In swine breeding, genetic selection is mostly performed in the nucleus to improve the performance of crossbred animals at the commercial level. However, the expression of immune-related traits is limited in the nucleus herds due to the high biosecurity. Thus, only when the animals are introduced to the commercial farms, they are mostly exposed to pathogens and other stressors, allowing the expression of such traits.
Therefore, selecting animals for immune-related traits depends on collecting crossbred data at the commercial level. The inclusion of phenotypic data from crossbred performance for selection of purebreds is known as combined crossbred and purebred selection. However, this requires crossbred animals to have pedigree information, which is not common in swine production. This issue could be overcome by genotyping individuals at the commercial level (Dekkers, 2007). This, in turn, allows producers to optimize the profitability by selecting key traits to improve crossbred performance and optimize strategic breeding decisions (Figure 1).
Reproductive performance is a key component contributing to the success of the swine industry. One strategy used in the industry to improve litter size is through the genetic selection for purebred performance. However, reproductive traits are very influenced by the environment and little explained by genetics (about 10%), creating challenges for rapid genetic improvement in litter size traits.
Also, animals that perform better at the nucleus environment do not necessarily have the best performance in the commercial herds. This happens because the genetic correlation between purebred (nucleus) and crossbred (commercial) performance is expected to be less than unit (Wientjes and Calus, 2017). Thus, an indicator trait of reproductive performance collected at the commercial level, such as antibody response to porcine reproductive and respiratory syndrome, could be used to obtain faster genetic progress on reproductive performance in sows.
Antibody response to PRRS virus has been shown to be a potential trait for selection for PRRS sows. Serão et al. (2014) reported that antibody response to PRRSV outbreak is highly heritable (45%). In addition, they showed that selection for increased antibody to PRRSV is expected to increase the number born alive and decrease number of stillborn in PRRSV-infected sows, since these two traits are highly genetically correlated with antibody response to PRRSV, with correlations of 0.72 and -0.73, respectively. However, waiting for PRRS outbreaks to occur in nucleus herds limits the use of this trait for routine selection purposes. Therefore, we have recently shown that if this relationship between antibody response to PRRSV and reproductive performance would also be favorable in non-infected commercial pigs by evaluating antibody response to PRRSV in vaccinated crossbred gilts (Sanglard et al., 2020).
To investigate the genetic relationship between antibody response to PRRSV vaccination and litter size in commercial sows, data from two commercial farms in North Carolina were used in the study, including 906 F1 (Landrace x Large White) replacement gilts. The animals were vaccinated (139 ± 17-days-old) with a commercial modified live PRRSV vaccine, and blood samples were collected (approximately 50 days after vaccination) for measurement of antibody response against PRRSV and genotyping.
A subset of 811 animals had reproductive performance recorded for up to three parities from January 2018 to December 2018 for the following litter size traits: number born alive, number still born, number born mummified, number of piglets weaned, number of pre-weaning mortality, number born dead (NSB + MUM), and total number born (NBA + NBD).
First, analysis was conducted to investigate the heritability of antibody response to PRRSV vaccination. Our results confirm that this trait is also highly heritable (34%) and, thus, a good candidate to be used as an indicator trait to obtain substantial genetic progress during selection.
Further, we investigated the relationship between the antibody response to PRRSV vaccination and litter size traits in commercial crossbred sows. High favorable genetic correlation was found with number born alive at Parity 1 (0.61), pre-weaning mortality at Parity 2 (-0.70), NSB at Parity 3 (-0.84) and MUM at Parity 3 (-0.83). The genetic correlation with the other traits were not as strong, although they were all in the favorable direction. In other words, increased antibody response to PRRSV would improve reproductive performance.
Finally, we demonstrated that indirect genetic gain on reproductive performance using antibody response to PRRSV vaccination was about 8% NBA, 72% PWM, 53% NSB and 49% MUM more efficient than the genetic gain when selecting directly over reproductive traits at the commercial level in 10 generations (Figure 2).
For this purpose, we performed simulations to estimate the genetic gain in both scenarios. It is important to note that the calculations of efficiency are simplistic, as in reality, animals are selected using an index with different economic weights. Therefore, a more comprehensive simulation is needed in follow-up studies including costs associated with vaccination, antibody measurements and genotyping.
In summary, combined crossbred and purebred selection has been shown to improve the performance of crossbred commercial individuals. Traits not expressed in the nucleus, such as immune-related traits, can be especially benefitted by this strategy. Thus, the measurement of antibody response to PRRSV vaccination in commercial herds could be a feasible strategy to obtain an indirect genetic improvement of reproductive performance in commercial crossbred and purebred sows.