Synapse formation

The goal of our work on synapse formation is to understand the molecular determinants that shape the function of synapses as the fundamental information processing units in neural circuits, with the overall aim of defining the molecular logic that constructs these circuits. Synapses exhibit a high degree of specificity. Neurons form synapses with only a select subset of other neurons, and the resulting synaptic connections exhibit an astounding diversity of properties that are specified by the pre- and postsynaptic neurons. Our laboratory is focusing on synaptic cell-adhesion molecules that form a dynamic network across the synaptic cleft to control for establishment and specification of synaptic connections between neurons. In addition, we are analyzing secreted synaptogenic molecules produced by neurons and glia or circulating in the blood that promote synapse formation and synaptic transmission.

Synaptic adhesion molecules

In examining synapse formation, our laboratory is studying a range of key synaptic adhesion molecules, such as neurexins, neuroligins, teneurins, Bai's and latrophilins. Together these adhesion molecules with their many intra- and extracellular binding partners make and shape synapses. They dictate the establishment of synapses, render synapses functional, and endow synapses with specific properties, such as the ability to undergo long-term potentiation. At present, we as neuroscientists do not understand how synapses are made, we do know why some synapses are competent to engage in long-term plasticity and others are not, and how the properties of synapses are specified. We hypothesize that all of these processes are controlled by signaling via synaptic adhesion molecules, and have obtained accumulating evidence in support of this hypothesis. 

In recent work, we discovered among others that specific splice variants of presynaptic neurexins instruct the postsynaptic glutamate receptor composition and enable postsynaptic spines to undergo LTP (Aoto et al., 2013; Dai et al., 2019). Furthermore, we found that at specific subsets of hippocampal synapses, different isoforms of latrophilins, postsynaptic adhesion-GPCRs, are essential for the formation of synapses by interacting with presynaptic teneurins and FLRTs (Sando et al., 2019). These examples illustrate the challenge we face: to deconvolve into simple individual processes the overall complex mechanism by which synaptic connections construct functional neural circuits. Moreover, mutations in neurexins and their 

ligands, chiefly neuroligins, are observed in autism spectrum disorders and in schizophrenia, suggesting that their role in shaping synaptic communication is impaired in these diseases. These mutations are physiologically significant, as we demonstrated using both isogenic engineered human neurons as well as patient-derived human neurons. 

Synaptotrophic signaling molecules

In addition to trans-synaptic adhesion molecules, proteins secreted by astrocytes or the choroid plexus, or proteins entering the brain from the systemic circulation, promote synapse formation by binding to specific cell-surface receptors. Such proteins include ApoE, SPARCL1, and thrombospondins. We are examining how these proteins work, and how their actions relate to aging and to neurodegenerative disorders such as Alzheimer’s disease. Understanding these proteins as metaplastic modulators of neural circuits will be essential to gain insight into how the brain’s circuits change as a function of environmental or pathogenic challenges. . 

Synapse modifications in memory

Long-term memories are thought to be mediated by changes in synapse numbers and synapse properties in defined neural circuits. How these changes are effected,  however, remains unclear. Our laboratory is using our expertise in the molecular definition of synapses and in monitoring synaptic connections for a better understanding of how synapse modifications mediate learning and memory. Our focus here is not on specific circuits, but on defining the molecular changes that are generally required for learning long-term memories and how they relate to synapses, using again a broad range of techniques that include single-cell RNAseq, imaging, behavior, and protein chemistry.

The Pathobiology of Synapses in neurodegenerative disorders

In extension of our work on synapses, we also examine how neurodegenerative disorders cause a loss of synapses. Here, we focus on ApoE and on APP, two proteins that are genetically linked to Alzheimer’s disease and other neurodegenerative disorders. ApoE is involved in lipid transport, but the relation of lipid transport to synapse formation and Alzheimer’s disease remains unclear and is a focus of our work. Despite decades of work, APP’s physiological and pathological roles remain incompletely understood, but APP has also been implicated in synaptic functions. Elucidating how ApoE and APP function at the synapse physiologically and how ther role at the synapse relates to neurodegenerative disorders thus is another general focus of our work.

Methods and approaches

To study synapses and their impairments in disease, we use a highly interdisciplinary approach that combines structural biology, mouse genetics, electrophysiology with optical imaging and behavior. We employ mice and human neurons trans-differentiated from stem cells as model systems. Our work is carried out in close collaborations with colleagues at Stanford and elsewhere, with the overall goal to not only describe how a neural circuit processes information, but to understand the underlying molecular events such explain information processing.