RESUBMISSION: Neural processing underlying auditory pattern recognition in an insect brain

神经科学的目标是了解感觉通路是如何组织的，刺激是如何处理的，以及如何提取特征来引发适当的运动反应。特别是在种内交流中，对特定刺激特征的识别起着重要作用，因为物种特有的嗅觉、视觉和/或声音信号被用于配偶吸引和竞争行为。由于声音总是瞬变的，在许多脊椎动物和无脊椎动物中，通过重复的脉冲模式发出信号是声音交流的核心。特别是青蛙和昆虫，如丛林蟋蟀和蟋蟀，它们使用简单的声音脉冲进行交流，这些脉冲是由物种特定的时间模式产生的。在接收者的听觉通路中，这些信号需要神经过滤机制来响应声音模式的时间结构。青蛙在描述选择性调谐到脉搏模式的脑神经元方面已经取得了相当大的进展。然而，尽管几十年来，蟋蟀是昆虫听觉的模型系统，但对蟋蟀声音模式的时间处理背后的神经元的了解仍然非常有限。雌性蟋蟀被雄性叫声的特定物种模式所吸引。在行为实验中，我们展示了它们的音向性行为对脉冲间隔、脉冲持续时间和啁啾间隔的变化的调节。例如，当周期较短或较长的脉冲不具吸引力时，它们的趋音行为就会根据男性歌曲的脉冲模式进行调整。我们现在的目标是在识别的神经元水平上了解大脑中允许识别特定物种歌曲模式的过滤机制。将叫声的脉冲模式从第一个胸神经节转发到大脑的上升听觉中间神经元没有过滤特性。关于大脑中时间过滤的假说已经被提出，然而，实际的神经机制还没有被揭示。在初步实验中，我们确定了局部脑神经元，它们在前脑形成环状听神经束，与上升中间神经元的轴突树枝相匹配。其中一些神经元对叫声的脉冲间隔表现出选择性反应，并接受抑制性和兴奋性突触输入。在其他系统中，抑制和激发之间的相互作用对于脉冲模式的选择性至关重要。使用与行为学研究相同的声学范式，我们将分析听觉脑神经元的活动。我们将重点解决四个问题：1)。哪些脑神经元参与了声脉冲模式的时间滤波？时间过滤的神经机制是什么？3)。听觉神经束的结构和功能是什么？4)哪些投射神经元将听神经束连接到其他脑区？这个项目将基于我们用尖锐的微电极在细胞内记录听觉脑神经元活动的经验。神经反应将与我们的行为数据进行比较，定量分析将揭示单个神经元的活动在多大程度上反映了行为调整。抑制性和兴奋性突触活动的模式将提供有关听觉网络中如何实现时间过滤的关键信息。我们将通过细胞内电流注入来操纵神经细胞的膜电位，以详细分析神经过滤机制的性质。最后，我们将使用共聚焦显微镜来揭示大脑中这些神经元的结构细节，并识别将听觉神经桩与大脑其他区域联系起来的神经元。对蟋蟀的模式识别网络的分析将在识别的神经元水平上提供对时间过滤的基本机制的洞察，并可以作为其他系统中时间选择性的模型。

Neuroscience aims at understanding how sensory pathways are organised, how stimuli are processed and how features are extracted to elicit an appropriate motor response. Especially in intraspecific communication the recognition of specific stimulus features plays an important role as species-specific olfactory, visual and/or acoustic signals are employed for mate attraction and rivalry behaviour. As sounds are always transient, signalling by repetitive patterns of pulses is central to acoustic communication in many vertebrates and invertebrates. Especially frogs and insects like bush-crickets and crickets use simple sound pulses generated with species-specific temporal patterns for communication. In the auditory pathway of the receiver, these signals require neural filter mechanisms that respond to the temporal structure of sound patterns. Considerable progress has been made in frogs in describing brain neurons with selective tuning to pulse patterns. However, there is still a very limited understanding of neurons underlying temporal processing of sound patterns in crickets although since decades they are a model system for insect hearing. Female crickets are attracted to the species-specific pattern of male calling song. In behavioural experiments we demonstrated the tuning of their phonotactic behaviour to changes in pulse interval, pulse duration and chirp interval. For example their phonotactic behaviour is tuned to the pulse pattern of the male song as pulses with shorter or longer periods are not attractive. We now aim to understand at the level of identified neurons the filter mechanisms in the brain that allow the recognition of the species-specific song pattern.The ascending auditory interneuron that forwards the pulse pattern of the calling song from the first thoracic ganglion to the brain exhibits no filter properties. Hypothesis for temporal filtering in the brain have been put forward, however, the actual neural mechanisms have not been revealed. In preliminary experiments we identified local brain neurons which form a ring-like auditory neuropil in the protocerebrum matching the axonal arborisations of the ascending interneuron. Some of these neurons showed a selective response to the pulse interval of the calling song and received inhibitory and excitatory synaptic inputs. In other systems the interaction between inhibition and excitation is crucial for selectivity to pulse patterns. Using the same acoustic paradigms as in the behavioural studies we will analyse the activity of auditory brain neurons. We will focus on four questions: 1). Which brain neurons are involved in temporal filtering of acoustic pulse patterns?2). What are the neural mechanisms underlying temporal filtering? 3). What is the structural and functional organisation of the auditory neuropil?4). Which projection neurons connect the auditory neuropil to other brain regions?The project will be based on our experience of recording intracellularly the activity of auditory brain neurons with sharp microelectrodes. The neural responses will be compared with our behavioural data and quantitative analysis will reveal to what degree the activity of single neurons mirrors the behavioural tuning. The pattern of inhibitory and excitatory synaptic activity will provide crucial information on how temporal filtering is achieved within the auditory network. We will manipulate the membrane potential of neurons by intracellular current injection to analyse the nature of the neural filter mechanisms in detail. Finally we will use confocal microscopy to reveal the structural details of these neurons in the brain and we will identify neurons which link the auditory neuropil to other areas of the brain. The analysis of the pattern recognition network in crickets will provide insight to principle mechanisms of temporal filtering at the level of identified neurons and can be a model for temporal selectivity in other systems.