In this experiment, three cats walked freely in four different conditions (walking on a flat surface in the dark, walking on a flat surface in the light, along a horizontal ladder, and a stone-cluttered pathway) while gaze was recorded. Four gaze behaviors were identified based upon head and eye velocity parameters relative to the walking velocity of the cat: constant gaze, fixation, gaze shift away, and gaze shift toward (see Methods). The objective of the study was to determine whether speed influences the phase that these gaze behaviors occur, where phase is defined as the degree from 0-360 of the step cycle. In the step cycle, 0 degrees is defined as the start of swing of the right forelimb. Additionally, speed’s influence on the uniformity of gaze behaviors to the step cycle was investigated in the three cats. The cats performed complex walking tasks, or conditions, as well as simple tasks to determine if speed has a greater effect on gaze behavior timing when walking terrain was difficult. I hypothesized that 1) gaze-stride coordination would be influenced by speed, 2) faster steps would show improved gaze behavior uniformity between subjects, and 3) fast steps during complex walking tasks would show further improvement of gaze behavior uniformity between subjects. To, this end, recorded steps were first split into fast and slow steps based upon step duration parameters (see Methods). These fast and slow steps were confirmed as significantly different from one another using a one-way ANOVA test on a linear mixed effects model (Table 3). Then, a linear mixed effects model was made per walking condition to account for subject effects, and a two-way ANOVA test was performed on the model to compare the phases of gaze behaviors to the speed when they occurred. It was found that speed does not influence the phase that gaze behaviors occur, except for walking on a flat surface in the dark. However, post-hoc tests could not be run to determine which behaviors were affected by speed. (see Discussion). The insignificance of speed suggests that speed is accounted for by the visual center responsible for the control of gaze behavior (see Discussion). Aside from speed’s influence on phase, uniformity was examined using standard deviation (Figure 3 ). It was found that faster steps tend to adopt a “gaze stepping” behavior described in a previous paper (Rivers et al. 2014). In future studies, it would be useful to increase the number of subjects for a similar experiment to improve the robustness of the results to determine if the relationship between speed and gaze behaviors reported in this paper is accurately depicted.

Intracellular voltage recordings from single neurons in vitro and in vivo have been fundamental to our understanding of neuronal function. Conventional electrodes and associated positioning systems for intracellular recording in vivo are large and bulky, which has largely restricted their use to single-channel recording from anesthetized animals. Further, intracellular recordings are very cumbersome, requiring a high degree of skill not readily achieved in a typical laboratory. This dissertation presents a robotic, head-mountable, MEMS (Micro-Electro-Mechanical Systems) based intracellular recording system to overcome the above limitations associated with form-factor, scalability and highly skilled and tedious manual operations required for intracellular recordings. This system combines three distinct technologies: 1) novel microscale, polycrystalline silicon-based electrode for intracellular recording, 2) electrothermal microactuators for precise microscale navigation of the electrode and 3) closed-loop control algorithm for autonomous movement and positioning of electrode inside single neurons. First, two distinct designs of polysilicon-based microscale electrodes were fabricated and tested for intracellular recordings. In the first approach, tips of polysilicon microelectrodes were milled to nanoscale dimensions (<300 nm) using focused ion beam (FIB) to develop polysilicon nanoelectrodes. Polysilicon nanoelectrodes recorded >1.5 mV amplitude, positive-going action potentials and synaptic potentials from neurons in the abdominal ganglion of Aplysia Californica. In the second approach, polysilicon microelectrodes were integrated with miniaturized glass micropipettes filled with electrolyte to fabricate glass-polysilicon microelectrodes. These electrodes consistently recorded high fidelity intracellular potentials from neurons in the abdominal ganglion of Aplysia Californica (Resting Potentials < -35 mV, Action Potentials > 60 mV) as well as the rat motor cortex (Resting Potentials < -50 mV). Next, glass-polysilicon microelectrodes were coupled with microscale electrothermal actuators and controller for autonomous intracellular recordings from single neurons in the abdominal ganglion. Consistent resting potentials (< -35 mV) and action potentials (> 60 mV) were recorded after each successful penetration attempt with the controller and microactuated glass-polysilicon microelectrodes. The success rate of penetration and quality of recordings achieved using electrothermal microactuators were comparable to that of conventional positioning systems. Finally, the feasibility of this miniaturized system to obtain intracellular recordings from single neurons in the motor cortex of rats in vivo is also demonstrated. The MEMS-based system offers significant advantages: 1) reduction in overall size for potential use in behaving animals, 2) scalable approach to potentially realize multi-channel recordings and 3) a viable method to fully automate measurement of intracellular recordings.

This study was conducted to examine the potential effects of exercise training on partial spinal cord injury on locomotor recovery in juvenile rats. Three groups were tested, where three female Long-Evans rats 10-12 weeks of age were studied for their locomotion. All animals underwent a T8-T9 laminectomy and two of the three in each group received a dorsal, partial spinal cord injury. Locomotion was then analyzed every week, over 8-10 weeks. One of the two injured animals was given open access to a wheel after 2 weeks for voluntary exercise training. The results of this study suggested that injured animals displayed more irregular stepping patterns, larger hindlimb bases of support, greater and more variable interpaw distances, slower hindlimb speed, and increased dependency of swing-phase duty cycle on hindlimb speed. Trained animals displayed quicker recovery of stepping patterns, stepping of the hindpaw in relation to the preceding ipsilateral forepaw, and higher swing-duty cycle dependency on hindlimb speed in comparison to injured animals that did not receive exercise training. Due to a small sample size, there was a large amount of variation between individual animals in most parameters. These results are considered to be potential effects that may be seen in further study with a larger sample size. The research team will continue the research project to examine changes in neural pathways in the spinal cord and the effects of exercise on recovery after injury.

The Erk/MAPK pathway plays a major role in cell growth, differentiation, and survival. Genetic mutations that cause dysregulation in this pathway can result in the development of Rasopathies, a group of several different syndromes including Noonan Syndrome, Costello Syndrome, and Neurofibromatosis Type-1. Since these mutations are germline and affect all cell types it is hard to differentiate the role that Erk/MAPK plays in each cell type. Previous research has shown that individual cell types utilize the Erk/MAPK pathway in different ways. For example, the morphological development of lower motor neuron axonal projections is Erk/MAPK-independent during embryogenesis, while nociceptive neuron projections require Erk/MAPK to innervate epidermal targets. Here, we tested whether Erk/MAPK played a role in the postnatal development of lower motor neurons during crucial periods of activity-dependent circuit modifications. We have generated Cre-dependent conditional Erk/MAPK mutant mice that exhibit either loss or gain of Erk/MAPK signaling specifically in ChAT:Cre expressing lower motor neurons. Importantly, we found that Erk/MAPK is necessary for the development of neuromuscular junction morphology by the second postnatal week. In contrast, we were unable to detect a significant difference in lower motor neuron development in Erk/MAPK gain-of-function mice. The data suggests that Erk/MAPK plays an important role in postnatal lower motor neuron development by regulating the morphological maturation of the neuromuscular junction.

Cell morphology and the distribution of voltage gated ion channels play a major role in determining a neuron's firing behavior, resulting in the specific processing of spatiotemporal synaptic input patterns. Although many studies have provided insight into the computational properties arising from neuronal structure as well as from channel kinetics, no comprehensive theory exists which explains how the interaction of these features shapes neuronal excitability. In this study computational models based on the identified Drosophila motoneuron (MN) 5 are developed to investigate the role of voltage gated ion channels, the impact of their densities and the effects of structural features.

First, a spatially collapsed model is used to develop voltage gated ion channels to study the excitability of the model neuron. Changing the channel densities reproduces different in situ observed firing patterns and induces a switch from resonator to integrator properties. Second, morphologically realistic multicompartment models are studied to investigate the passive properties of MN5. The passive electrical parameters fall in a range that is commonly observed in neurons, MN5 is spatially not compact, but for the single subtrees synaptic efficacy is location independent. Further, different subtrees are electrically independent from each other. Third, a continuum approach is used to formulate a new cable theoretic model to study the output in a dendritic cable with many subtrees, both analytically and computationally. The model is validated, by comparing it to a corresponding model with discrete branches. Further, the approach is demonstrated using MN5 and used to investigate spatially distributions of voltage gated ion channels.