Artificial light at night amplifies seasonal relapse of haemosporidian parasites in a widespread songbird

Abstract

Urban habitats can shape interactions between hosts and parasites by altering not only exposure rates but also within-host processes. Artificial light at night (ALAN) is common in urban environments, and chronic exposure can impair host immunity in ways that may increase infection. However, studies of causal links between this stressor, immunity, and infection dynamics are rare, particularly in migratory animals. Here, we experimentally tested how ALAN affects cellular immunity and haemosporidian parasite intensity across the annual cycle of migrant and resident subspecies of the dark-eyed junco (Junco hyemalis). We monitored an experimental group exposed to light at night and a control group under natural light/dark cycles as they passed through short days simulating early spring to longer days simulating the breeding season, followed by autumn migration. Using generalized additive mixed models, we show that ALAN increased inflammation, and leucocyte counts were greatest in early spring and autumn. At the start of the experiment, few birds had active infections based on microscopy, but PCR revealed many birds had chronic infections. ALAN increased parasitaemia across the annual cycle, with strong peaks in spring and autumn that were largely absent in control birds. As birds were kept in indoor aviaries to prevent vector exposure, this increased parasitaemia indicates relapse of chronic infection during costly life-history stages (i.e. reproduction). Although the immunological and parasitological time series were in phase for control birds, cross-correlation analyses also revealed ALAN desynchronized leucocyte profiles and parasitaemia, which could suggest a general exaggerated inflammatory response. Our study shows how a common anthropogenic influence can shape within-host processes to affect infection dynamics.

Keywords: Junco hyemalis; avian malaria; ecoimmunology; generalized additive models; photoperiod; urbanization.

Figure 1.

Effects of ALAN on (a) total leucocytes and (b) HL ratios across the simulated annual cycle (i.e. photoperiod, L) of resident and migrant dark-eyed juncos. Points show medians with jackknife estimates of standard error and are jittered to reduce overlap. Lines and shaded bands show the fitted values and 95% confidence interval from each GAMM. The vertical axes are displayed with a modulus transformation to accommodate lower bounds at zero. (Online version in colour.)

Figure 2.

Effects of ALAN on haemosporidian infection dynamics across the simulated annual cycle (i.e. photoperiod, L) of resident and migrant dark-eyed juncos. (a) Prevalence of microscopy-detectable infection is displayed with points and 95% confidence intervals. (b) Parasite intensity is shown as medians with jackknife estimates of standard error. Lines and shaded bands show the fitted values and 95% confidence interval from the GAMM. The vertical axes are displayed with a modulus transformation to accommodate lower bounds at zero. All points are jittered to reduce overlap. (Online version in colour.)

Figure 3.

Cross-correlation between immune measures (total leucocytes and HL ratios) and haemosporidian intensity throughout the duration of the experiment. Points display correlation coefficients (r) at each photoperiod lag and migratory strategy and are coloured by treatment. Lines and shaded bands display the fitted values and 95% confidence interval from each GAM. (Online version in colour.)