Preliminary Study On Plasticity Of Cultured Neuronal Networks

Neuronal plasticity is essential for learning and memory, and is one of the hot issuesin neuroscience. Studies on network plasticity have more significance as integration ofinformation occurs at network-level. Long-term potentiation and long-term depressionhave both been found by single-cell patch clamp. However, due to space limits, it isdifficult to record multi-site activities simultaneously using the patch clamp technique. Onthe other hand, studies employing EEG, MRI, etc., show the macroscopical activities. Butthese results cannot reveal the cell and molecular mechanism of neural informationprocessing. In this thesis, network-level plasticity was investigated on dissociatedneuronal networks cultured on multi-electrode arrays. Plasticity model was induced byelectrical stimulation, and then evaluated.Based on learning and establishing long time cultured hippocampal neuronalnetworks in vitro, two active sites with synchronous firing activities, named S1and S2,were selected by analyzing firing rate and cross correlation of spontaneous activity. Theywere then stimulated by varied amplitudes pulses at low frequency in a random order todetermine suitable stimulation parameters for Dual-site Plasticity experiments. After that,S1and S2were trained in turn with a5ms interval, at high frequency repeatedly, calledDual-site Time-interval High-frequency Training. Firing activities on S2within100msinduced by low-frequency stimulation on S1were analyzed before and after training.Transfer pathway was constructed by spatio-temporal characteristics of responses. Wefound that, in56percent of experiments, the onset of the first spike on S2became earlier,and moreover, spikes on S2were more when S1was tested after training than those before,and also the transfer pathway was changed. Actually, the results were found to be variableamong experiments on different cultures or with different electrodes trained. Furthermore,Triple-site Plasticity Model was constructed based on Time-interval High-frequencyTraining on three sites (S1, S2and S3). Induced firing activities on S3by low-frequencystimulation on S1were analyzed. Results were discovered to be similar with changes ofS2in Dual-site Plasticity Model in53percent of experiments.In summary, neuronal plasticity on the network-level was investigated on cultured neuronal networks. Construction and evaluation of the plasticity model were realizedpreliminary by practical design of experimental scheme and data processing protocol.Results showed that information transmission among two or three sites were refined by theinterval-mode repetitive training. This provides a simple and reliable model for studies ofnetwork plasticity, and would also be potentially important for studies of plasticitymechanism and therapy of nerve diseases.