Dynamic properties of a locomotory muscle of the tobacco hornwormManduca sextaduring strain cycling and simulated natural crawling

Abstract

SUMMARYCaterpillars are soft-bodied terrestrial climbers that perform a wide variety of complex movements with several hundred muscles and a relatively small number of neurons. Control of movements is therefore expected to place unusual demands on the mechanical properties of the muscles. The muscles develop force slowly (1–6 s to peak) yet over a strain range extending from under 60% to more than 160% of resting length, with a length-tension relationship resembling that of supercontracting or cross-striated muscle. In passive and active sinusoidal strain cycling, muscles displayed viscoelastic qualities, with very low and stretch-velocity dependent resilience; there was a positive linear relationship between stretch velocity and the fraction of work dissipation attributable to passive muscle properties (20–80%). In linear stretches of unstimulated muscles at velocities bracketing those encountered in natural crawling, the rise in tension showed a distinct transition to a lower rate of increase, with transition tension dependent upon stretch velocity; peak force was exponentially related to stretch velocity. When stretching ceased, force decayed exponentially, with slower decay associated with lower stretch velocities; the decay time constant was exponentially related to stretch velocity. From the kinematics of caterpillars crawling horizontally we determined that the ventral interior lateral muscle(VIL) of the third abdominal segment (A3) is at or near resting length for most of the crawl cycle, with a fairly linear shortening by 25–30% and re-lengthening occupying about 45% of cycle duration. Synchronized kinematic and EMG recordings showed that during horizontal crawling A3 VIL is stimulated as the muscle shortens from about 95% to 75% of its resting length. We subjected in vitro VIL preparations to strain cycling and stimulus phase and duration similar to that of natural crawling. The resulting work loops were figure-eight shaped, with the muscle performing work during the shortest 45–65% of the strain cycle but dissipating work during the rest of the cycle. The muscle remained in the ascending limb of its length-tension relationship throughout the crawl cycle. Peak force occurred at the end of re-lengthening, nearly a full second after stimulation ceased, underscoring the importance of understanding passive muscle properties to explain caterpillar locomotion. Whether A3 VIL functions as an actuator at all during simulated natural strain cycling is highly sensitive to stimulus timing but far less so to stimulus duration. The muscle's elastomer-like properties appear to play a major role in its function.