What are the neurobiological mechanisms mediating operant self-learning?

在操作性学习中，涉及两种进化上保守的机制，一种是主体学习环境中的因果关系（世界学习），另一种是主体学习自己行为的结果（自我学习）。介导世界学习的神经生物学机制已被很好地理解：它们在进化上是保守的，因此几乎与许多其他学习实验（如经典条件反射）中的机制相同。然而，我们对操作性自我学习的神经生物学知之甚少。在前一个资助期成功结合分子生物学和分离操作性自学习成分的行为实验的基础上，我们提出了一个类似的项目，以进一步阐明涉及操作性自学习的基因和神经元。先前的资助期提供的结果表明，foxp依赖的可塑性并不需要在大脑中，而是在腹侧神经索的运动神经元中，这些运动神经元支配着特定的直接翼操纵肌。我们还发现非典型蛋白激酶C （aPKC）的基因产物在foxp阳性神经元中是必需的。这个资助期的一个核心研究问题将是确定FoxP/aPKC通路的更多成分，这些成分介导运动神经元的可塑性，这是操作性自我学习的基础。为此，我们将使用DamID和TurboID两种分子技术。使用DamID，我们将确定FoxP结合和调节的靶基因。使用TurboID，我们将识别aPKC与之相互作用的靶蛋白。这两种技术中最有希望的候选基因将被操纵和测试用于操作性自学习。除了这些分子技术，我们还从文献中生成了候选基因列表，以确定FoxP/aPKC以外的可塑性机制。最后，我们还将与北京的李刘教授合作，验证他在屏幕中确定的参与操作活动的脑神经元，这是导致自我学习的过程。刘的实验室已经在大脑中发现了操作性自我学习所必需的神经元，无论它们是否表达FoxP。在我们的合作中，我们将独立地复制这个实验室的结果，而北京的实验室将复制我们最有希望的候选系实验。

In operant learning, two evolutionarily conserved mechanisms are engaged, one where the subject learns about causal relations in the environment (world-learning) and another where the subject learns about the outcomes of their own behavior (self-learning). The neurobiological mechanisms mediating world-learning are well understood: they are evolutionarily conserved and thus nearly identical to those in many other learning experiments such as classical conditioning. However, we know very little about the neurobiology of operant self-learning. Building on the previous funding period’s successful combination of molecular biology and a behavioral experiment that isolates the operant self-learning component, we propose an analogous project to further elucidate the genes and neurons involved in operant self-learning. The previous funding period provided results that suggest that FoxP-dependent plasticity is not required in the brain but in motor neurons in the ventral nerve cord that innervate specific direct wing steering muscles. We also discovered that the gene product of the atypical protein kinase C (aPKC) is required in FoxP-positive neurons. One core research question for this funding period will be to identify more components of the FoxP/aPKC pathway mediating the plasticity in motor neurons that underlies operant self-learning. To this end, we will use two molecular techniques, DamID and TurboID. With DamID we will identify the target genes that FoxP binds to and regulates. With TurboID, we will identify the target proteins that aPKC interacts with. The most promising candidate genes from these two techniques will be manipulated and tested for operant self-learning. In addition to these molecular techniques, we have generated a list of candidate genes from the literature, in order to identify plasticity mechanisms beyond FoxP/aPKC. Finally, in cooperation with Prof. Dr. Li Liu in Beijing, we will also validate the candidates identified in his screen for brain neurons involved in operant activity, the process that leads to self-learning. The Liu laboratory has already discovered neurons in the brain that are required for operant self-learning, irrespective of whether they express FoxP or not. In our collaboration, we will independently replicate the results from this laboratory and the laboratory in Beijing will replicate our most promising candidate line experiments.