This article investigates bioinspired solutions for achieving stable dynamic gaits in legged robots through leg coordination and foot trajectories. In this study, we recorded the kinematics of underwater running of the crab, Pachygrapsus marmoratus, and implemented the parameterized foot trajectories and inter-leg coordination on an underwater legged robot, SILVER 2.0. The robot's design parameters like legs’ stiffness, leg length, and body mass are based on the Underwater Spring Loaded Inverted Pendulum (USLIP), a model that describes underwater running in animals. With this implementation, we observed the spontaneous emergence of USLIP dynamics in 20% of the strides in the robot. This approach allowed SILVER 2.0 to leverage the advantages of stable dynamic gaits while optimizing the foot trajectory and inter-leg coordination, resulting in improved locomotion performances. The robot achieved a forward velocity of 0.16 m/s, twice the value obtained in previous gaits. Our study presents a promising approach for improving the locomotion performance of legged robots, enabling their effective use in various field applications, and further confirms a broad embedding of controllers generating template dynamics.