Genetic parameters for novel climatic resilience indicators derived from automatically-recorded vaginal temperature in lactating sows under heat stress conditions

BACKGROUND

Longitudinal records of automatically-recorded vaginal temperature (TV) could be a key source of data for deriving novel indicators of climatic resilience (CR) for breeding more resilient pigs, especially during lactation when sows are at an increased risk of suffering from heat stress (HS). Therefore, we derived 15 CR indicators based on the variability in TV in lactating sows and estimated their genetic parameters. We also investigated their genetic relationship with sows' key reproductive traits.

RESULTS

The heritability estimates of the CR traits ranged from 0.000 ± 0.000 for slope for decreased rate of TV (SlopeDe) to 0.291 ± 0.047 for sum of TV values below the HS threshold (HSUB). Moderate to high genetic correlations (from 0.508 ± 0.056 to 0.998 ± 0.137) and Spearman rank correlations (from 0.431 to 1.000) between genomic estimated breeding values (GEBV) were observed for five CR indicators, i.e. HS duration (HSD), the normalized median multiplied by normalized variance (Nor_medvar), the highest TV value of each measurement day for each individual (MaxTv), and the sum of the TV values above (HSUA) and below (HSUB) the HS threshold. These five CR indicators were lowly to moderately genetically correlated with shoulder skin surface temperature (from 0.139 ± 0.008 to 0.478 ± 0.048) and respiration rate (from 0.079 ± 0.011 to 0.502 ± 0.098). The genetic correlations between these five selected CR indicators and sow reproductive performance traits ranged from - 0.733 to - 0.175 for total number of piglets born alive, from - 0.733 to - 0.175 for total number of piglets born, and from - 0.434 to - 0.169 for number of pigs weaned. The individuals with the highest GEBV (most climate-sensitive) had higher mean skin surface temperature, respiration rate (RR), panting score (PS), and hair density, but had lower mean body condition scores compared to those with the lowest GEBV (most climate-resilient).

CONCLUSIONS

Most of the CR indicators evaluated are heritable with substantial additive genetic variance. Five of them, i.e. HSD, MaxTv, HSUA, HSUB, and Nor_medvar share similar underlying genetic mechanisms. In addition, individuals with higher CR indicators are more likely to exhibit better HS-related physiological responses, higher body condition scores, and improved reproductive performance under hot conditions. These findings highlight the potential benefits of genetically selecting more heat-tolerant individuals based on CR indicators.

Disentangling the dynamics of energy allocation to develop a proxy for robustness of fattening pigs

BACKGROUND

There is a growing need to improve robustness of fattening pigs, but this trait is difficult to phenotype. Our first objective was to develop a proxy for robustness of fattening pigs by modelling the longitudinal energy allocation coefficient to growth, with the resulting environmental variance of this allocation coefficient considered as a proxy for robustness. The second objective was to estimate its genetic parameters and correlations with traits under selection and with phenotypes that are routinely collected. In total, 5848 pigs from a Pietrain NN paternal line were tested at the AXIOM boar testing station (Azay-sur-Indre, France) from 2015 to 2022. This farm is equipped with an automatic feeding system that records individual weight and feed intake at each visit. We used a dynamic linear regression model to characterize the evolution of the allocation coefficient between the available cumulative net energy, which was estimated from feed intake, and cumulative weight gain during the fattening period. Longitudinal energy allocation coefficients were analysed using a two-step approach to estimate both the genetic variance of the coefficients and the genetic variance in their residual variance, which will be referred to as the log-transformed squared residual (LSR).

RESULTS

The LSR trait, which could be interpreted as an indicator of the response of the animal to perturbations/stress, showed a low heritability (0.05 ± 0.01), a high favourable genetic correlation with average daily growth (- 0.71 ± 0.06), and unfavourable genetic correlations with feed conversion ratio (- 0.76 ± 0.06) and residual feed intake (- 0.83 ± 0.06). Segmentation of the population in four classes using estimated breeding values for LSR showed that animals with the lowest estimated breeding values were those with the worst values for phenotypic proxies of robustness, which were assessed using records routinely collected on farm.

CONCLUSIONS

Results of this study show that selection for robustness, based on estimated breeding values for environmental variance of the allocation coefficients to growth, can be considered in breeding programs for fattening pigs.

Embryo survival and fertility differ in lines divergently selected for birth weight homogeneity in mice

The selection of animals for lower environment sensitivity around the optimum trait value can also provide benefits in productivity and welfare. A divergent selection experiment for birth weight environmental variability in mice was successfully conducted over 17 generations. Animals from low variability selected line (L-line) were more robust by having a higher litter size and survival at weaning in a common breeding environment, than those from high variability line (H-line). The objective of this study was to analyze the differences between those divergently selected lines for embryo and fetal survival and for fertility and prolificacy rate. To study embryo survival and ovulation rate, a total of 98 females (34 H-line and 50 L-line) were studied in four generations of the divergent experiment. To analyze fetal survival and fertility rate, 378 female mice (138 H-line and 240 L-line) in 10 generations or the divergent experiment were studied. Ultrasound scans were performed at day 14 of gestation to establish the number of total fetal and the embryo absorptions. Mortality was addressed as the difference between litter size at birth and the number of fetuses at 14 days of gestation. The number of pregnant females in the first 3 days after mating was used to measure fertility. A linear model was also fitted to analyze embryo mortality, litter size, and the number of embryos at 14 days of gestation. A categorical model was then used to study fertility, including line, generation, and its interaction as effects. Despite the fact that there were no significant differences in the ovulation rate, litter size at birth was significantly higher in the L-line than in the H-line (9.82 vs. 8.36 pups, p < 0.001). Moreover, embryo mortality was significantly lower in the L-line than in the H-line (1.39 vs. 2.87 fetuses, p < 0.001). L-line females were more fertile (53.49% vs. 23.26% for the H-line). According to these results, the line selected for low environmental variance would be preferable for robustness and animal welfare.

The impact of Box-Cox transformation on phenotypic and genomic characteristics of litter size variability in Landrace pigs

The genetic background of variability remains of interest especially in traits of high economic importance, e.g. litter size in pigs. It has been indicated that the data transformation can affect the variability phenotype. This study aims to evaluate the phenotypic and genomic background of variability of litter size obtained from data before and after the Box-Cox transformation. In total, 67 500 records on the total number born (TNB) in Landrace pig population were used. Since the data presented skewness, the decision was made to perform Box-Cox transformation on TNB and obtain bcTNB. Next, the phenotypic variability was estimated as log-transformed variance of residuals (LnVar) for both TNB (LnVar_TNB) and bcTNB (LnVar_bcTNB). The variability traits were further used in the genome-wide association study (GWAS) performed on 10 688 sows genotyped with Axiom porcine 660 K or imputed to 660 K SNP-chip. The substantial difference in skewness was observed after data transformation, represented as a change from -0.46 to -0.02. Heritability for TNB was 0.118 vs 0.125 for bcTNB. The heritability for LnVar_TNB was 0.0025 vs 0.0037 for LnVar_bcTNB. The change in the genetic variance was confirmed when genetic coefficients on SD level were compared: 2% for LnVar_TNB vs 4% for LnVar_bcTNB. In bivariate analysis, the genetic correlation between the additive genetic effects of the mean TNB and its variability changed from 0.38 to 0.63. The observed positive genetic correlations indicated that selection focused on increasing the litter size will simultaneously cause an increase in litter size variability. Based on GWAS, 14 SNPs were detected for LnVar_TNB and eight for LnVar_bcTNB, with two of them indicating the most promising candidate genes. First candidate gene located on Sus scrofa chromosome (SSC) 3 is STAG3, which plays an essential role in gametogenesis. Second gene located on SSC 10 is ESRRG, which affects placenta development. The additional post-GWAS analysis indicated even more candidate genes for LnVar_TNB and LnVar_bcTNB. The most promising candidate gene was located on SSC 13 - MFN1, which is involved in embryonic development. The results of this study indicated a substantial change in variance components for variability when the Box-Cox transformation was applied to data presenting skewness. Moreover, the data transformation changed the phenotype substantially enough that only part of SNP overlapped between two variability traits. Our investigation shows that it is essential to perform Box-Cox transformation for skewed data in order to properly describe phenotypic and genomic properties of litter size variability in Landrace pigs.

Traits Defining Sow Lifetime Maternal Performance

Declining sow performance with increasing parity or an increase in the number of poor- quality pigs potentially impacts on farm productivity. This study investigated the phenotypic and genetic background of the sow's influence on (i) the number of pigs not meeting the industry standards (tail-enders) and (ii) changes in performance with parity. Data were available for 3592 sows and their litters (13,976 litters) from a pig production system in NSW, Australia. The mean, standard deviation (SD), and slope for trait values over time were estimated for the sow characteristic traits: number of born-alive (NBA) and stillborn (SB) piglets and body condition of sow recorded with a caliper (CAL), along with maternal effects on piglet performance, represented by: average piglet birth weight (APBW), number of weaned piglets (WEAN), and tail-enders (TEND). Traits were analyzed in ASReml 4.2, by using an animal model. The number of tail-enders produced by a sow is a heritable trait, with a heritability estimate of 0.14 ± 0.04. Sow characteristics and maternal effects on piglet performance expressed by mean and slope had similar heritability estimates, ranging from 0.10 ± 0.03 to 0.38 ± 0.05, whereas estimates for SD traits were generally not different from zero. The latter suggests individual variability in sow characteristics or maternal performance between parities is largely not genetic in origin. This study demonstrated that more attention is required to identify contributions to the problem of tail-enders, and that slope traits could potentially be useful in the breeding program to maximize lifetime performance.

Heterozygosity and homozygosity regions affect reproductive success and the loss of reproduction: a case study with litter traits in pigs

Runs of heterozygosity (ROHet) and homozygosity (ROH) harbor useful information related to traits of interest. There is a lack of investigating the effect of ROHet and ROH on reproductive success and the loss of reproduction in mammals. Here, we detected and characterized the ROHet and ROH patterns in the genomes of Chinese indigenous pigs (i.e., Jinhua, Chun’an, Longyou Black, and Shengxian Spotted pigs), revealing the similar genetic characteristics of indigenous pigs. Later, we highlighted the underlying litter traits-related ROHet and ROH using association analysis with linear model in these four indigenous pig breeds. To pinpoint the promising candidate genes associated with litter traits, we further in-depth explore the selection patterns of other five pig breeds (i.e., Erhualian, Meishan, Minzhu, Rongchang, and Diqing pigs) with different levels of reproduction performance at the underlying litter traits-related ROHet and ROH using F_ST and genetic diversity ratio. Then, we identified a set of known and novel candidate genes associated with reproductive performance in pigs. For the novel candidate genes (i.e., CCDC91, SASH1, SAMD5, MACF1, MFSD2A, EPC2, and MBD5), we obtained public available datasets and performed multi-omics analyses integrating transcriptome-wide association studies and comparative single-cell RNA-seq analyses to uncover the roles of them in mammalian reproductive performance. The genes have not been widely reported to be fertility-related genes and can be complementally considered as prior biological information to modify genomic selections models that benefits pig genetic improvement of litter traits. Besides, our findings provide new insights into the function of ROHet and ROH in mammals.

Evaluation of the phenotypic and genomic background of variability based on litter size of Large White pigs

BACKGROUND

The genetic background of trait variability has captured the interest of ecologists and animal breeders because the genes that control it could be involved in buffering various environmental effects. Phenotypic variability of a given trait can be assessed by studying the heterogeneity of the residual variance, and the quantitative trait loci (QTL) that are involved in the control of this variability are described as variance QTL (vQTL). This study focuses on litter size (total number born, TNB) and its variability in a Large White pig population. The variability of TNB was evaluated either using a simple method, i.e. analysis of the log-transformed variance of residuals (LnVar), or the more complex double hierarchical generalized linear model (DHGLM). We also performed a single-SNP (single nucleotide polymorphism) genome-wide association study (GWAS). To our knowledge, this is only the second study that reports vQTL for litter size in pigs and the first one that shows GWAS results when using two methods to evaluate variability of TNB: LnVar and DHGLM.

RESULTS

Based on LnVar, three candidate vQTL regions were detected, on Sus scrofa chromosomes (SSC) 1, 7, and 18, which comprised 18 SNPs. Based on the DHGLM, three candidate vQTL regions were detected, i.e. two on SSC7 and one on SSC11, which comprised 32 SNPs. Only one candidate vQTL region overlapped between the two methods, on SSC7, which also contained the most significant SNP. Within this vQTL region, two candidate genes were identified, ADGRF1, which is involved in neurodevelopment of the brain, and ADGRF5, which is involved in the function of the respiratory system and in vascularization. The correlation between estimated breeding values based on the two methods was 0.86. Three-fold cross-validation indicated that DHGLM yielded EBV that were much more accurate and had better prediction of missing observations than LnVar.

CONCLUSIONS

The results indicated that the LnVar and DHGLM methods resulted in genetically different traits. Based on their validation, we recommend the use of DHGLM over the simpler method of log-transformed variance of residuals. These conclusions can be useful for future studies on the evaluation of the variability of any trait in any species.

Selection for litter size and litter birthweight in Large White pigs: Maximum, mean and variability of reproduction traits

Gradually increasing trend of litter size poses a challenge to pig farmers in terms of managing larger litters. Therefore, it seems that a balanced approach that optimises litter size, litter birthweight, and uniformity of those traits is needed in order to address animal welfare and farm management concerns. This study aimed to investigate this issue by defining several traits for total number born (TNB), number born alive (NBA) and litter birthweight (LW). First, the highest value from at least five records per sow was selected as maximum (max) value for each reproduction trait. Second, a mean (mean) for each reproduction trait was calculated per sow. Last, the variability of reproduction traits between parities of the sow was calculated as log-transformed variance of residuals of all observations per sow for each reproduction trait (LnVar). In total, 23 193 Large White sows from Topigs Norsvin with 152 282 litter records were used for analysis in ASReml 4.1. Also, a simulation of breeding schemes was performed with the use of SelAction 2.1 and estimates from genetic analysis. Maximum value of reproductive traits had a much higher heritability than repeated observations or mean of reproduction traits, e.g., 0.31 for maxTNB vs. 0.12 for TNB and 0.07 for meanTNB, which allows for a faster response under selection. The maximum value traits, however, were found to carry more risks, i.e. higher ratio of stillborn (not for maxNBA) and increased variability of traits. Thus, using them in breeding programme should be carefully considered. The genetic coefficient of variation on SD level estimated to indicate the genetic magnitude for variability phenotypes indicated a maximum change of 6-9% in genetic SD of TNB, NBA and LW. The genetic correlations between mean and corresponding variability traits varied from 0.66 to 0.74, whereas the correlation between other mean and variability traits ranged from 0.33 to 0.99. The simulation indicated that even with selection targeted against the variability of reproduction traits, a very limited change should be expected due to a complex genetic and phenotypic relationship between the traits. In the scenarios with selection against LnVarTNB and LnVarLW, this was a decrease of 0.1-0.6% per year, whereas in scenario with selection against LnVarNBA, the range was 0.6-1.1% per year. It is still possible to increase litter size and birthweight further, however, a balance between mean and variability of reproduction traits is required, which can be obtained only by a very well designed breeding programme.

Genetic analysis for sow stayability at different parities in purebred Landrace and Large White pigs

Genetic parameters for sow stayability were estimated from farrowing records of 10,295 Landrace sows and 8192 Large White sows. The record for sow stayability from parity k to parity k + 1 (k = 1, …, 6) was 0 when a sow had a farrowing record at parity k but not at parity k + 1, and 1 when a sow had both records. Heritability was estimated by using single‐trait linear and threshold animal models. Genetic correlations among parities were estimated by using two‐trait linear–linear and single‐trait random regression linear animal models. Genetic correlations with litter traits at birth were estimated by using a two‐trait linear–linear animal model. Heritability estimates by linear model analysis were low (0.065–0.119 in Landrace & 0.061–0.157 in Large White); those by threshold model analysis were higher (0.136–0.200 & 0.110–0.283). Genetic correlations among parities differed between breeds and models. Genetic correlation between sow stayability and number born alive was positive in many cases, implying that selection for number born alive does not reduce sow stayability. The results seem to be affected by decisions on culling made by farmers.

Effects of Low ω6:ω3 Ratio in Sow Diet and Seaweed Supplement in Piglet Diet on Performance, Colostrum and Milk Fatty Acid Profiles, and Oxidative Status

Simple Summary

Feeding maternal animals divergent ratios of omega-6 (ω6) and omega-3 (ω3) fatty acids can change not only their health, physiological condition, and performance but also do the same for their offspring. In swine production, various ω6:ω3 ratios have been tested, but the search for an optimal proportion in the sow diet is still in progress. For piglets, weaning oxidative stress has been alleviated by supplementing with abundant sources of bioactive compounds. In this case, brown seaweed, a rich source of natural antimicrobials and antioxidants, can be a good candidate, but its supplementation in piglet diet is limited. This study explores the hypothesis that feeding a low ω6:ω3 ratio diet to sows during gestation and lactation, together with the supplementation of Ascophyllum nodosum for piglets during the post-weaning period, could benefit piglets’ performance and oxidative status more than the respective single treatment provided to the mother or the piglet. Results showed that the low dietary ω6:ω3 ratio (4:1) and seaweed supplement did not affect the post-weaning piglets’ growth rate and oxidative status. However, a low ω6:ω3 ratio diet alone improved weaning survival rate, suckling piglets’ weight gain, and total ω3 fatty acids in colostrum and milk.

Abstract

The ratio of omega-6 (ω6) to omega-3 (ω3) polyunsaturated fatty acids (PUFAs) in the diet contributes to animal health and performance modulations because they have mostly opposite physiological functions. Increasing ω3 PUFAs content in the maternal diet can stimulate antioxidative capacity in sow and piglets; however, the optimal ratio of ω6 and ω3 PUFAs in the sow diet is still under discussion. Rich sources of bioactive constituents such as brown seaweed are an excellent supplementation to promote animal health and antioxidant status. However, the knowledge of the effects of this compound, specifically in post-weaning piglets, is still limited. Moreover, the combined effect of a low ω6:ω3 PUFAs ratio in sow diet and seaweed supplementation in post-weaning piglets’ diet has never been studied. This research aims to assess the combined effect of a low ω6:ω3 ratio in sow diets and seaweed supplementation in piglet diets on their growth and oxidative status. We also assessed the impact of a low ω6:ω3 ratio in the maternal diet on reproduction, milk fatty acid (FA) profile, and plasma leptin concentration. Two sow diets (n = 8 each) contained either a control ratio (CR, 13:1 during gestation, starting from day 28 (G28) and 10:1 during lactation) or a low ratio (LR, 4:1 from G28 until the end of lactation (L-End)) of ω6:ω3 FA by adding soybean oil or linseed oil, respectively. Reproductive performance was evaluated. Colostrum and milk at lactation day 7 (L7) and L-End were collected to analyze FA profile. Plasma was collected at G28, G79, G108, L7, L14, and L-End for determination of leptin and oxidative status. At weaning, 20 male piglets were selected per sow group to form 4 diet treatments (n = 10 each), which were supplemented with or without 4 g/kg seaweed. Recording of growth performance and collection of blood were performed at days 0, 7, 15, and 21 of post-weaning for oxidative status. LR diet increased (p < 0.05) the survival rate of piglets at weaning, and individual and litter weight gains. Colostrum and milk at L7 and L-End had lower (p < 0.05) ω6:ω3 ratio in LR sows. Interaction between dietary treatments on sows and piglets was revealed for all examined growth parameters at most time points (p < 0.05). LR diet did not affect plasma leptin levels and oxidative status. These findings suggest that the seaweed supplement during post-weaning could not improve growth rate and oxidative status of piglets born from mothers receiving a low dietary ω6:ω3 ratio (4:1) during gestation and lactation. However, this low ratio was beneficial for weaning survival rate, sucking piglets’ weight gain, and ω3 enrichment in colostrum and milk.