Hypoxia increases the behavioural activity of schooling herring: a response to physiological stress or respiratory distress?

Abstract

Atlantic herring, Clupea harengus, increase their swimming speed during low O₂ (hypoxia) and it has been hypothesised that the behavioural response is modulated by the degree of “respiratory distress” (i.e. a rise in anaerobic metabolism and severe physiological stress). To test directly whether a deviation in physiological homeostasis is associated with any change in behavioural activity, we exposed C. harengus in a school to a progressive stepwise decline in water oxygen pressure ${\left( {P_{{{\text{O}}_{2} }} = 20.4,15.2,12.7,10.6,8.5,6.4\;{\text{and}}\;4.2\,{\text{kPa}}} \right)}$ and measured fish swimming speed and valid indicators of primary and secondary stress (i.e. blood cortisol, lactate, glucose and osmolality). Herring in hypoxia increased their swimming speed by 11–39% but only when $P_{{{\text{O}}_{2} }}$ was <8.5 kPa and in an unsteady (i.e. declining) state. In parallel with the shift in behaviour, plasma cortisol also exhibited an increase with $P_{{{\text{O}}_{2} }} \le 8.5\,{\text{kPa,}}$ plasma osmolality was subject to a transient rise at 8.5 kPa and plasma glucose was generally reduced at $P_{{{\text{O}}_{2} }} \le 12.7\,{\text{kPa}}{\text{.}}$ However, without any rise in anaerobically derived lactate levels, there was no evidence of respiratory distress at any set $P_{{{\text{O}}_{2} }} .$ We show that a shift in physiological homeostasis is indeed linked with an increase in the swimming speed of herring but the physiological response reflects a hypoxia-induced shift in metabolic fuel-use rather than respiratory distress per se. The significance of this behavioural–physiological reaction is discussed in terms of behavioural-energetic trade-offs, schooling dynamics and the hypoxia tolerance of herring.