Lecture Summaries

• Getting to know each other
• Overview and discussion of syllabus
• How to efficiently search the scientific literature
• Brief tips on how to read and analyze primary scientific literature
• Introductory lecture from the instructors about the course in general and the topic of the next session

Information processing in the brain requires precise synaptic connections between neurons. What mechanisms underlie synapse formation, and how are synaptic functional identities established? This week we will compare and contrast various experimental approaches to address these questions. The first paper examines how transynaptic adhesion molecules contribute to synapse formation and identity. Using cultured neurons, Chih et al. demonstrate that the transynaptic adhesion complex neurexin / neuroligin regulates synapse formation and controls the functional balance between excitatory and inhibitory synapse. The second paper uses the genetically tractable invertebrate C. elegans to identify a family of secreted factors that contribute to synapse formation. Interestingly, Pinan-Lucarre et al. demonstrate that specific combinations of these factors contribute to the functional identity of synapses by differentially clustering excitatory and inhibitory postsynaptic receptors. Surprisingly, the ability of these factors to dictate postsynaptic identity is independent of the identity of the presynaptic partner. What are the implications of this finding in how we think about synapse formation? We will also discuss how imbalances in excitatory / inhibitory input might contribute to brain disorders. We will end the session by considering the advantages and disadvantages of using simpler invertebrate organisms to investigate mechanisms of synaptic neurobiology.

In a landmark paper in 1973, Bliss and Lomo discovered long-term potentiation (LTP) in the rabbit hippocampus, a region of the brain that plays an important role in memory formation. This activity-dependent long-term enhancement of neurotransmission as well as activity-dependent long-term synaptic depression (LTD) are forms of synaptic plasticity thought to underlie development of brain connectivity and the encoding and storage of memory. What is the cellular locus (pre or post synapse) of this change in synaptic function, and what are the molecular mechanisms that mediate this synaptic plasticity? Our first paper investigates the molecular correlates of LTP. Hayashi et al. use cultured hippocampal slices to exogenously express the AMPA-type glutamate receptor (AMPAR) subunit GluRA1. AMPARs are neurotransmitter (in this case glutamate)-gated ion channels that mediate fast excitatory transmission at synapses. The authors demonstrate that LTP induction leads to synaptic membrane insertion of GluRA1-containing receptors, supporting a postsynaptic mechanism for neurotransmission enhancement. This insertion requires the C-terminal PDZ-ligand, a domain that associates with PDZ-domain containing proteins enriched in the postsynaptic density, such as membrane-associated guanylate kinases (MAGUKs). These results suggest a molecularly defined postsynaptic mechanism by which the membrane insertion of AMPARs, regulated by protein interactions between its C-terminal tail and intracellular proteins such as MAGUKs, leads to long-lasting changes in the efficacy of synapse function. In our second paper, Bender et al. investigate mechanisms of a form of synaptic plasticity that depends on the precise coincident timing of presynaptic and postsynaptic neuronal activity. In spike-time-dependent plasticity (STDP), synaptic plasticity can be expressed depending on the millisecond-scale timing and sequence of the pre- and postsynaptic neuronal action potentials (or spikes). More specifically, LTP is expressed when a presynaptic spike within a 20–40 millisecond time window precedes a postsynaptic spike, whereas LTD is expressed when the order of these events is reversed. Rather than a single synaptic locus for plasticity expression (e.g. postsynaptic insertion or removal of AMPARs for LTP and LTD, respectively), the authors demonstrate that the cellular expression mechanisms of LTP and LTD in STDP can be decoupled—one with origins that are presynaptic and the other postsynaptic. We will compare and contrast this form of synaptic plasticity to other forms of plasticity, which will be covered later in the course.

A fine balance of synaptic excitation and inhibition controls dynamic neuronal processes and behavior. It is, however, unclear which circuit elements, cell types and brain states control this balance. In this class, we will first explore how highly active interconnected cortical neuronal ensembles process sensory-driven information. Measurements of synaptic parameters and paired-recording between synaptically-connected cells to measure changes in synaptic excitation and inhibition will be introduced.

Feedforward excitation and feedback inhibition are regulated by distinct circuit elements that, in turn, control behavior. Donato et al. show how cellular subtypes that are altered by recent experience distinctly influence learning and memory processes. The authors focus on the emerging circuits and connectivities among different interneuron subtypes, their functional interactions, and how these interactions influence behavior.

One of the challenges of the neuroscience field is to establish a causal link between synaptic or cellular functions and behavior. How does one begin to address this challenge? This week we investigate two behaviors, learning and feeding, and explore how synaptic plasticity and neuronal circuits lead to these complex behaviors. Our first paper tests long-standing hypotheses that activity-dependent modifications of synaptic strengths (LTD and LTP) are cellular correlates of memory processes. Nabavi et al. propose a causal link between inactivation and reactivation of a memory using LTP and LTD. We will discuss how the authors came to this conclusion and the validity of this claim. Our second paper examines the neural circuit basis for feeding decisions. Studying Drosophila, Pool et al. identify specific inhibitory interneurons that regulate feeding behavior and propose that these neurons, as one element of a neuronal feeding circuit, acts to establish a threshold to "gate" feeding behavior. We will discuss what such a feeding circuit might look like and how this circuit could be modulated by external sensory cues and metabolic states to regulate consumption.

Sophisticated genomic technologies have identified human gene variants associated with neurodevelopmental diseases, such as Autism Spectrum Disorder (ASD). A current challenge is to generate experimentally accessible animal model systems that mimic human disease states, behaviorally and / or neuronally, and to identify specific pathways that lead to these conditions to facilitate the identification of potential therapeutic targets. Our first paper analyzes genetic mouse models harboring specific neuroligin-3 mutations that have been associated with ASD in humans. Rothwell et al. report an interesting behavioral motor learning phenotype and identify a subset of neurons in the nucleus accumbens that are selectively sensitive to these mutations. The nucleus accumbens, located in the midbrain, is part of the reward circuit that mediates behaviors such as addiction, motivation, and fear. Therefore, the authors identify a potential circuit that contributes the ASD etiology as well as a potential therapeutic target for patients. Our second paper examines FMR1 knockout mice. Goncalves et al. study genetically manipulated FMR1 mice to model the reduced expression of FMR1 gene products in humans with Fragile X syndrome, a severe form of ASD. The authors report altered cortical network dynamics in mice lacking FMR1, providing potential insights into the etiology and potential therapeutic targets to ameliorate ASD. We will discuss to what degree the mouse models in both papers faithfully replicate ASD. We will also question to what degree replication of human behaviors by these models is necessary and discuss how network-level disruptions, possibly by excitation / inhibition imbalances, reported in both these models might point toward therapeutic targets in ASD patients.

Students will visit Pfizer, a leading pharmaceutical company actively investigating and developing neurodevelopmental therapeutics. Having spent the course examining how synapses work and how their normal function is disrupted at the molecular and circuit levels in neurological disorders, students will learn how a large pharmaceutical company is developing strategies to combat neurological diseases. Students will be hosted by Dr. John Allen, a pioneer in the field of molecular therapeutics (recommended reading below) and a member of the Drug Discovery unit at Pfizer. Dr. Allen will give us an overview of the pressing societal and scientific challenges of modern drug discovery for treating neurological disorders, how drug targets are selected and evaluated, and how clinical trials are performed. A brief tour of the labs will follow.