Functional and Behavioral Trade-Offs between Terrestrial and Aquatic Locomotion in the Amphibious Fish Kryptolebias marmoratus

Abstract

Synopsis The mangrove rivulus (Kryptolebias marmoratus) is a phenotypically plastic teleost fish that can spend considerable time on land and traverse the terrestrial realm through a behavior termed the tail-flip jump. The tail-flip jump is a transitional stage between fully aquatic and terrestrial lifestyles. Therefore, understanding this behavior can provide insight into how organisms adapt to new environments over evolutionary time. Studies of K. marmoratus show that terrestrial acclimation and exercise improve tail-flip jumping performance due to muscle remodeling, but the implications of these muscular changes on aquatic locomotion are unknown. In the present study, we hypothesized that (1) terrestrial acclimation and exercise lead to physiological changes, such as changes to muscle fiber type, muscle mass distribution, or body shape, that optimize tail-flip jump distance and endurance while negatively impacting swimming performance in K. marmoratus, and (2) plasticity of the brain (which has been demonstrated in response to a variety of stimuli in K. marmoratus) allows terrestrial emersion and exercise to cause behavioral changes that promote survival and long-term reproductive success. To test these hypotheses, we measured the critical swimming speed (Ucrit), tail-flip jump distance, terrestrial endurance, and undisturbed aquatic behavior of age- and size-matched K. marmoratus before and after a terrestrial exercise period. This period consisted of six 3-min exercise sessions spread over 12 days, during which the fish were prompted to jump continuously. To isolate the effects of air exposure, a separate group was exposed to air for an equivalent period but not allowed to jump. Air exposure improved maximum jump distance but negatively affected swimming performance (Ucrit). Terrestrial endurance (number of jumps) improved in the exercised group, but Ucrit showed no significant change. Contrary to our first hypothesis, a trade-off exists between jump distance and Ucrit but not between jump endurance and Ucrit. Exercised individuals were more active following exercise, resulting either from the onset of dispersion behavior or a heightened stress response.