HHMI-NIH Research Scholars
Learn about the HHMI-NIH Research Scholars Program, also known as the Cloister Program. More

Janelia Farm Research Campus
Learn about the new HHMI research campus located in Virginia. More

Summary: Thomas Südhof is interested in how presynaptic terminals are formed during synaptogenesis, how presynaptic terminals release neurotransmitters, and how presynaptic terminals degenerate in neurodegenerative disease. To address these questions, Südhof's laboratory employs approaches ranging from biophysical studies to the physiological and behavioral analyses of mutant mice.

Neurons communicate with each other by synaptic transmission at specialized intercellular junctions called synapses. During synaptic transmission, a presynaptic neuron releases a chemical neurotransmitter that is then recognized by the postsynaptic neuron. Neurotransmitter release is triggered when an action potential opens voltage-gated Ca²⁺ channels and Ca²⁺ flows into the presynaptic terminal. Ca²⁺ triggers neurotransmitter release by stimulating the fusion of synaptic vesicles (abundant secretory vesicles that are filled with neurotransmitters) with the presynaptic plasma membrane at the active zone (a specialized plasma membrane section that marks the synapse). Released neurotransmitters then elicit a postsynaptic signal. Synaptic transmission occurs by this mechanism in all synapses, but the specific properties of synaptic transmission vary among synapses.

Work in my laboratory focuses on the molecular mechanisms that mediate the functions of presynaptic nerve terminals in synaptic transmission. Specifically, we ask three questions: (1) How does a presynaptic nerve terminal form a targeted synaptic junction at a defined position on a postsynaptic neuron, i.e., contribute to determining synaptic connectivity? (2) How does a nerve terminal release neurotransmitters in an exquisitely regulated manner? (3) How do presynaptic functions become abnormal in neurodegenerative diseases? Our overall goal is to identify the relevant molecules involved in these processes, to describe their properties and atomic structures, and to understand their physiological functions.

	The synaptic vesicle cycle... Synaptic vesicle recycling pathways... Structures of synaptic vesicle trafficking proteins... SNARE proteins, complexins, and synaptotagmins... Protein-protein interactions...

Synapse Formation
At a synapse, pre- and postsynaptic compartments are linked by trans-synaptic cell adhesion molecules that in turn are coupled to the presynaptic release machinery or to postsynaptic receptors. A decade ago, we identified candidate trans-synaptic cell adhesion molecules called neurexins and neuroligins. Neurexins are thought to be presynaptic, are expressed in two principal forms, α- and β-neurexins, and are highly polymorphic due to extensive alternative splicing. Neuroligins are primarily postsynaptic and are also alternatively spliced; their importance is highlighted by the finding that neuroligin mutations are pathogenic in a subset of patients with familial autistic syndrome.

Neurexins bind to neuroligins to form heterophilic intercellular junctions at the synapse; this binding is tightly regulated by alternative splicing of neurexins and—as we recently observed—of neuroligins, resulting in a trans-synaptic splice code. Moreover, recent experiments with mutant mice revealed that neurexins and neuroligins are not required for the establishment of initial synaptic contacts, but are both essential for the regular function of synapses. Mice with mutant neurexin or neuroligin genes form ultrastructurally normal synapses. In the mutants, however, synaptic transmission is severely impaired, such that the mutant mice die at birth. Overall, these studies showed that neurexins and neuroligins are genuine synaptic cell adhesion molecules that guide synapse formation not by triggering the assembly of synaptic junctions but by recruiting crucial components of the pre- and postsynaptic machinery.

Neurotransmitter Release
To identify key molecules involved in release, we initially set out to characterize the proteins of synaptic vesicles and of the active zone. We then used biochemical and biophysical methods to determine the properties and atomic structures of these proteins, and targeted mouse mutants to examine their functions. Although many questions remain, this combination of approaches has elucidated fundamental mechanisms underlying key aspects of neurotransmitter release (e.g., Ca²⁺ triggering of fast synaptic vesicle fusion).

Synaptic vesicle fusion is at least in part mediated by the assembly of three synaptic SNARE proteins—the vesicle protein synaptobrevin/VAMP and the plasma membrane proteins SNAP-25 and syntaxin/HPC-1—into a tight complex. Syntaxin also interacts with another essential fusion protein, Munc18-1. SNARE proteins and Munc18-1 perform multiple functions that are exquisitely regulated. For example, during fusion, syntaxin changes from a closed into an open conformation. We recently showed in mice that permanent "opening" of syntaxin by mutagenesis destabilizes synapses, leading to massive epilepsy. As another example, synaptobrevin is required for both spontaneous and evoked synaptic vesicle fusion, but recent data reveal that the mechanisms by which synaptobrevin acts in these two forms of fusion differ. In addition to this dual function in exocytosis, synaptobrevin is also required for normal fast endocytosis of vesicles. These findings suggest an unexpectedly economical organization of the secretory machinery in which different steps are mediated by the same molecules via different mechanisms.

At the synapse, Ca²⁺ triggers both a fast component and a slow component of release. The fast component is induced by Ca²⁺ binding to the synaptic vesicle proteins synaptotagmin 1 and 2. Recent data show that these two synaptotagmins act similarly in stabilizing resting synapses and in triggering fast release upon Ca²⁺ binding. Synaptotagmins contain two C₂-domains that bind Ca²⁺ in a ternary complex with phospholipids and may also interact with SNAREs. An emerging model suggests that before release is triggered, the Ca²⁺-independent interaction of synaptotagmin with SNARE complexes positions synaptotagmin for subsequent Ca²⁺ sensing. When Ca²⁺ flows into the terminal during an action potential, Ca²⁺ binding to the synaptotagmin C₂-domains induces their association with phospholipids and opens the fusion pore to trigger release.

In addition to synaptotagmin 1 and 2, 13 other synaptotagmins are expressed in brain. Current results suggest that some of these other synaptotagmins may also contribute to the Ca²⁺ triggering of release, either by modulating fast release induced by synaptotagmin 1 and 2, and/or by stimulating slow release. Although synaptotagmin 1 and 2 are exclusively presynaptic, some of the other synaptotagmins appear to be postsynaptic and may regulate membrane traffic of neurotransmitter receptors or other signal transduction molecules.

The amount of release triggered by Ca²⁺ is regulated in a use-dependent manner during presynaptic plasticity. Active-zone proteins not only serve as a receptacle of synaptic vesicles for fusion but also regulate synaptic vesicle fusion during plasticity. Key active-zone proteins for both functions are Munc13s and RIMs, multidomain proteins that bind to each other and to other active-zone and synaptic vesicle proteins. Many isoforms of Munc13s and RIMs with differential properties are expressed, but all isoforms appear to function in release. Recent results, for example, demonstrated that deleting one RIM isoform, RIM1α, leads to discrete changes in both vesicle priming and the plasticity of release, whereas deletion of multiple isoforms almost completely abolishes release. These and other findings have led to the view of the active zone as a protein mosaic whose precise composition is a key determinant of the properties (for example, strength and plasticity) of a synapse.

Neurodegenerative Disease
Finally, we are interested in understanding how the normal functions of a synapse relate to pathological changes observed in neurodegenerative disorders such as Alzheimer's and Parkinson's disease. Several facts—for example, the critical role of the presynaptic protein α-synuclein in the pathogenesis of Parkinson's disease, or the transport of the cell-surface protein APP (the amyloid-β precursor protein that is critical for the pathogenesis of Alzheimer's disease) into presynaptic nerve terminals—point to a presynaptic dysfunction in at least some neurodegenerative disorders. To approach this problem, we have begun to study the normal synaptic function of APP and synuclein. Our studies have uncovered an unexpected role for APP in gene expression, and led to a description of a powerful activity of synucleins in suppressing neurodegeneration. Strikingly, we found that synucleins genetically interact with CSPα (cysteine string protein α), a synaptic vesicle cochaperone protein. CSPα acts as a proprietary chaperone on synaptic vesicles whose deletion causes massive neurodegeneration. Moderate overexpression of synucleins abolishes the neurodegeneration in CSPα-deficient mice, while deletion of endogenous synucleins accelerates the neurodegeneration. A better understanding of how normal chaperone activities in presynaptic terminals protect the terminals against use-dependent degeneration should uncover molecular pathogenesis pathways that may provide clues to novel therapeutic approaches.

These studies, particularly those related to the mechanism of synaptic cell adhesion and to neurodegenerative diseases, were funded in part by grants from the National Institutes of Health.

Last updated: May 6, 2008