Immune challenge reduces daily activity period in free-living birds for three weeks

Abstract

Non-lethal infections are common in free-living animals and the associated sickness behaviours can impact crucial life-history trade-offs. However, little is known about the duration and extent of such sickness behaviours in free-living animals, and consequently how they affect life-history decisions. Here, free-living Eurasian blackbirds, Turdus merula, were immune-challenged with lipopolysaccharide (LPS) to mimic a bacterial infection and their behaviour was monitored for up to 48 days using accelerometers. As expected, immune-challenged birds were less active than controls within the first 24 h. Unexpectedly, this reduced activity remained detectable for 20 days, before both groups returned to similar activity levels. Furthermore, activity was positively correlated with a pre-experimental index of complement activity, but only in immune-challenged birds, suggesting that sickness behaviours are modulated by constitutive immune function. Differences in daily activity levels stemmed from immune-challenged birds resting earlier at dusk than control birds, while activity levels between groups were similar during core daytime hours. Overall, activity was reduced by 19% in immune-challenged birds and they were on average almost 1 h less active per day for 20 days. This unexpected longevity in sickness behaviour may have severe implications during energy-intense annual-cycle stages (e.g. breeding, migration, winter). Thus, our data help to understand the consequences of non-lethal infections on free-living animals.

Keywords: eco-immunology; eco-physiology; immunity; inflammation; life-history ecology; pathogen.

Figure 1.

(a) Mean hourly activity scores of blackbirds for daytime hours (calculated from raw data), with standard error bars. Data displayed for control (C; birds that received no injection) and experimental (E; birds injected with LPS) groups during the first 48 days after treatment. The follow-up period (days 25–48) is to the right of the dashed line, whereas all other analysed time frames (up to 24 days) are to the left. (b) Predicted marginal means for the activity score of control and experimental blackbirds, calculated from generalized linear mixed models for each time frame analysed (see Methods). Error bars represent the lower and upper confidence intervals for the predicted mean (95%). Asterisks denote significant effect estimates for activity score between the two treatment groups with ***p < 0.001; *p < 0.05; ^.p < 0.1.

Figure 2.

Distribution of the first hour of activity at (a) dawn or last hour of activity at (b) dusk. Data displayed for immune-challenged (E: birds injected with LPS) and control (C: birds that received no injection) blackbirds. Dawn/dusk are represented by dashed vertical lines; violin plots depict the density of the first or last hour of activity before (to the left of the dashed lines) or after (to the right of the dashed lines) dawn/dusk. Filled points are the median first or last hour of activity for either treatment group. Each panel represents a different time window after the experiment started: the first evening/morning, and the first 2, 5, 20 and 24 mornings/evenings; data were analysed for 22 birds for each time period. An asterisk indicates where the first/last hour of activity significantly differs between the two treatment groups, whereas a ‘^.’ indicates if there was an observed trend (p > 0.05 and less than 0.1).