Studies in Trypsin as an Alarm Substance in Zebrafish

Previous studies have shown that fish release alarming substances into the water to alert their kin to escape from danger. In our laboratory, we found that zebrafish produce trypsin and release it from their gills into the environment when they are under stress. By placing the zebrafish larvae in the middle of a small tank and then placing trypsin at one end of the tank, we observed that the larvae moved away from the trypsin zone and almost to the opposite end of the tank. This escape response was significant and did not occur in response to the control substances, bovine serum albumin (BSA), Russell's viper venom (RVV), and collagen. Also, previously, we had shown that the trypsin could act via a protease-activated receptor-2 (PAR2) on the surface of the cells. Therefore, we hypothesized that trypsin would induce a change in neuronal activity in the brain via PAR2-mediated signaling in cells on the surface of the fish body. To investigate whether the trypsin-responsive cells were surface cells, we generated a primary cell culture of zebrafish keratinocytes, confirmed these cells' identity by specific marker expression, and then incubated these cells with the calcium indicator Fluo-4 and exposed them to trypsin. By using calcium flux assay in a flow-cytometer, we found that trypsin-treated keratinocytes showed an increase in intracellular calcium release. To test whether PAR2 mediates the escape response to trypsin, we treated larvae with a PAR2 antagonist and showed that the trypsin-initiated escape response was abrogated. Furthermore, par2a mutants with knockdown of par2a by the piggyback knockdown method failed to respond to trypsin. Trypsin treatment of adult fish led to an approximately 2-fold increase in brain c-fos mRNA levels 45 mins after trypsin treatment, suggesting that trypsin signals may have reached the brain, probably via a spinothalamic pathway. Taken together, our results reveal a novel trypsin-initiated escape response in fish. These studies should enhance our understanding of fish communication in general and alarm behavior in particular. Furthermore, since pain receptors in other animals are also PAR2, our finding may be useful in exploring pathways of pain reception