Development of optogenetic approaches to modulate synaptic plasticity in the postsynaptic cells

Over the last few decades, learning and memory processes have been extensively studied from a variety of perspectives. Long-term potentiation (LTP), the long lasting strengthening of synaptic communications, is a key cellular event associated with learning at the organism level. Even now, 45 years since the first reported LTP recording, a precise molecular mechanism of LTP is still not fully elucidated. Understanding the cellular basis of learning and memory formation is important for the development of novel therapeutic approaches for patients with Alzheimer's disease or post-traumatic stress disorder, or even development of effective education curricula. LTP and its connection with learning have been studied using genetic, pharmaceutical and electrophysiological methods that provided valuable understanding of these processes. An optogenetic approach, that can manipulate the specific biochemical signalling associated with LTP with higher temporal and spatial precision in a genetically defined neuronal population, can provide a better causal and correlative link between LTP and learning. The main goal of this thesis is to generate a new optogenetic tool, which can be ultimately used to influence LTP, utilizing our current knowledge of the molecular mechanisms involved in LTP. The cellular signalling, targeted in this thesis optogenetically, is the one associated with receptor tyrosine kinase B (TrkB), the receptor for brain-derived neurotrophic factor (BDNF). BDNF and TrkB are implicated in LTP and learning; therefore, optogenetic manipulation of these proteins could directly alter LTP and possibly animal behaviour. TrkB is a typical tyrosine kinase receptor that is located in the plasma membrane and activated through the receptor dimerization (mediated by the extracellular domain upon BDNF binding) and autophosphorylation (mediated by the intracellular kinase domain). TrkB activates three main signalling enzymes: phospholipase C-˜í‚â•1 (PLC˜í‚â•1), extracellular signal-regulated kinase (ERK) and AKT kinase. To trigger TrkB signalling in living cells upon blue light stimulation, the kinase domain of TrkB (TrkB(kd)) was attached to a photoreceptor, cryptochrome 2 (Cry2), as a freely diffusive cytosolic recombinant protein. Upon blue light illumination, Cry2 forms homo-oligomers and hetero-oligomers with its binding partner, CIB1. CIB1 was tethered to the plasma membrane. In this design, the intention was to trigger the oligomerization of TrkB(kd)-Cry2 and its recruitment to the membrane via Cry2/CIB1 heterodimerization upon blue light exposure. This led to the activation of TrkB(kd) and consequently initiated biochemical signalling cascades at the plasma membrane. In this thesis, the designed two-component TrkB system was shown to trigger PLC˜í‚â•1, ERK and AKT signalling pathways in non-neuronal HEK293 cells as well as in cultured neurons in a light-dependent manner. During the validation process, I discovered that, without the membrane-tethered CIB1 component, TrkB(kd)-Cry2 alone activated ERK and AKT signalling without elevation of intracellular Ca2\\(^+\\) that is related to the functional activation of PLC˜í‚â•1. Additionally, a mutation was introduced in TrkB(kd) that prevented activation of PLC˜í‚â•1, but still allowed ERK and AKT pathways to be triggered upon light stimulation. These two observations might have a practical application for dissecting biochemical signalling by TrkB, which is still not completely understood in LTP. Preliminary experiments in DRG neurons demonstrated that varying the amount of light caused contradictory responses of a growth cone, attraction or repulsion, when TrkB(kd)-Cry2 and CIB1 were used. The TrkB(kd)-Cry2 component alone was not able to change the growth cone turning in DRG neurons. These results suggest important roles of PLC˜í‚â•1 and Ca2\\(^+\\) in axon steering over short timescales (minutes), but minimal effects of ERK and AKT on these timescales. Finally, the TrkB system was successfully introduced into the amygdala region of a freely behaving rat. In this study, the process of designing and validating the optogenetic TrkB tool is presented as a part of broad project that ultimately aims to manipulate LTP and learning. This new photoactivatable TrkB system has several advantages over previously published optogenetic tools, which make it unique for in vivo applications in subsequent studies. This tool (1) can be packaged into recombinant adeno-associated viral vectors (AAV), (2) cannot be activated by endogenous BDNF due to the lack of TrkB extracellular domain, (3) provide separate controls of different biochemical pathways when used without membrane-tethered CIB1 or when specific mutation is introduced into TrkB(kd). Together this provides an exciting new approach to investigate how manipulation of BDNF/TrkB signalling affects learning and memory in awake behaving animals.