Dynamin isoforms decode action potential firing for synaptic vesicle recycling

A centronuclear myopathy-causing mutation in dynamin-2 disrupts neuronal morphology and excitatory synaptic transmission in a murine model of the disease

AIMS

Dynamin-2 is a large GTPase, a member of the dynamin superfamily that regulates membrane remodelling and cytoskeleton dynamics. Mutations in the dynamin-2 gene (DNM2) cause autosomal dominant centronuclear myopathy (CNM), a congenital neuromuscular disorder characterised by progressive weakness and atrophy of the skeletal muscles. Cognitive defects have been reported in some DNM2-linked CNM patients suggesting that these mutations can also affect the central nervous system (CNS). Here we studied how a dynamin-2 CNM-causing mutation influences the CNS function.

METHODS

Heterozygous mice harbouring the p.R465W mutation in the dynamin-2 gene (HTZ), the most common causing autosomal dominant CNM, were used as disease model. We evaluated dendritic arborisation and spine density in hippocampal cultured neurons, analysed excitatory synaptic transmission by electrophysiological field recordings in hippocampal slices, and evaluated cognitive function by performing behavioural tests.

RESULTS

HTZ hippocampal neurons exhibited reduced dendritic arborisation and lower spine density than WT neurons, which was reversed by transfecting an interference RNA against the dynamin-2 mutant allele. Additionally, HTZ mice showed defective hippocampal excitatory synaptic transmission and reduced recognition memory compared to the WT condition.

CONCLUSION

Our findings suggest that the dynamin-2 p.R465W mutation perturbs the synaptic and cognitive function in a CNM mouse model and support the idea that this GTPase plays a key role in regulating neuronal morphology and excitatory synaptic transmission in the hippocampus.

Mechanisms of Synaptic Vesicle Exo- and Endocytosis

Within 1 millisecond of action potential arrival at presynaptic terminals voltage-gated Ca²⁺ channels open. The Ca²⁺ channels are linked to synaptic vesicles which are tethered by active zone proteins. Ca²⁺ entrance into the active zone triggers: (1) the fusion of the vesicle and exocytosis, (2) the replenishment of the active zone with vesicles for incoming exocytosis, and (3) various types of endocytosis for vesicle reuse, dependent on the pattern of firing. These time-dependent vesicle dynamics are controlled by presynaptic Ca²⁺ sensor proteins, regulating active zone scaffold proteins, fusion machinery proteins, motor proteins, endocytic proteins, several enzymes, and even Ca²⁺ channels, following the decay of Ca²⁺ concentration after the action potential. Here, I summarize the Ca²⁺-dependent protein controls of synchronous and asynchronous vesicle release, rapid replenishment of the active zone, endocytosis, and short-term plasticity within 100 msec after the action potential. Furthermore, I discuss the contribution of active zone proteins to presynaptic plasticity and to homeostatic readjustment during and after intense activity, in addition to activity-dependent endocytosis.

Neurotransmitter Release Site Replenishment and Presynaptic Plasticity

An action potential (AP) triggers neurotransmitter release from synaptic vesicles (SVs) docking to a specialized release site of presynaptic plasma membrane, the active zone (AZ). The AP simultaneously controls the release site replenishment with SV for sustainable synaptic transmission in response to incoming neuronal signals. Although many studies have suggested that the replenishment time is relatively slow, recent studies exploring high speed resolution have revealed SV dynamics with milliseconds timescale after an AP. Accurate regulation is conferred by proteins sensing Ca²⁺ entering through voltage-gated Ca²⁺ channels opened by an AP. This review summarizes how millisecond Ca²⁺ dynamics activate multiple protein cascades for control of the release site replenishment with release-ready SVs that underlie presynaptic short-term plasticity.

Dynamin Superfamily at Pre- and Postsynapses: Master Regulators of Synaptic Transmission and Plasticity in Health and Disease

Dynamin superfamily proteins (DSPs) comprise a large group of GTP-ases that orchestrate membrane fusion and fission, and cytoskeleton remodeling in different cell-types. At the central nervous system, they regulate synaptic vesicle recycling and signaling-receptor turnover, allowing the maintenance of synaptic transmission. In the presynapses, these GTP-ases control the recycling of synaptic vesicles influencing the size of the ready-releasable pool and the release of neurotransmitters from nerve terminals, whereas in the postsynapses, they are involved in AMPA-receptor trafficking to and from postsynaptic densities, supporting excitatory synaptic plasticity, and consequently learning and memory formation. In agreement with these relevant roles, an important number of neurological disorders are associated with mutations and/or dysfunction of these GTP-ases. Along the present review we discuss the importance of DSPs at synapses and their implication in different neuropathological contexts.

A Tripartite Interaction Among the Calcium Channel α1- and β-Subunits and F-Actin Increases the Readily Releasable Pool of Vesicles and Its Recovery After Depletion

Neurotransmitter release is initiated by the influx of Ca²⁺via voltage-gated calcium channels. The accessory β-subunit (Ca_Vβ) of these channels shapes synaptic transmission by associating with the pore-forming subunit (Ca_Vα₁) and up-regulating presynaptic calcium currents. Besides Ca_Vα_1, Ca_Vβ interacts with several partners including actin filaments (F-actin). These filaments are known to associate with synaptic vesicles (SVs) at the presynaptic terminals and support their translocation within different pools, but the role of Ca_Vβ/F-actin association on synaptic transmission has not yet been explored. We here study how Ca_Vβ₄, the major calcium channel β isoform in mamalian brain, modifies synaptic transmission in concert with F-actin in cultured hippocampal neurons. We analyzed the effect of exogenous Ca_Vβ₄ before and after pharmacological disruption of the actin cytoskeleton and dissected calcium channel-dependent and -independent functions by comparing the effects of the wild-type subunit with the one bearing a double mutation that impairs binding to Ca_Vα₁. We found that exogenously expressed wild-type Ca_Vβ₄ enhances spontaneous and depolarization-evoked excitatory postsynaptic currents (EPSCs) without altering synaptogenesis. Ca_Vβ₄ increases the size of the readily releasable pool (RRP) of SVs at resting conditions and accelerates their recovery after depletion. The enhanced neurotransmitter release induced by Ca_Vβ₄ is abolished upon disruption of the actin cytoskeleton. The Ca_Vα₁ association-deficient Ca_Vβ₄ mutant associates with actin filaments, but neither alters postsynaptic responses nor the time course of the RRP recovery. Furthermore, this mutant protein preserves the ability to increase the RRP size. These results indicate that the interplay between Ca_Vβ₄ and F-actin also support the recruitment of SVs to the RRP in a Ca_Vα₁-independent manner. Our studies show an emerging role of Ca_Vβ in determining SV maturation toward the priming state and its replenishment after release. We envision that this subunit plays a role in coupling exocytosis to endocytosis during the vesicle cycle.

Optical excitation and detection of neuronal activity

Optogenetics has emerged as an exciting tool for manipulating neural activity, which in turn, can modulate behavior in live organisms. However, detecting the response to the optical stimulation requires electrophysiology with physical contact or fluorescent imaging at target locations, which is often limited by photobleaching and phototoxicity. In this paper, we show that phase imaging can report the intracellular transport induced by optogenetic stimulation. We developed a multimodal instrument that can both stimulate cells with subcellular spatial resolution and detect optical pathlength (OPL) changes with nanometer scale sensitivity. We found that OPL fluctuations following stimulation are consistent with active organelle transport. Furthermore, the results indicate a broadening in the transport velocity distribution, which is significantly higher in stimulated cells compared to optogenetically inactive cells. It is likely that this label-free, contactless measurement of optogenetic response will provide an enabling approach to neuroscience.

Proteomic Analysis of Mouse Brain Subjected to Spaceflight

There is evidence that spaceflight poses acute and late risks to the central nervous system. To explore possible mechanisms, the proteomic changes following spaceflight in mouse brain were characterized. Space Shuttle Atlantis (STS-135) was launched from the Kennedy Space Center (KSC) on a 13-day mission. Within 3–5 h after landing, brain tissue was collected to evaluate protein expression profiles using quantitative proteomic analysis. Our results showed that there were 26 proteins that were significantly altered after spaceflight in the gray and/or white matter. While there was no overlap between the white and gray matter in terms of individual proteins, there was overlap in terms of function, synaptic plasticity, vesical activity, protein/organelle transport, and metabolism. Our data demonstrate that exposure to the spaceflight environment induces significant changes in protein expression related to neuronal structure and metabolic function. This might lead to a significant impact on brain structural and functional integrity that could affect the outcome of space missions.

SAD-B Phosphorylation of CAST Controls Active Zone Vesicle Recycling for Synaptic Depression

Short-term synaptic depression (STD) is a common form of activity-dependent plasticity observed widely in the nervous system. Few molecular pathways that control STD have been described, but the active zone (AZ) release apparatus provides a possible link between neuronal activity and plasticity. Here, we show that an AZ cytomatrix protein CAST and an AZ-associated protein kinase SAD-B coordinately regulate STD by controlling reloading of the AZ with release-ready synaptic vesicles. SAD-B phosphorylates the N-terminal serine (S45) of CAST, and S45 phosphorylation increases with higher firing rate. A phosphomimetic CAST (S45D) mimics CAST deletion, which enhances STD by delaying reloading of the readily releasable pool (RRP), resulting in a pool size decrease. A phosphonegative CAST (S45A) inhibits STD and accelerates RRP reloading. Our results suggest that the CAST/SAD-B reaction serves as a brake on synaptic transmission by temporal calibration of activity and synaptic depression via RRP size regulation.

Millisecond Ca2+ dynamics activate multiple protein cascades for synaptic vesicle control

For reliable transmission at chemical synapses, neurotransmitters must be released dynamically in response to neuronal activity in the form of action potentials. Stable synaptic transmission is dependent on the efficacy of transmitter release and the rate of resupplying synaptic vesicles to their release sites. Accurate regulation is conferred by proteins sensing Ca2+ entering through voltage-gated Ca2+ channels opened by an action potential. Presynaptic Ca2+ concentration changes are dynamic functions in space and time, with wide fluctuations associated with different rates of neuronal activity. Thus, regulation of transmitter release includes reactions involving multiple Ca2+-dependent proteins, each operating over a specific time window. Classically, studies of presynaptic proteins function favored large invertebrate presynaptic terminals. I have established a useful mammalian synapse model based on sympathetic neurons in culture. This review summarizes the use of this model synapse to study the roles of presynaptic proteins in neuronal activity for the control of transmitter release efficacy and synaptic vesicle recycling.

Neural activity selects myosin IIB and VI with a specific time window in distinct dynamin isoform-mediated synaptic vesicle reuse pathways

Presynaptic nerve terminals must maintain stable neurotransmissions via synaptic vesicle (SV) resupply despite encountering wide fluctuations in the number and frequency of incoming action potentials (APs). However, the molecular mechanism linking variation in neural activity to SV resupply is unknown. Myosins II and VI are actin-based cytoskeletal motors that drive dendritic actin dynamics and membrane transport, respectively, at brain synapses. Here we combined genetic knockdown or molecular dysfunction and direct physiological measurement of fast synaptic transmission from paired rat superior cervical ganglion neurons in culture to show that myosins IIB and VI work individually in SV reuse pathways, having distinct dependency and time constants with physiological AP frequency. Myosin VI resupplied the readily releasable pool (RRP) with slow kinetics independently of firing rates but acted quickly within 50 ms after AP. Under high-frequency AP firing, myosin IIB resupplied the RRP with fast kinetics in a slower time window of 200 ms. Knockdown of both myosin and dynamin isoforms by mixed siRNA microinjection revealed that myosin IIB-mediated SV resupply follows amphiphysin/dynamin-1-mediated endocytosis, while myosin VI-mediated SV resupply follows dynamin-3-mediated endocytosis. Collectively, our findings show how distinct myosin isoforms work as vesicle motors in appropriate SV reuse pathways associated with specific firing patterns.

Differential targeting of dynamin-1 and dynamin-3 to nerve terminals during chronic suppression of neuronal activity

Neurons express three closely related dynamin genes. Dynamin 1 has long been implicated in the regulation of synaptic vesicle recycling in nerve terminals, and dynamins 2 and 3 were more recently shown also to contribute to synaptic vesicle recycling in specific and distinguishable ways. In cultured hippocampal neurons we found that chronic suppression of spontaneous network activity differentially regulated the targeting of endogenous dynamins 1 and 3 to nerve terminals, while dynamin 2 was unaffected. Specifically, when neural activity was chronically silenced for 1-2weeks by tetrodotoxin (TTX), the clustering of dynamin 1 at nerve terminals was reduced, while the clustering of dynamin 3 significantly increased. Moreover, dynamin 3 clustering was induced within hours by the sustained blockade of AMPA receptors, suggesting that AMPA receptors may function to prevent Dyn3 accumulation within nerve terminals. Clustering of dynamin 3 was induced by an antagonist of the calcium-dependent protein phosphatase calcineurin, but was not dependent upon intact actin filaments. TTX-induced clustering of Dyn3 occurred with a markedly slower time-course than the previously described clustering of synapsin 1. Potassium-induced depolarization rapidly de-clustered dynamin 3 from nerve terminals within minutes. These results, which have implications for homeostatic synapse restructuring, indicate that the three dynamins have evolved different regulatory mechanisms for trafficking to and from nerve terminals in response to changes in neural activity.

Late stages of the synchronized macrophage fusion in osteoclast formation depend on dynamin

Macrophage fusion that leads to osteoclast formation is one of the most important examples of cell-cell fusion in development, tissue homoeostasis and immune response. Protein machinery that fuses macrophages remains to be identified. In the present study, we explored the fusion stage of osteoclast formation for RAW macrophage-like murine cells and for macrophages derived from human monocytes. To uncouple fusion from the preceding differentiation processes, we accumulated fusion-committed cells in the presence of LPC (lysophosphatidylcholine) that reversibly blocks membrane merger. After 16 h, we removed LPC and observed cell fusion events that would normally develop within 16 h develop instead within 30-90 min. Thus, whereas osteoclastogenesis, generally, takes several days, our approach allowed us to focus on an hour in which we observe robust fusion between the cells. Complementing syncytium formation assay with a novel membrane merger assay let us study the synchronized fusion events downstream of a local merger between two plasma membranes, but before expansion of nascent membrane connections and complete unification of the cells. We found that the expansion of membrane connections detected as a growth of multinucleated osteoclasts depends on dynamin activity. In contrast, a merger between the plasma membranes of the two cells was not affected by inhibitors of dynamin GTPase. Thus dynamin that was recently found to control late stages of myoblast fusion also controls late stages of macrophage fusion, revealing an intriguing conserved mechanistic motif shared by diverse cell-cell fusion processes.

Synaptic vesicle glycoprotein 2A modulates vesicular release and calcium channel function at peripheral sympathetic synapses

Synaptic vesicle glycoprotein (SV)2A is a transmembrane protein found in secretory vesicles and is critical for Ca(2+) -dependent exocytosis in central neurons, although its mechanism of action remains uncertain. Previous studies have proposed, variously, a role of SV2 in the maintenance and formation of the readily releasable pool (RRP) or in the regulation of Ca(2+) responsiveness of primed vesicles. Such previous studies have typically used genetic approaches to ablate SV2 levels; here, we used a strategy involving small interference RNA (siRNA) injection to knockdown solely presynaptic SV2A levels in rat superior cervical ganglion (SCG) neuron synapses. Moreover, we investigated the effects of SV2A knockdown on voltage-dependent Ca(2+) channel (VDCC) function in SCG neurons. Thus, we extended the studies of SV2A mechanisms by investigating the effects on vesicular transmitter release and VDCC function in peripheral sympathetic neurons. We first demonstrated an siRNA-mediated SV2A knockdown. We showed that this SV2A knockdown markedly affected presynaptic function, causing an attenuated RRP size, increased paired-pulse depression and delayed RRP recovery after stimulus-dependent depletion. We further demonstrated that the SV2A-siRNA-mediated effects on vesicular release were accompanied by a reduction in VDCC current density in isolated SCG neurons. Together, our data showed that SV2A is required for correct transmitter release at sympathetic neurons. Mechanistically, we demonstrated that presynaptic SV2A: (i) acted to direct normal synaptic transmission by maintaining RRP size, (ii) had a facilitatory role in recovery from synaptic depression, and that (iii) SV2A deficits were associated with aberrant Ca(2+) current density, which may contribute to the secretory phenotype in sympathetic peripheral neurons.

Kinetic organization of Ca2+ signals that regulate synaptic release efficacy in sympathetic neurons

Calcium regulation of neurotransmitter release is essential for maintenance of synaptic transmission. However, the temporal and spatial organization of Ca(2+) dynamics that regulate synaptic vesicle (SV) release efficacy in sympathetic neurons is poorly understood. Here, we investigate the N-type Ca(2+) channel-mediated kinetic structure of Ca(2+) regulation of cholinergic transmission of sympathetic neurons. We measured the effect of Ca(2+) chelation with fast 1,2-bis(2-aminophenoxy) ethane-tetraacetic acid (BAPTA) and slow ethyleneglycol-tetraacetic acid (EGTA) buffers on exocytosis, synaptic depression, and recovery of the readily releasable vesicle pool (RRP), after both single action potential (AP) and repetitive APs. Surprisingly, postsynaptic potentials peaking at ~12 milliseconds after the AP was inhibited by both rapid and slow Ca(2+) buffers suggests that, in addition to the well known fast Ca(2+) signals at the active zone (AZ), slow Ca(2+) signals at the peak of Ca(2+) entry also contribute to paired-pulse or repetitive AP responses. Following a single AP, discrete Ca(2+) transient increase regulated synaptic depression in rapid (<30-millisecond) and slow (<120-millisecond) phases. In contrast, following prolonged AP trains, synaptic depression was reduced by a slow Ca(2+) signal regulation lasting >200 milliseconds. Finally, after an AP burst, recovery of the RRP was mediated by an AP-dependent rapid Ca(2+) signal, and the expansion of releasable SV number by an AP firing activity-dependent slow Ca(2+) signal. These data indicate that local Ca(2+) signals operating near Ca(2+) sources in the AZ are organized into discrete fast and slow temporal phases that remodel exocytosis and short-term plasticity to ensure long-term stability in acetylcholine release efficacy.