Do Real Stimuli Better Evoke Backpropagation and LTP?

神经元动作电位是沿着轴突传播以诱导突触释放化学神经递质的电信号。然而，动作电位也会反向传播到树突，即神经元的分支输入结构。尽管反向传播的作用尚未完全理解，但它无疑对树突信号处理产生重要影响。了解树突信号处理对于充分理解正常学习和记忆至关重要，其中包括树突和树突输入的变化，以及涉及树突功能障碍的疾病，例如癫痫、某些形式的智力低下、帕金森病等。该提案探讨了一种可能性，即与通常用于实验诱导的人工规则尖峰训练相比，具有现实、不规则时序的尖峰训练在生成反向传播和长期增强（LTP）（学习和记忆的主要模型）方面可能更有效。最初的实验将测试海马 CAI 锥体神经元中刺激的哪种类型的尖峰序列能够更有效地反向传播到树突：取自体内实验的真实的、不规则的尖峰序列或具有相同总体频率的规则尖峰序列。将测试尖峰序列的各种频率分量在反向传播效率中的作用。接下来，CA1 树突的钙成像将确定实际输入是否会比人工输入导致更多的钙流入，这是 LTP 诱导所必需的。最后，通过将阈下突触刺激序列与 CA1 锥体体中匹配的尖峰序列配对，在 Schaffer 侧支 CA1 突触处生成 LTP。将比较真实的、不规则的和人工的、规则的尖峰训练引起的 LTP 量。这些实验将检验生理反循环比常用的常规刺激更有效并且更容易支持 LTP 的假设。

Neuronal action potentials are electrical signals that travel along the axon to induce release of chemical neurotransmitters at synapses. However, action potentials also backpropagate into dendrites, the branching input structures of neurons. Though its role is not fully understood, backpropagation undoubtedly affects dendritic signal processing in important ways. Understanding dendritic signal processing is critical to a full understanding of normal learning and memory, which includes changes in dendrites and dendritic inputs, and diseases involving dendritic dysfunction such as epilepsy, some forms of mental retardation, Parkinson's disease, and others. This proposal explores the possibility that spike trains with realistic, irregular timing may be more effective at generating backpropagation and long term potentiation (LTP), the primary model of learning and memory, than the artificial regular spike trains that are generally used to induce it experimentally. Initial experiments will test which type of spike train stimulated in hippocampal CAI pyramidal neurons backpropagates more efficiently into the dendrite: real, irregular spike trains taken from in vivo experiments or regular spike trains of the same overall frequency. The role of various frequency components of the spike trains in backpropagation efficiency will be tested. Next calcium imaging of the CA1 dendrites will determine if real inputs result in more calcium influx, necessary to LTP induction, than artificial inputs. Lastly, LTP will be generated at the Schaffer collateral-CA1 synapse by pairing trains of subthreshold synaptic stimuli with matched spike trains in the CA1 pyramidal soma. The amount of LTP induced by real, irregular and artficial, regular spike trains will be compared. These experiments will test the hypothesis that physiological back ro a ation is more effective and supports LTP more readil than commonly-used regular stimuli.