In a context of growing interest in breeding more resilient animals, a noninvasive indicator of resilience would be very valuable. We hypothesized that the time-course of concentrations of several milk metabolites through a short-term underfeeding challenge could reflect the variation of resilience mechanisms to such a challenge. We submitted 138 one-year-old primiparous goats, selected for extreme functional longevity (i.e., productive longevity corrected for milk yield [60 low longevity line goats and 78 high longevity line goats]), to a 2-d underfeeding challenge during early lactation. We measured the concentration of 13 milk metabolites and the activity of 1 enzyme during prechallenge, challenge, and recovery periods. Functional principal component analysis summarized the trends of milk metabolite concentration over time efficiently without preliminary assumptions concerning the shapes of the curves. We first ran a supervised prediction of the longevity line of the goats based on the milk metabolite curves. The partial least square analysis could not predict the longevity line accurately. We thus decided to explore the large overall variability of milk metabolite curves with an unsupervised clustering. The large year × facility effect on the metabolite concentrations was precorrected for. This resulted in 3 clusters of goats defined by different metabolic responses to underfeeding. The cluster that showed higher β-hydroxybutyrate, cholesterol, and triacylglycerols increase during the underfeeding challenge was associated with poorer survival compared with the other 2 clusters. These results suggest that multivariate analysis of noninvasive milk measures show potential for deriving new resilience phenotypes.
Today, there is growing interest in selecting for resilience, as livestock are expected to face increasingly harsh environmental and climatic conditions. Animal resilience is defined as the ability to overcome short-term environmental disturbances and quickly return to a predisturbance state (Colditz and Hine, 2016). In this context, resilience can be seen as an underlying component of longevity since it corresponds to the ability to cope with and recover from challenges to allow the animal to carry on its productive life (Friggens et al., 2017; Scheffer et al., 2018). Longevity corresponds to true longevity (all culling reasons) and functional longevity that includes all culling reasons, except productivity (Sasaki, 2013). Several studies estimated heritability of functional longevity to be around 10% in cattle and goats (Castañeda-Bustos et al., 2017; Nayeri et al., 2017; Palhière et al., 2018). Ithurbide et al. (2022) showed that selection on functional longevity in a commercial population of dairy goats translated into significant differences in longevity and resilience-related traits such as better mammary health and lower body fat mobilization during the beginning of the first lactation for goats selected for longer functional longevity. Selection seems to be possible; however, improvements are expected to be slow due to low heritability. This low heritability could be explained by the fact that longevity is a multifactorial trait, that is, there are factors other than resilience contributing to longevity, and strong genetic × environmental interactions can be involved (Tsartsianidou et al., 2021). Thus, there is a need to find more direct resilience indicators. Being less multifactorial, more direct resilience indicators could have a higher heritability than functional longevity, and allow a more efficient selection and for instance select animals for longevity at an early stage of productive life.
We hypothesized that the metabolic response to short-term feed restriction could provide information about some genetic characteristics of goat resilience. The objective of this study is to explore the existence of underlying resilience components within the time-course of 13 milk metabolites and 1 enzyme activity during an underfeeding challenge imposed on 2 divergent lines of goats for functional longevity. We propose a new statistical approach to model and explore multivariate longitudinal data.
The experiment was carried out in agreement with French National Regulations for the humane care and use of animals for research purposes. Animals were bred at 2 experimental INRAE Farms: P3R Bourges (UE0332, La Sapinière, Osmoy, France, license to carry out animal experiments: C18–174–01) and Experimental Installation, UMR MoSAR (Route de la Ferme, Thiverval-Grignon, France) close to Paris (license to carry out animal experiments: A 78 615 1002). This article followed the STROBE-Vet guidelines (O’Connor et al., 2016). All procedures performed on animals were approved by the Ethics Committee on Animal Experimentation and the French Ministry of Higher Education, Research and Innovation (APAFIS#8613–2017012013585646 V4 and APAFIS#24314–2019120915403741).
Following the method developed by Palhière et al. (2018) and described by Ithurbide et al. (2022), we created 2 functional longevity lines of Alpine goats. Since 2017, we have run the genetic evaluation for functional longevity over 8,787 Alpine AI bucks based on the productive longevity of their daughters (time difference between first kidding and culling) corrected for milk yield. We selected the 16 bucks who had the highest EBV and the 19 bucks who had the lowest EBV among the whole AI buck population to find the low longevity line (Low_LGV) and high longevity line (High_LGV), respectively. From 2019 to 2022, 138 goats were bred: 60 Low_LGV goats and 78 High_LGV goats. Among them, 69 were bred in the INRAE P3R Bourges facility and 69 in the INRAE Paris facility (Table 1). Within each facility, Low_LGV and High_LGV goats were housed in common pens.