目的:探讨培养的海马神经元网络对外界刺激信息的群体编码机制。方法:本文利用多电极阵列对培养海马神经元网络进行多位点刺激及同步记录,运用线性统计方法分析网络中各点对刺激的响应规律,以及聚类算法分辨网络对不同位置刺激的响应。结果:各记录位点在刺激后100ms内的响应放电频率与距刺激点的空间距离线性无关、与记录/刺激位点的自发活动相关系数呈弱线性相关,与刺激前100ms内的自发放电频率呈线性关系。进一步实验结果表明刺激不同位点时自发放电频率与响应放电频率线性关系的斜率和截距不同。不同位点刺激时的自发-响应数据能够通过聚类进行区分,并且自发-响应放电频率的线性相关系数越大分类的正确率越高。药理实验结果表明,该线性相关系数在APV阻断NMDA受体后减小,而在CNQX阻断AMPA受体后相关系数增大。结论:培养的海马神经元网络中的神经元对刺激的响应放电频率与其自发放电频率线性相关,该群体响应特征可以用来实现对刺激位置的编码,并且NMDA受体的存在是维系该群体响应特征的因素之一。
Objective:To investigate the mechanism of population coding underlying the stimulus information processing in hippocampal neuronal network. Method: Cultured hippocampal neuronal networks were stimulated and recorded from different sites on multi-electrode arrays. The response characterization of neuronal network was analyzed by linear statistic method, and the responses of network to stimuli of different location were analyzed by clustering algorithm. Results: The response firing rate within 100 ms after a stimulus was linearly uncorrelated with the distance, weakly linearly correlated with the correlation coefficient between spontaneous activities of response and stimulated sites, and linearly correlated with the spontaneous firing rate 100 ms before a stimulus. Furthermore, the slope and intercept of the linear correlation between response and spontaneous firing rate varied when the location of the stimulation changed, and spontaneous-response data could be districted by clustering algorithm. The correct rate of cluster was determined by the linear correlation coefficient of spontaneous-response firing rate, and the greater the coefficient, the higher the rate. The results of pharmacological experiments indicated that the coefficient of such correlation was decreased when NMDA receptors blocked by APV, but increased when AMPA receptors blocked by CNQX. Conclusions: The response of cultured hippocampal neurons represented mutual characterization, which could be used to code the stimulus location and was influenced by NMDA and AMPA receptors.