of successful parities, the total number of NBA and NTB during the productive life of the sows (NBA_life and NTB_life, respectively) and the stayability of the sow at farm level (A, B, C, and D). Moreover, data from all farms were merged, and also distributed into farms with (endemic/epidemic-AB) and without (endemic-CD) PRRSV outbreaks. The coefficient of variation is extraordinarily variable between the parameters studied. For example, it is <30% for NTB and NBA; <60% for stayability days, number of successful parities, NBA_life and NTB_life; close to 100% for NSB, NLP, and %NLP/NTB and well over 100% for NMU. Finally, the %NLP/NTB was 11.4 (SD 0.4)% in the endemic situation compared to 16.8 (SD 0.8)% in the epidemic situation when all farms are merged (A, B, C, and D) and 11.7 (SD 0.5)% vs 16.7 (SD 0.8)% when only the two farms with PRRSV outbreaks are combined (AB).
Table 2.
Mean (standard deviation, SD) for the number of piglets born alive (NBA), stillborn (NSB), mummified (NMU), lost (NLP = NSB + NMU), total born (NTB = NBA + NLP), %NLP/NTB, the number of successful parities (Parities), total number of NBA and NTB during the productive life of the sow(NBA_life and NTB_life, respectively) and the stayability of the sow at farm level (A, B, C, and D), merging all the farms together (merged) divided according to the presence or absence of PRRSV outbreaks (AB and CD), respectively
| N of litters | NTB | NBA | NSB | NMU | NLP | %NLP/NTB | Stayability (days) | Parities | NBA_life | NTB_life | |
|---|---|---|---|---|---|---|---|---|---|---|---|
| All farms merged | 3,428 | 14.6 (3.6) | 12.8 (3.4) | 1.6 (2.0) | 0.2 (0.7) | 1.8 (2.2) | 11.4 (13.5) | 546.3 (313.7) | 4.4 (2.1) | 56.9 (31.7) | 64.7 (35.0) |
| A | 1,464 | 13.7 (3.8) | 11.7 (3.3) | 1.8 (2.1) | 0.2 (0.8) | 2.0 (2.3) | 13.2 (14.3) | 459.1 (263.3) | 3.8 (1.8) | 45.0 (23.3) | 52.5 (27.1) |
| B | 611 | 15.0 (3.5) | 13.6 (3.3) | 1.4 (1.8) | 0.0 (0.3) | 1.4 (1.9) | 8.8 (11.6) | 609.5 (349.0) | 4.8 (2.3) | 66.0 (34.3) | 72.8 (37.2) |
| C | 537 | 14.8 (3.2) | 13.7 (3.2) | 0.9 (1.5) | 0.2 (0.6) | 1.1 (1.6) | 7.0 (11.0) | 683.1 (342.4) | 5.5 (2.4) | 75.9 (35.2) | 81.8 (35.5) |
| D | 816 | 15.7 (3.1) | 13.5 (3.1) | 1.8 (2.1) | 0.3 (0.9) | 2.1 (2.4) | 13.1 (14.1) | 618.3 (318.9) | 4.9 (2.1) | 66.2 (33.5) | 76.6 (37.2) |
| Farms with endemic/epidemic PRRSV situation (AB) | 2,075 | 14.1 (3.7) | 12.3 (3.4) | 1.6 (2.0) | 0.2 (0.7) | 1.8 (2.2) | 11.9 (13.7) | 496.4 (293.9) | 4.1 (2.0) | 50.2 (27.9) | 57.6 (31.1) |
| Farms with endemic PRRSV situation (CD) | 1,353 | 15.3 (3.2) | 13.6 (3.1) | 1.5 (2.0) | 0.3 (0.8) | 1.7 (2.2) | 10.6 (13.2) | 642.1 (328.6) | 5.1 (2.3) | 69.7 (34.4) | 78.5 (37.7) |
PRRSV, porcine reproductive and respiratory syndrome virus.
Univariate analysis
The main results of the univariate analysis (by farm or merged, as previously detailed) are presented in Figures 2–4 and in Supplementary Tables S1-S6 and Figures S1-S3. In brief, the NTB and %NLP/NTB were significantly associated with farm, resilience criteria (R/S), sow parity, and global PRRSV status (endemic vs endemic/epidemic) when data were analyzed by farm (except for farms C and D), pooled considering the PRRSV status (endemic/epidemic [AB] vs endemic [CD]) and merged (A, B, C, and D). In the case of the number of successful parities and stayability, a significant association (P < 0.05) with farm was observed when merging all data (A, B, C, and D) and when considering the PRRSV status, except for stayability in farms (CD) without PRRSV outbreaks (P = 0.12). However, a statistical trend (P < 0.10) was observed for both merging situations for resilient criteria (R/S) and for stayability and number successful of parities, but a significant association (P = 0.03) was observed for stayability and resilient criteria in farms with PRRSV outbreaks (AB) and no association (P = 0.30) for the number of successful parities and farms without PRRSV outbreaks (CD). Finally, NBA_life and NTB_life were only significantly associated with farm consolidation data as described above except for farms without PRRSV outbreaks (CD; P = 0.27). In addition, no significant association was observed between resilience criteria (R/S) and NBA_life and NTB_life, but a statistical trend (P = 0.08) for NBA_life and the R/S classification was observed for farms with PRRSV outbreaks (AB).
Figure 2.
(A) Mean total number born (NTB), (B) number of lost piglets (NLP) and (C) percentage of lost piglets (%NLP/NTB) by resilience phenotype (R—resilient / S—susceptible) in the total population (left bars) and in farms grouped by porcine reproductive and respiratory syndrome virus (PRRSV) status (right bars). AB—farms with endemic and epidemic PRRSV phases; CD—farms with endemic PRRSV situation only. Error bars indicate SEM.
Figure 4.
(A) Sow’s stayability (days) and (B) number of successful parities during the entire productive life of the sow, by resilience phenotype (R, resilient / S, susceptible) in the total population (left bars) and in farms grouped by porcine reproductive and respiratory syndrome virus (PRRSV) status (right bars). AB—farms with endemic and epidemic PRRSV phases; CD—farms with endemic PRRSV situation only. Error bars indicate SEM.
Figure 3.
(A) Cumulative total number born (NTB_life) and (B) of piglets born alive (NBA_life) during the entire productive life of the sow, by resilience phenotype (R—resilient / S—susceptible) in the total population (left bars) and in farms grouped by porcine reproductive and respiratory syndrome virus (PRRSV) health (right bars). AB—farms with endemic and epidemic PRRSV phases; CD—farms with endemic PRRSV situation only. Error bars indicate SEM.
Multivariable analysis
The most relevant results in connection with the multivariable analysis (by farm or merged, as previously detailed) are shown in Supplementary Tables S7-S13. A significant association was observed between farm and PRRSV global status, which did not allow both to be included in the final model. The predictors (P < 0.05) included in the final multivariable model for NTB and %NLP/NTB were resilience criterion (R/S), farm, and parity, merged them considering the PRRSV health status (AB vs CD) and merging all of them (A, B, C, and D), except for resilience criteria in the case of farms without PRRSV outbreaks (CD). The predictors (P < 0.05) included in the final multivariable model for the number of successful parities and stayability were resilience criteria (R/S) and the farm merging them considering the PRRS status (AB vs CD) and merging them all (A, B, C, and D). In any case, no clear pattern in the distribution of residuals vs predicted values was observed suggesting homoscedasticity. The outcomes from the final models are explained in detail below.
Effect of the resilience classification on piglets born and lost
After running the model, NTB and %NLP/NTB were significantly higher in susceptible sows than in resilient sows (P < 0.05) in both combining conditions except in farms without PRRSV outbreaks (CD; Table 3). Furthermore, significant differences between resilient and susceptible sows were observed in farms AB, where clinical outbreaks of PRRSV were diagnosed (Table 4). On average, susceptible sows gave birth to half a piglet more than resilient sows but had about 2% more losses than resilient sows. During the outbreaks, the differences in %NLP/NTB maximized to around 4%. However, there was no significant interaction between PRRSV status (endemic/epidemic) and the resilience criteria (R/S).
Table 3.
LS Means (±SEM) of the effect of resilience phenotype (R—resilient / S—susceptible) on the total number of born piglets (NTB) and the percentage of lost piglets (%NLP/NTB) in the total population of sows and in farms pooled by PRRSV status
| NTB | |||
|---|---|---|---|
| R | S | P-value | |
| All farms | 14.3 ± 0.2 | 14.9 ± 0.1 | 0.004 |
| Farms AB | 14.0 ± 0.2 | 14.5 ± 0.2 | 0.01 |
| Farms CD | 14.8 ± 0.3 | 15.2 ± 0.2 | 0.25 |
| %NLP/NTB | |||
|---|---|---|---|
| R | S | P-value | |
| All farms | 10.0 ± 0.5 | 12.2 ± 0.4 | 0.0003 |
| Farms AB | 12.2 ± 0.7 | 14.6 ± 0.6 | 0.0004 |
| Farms CD | 10.0 ± 1.0 | 11.5 ± 0.6 | 0.24 |
AB—farms with endemic and epidemic PRRS phases; CD—farms with only endemic PRRS situation. PRRSV, porcine reproductive and respiratory syndrome virus.
Table 4.
LS Means (± SEM) of the interaction between the resilience phenotype (R—resilient / S—susceptible) and the PRRSV status of the farm for the percentage of total lost piglets (%NLP/NTB) in the total sow population and in farms that experienced a PRRS outbreak during the study period
| %NLP/NTB | |||||
|---|---|---|---|---|---|
| Endemic | Epidemic | P-value interaction | |||
| R | S | R | S | ||
| Farms AB | 10.5 ± 0.7c | 12.9 ± 0.6b | 14.5 ± 1.2b | 18.5 ± 1.0a | 0.30 |
| Farm A | 11.7 ± 1.0c | 13.8 ± 0.9bc | 17.9 ± 1.7ab | 22.4 ± 1.3a | 0.27 |
| Farm B | 7.8 ± 1.0b | 10.4 ± 0.9a | 6.8 ± 1.9 ab | 9.6 ± 1.6ab | 0.95 |
The superscript letter abc within each row indicates differences in LS Means (P < 0.05). PRRSV, porcine reproductive and respiratory syndrome virus.
Effect of the resilience classification on sow’s productive life
After running the model, the number of successful parities were significantly higher in resilient than in susceptible sows in the merged data (P < 0.05) and differences reached a statistical trend (P = 0.09) in farms with PRRSV outbreaks (AB; Table 5). The sow’s stayability was about 50 days higher (P < 0.05) in resilient than in susceptible sows in the merged data and in farms affected by epidemic outbreaks (AB) but not in farms only with PRRSV endemic status (CD). Finally, although NBA_life and NTB_life were higher in the resilient sows, the differences between the two groups were not significant (P > 0.05; Table 5).
Table 5.
LS Means (±SEM) of the effect of the resilience phenotype (R—resilient / S—susceptible) over the number of successful parities, total number of piglets born (NTB_life) and born alive (NBA_life) throughout sow´s productive life and the sow’s stayability in the total population of sows and in farms pooled by PRRSV status
| Number of successful parities | |||
|---|---|---|---|
| R | S | P-value | |
| All farms | 5.0 ± 0.1 | 4.7 ± 0.1 | 0.04 |
| Farms AB | 4.5 ± 0.1 | 4.2 ± 0.1 | 0.09 |
| Farms CD | 5.6 ± 0.3 | 5.1 ± 0.2 | 0.23 |
| Stayability (days) | |||
|---|---|---|---|
| R | S | P-value | |
| All farms | 631.1 ± 20.8 | 574.4 ± 14.6 | 0.02 |
| Farms AB | 568.8 ± 22.3 | 514.7 ± 17.6 | 0.04 |
| Farms CD | 698.5 ± 46.5 | 643.6 ± 26.1 | 0.25 |
| NTB_life | |||
|---|---|---|---|
| R | S | P-value | |
| All farms | 72.6 ± 2.3 | 70.2 ± 1.6 | 0.37 |
| Farms AB | 63.9 ± 2.3 | 62.0 ± 1.8 | 0.50 |
| Farms CD | 82.5 ± 5.3 | 78.0 ± 3.0 | 0.50 |
| NBA_life | |||
|---|---|---|---|
| R | S | P-value | |
| All farms | 65.7 ± 2.0 | 62.1 ± 1.4 | 0.13 |
| Farms AB | 57.5 ± 2.1 | 54.4 ± 1.6 | 0.19 |
| Farms CD | 74.8 ± 4.9 | 69.6 ± 2.7 | 0.38 |
PRRSV, porcine reproductive and respiratory syndrome virus.
Discussion
Two PRRSV-negative multiplication farms and four PRRSV-positive production farms were voluntarily enrolled in this 3-yr validation study. All samples from the multiplication farms were negative throughout the study, either by qRT-PCR (shedding status) or antibody detection (exposure to the virus). This result is critical because it guarantees that only naïve animals were vaccinated with the PRRSV MLV vaccine to carry out the same phenotyping procedure as previously published (Abella et al., 2019). In two of the four sow farms, the PRRSV health status changed over time, from endemic to epidemic according to the detection of PRRSV by qRT-PCR whereas in the other two production farms, the shedding was negative (no detection of PRRSV by qRT-PCR) throughout the study. However, most of the sera tested were positive for PRRSV antibodies in adult breeding animals, weaners, breeding replacements, and growing pigs on the four production farms. In particular, the prevalence of PRRSV by antibody detection was above 80% throughout the study (data not shown). Thus, the farms selected to validate this phenotyping method are representative of the most common epidemiological situation with PRRSV under field conditions (Montaner-Tarbes et al., 2019). Finally, the PRRSV strain used in the phenotyping method (Porcilis PRRS) was different, based on ORF5 sequence, with the field PRRSV strains detected in the production farms A and B, respectively, during the PRRSV outbreak (Figure 1). In conclusion, our results could probably be extrapolated to many other pig-producing farms affected by this devastating viral disease, thus guaranteeing their external validity.
We were able to associate the resilient phenotype (R) with lower losses in sow farms, both in endemic and endemic/epidemic situations. This is in agreement with previous observational (Rashidi et al., 2014; Serão et al., 2014) and experimental studies (Ladinig et al., 2014) with PRRSV-2 infected gilts and sows, where the spread of the disease in fetuses was highly variable both between and often within litters. Our results confirmed that PRRSV resilient (R) sows had a lower percentage of stillborn and mummified piglets (NLP) relative to total piglets born (%NLP/NTB) and a longer sow stayability than PRRSV susceptible (S) sows in four sow farms infected with different wild-type strains of PRRSV-1. To the best of our knowledge, this is the first time that a phenotyping method to identify PRRSV-resilient sows (Abella et al., 2019) has been validated under practical conditions in different farms. A sow’s reproductive performance and the length of its productive life depend on infectious and noninfectious factors. For instance, the sire line used in the farm can also influence the reproductive performance of the sow. Also, as the gilts in our study were allocated to four conventional PRRSV-positive farms, they were exposed to PRRSV but potentially also to other porcine pathogens, which are likely to differ between farms. Thus, their reproductive outcome and stayability are the results of their intrinsic reproductive ability, their resilience to PRRSV and other diseases such as PCV2 infection among others, and their resilience to noninfectious factors. In this study, these three components cannot be disentangled because each individual pathogen burden and other noninfectious factors are not known and cannot be robustly quantified. However, as R and S gilts were randomly assigned in contemporary production batches, they had the same probability of being infected with PRRSV or any other reproductive pathogen, thus avoiding group bias. Finally, the resilient phenotypes are not strain-specific because a similar response between resilient and susceptible sows was observed on farms with different PRRSV field strains. This result supports that the proposed phenotyping procedure appears to target the innate rather than the acquired immune response capacity of the animals. However, the classification could be influenced by the vaccine selected for the experiment, i.e., with the information at hand we cannot anticipate the response achieved with other PRRSV vaccines with different levels of virulence and the experiment (dose and days to control viremia levels) might need to be finetuned for each commercial PRRSV vaccine.
It is complex to design effective vaccines against PRRSV, especially in heterologous situations (Nan et al., 2017), due to its high variability and ability to evade the immune system. Therefore, it is necessary to develop additional tools other than vaccination to control this disease. Among these, resilient pigs could be an effective option as they are likely to show fewer clinical signs and shed less virus after PRRSV infection (Rowland et al., 2012). As a result, subsequent improvements in health and productive performance (Reiner, 2016) are expected due to the reduced infection pressure within and between herds. However, there is no established protocol for identifying PRRSV-resilient pigs. Overall, the results obtained with our proposed protocol indicate that sows phenotyped as resilient are more efficient in environments where PRRSV is circulating, as they can cope with the same number of viable offspring as susceptible sows, but at a lower biological cost (i.e., 2% to 4% lower %NLP/NTB). Conversely, susceptible sows produce larger litters (i.e., half a piglet more in NTB), but their stayability is lower than that of resilient sows. In summary, sows that are less sensitive to unfavorable conditions are also less productive under favorable conditions. It can therefore be hypothesized that there may be a tradeoff between reproductive performance and the phenotyping method for PRRSV resilience. To date, other authors have described that PRRSV antibody response (specific acquired immune response) shows strong genetic correlations with many reproductive traits in sows (number of mummies and piglets born alive) in some studies (Serão et al., 2014; Hickmann et al., 2021) but with fewer traits (number of mummies) in others (Putz et al., 2019). Selection for improved production may affect the so-called functional traits associated with resilience (survivability, longevity, and immunocompetence). Therefore, selective breeding for resilience, which aims to select animals for stable production in changing environments, could compromise productive performance. In these lines, tradeoffs between growth and immunity have been demonstrated in intensive poultry systems (Zerjal et al., 2021), sheep (Douhard et al., 2022), and pigs (Clapperton et al., 2006), but less information is available between reproductive performance and resilience. Our results pave the way for further studies to unravel the mechanisms that might explain this tradeoff.
Conclusion
We have further validated the usefulness of a PRRSV vaccine-based protocol to classify sows as resilient or susceptible to future encounters with PRRSV. On average, resilient sows produce half a piglet less per parity than susceptible sows. This drawback is offset by lower litter losses in both PRRSV endemic and epidemic situations, as well as a longer productive life, resulting in more successful parities per sow. Resilient sows thus contribute to a more sustainable production system, reducing sow replacement and piglet mortality. The validation of this protocol on four independent production farms paves the way for the study of the genetic variation underlying the R/S classification, with a view to incorporating this information into selection programs in the future.
Supplementary Material
Acknowledgments
We would like to acknowledge the support of Pinsos del Segre SA and Vall Companys SA in carrying out this experiment and the generosity of the veterinarians and farmers involved in this study.
Glossary
Abbreviations
- Ct
the cycle threshold value on a qRT-PCR analysis
- DPV
days post-vaccination
- MLV-PRRSV
PRRSV modified live vaccine
- NBA
number of piglets born alive per litter
- NBA_life
total number of piglets born alive throughout sow´s productive life
- NLP
number of lost piglets per litter
- NMU
number of mummified piglets per litter
- NSB
number of stillborn piglets per litter
- NTB
the total number of piglets born per litter
- NTB_life
total number of piglets born throughout sow´s productive life
- PRRS
Porcine reproductive and respiratory syndrome
- PRRSV
Porcine reproductive and respiratory syndrome virus
- qRT-PCR
quantitative reverse-transcriptase PCR
- R
resilient phenotype
- S
susceptible phenotype
Contributor Information
Lorenzo Fraile, Departament de Ciència Animal, Universitat de Lleida – AGROTECNIO-CERCA Centre, 25198 Lleida, Spain.
Albert Vidal, Grupo Vall companys, 25191, Lleida, Spain.
Javier Romero, Grupo Vall companys, 25191, Lleida, Spain.
Gloria Abella, Grupo Vall companys, 25191, Lleida, Spain.
Jordi Gracia, Pinsos del Segre, 25600, Balaguer, Spain.
Isabel Blanco-Penedo, Departament de Ciència Animal, Universitat de Lleida – AGROTECNIO-CERCA Centre, 25198 Lleida, Spain; Clinical Sciences, Swedish University of Agricultural Sciences (SLU), Uppsala SE-75007, Sweden.
Ramona N Pena, Departament de Ciència Animal, Universitat de Lleida – AGROTECNIO-CERCA Centre, 25198 Lleida, Spain.
Funding
This research and the APC were partially funded by MCIN/AEI 10.13039/501100011033 and by “ERDF A way of making Europe”, projects COMRDI16-1-0035-03 and PID2021-124149OB-I00.
Conflict of interest statement
All authors declare that they have no conflicts of interest.
Literature Cited
- Abella, G., Pena R. N., Nogareda C., Armengol R., Vidal A., Moradell L., Tarancon V., Novell E., Estany J., and Fraile L... 2016. A WUR SNP is associated with European Porcine Reproductive and Respiratory Virus Syndrome resistance and growth performance in pigs. Res. Vet. Sci. 104:117–122. doi: 10.1016/j.rvsc.2015.12.014 [DOI] [PubMed] [Google Scholar]
- Abella, G., Novell E., Tarancon V., Varona L., Pena R. N., Estany J., and Fraile L... 2019. Identification of resilient sows in porcine reproductive and respiratory syndrome virus-infected farms. J. Anim. Sci. 97:3228–3236. doi: 10.1093/jas/skz192 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Abella, G., Pagès-Bernaus A., Estany J., Pena R. N., Fraile L., and Plà-Aragonés L. M... 2021. Using PRRSV-resilient sows improve performance in endemic infected farms with recurrent outbreaks. Animals (Basel). 11:740. doi: 10.3390/ani11030740 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Clapperton, M., Bishop S. C., and Glass E. J... 2006. Selection for lean growth and food intake leads to correlated changes in innate immune traits in Large White pigs. Anim. Sci. 82:867–876. doi: 10.1017/asc200691 [DOI] [Google Scholar]
- Doeschl-Wilson, A. B., and Kyriazakis I... 2012. Should we aim for genetic improvement in host resistance or tolerance to infectious pathogens? Front. Gene 3:272. doi: 10.3389/fgene.2012.00272 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Dohoo, I. R., Martin S. W., and Stryhn H... 2012. Methods in Epidemiologic Research. Canada: VER Incorporated, Charlottetown, Prince Edward Island. [Google Scholar]
- Douhard, F., Doeschl-Wilson A. B., Corbishley A., Hayward A. D., Marcon D., Weisbecker J. -L., Aguerre S., Bordes L., Jacquiet P., McNeilly T. N.,. et al. 2022. The cost of host genetic resistance on body condition: Evidence from divergently selected sheep. Evol. Appl. 15:1374–1389. doi: 10.1111/eva.13442 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Fraile, L., Alegre A., López-Jiménez R., Nofrarías M., and Segalés J... 2010. Risk factors associated with pleuritis and cranio-ventral pulmonary consolidation in slaughter-aged pigs. Vet. J. 184:326–333. doi: 10.1016/j.tvjl.2009.03.029 [DOI] [PubMed] [Google Scholar]
- Fraile, L., Fernández N., Pena R. N., Balasch S., Castellà G., Puig P., Estany J., and Valls J... 2020. A probabilistic Poisson-based model to detect PRRSV recirculation using sow production records. Prev. Vet. Med. 177:104948. doi: 10.1016/j.prevetmed.2020.104948 [DOI] [PubMed] [Google Scholar]
- Guzmán, M., Meléndez R., Jiménez C., Piche M., Jiménez E., León B., Cordero J. M., Ramirez-Carvajal L., Uribe A., Van Nes A.,. et al. 2021. Analysis of ORF5 sequences of Porcine Reproductive and Respiratory Syndrome virus (PRRSV) circulating within swine farms in Costa Rica. BMC Vet. Res. 17:217. doi: 10.1186/s12917-021-02925-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Harding, J. C. S., Ladinig A., Novakovic P., Detmer S. E., Wilkinson J. M., Yang T., Lunney J. K., and Plastow G. S... 2017. Novel insights into host responses and reproductive pathophysiology of porcine reproductive and respiratory syndrome caused by PRRSV-2. Vet. Microbiol. 209:114–123. doi: 10.1016/j.vetmic.2017.02.019 [DOI] [PubMed] [Google Scholar]
- Hickmann, F. M. W., Braccini Neto J., Kramer L. M., Huang Y., Gray K. A., Dekkers J. C. M., Sanglard L. P., and Serão N. V. L... 2021. Host genetics of response to porcine reproductive and respiratory syndrome in sows: Antibody response as an indicator trait for improved reproductive performance. Front. Genet. 12:707873. doi: 10.3389/fgene.2021.707873 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Holtkamp, D. J., Polson D. D., Torremorell M., Morrison B., Classen D. M., Becton L., Henry S., Rodibaugh M. T., Rowland R. R., Snelson H.,. et al. 2011. Terminology for classifying swine herds by porcine reproductive and respiratory syndrome virus status. J. Swine Health Prod. 19:44–56. [PubMed] [Google Scholar]
- Ladinig, A., Wilkinson J., Ashley C., Detmer S. E., Lunney J. K., Plastow G., and Harding J. C. S... 2014. Variation in fetal outcome, viral load and ORF5 sequence mutations in a large scale study of phenotypic responses to late gestation exposure to type 2 porcine reproductive and respiratory syndrome virus. PLoS One 9:e96104. doi: 10.1371/journal.pone.0096104 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Lunney, J. K., and Chen H... 2010. Genetic control of host resistance to porcine reproductive and respiratory syndrome virus (PRRSV) infection. Virus Res. 154:161–169. doi: 10.1016/j.virusres.2010.08.004 [DOI] [PubMed] [Google Scholar]
- Montaner-Tarbes, S., Del Portillo H. A., Montoya M., and Fraile L... 2019. Key gaps in the knowledge of the porcine respiratory reproductive syndrome virus (PRRSV). Front. Vet. Sci. 6:38. doi: 10.3389/fvets.2019.00038 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Nan, Y., Wu C., Gu G., Sun W., Zhang Y. -J., and Zhou E. -M... 2017. Improved Vaccine against PRRSV: current progress and future perspective. Front. Microbiol. 8:1635. doi: 10.3389/fmicb.2017.01635 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Nathues, H., Alarcon P., Rushton J., Jolie R., Fiebig K., Jimenez M., Geurts V., and Nathues C... 2017. Cost of porcine reproductive and respiratory syndrome virus at individual farm level – An economic disease model. Prev. Vet. Med. 142:16–29. doi: 10.1016/j.prevetmed.2017.04.006 [DOI] [PubMed] [Google Scholar]
- Prunier, A., Heinonen M., and Quesnel H... 2010. High physiological demands in intensively raised pigs: impact on health and welfare. Animal. 4:886–898. doi: 10.1017/S175173111000008X [DOI] [PubMed] [Google Scholar]
- Putz, A. M., Harding J. C. S., Dyck M. K., Fortin F., Plastow G. S., Dekkers J. C. M., and PigGen C... 2019. Novel resilience phenotypes using feed intake data from a natural disease challenge model in wean-to-finish pigs. Front. Genet. 9:660. doi: 10.3389/fgene.2018.00660 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Rashidi, H., Mulder H. A., Mathur P., Van Arendonk J. A. M., and Knol E. F... 2014. Variation among sows in response to porcine reproductive and respiratory syndrome. J. Anim. Sci. 92:95–105. doi: 10.2527/jas.2013-6889 [DOI] [PubMed] [Google Scholar]
- Reiner, G. 2016. Genetic resistance - an alternative for controlling PRRS? Porcine Health Manag. 2:27. doi: 10.1186/s40813-016-0045-y [DOI] [PMC free article] [PubMed] [Google Scholar]
- Renken, C., Nathues C., Swam H., Fiebig K., Weiss C., Eddicks M., Ritzmann M., and Nathues H... 2021. Application of an economic calculator to determine the cost of porcine reproductive and respiratory syndrome at farm-level in 21 pig herds in Germany. Porcine Health Manag. 7:3. doi: 10.1186/s40813-020-00183-x [DOI] [PMC free article] [PubMed] [Google Scholar]
- Rowland, R. R. R., Lunney J., and Dekkers J... 2012. Control of porcine reproductive and respiratory syndrome (PRRS) through genetic improvements in disease resistance and tolerance. Front. Genet. 3:260. doi: 10.3389/fgene.2012.00260 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Serão, N. V. L., Matika O., Kemp R. A., Harding J. C. S., Bishop S. C., Plastow G. S., and Dekkers J. C. M... 2014. Genetic analysis of reproductive traits and antibody response in a PRRS outbreak herd. J. Anim. Sci. 92:2905–2921. doi: 10.2527/jas.2014-7821 [DOI] [PubMed] [Google Scholar]
- Zerjal, T., Härtle S., Gourichon D., Guillory V., Bruneau N., Laloë D., Pinard-van der Laan M. -H., Trapp S., Bed’hom B., and Quéré P... 2021. Assessment of trade-offs between feed efficiency, growth-related traits, and immune activity in experimental lines of layer chickens. Genet. Sel. Evol. 53:44. doi: 10.1186/s12711-021-00636-z [DOI] [PMC free article] [PubMed] [Google Scholar]
- Zhang, Z., Li Z., Li H., Yang S., Ren F., Bian T., Sun L., Zhou B., Zhou L., and Qu X... 2022. The economic impact of porcine reproductive and respiratory syndrome outbreak in four Chinese farms: Based on cost and revenue analysis. Front. Vet. Sci. 9:1024720. doi: 10.3389/fvets.2022.1024720 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.