Influence of lighting on sleep behaviour, circadian rhythm and spontaneous blink rate in stabled riding school horses (Equus caballus)

Abstract

Modern horse husbandry involves significant time spent indoors, often in suboptimal lighting conditions and with frequent night-time disturbances by humans for management purposes. The aim of this study was to investigate the influence of a customised light-emitting diode (LED) lighting system and a standard fluorescent lighting fixture on equine sleep behaviours, circadian rhythmicity and spontaneous blink rates in horses. Ten riding school horses experienced two stable lighting conditions for four weeks each in a cross-over study running from January to March, 2023. The treatment lighting consisted of an LED system that provided timed, blue-enriched white polychromatic light by day and dim red light at night, and control lighting was a fluorescent tube that was turned on and off manually morning and evening. During week 4 of each experimental period, spontaneous blink rate was recorded twice for 30 min, behaviour of horses in their stables was recorded continuously for 72 h, and hair samples for circadian clock gene analysis were collected at 4-h intervals for 52 h. No differences were detected for total sleep, lateral or sternal recumbency, wakefulness, standing, standing sleep, or spontaneous blink rate (P > 0.05), between lighting conditions. The lighting period (Day versus Night) influenced total sleep (P < 0.01), total recumbency (P < 0.01), wakefulness (P < 0.01), and standing sleep (P < 0.05) in both conditions. For the treatment condition only, higher wakefulness was recorded during Day (P < 0.05). An overall effect of time for clock genes PER2 and DBP was detected (P < 0.01), but there was no effect of treatment, or time by treatment interaction. Cosinor analysis detected significant 24-h rhythmicity for PER2 and DBP (P < 0.01) in both lighting conditions. Results imply that dim red light at night does not negatively impact normal sleep patterns or circadian rhythmicity, and provide evidence supporting further research to better understand the role of blue-enriched LED light at promoting increased wakefulness during daytime in stabled horses.

Fig 1. Schematic of the experimental barn layout and horse assignment to individual stables and groups.

Fig 2. Images of the stable lighting.

(A) Treatment nighttime (Equilume® Stable Light), (B) Treatment daytime (Equilume® Stable Light), and (C) Control daytime (fluorescent strip light).

Fig 3. Spectral light profiles.

(A) Blue-enriched white polychromatic light by day, (B) red light by night and (C) fluorescent light. Light measurements were taken using a CRI Illuminance Meter CL-70F at horse-eye level, with the light sensor facing upwards.

Fig 4. Video recording within stables.

(A) Network Video Recorder visual of a stable and (B) placement of camera (denoted by white arrow) within the stable.

Fig 5. Spontaneous blink rate.

(A) Example of Go-Pro Hero 10 camera (7 x 5 x 4 cm) attached to the headcollar of a horse for collection of spontaneous blink rate data. (B) Comparison of Full Blinks between Fluorescent (control) and LED (treatment) lighting conditions presented as means ± SEM.

Fig 6. Total sleep duration during Day and Night for three consecutive 24 h periods for each individual horse.

Data are presented as means ± SEM.

Fig 7. Total recumbency duration of horses maintained under customised LED lighting (treatment) and fluorescent lighting (control) for three consecutive days.

(A) Total recumbency duration during Day (B) and Night (C) for each lighting condition (control and treatment). Data are presented as means ±SEM for three consecutive 24-h periods.

Fig 8. Day versus night durations in wakefulness for three consecutive days.

(A) Day versus night duration under treatment (LED) and B) Day versus night duration under control (fluorescent) lighting conditions Data are presented as means + /-SEM for three consecutive 24 h periods. * indicates P < 0.05.

Fig 9. Profiles of clock gene expression in hair follicles from horses (n = 10) maintained under customised LED lighting (treatment) or fluorescent (control) lighting conditions.

(A) PER2, (B) DBP and C) NR1D2. Data are presented as means ± SEM. There was a significant overall effect of time for PER2 and DBP (P < 0.01; P < 0.001), but no effect of treatment or time by treatment interaction (P > 0.05). No effect of time, treatment or time by treatment interaction was detected for NR1D2 (P > 0.05). White and red or black bars beneath the graphs represent periods of daytime light and red light or darkness at night, respectively.