Synaptic and circuit mechanisms of olfactory processing

描述（由申请人提供）：感觉处理的早期阶段对各种感觉模式提出了类似的问题，例如，控制增益和使噪声最小化。目前还不清楚解决这些问题的计算是如何在细胞和突触层面实现的。早期的嗅觉系统是一个有用的准备调查这些问题，因为它的隔室结构。所有表达相同气味受体的嗅觉受体神经元（ORN）聚集在同一个隔室（肾小球），并且肾小球通过抑制性和兴奋性侧连接连接。这种架构提出了具体的问题，在体内功能的突触相互作用在这个电路。具体来说，为什么这么多的ORN聚集在每个肾小球上？突触后神经元如何在时域中整合收敛的ORN尖峰？当肾小球之间的连接被破坏时会发生什么？不同的肾小球处理通道对它们的前馈输入执行不同的计算吗？为什么局部神经元（local neurons，LN）之间存在如此大的差异？为什么同时具有兴奋性和抑制性淋巴结会有用？这些问题将使用有针对性的遗传操作，在体内全细胞记录，钙成像在果蝇触角叶。果蝇触角叶是解决这些问题的一个很好的模型，因为它与脊椎动物的同源物嗅球非常相似。此外，它可以进行目前在其他制剂中不可能进行的实验。具体而言，它是可能的遗传操纵特定的突触，验证这些扰动的细胞相关性，使用在体内细胞内电生理测量，并检查这些扰动的功能后果的完整电路。在这些研究中，将使用遗传工具（1）操纵ORN的会聚和一致性，（2）选择性地消除侧向兴奋、侧向突触前抑制和侧向突触后抑制，（3）刺激特定的LN群体，以及（4）监测和操纵侧向抑制的空间扩散。这些操作中的大多数都是由于最近对该系统生物学的发现而成为可能的。这些研究中的所有技术都在实验室中常规使用，因此证明了其可行性。这些研究的结果应阐明在体内感觉回路突触整合的一般原则。 也就是说，这些研究应该有助于阐明神经回路如何最大化其信噪比，不同的电路模块如何执行专门的计算，为什么局部中间神经元如此多样化，以及为什么侧兴奋和侧抑制经常共存。更具体地说，这些研究应该澄清嗅觉处理的突触基础，在脊椎动物以及在简单的模式生物。了解大脑如何处理气味有助于设计所谓的“人工鼻子”，即设计用于分析有机挥发物的传感器，这些传感器在医学诊断中有重要应用。

DESCRIPTION (provided by applicant): The early stages of sensory processing pose similar problems for various sensory modalities-for example, controlling gain and minimizing noise. It is not clear how the computations that solve these problems are implemented at the level of cells and synapses. The early olfactory system is a useful preparation for investigating these issues because of its compartmental architecture. All the olfactory receptor neurons (ORNs) that express the same odorant receptor converge on the same compartment (glomerulus), and glomeruli are linked by both inhibitory and excitatory lateral connections. This architecture raises specific questions about the in vivo function of synaptic interactions in this circuit. Specifically, why do so many ORNs converge on each glomerulus? How do postsynaptic neurons integrate converging ORN spikes in the time domain? What happens when the connections between glomeruli are abolished? Do different glomerular processing channels perform different computations on their feedforward inputs? Why is there such diversity among local neurons (LNs)? Why would it be useful to have both excitatory and inhibitory LNs? These questions will be addressed using targeted genetic manipulations, in vivo whole-cell recordings, and calcium imaging in the Drosophila antennal lobe. The Drosophila antennal lobe is a good model for addressing these questions because it bears a strong similarity to its vertebrate homolog, the olfactory bulb. Moreover, it enables experiments that are currently not possible in other preparations. Specifically, it is possible to genetically manipulate specific synapses, validate the cellular correlates of these perturbations using in vivo intracellular electrophysiological measurements, and examine the functional consequences of these perturbations for the intact circuit. In these studies, genetic tools will be used (1) to manipulate ORN convergence and coherence, (2) to selectively abolish lateral excitation, lateral presynaptic inhibition, and lateral postsynaptic inhibition, (3) to stimulate specific LN populations, and (4) to monitor and manipulate the spatial spread of lateral inhibition. Most of these manipulations were made possible by recent discoveries about the biology of this system. All the techniques in these studies are routinely used in the laboratory, and thus their feasibility is proven. The results of these studies should illuminate general principles underlying synaptic integration in sensory circuits in vivo. Namely, these studies should help clarify how neural circuits can maximize their signal-to-noise ratio, how different circuit modules might perform specialized computations, why local interneurons are so diverse, and why lateral excitation and lateral inhibition often co-exist. More specifically, these studies should clarify the synaptic basis of olfactory processing, in vertebrates as well as in simpler model organisms. Understanding how the brain processes odors should aid the design of so-called "artificial noses", sensors designed to analyze organic volatiles which have important applications in medical diagnosis.