Stress response, peripheral serotonin and natural antibodies in feather pecking genotypes and phenotypes and their relation with coping style

Abstract

Feather pecking (FP), a serious welfare and economic issue in the egg production industry, has been related to coping style. Proactive and reactive coping styles differ in, among others, the stress response, serotonergic activity and immune activity. Yet, it is unknown whether genetic lines divergently selected on FP (i.e. FP genotypes) or individuals differing in FP (i.e. FP phenotypes) can be categorized into coping styles. Therefore, we determined peripheral serotonin (5-HT) levels, natural antibody (NAb) titers, behavioral and corticosterone (CORT) responses to manual restraint (MR) in FP genotypes (high FP (HFP), low FP (LFP) and unselected control (CON) line) and FP phenotypes (feather pecker, feather pecker-victim, victim and neutral). We further examined the consistency of and relationships between behavioral and physiological measures. FP genotypes differed in behavioral responses to MR, 5-HT levels and NAb titers, but not in CORT levels after MR. HFP birds had less active responses at adolescent age, but more active responses at adult age compared to LFP and CON birds. The CON line had higher 5-HT levels at adolescent age, while the HFP line had lower 5-HT levels than the other lines at adult age. Overall, the HFP line had lower IgM NAb titers, while the LFP line had lower IgG NAb titers compared to the other lines. FP phenotypes differed in behavioral responses to MR and 5-HT levels, but not in CORT levels after MR or NAb titers. Within the HFP line, feather peckers tended to have less active responses compared to neutrals at adolescent age, while victims had more active responses compared to the other phenotypes at adult age. Feather peckers had higher 5-HT levels than neutrals at adult age. Behavioral and CORT responses to MR were not consistent over time, suggesting that responses to MR might not reflect coping style in this study. Furthermore, proactive behavioral responses were correlated with reactive physiological measures and vice versa. Thus, it was not possible to categorize FP genotypes or FP phenotypes into specific coping styles.