Male urine signals social rank in the Mozambique tilapia (Oreochromis mossambicus)

Abstract

Background: The urine of freshwater fish species investigated so far acts as a vehicle for reproductive pheromones affecting the behaviour and physiology of the opposite sex. However, the role of urinary pheromones in intra-sexual competition has received less attention. This is particularly relevant in lek-breeding species, such as the Mozambique tilapia (Oreochromis mossambicus), where males establish dominance hierarchies and there is the possibility for chemical communication in the modulation of aggression among males. To investigate whether males use urine during aggressive interactions, we measured urination frequency of dye-injected males during paired interactions between size-matched males. Furthermore, we assessed urinary volume stored in the bladder of males in a stable social hierarchy and the olfactory potency of their urine by recording of the electro-olfactogram.

Results: Males released urine in pulses of short duration (about one second) and markedly increased urination frequency during aggressive behaviour, but did not release urine whilst submissive. In the stable hierarchy, subordinate males stored less urine than males of higher social rank; the olfactory potency of the urine was positively correlated with the rank of the male donor.

Conclusion: Dominant males store urine and use it as a vehicle for odorants actively released during aggressive disputes. The olfactory potency of the urine is positively correlated with the social status of the male. We suggest that males actively advertise their dominant status through urinary odorants which may act as a 'dominance' pheromone to modulate aggression in rivals, thereby contributing to social stability within the lek.

Figure 1

Urination frequency of male tilapia. Frequency of urination (pulses per minute every 5 min; mean ± SEM; n = 4) by dye-injected resident males in social isolation (open horizontal bar) followed by the presence of a saline-injected intruder male of similar size (dark horizontal bar).

Figure 2

Urination frequency and behaviour of two neighbouring tilapia males. (A) Frequency of urination (pulses per minute every 5 min; mean ± SEM; n = 16) by males in social isolation (open horizontal bar) followed by contact with a neighbour male (dark horizontal bar). (B) Examples (1–3) of behaviour (submissive: white; not aggressive: light grey; aggressive displays: dark grey; highly aggressive: black) and release of urine pulses (urination), of around 1 s, during 45 min of interaction between two territorial male tilapia (a) and (b). In (1), male (a) increased its urination frequency 25 min after coming into contact with male (b) and initiated aggressive displays which escalated to symmetrical high aggression (circle fight); in turn, male (b) changed from submissive, not urinating, to aggressive displays and its urination frequency increased as the agonistic interaction escalated to high symmetrical aggression. In (2), both males increased their urination frequency within 10–15 min when both initiated aggressive displays which further escalated to high symmetrical aggression; although the two males maintained this level of aggression throughout the observation period, their urination frequency dropped significantly. In (3), we see the only case of high urination frequency observed during non-aggressive behaviours, before aggressive behaviour escalation, during which the frequency of urination was also high; male (a) became submissive within a few minutes after the symmetrical high aggression and stopped urinating, whereas male (b) continued with aggressive displays and stopped urine release as well.

Figure 3

Urination and behaviour of neighbouring tilapia males. Urination frequency (mean ± SEM, n = 16) during non-aggressive behaviour, aggressive displays and high aggression. Different letters over the bars indicate significant differences (repeated-measures analysis of variance (ANOVA), F_1,15 = 125.6, p < 0.001 followed by least significant difference (LSD) pairwise comparisons, p < 0.05).

Figure 4

Urine volume, urine olfactory potency and dominance index of donor male tilapia in a community tank. (A) Scatter plot of urine volume (mean ± SEM, n = 5 days) collected from males (n = 12) and their Di (mean ± SEM, n = 5 days) during 5 days in a community tank with females. The mean urine volume collected from the five subordinate males (Di between 0.0 and 0.21) was significantly smaller (0.39 ± 0.09 ml) than that collected from males of higher social rank (0.83 ± 0.06 ml; Student's t test, t₁₀ = 4.330, p < 0.005). (B) Scatter plot of EOG amplitudes (mean ± SEM, n = 30) elicited by male urine (1:10⁴ v/v in water) and the dominance index (mean ± SEM, n = 5 days) of donor males (n = 12). The urine of subordinate males evoked significantly smaller normalized EOG responses (0.87 ± 0.14) than that of higher rank males (1.92 ± 0.12; t₁₀ = 5.715, p < 0.001).