William J. Brown Professor Carnegie Mellon University, United States
Introduction: Plants are intriguing living systems, possessing remarkably different morphology to mammals, yet they have still evolved ways to defend themselves, consume prey, communicate, and in the case of plants like Mimosa pudica even move in response to a variety of stimuli. The complex physiological pathways driving this are of great interest, though many questions remain. In this work we mechanically stimulated a known responsive plant, M. pudica, in terms of wounding via removal of pinnae, nonwounding mechanical poking, and nonwounding pulses of air through a designed small nozzle approach. A kinematic model was developed to quantify leaflet folding, and measured data was fed into this model. The results revealed distinct differences in the behavior of leaflets across the pinna between stimuli. Removal of clusters called pinnae resulted in rapid, asymmetric response in the adjacent pinnae, while mechanical poking and air pulse responses were slower and more localized. Additionally, while the response from poking propagated across the plant, wind stimuli consistently resulted in the actuation of only the leaflets directly stimulated, suggesting unique sensing mechanisms. These results suggest an evolutionary benefit to having unique responses to different stimuli. Mechanical damage may imply a potential predator, while mechanical stimulation from air flow may be processed as wind, which is of little danger. These findings demonstrate an intricate, stimulant-dependent mechanical sensing process, which is important in plant physiology, mechanobiology, and future biohybrid soft robotic designs.
Materials and
Methods: In this work we captured video footage of M. Pudica’s response to wounding via removal of pinnae (lab scissors), nonwounding mechanical poking by blunt metal tip (linear actuation), and nonwounding pulses of air through a small nozzle (pneumatic pulse). The ends and base of each leaflet were optically tracked in the Physlets Tracker software to measure the motion of response. A kinematic model was developed to quantify leaflet folding from this tracked data.
Removal was done using a small pair of stainless-steel scissors, with the cut performed at the base of the rachis near the secondary pulvinus. One of the two central pinnae was removed, and then the adjacent exterior pinnae, left isolated, was analyzed. Poking was accomplished using a custom linear actuator mounted to the end of a heavy-duty gooseneck tablet holder for stability and free positioning. An AWG23 jumper wire was used for the prodding tip, secured with adhesive. The tip was positioned ~1cm from the rear of the distal half of leaflet. Once positioned, we moved the tip forward at a rate of 4.3mm/s for ~10mm to poke the leaflet. After this, another unactuated leaflet was poked until a cascade response was observed. Air pulse was performed using a solenoid valve connected to an air compressor and controlled with an Arduino microcontroller. 50ms duration pulses of air at 6kPa were passed through a 25Ga blunt-tipped syringe positioned ~1cm from the rear of the distal half of leaflet. Leaflets were stimulated until most were fully folded.
Results, Conclusions, and Discussions: Our findings allowed us to consider plants’ adaptations to predators relative to mechanical stimulation. Comparing blunt-tip and air-pulse stimuli revealed dramatically different behaviors—while poking contact produced an orderly, regular propagation of signal along the plant. Air pulses results were much more irregular and never achieved such propagation with only directly stimulated leaflets actuating. This suggests unique triggering mechanisms for the two stimuli. One explanation of this behavior could be that it provides an evolutionary advantage, allowing M. Pudica to only close the parts of itself that are being exposed to strong winds, as the parts that are not experiencing wind are not in danger unless the wind changes. This contrasts with wounding and solid touch stimulus, which may damage or touch the plant again.
Experiments also revealed new behaviors or intricacies of existing behaviors previously unrecorded. The response observed in the poking experiment is interesting as it resembled the variation potential, or “wounding” signal, however no samples were wounded to produce a similar response. This suggests that variation potential generation could occur from unique sensory organs, or from buildup of signaling molecules/ions from stimulation of multiple proximate leaflets in rapid succession. Additionally, the removal experiment demonstrated an asymmetrical response across the halves of leaflet clusters that suggests additional complexity of interaction between the proximate halves of adjacent leaflet clusters. While independent behavior across halves of these clusters has been observed (Bose Nature 1926), our work shows a possible connection between halves of entirely separate clusters.
Observations of spatiotemporal responses help support findings related to signaling mechanisms in M. Pudica due to mechanical stimulation. Houwink characterizes signaling in M. Pudica into three pathways: 1)The fast, non-wounding, location action potential 2)The slower, wound-triggered, propagating variation potential and 3)The rapid, long distance, extreme wound-triggered “unknown third signal” (Houwink 1935). Our experiments reproduce these described behaviors e.g., localized actuation followed by propagation in the poking experiments. Additionally supported by our removal experiments is the rapid response of the unknown third signal pathway, as well as this pathway’s ability to bypass parts of the plant and send long-distance signals.
.jpg "Philip R. R. LeDuc, PhD photo")