Release-dependent feedback inhibition by a presynaptically localized ligand-gated anion channel

Presynaptic ligand-gated ion channels (LGICs) have long been proposed to affect neurotransmitter release and to tune the neural circuit activity. However, the understanding of their in vivo physiological action remains limited, partly due to the complexity in channel types and scarcity of genetic models. Here we report that C. elegans LGC-46, a member of the Cys-loop acetylcholine (ACh)-gated chloride (ACC) channel family, localizes to presynaptic terminals of cholinergic motor neurons and regulates synaptic vesicle (SV) release kinetics upon evoked release of acetylcholine. Loss of lgc-46 prolongs evoked release, without altering spontaneous activity. Conversely, a gain-of-function mutation of lgc-46 shortens evoked release to reduce synaptic transmission. This inhibition of presynaptic release requires the anion selectivity of LGC-46, and can ameliorate cholinergic over-excitation in a C. elegans model of excitation-inhibition imbalance. These data demonstrate a novel mechanism of presynaptic negative feedback in which an anion-selective LGIC acts as an auto-receptor to inhibit SV release.

Position of UNC-13 in the active zone regulates synaptic vesicle release probability and release kinetics

The presynaptic active zone proteins UNC-13/Munc13s are essential for synaptic vesicle (SV) exocytosis by directly interacting with SV fusion apparatus. An open question is how their association with active zones, hence their position to Ca²⁺ entry sites, regulates SV release. The N-termini of major UNC-13/Munc13 isoforms contain a non-calcium binding C₂A domain that mediates protein homo- or hetero-meric interactions. Here, we show that the C₂A domain of Caenorhabditis elegans UNC-13 regulates release probability of evoked release and its precise active zone localization. Kinetics analysis of SV release supports that the proximity of UNC-13 to Ca²⁺ entry sites, mediated by the C₂A-domain containing N-terminus, is critical for accelerating neurotransmitter release. Additionally, the C₂A domain is specifically required for spontaneous release. These data reveal multiple roles of UNC-13 C₂A domain, and suggest that spontaneous release and the fast phase of evoked release may involve a common pool of SVs at the active zone.

DOI:http://dx.doi.org/10.7554/eLife.01180.001

Neurons are connected to each other by junctions called synapses. When an electrical signal travelling along a neuron arrives at a synapse, it causes the release of bubble-like structures called synaptic vesicles that contain chemicals called neurotransmitters. When released by the vesicles these neurotransmitters bind to receptors on a second neuron and allow the signal to continue on its way through the nervous system.

The release of synaptic vesicles from the neuron depends largely on the number of calcium ions that enter this neuron via structures called ion channels, and also on the rate at which they enter. Vesicles are released in one of three ways: they can be released quickly (within a few milliseconds) in response to the influx of calcium ions; they can be released slowly (over a period of tens or hundreds of milliseconds) in response to the influx; or they can be released at random times that are not related to the influx.

It is known that the sensitivity of certain calcium sensors near the synapse influences the release of the vesicles. It had been thought that the distance between the “active zone” where the calcium ions enter the neuron and the region where the vesicles reside might also influence rate of release, but the molecular mechanism underlying this hypothesis is poorly understood.

Zhou et al. have now shed new light on this question by performing a series of experiments that involved manipulating a protein called UNC-13 – which is known to be involved in the release of vesicles – in neurons from C. elegans, a nematode worm. First it was shown that the precise position of UNC-13 in the active zone depended on a domain within the protein called the C₂A domain. Next it was shown that the distance between the UNC-13 protein and the calcium ion channels strongly influences the quick mode of vesicle release. Finally, Zhou et al. showed that the C₂A domain also had a significant influence on the spontaneous release of vesicles, which suggests that a common fleet of vesicles might be used for both the quick and the spontaneous modes of vesicle release. Zhou et al. also generated mutant worms that mimicked a neurological disease, epileptic seizure, and showed that eliminating the C₂A domain can relieve some of the symptoms associated with the disease.

Many neurological diseases are caused by signals not being transmitted properly at synapses, so in addition to providing insights into the basic mechanism underlying synaptic action, these results could also assist with the development of new strategies for managing neurological diseases.

DOI:http://dx.doi.org/10.7554/eLife.01180.002