Chemical communication in tilapia: A comparison of Oreochromis mossambicus with O. niloticus

•Mozambique and Nile tilapia have high olfactory sensitivity to each-other's urine.•Male Mozambique and Nile tilapia have the same urinary steroid 3α-glucuronides.•Mozambique and Nile tilapia have similar olfactory sensitivity to these steroids.•Mozambique and Nile tilapia are probably using the same chemical cues.•Geographical isolation does not select for differentiation of chemical cues.

In allopatric speciation species differentiation generally results from different selective pressures in different environments, and identifying the traits responsible helps to understand the isolation mechanism(s) involved. Male Mozambique tilapia (Oreochromis mossambicus) use urine to signal dominance; furthermore, 5β-pregnane-3α,17,20β-triol-3α-glucuronide (and its α-epimer, 5β-pregnane-3α,17,20α-triol-3α-glucuronide), in their urine is a potent pheromone, the concentration of which is correlated with social status. The Nile tilapia (Oreochromis niloticus) is a close relative; species divergence probably resulted from geographical separation around 6 million years ago. This raises the question of whether the two species use similar urinary chemical cues during reproduction. The olfactory potency of urine, and crude extracts, from either species was assessed by the electro-olfactogram and the presence of the steroid glucuronides in urine from the Nile tilapia by liquid-chromatography/mass-spectrometry. Both species showed similar olfactory sensitivity to urine and respective extracts from either species, and similar sensitivity to the steroid glucuronides. 5β-Pregnan-3α,17α,20β-triol-3α-glucuronide was present at high concentrations (approaching 0.5 mM) in urine from Nile tilapia, with 5β-pregnan-3α,17α,20α-triol-3α-glucuronide present at lower concentrations, similar to the Mozambique tilapia. Both species also had similar olfactory sensitivity to estradiol-3-glucuronide, a putative urinary cue from females. Together, these results support the idea that reproductive chemical cues have not been subjected to differing selective pressure. Whether these chemical cues have the same physiological and behavioural roles in O. niloticus as O. mossambicus remains to be investigated.