| Popis: |
Animals exhibit a rich repertoire of locomotive behaviors. In the context of legged locomotion, i.e. walking, animals can change their heading direction, traverse diverse substrates with different speeds, or can even compensate for the loss of a leg. This versatility emerges from the fact that biological limbs have more joints and/or more degrees of freedom (DOF), i.e. independent directions of motions, than required for any single movement task. However, this further entails that multiple, or even infinitely many, joint configuration can result in the same leg stepping pattern during walking. How the nervous system deals with such kinematic redundancy remains still unknown. One proposed hypothesis is that the nervous system does not control individual DOFs, but uses flexible combinations of groups of anatomical or functional DOFs, referred to as motor synergies. Drosophila melanogaster represents an excellent model organism for studying the motor control of walking, not least because of the extensive genetic toolbox available, which, among others, allows the identification and targeted manipulation of individual neurons or muscles. However, their tiny size and ability for relatively rapid leg movements hampered research on the kinematics at the level of leg joints due to technical limitations until recently. Hence, the main objective of this dissertation was to investigate the three-dimensional (3D) leg joint kinematics of Drosophila during straight walking. For this, I first established a motion capture setup for Drosophila which allowed the accurate reconstruction of the leg joint positions in 3D with high temporal resolution (400 Hz). Afterwards, I created a kinematic leg model based on anatomical landmarks, i.e. joint condyles, extracted from micro computed-tomography scan data. This step was essential insofar that the actual DOFs of the leg joints in Drosophila were currently unknown. By using this kinematic model, I have found that a mobile trochanter-femur joint can best explain the leg movements of the front legs, but is not mandatory in the other leg pairs. Additionally, I demonstrate that rotations of the femur-tibia plane in the middle legs arise from interactions between two joints suggesting that the natural orientation of joint rotational axes can extent the leg movement repertoire without increasing the number of elements to be controlled. Furthermore, each leg pair exhibited distinct joint kinematics in terms of the joint DOFs employed and their angle time courses during swing and stance phases. Since it is proposed that the nervous system could use motor synergies to solve the redundancy problem, I finally aimed to identify kinematic synergies based on the obtained joint angles from the kinematic model. By applying principal component analysis on the mean joint angle sets of leg steps, I found that three kinematic synergies are sufficient to reconstruct the movements of the tarsus tip during stepping for all leg pairs. This suggests that the problem of controlling seven to eight joint DOFs can be in principle reduced to three control parameters. In conclusion, this dissertation provides detailed insights into the leg joint kinematics of Drosophila during forward walking which are relevant for deciphering motor control of walking in insects. When combined with the extensive genetic toolbox offered by Drosophila as model organism, the experimental platform presented here, i.e. the 3D motion capture setup and the kinematic leg model, can facilitate investigations of Drosophila walking behavior in the future. |
