Kin Recognition in Juvenile Zebrafish (Danio rerio) Based on Olfactory Cues

Abstract

Genetic analyses of numerous fish species have shown that shoals formed by larvae often consist of closely related kin (1). Aggregating with kin may be an altruistic trait that evolved through kin selection (2). Individuals would increase their inclusive fitness by sharing the benefits of shoaling among related individuals (3). Laboratory experiments on recognition of kin vs. non-kin groups of Atlantic Salmon (Salmo salar) (4) demonstrated possible advantages: kin groups had fewer aggressive interactions, used a greater proportion of “threat” behavior as opposed to fighting, and subordinates especially had improved growth. The mechanisms by which these kin groups develop and stay separate from each other are not known. The genes of the major histocompatibility complex (MHC) are a source of individual odors released into the water via urine (5). Such pheromones might be involved in olfactory kin recognition. Here, we tested the hypothesis that zebrafish can recognize kin based on olfactory cues. Zebrafish (Danio rerio) live in freshwater streams and rice paddies in the Ganges River of East India, Bangladesh, and Burma. Although this species is widely used as a model in genetic and developmental research, little is known about its natural behavior. Zebrafish spawn up to several hundred eggs at a time, and these develop in the substrate without any parental care. Larvae (G.G. pers. obs.), and sometimes adults, can be observed in shoals (6, 7), but their genetic relatedness is unknown. We observed wild-type juvenile zebrafish, aged 6–8 weeks; the fish were kept in 2.5-1 aquaria under a day/night cycle of 14/10 h and fed on a standard diet of brine shrimp nauplii and dry fish food. Twenty-four hours before an experiment started, 2 separate kin groups consisting of 12 full siblings each were placed into two 9-1 aquaria with standing water. From each kin group, 3 fish were tested in the flume, one at a time, for a total of 6 fish. This procedure was repeated 3 times for a total of 6 kin groups and 18 test animals. Water from each of the two aquaria was used as the two stimuli in an olfactory preference test. Single individuals of either group were used as test fish and were placed into a choice flume (20 cm long 4 cm wide, water level 2.5 cm) that maintained two separate water columns (Fig. 1) (8). Uniform and unidirectional water flow was maintained at a constant rate of 40 ml/min ( 3.5 mm/s). Periodic dye tests showed that the two water columns remained well separated. Prior to each trial, formulated fresh water was run through both channels of the flume for 5 min to allow the subject to acclimate. Each trial consisted of four 3-min periods, during which water from kin and non-kin aquaria was presented on alternate sides of the flume to correct for the possibility of unrelated side bias. Every 10 s, we recorded which side of the flume (A or B, Fig. 1) the fish was swimming on. The number of times each animal was recorded on the side with kin stimulus was expressed as a percentage of the total number of recorded observations (i.e., kin plus non-kin). A score greater than 50% (random distribution) indicated a preference for the kin stimulus; and when the percentages of all the fish were compared using a Wilcoxon matched-pairs signed-ranks (WSR) test, the preference for kin was significant (WSR 45.0, P 0.050) (Fig. 1). Our study is the first to show that juvenile zebrafish can recognize and prefer their siblings to unrelated conspecifics based on olfactory cues. There are two general categories of kin recognition mechanisms, both based on learning processes as Tang-Martinez (9) emphasized. The first (“indirect”) mechanism is based on familiarity, where individuals behave nepotistically to conspecifics with whom they grow up. The second (“direct”) kin recognition mechanism allows individuals to identify even unfamiliar kin. Direct recognition is thought to be based on ‘phenotype matching’, in which an individual must learn cues, either from the phenotypes of close relatives (familial imprinting) (10), or from itself, to form a template for comparison with the phenotype of other individuals. Our results cannot distinguish between these recognition mechanisms.