Calcium microdomains in regulated exocytosis
Introduction
“This brings us to the possibility of an even wider generalization: for it appears that the crucial role of calcium is shared by many other types of secretory mecanism in which quantal release of a specific substance is evoked, whether through nerve impulses or through changes in the chemical environment of the cell.”
(Sir Bernard Katz, ‘The Release of Neural Substances’, Sherrington Lecture, p. 47, Liverpool University Press, 1969)
Almost 40 years ago, Douglas and Rubin [1] and Douglas [2] proposed that intracellular Ca2+ controls stimulus–secretion coupling in endocrine cells. Katz and Miledi1 [3], [4] suggested that intracellular Ca2+ ions control the rapid release of neurotransmitters from synapses. During their analysis of acetylcholine release from the frog neuromuscular junction Katz and collaborators observed that the presence of extracellular Ca2+ was required to enable the depolarizing pulses to become effective and evoke a postsynaptic response [5], [6], [7]. If the extracellular Ca2+ was replaced by Mg2+ or if they delayed the iontophoretic application of Ca2+ until the end of the depolarization, no release occurred. At the time, Katz concluded that “the ion which is indispensable for the electrically evoked transmitter release is Ca2+” (Katz, op.cit., p. 33) and hypothesized that “depolarization somehow changes the properties of the terminal axon membrane so that it represents more reactive sites to the colliding vesicles and thereby raises the statistical changes for quantal release”, which he and Miledi later modified to the more specific suggestion: “namely that depolarization opens a gate to calcium ions (…). As a consequence, of this increased ‘calcium conductance’, Ca ions can move – down a very high concentration gradient – towards the inside of the axon membrane and thus reach the critical sites of the release action. We are suggesting that, at these sites, calcium is essential for the process that causes a transient fusion of axon and vesicular membranes and which leads to the release of a quantal packet of transmitter” (Katz, op.cit., p. 34).
This stunningly modern picture resumes the main elements that are still employed today to explain how Ca2+ triggers synchronous release of neurotransmitters from the presynaptic nerve terminal (Fig. 1A) (see, e.g., [8], [9], [10] for recent reviews) and how Ca2+ modulates synaptic plasticity [11]. The depolarization caused by the action potential (AP) transiently opens voltage-gated Ca2+ ion channels that mediate a brief Ca2+ influx across the terminal membrane. Inside the cytoplasm, Ca2+ can only move by passive diffusion, a process that is further modified by the presence of mobile Ca2+ buffers [12], [13], [14]. Within microseconds, buffered diffusion sets up a standing Ca2+ domain around the mouth of the open ion channel [15], [16], [17], [18], with peak concentrations of hundreds of micromolars. These high levels of Ca2+ attained locally activate the Ca2+ sensor that trigger membrane fusion (see [19] for a recent review). Both the exact amplitude and time-course of [Ca2+]i govern the transmission characteristics of the synapse.
At mammalian central synapses, transmitter release occurs within less than a millisecond after Ca2+ channel opening. To reach this degree of synchrony, the vesicle and the Ca2+ binding protein must be located sufficiently close to one or more Ca2+ channels. The exact distance between the vesicle and Ca2+ channel(s) critically determines both the amplitude and kinetics of transmitter release. Thus, without knowledge of the exact spatial interrelationship of Ca2+ channels and release sites, without knowledge of the affinity, kinetics and spatial localization of the involved Ca2+ binding sites, it is impossible to understand the interaction between voltage-gated Ca2+ entry and the target Ca2+ sensor for exocytosis in quantitative terms. Despite the obvious difficulties to directly study sub-millisecond nanometric [Ca2+]i domains, remarkable advances have been made during the last few years, by combining pre- and postsynaptic recordings, making clever use of exogenous Ca2+ buffers to localize spatial gradients, measuring the near-membrane [Ca2+] with unprecedented spatial and temporal resolution by the use of new microscopies, and by a more and more precise modeling of the buffered diffusion of Ca2+ on a length-scale inaccessible to optical measurements. Modeling studies have benefited from the efforts to constrain their parameters, based on a careful 3D reconstruction of ultrastructural data.
In this review, we survey the recent literature and compare the Ca2+ signals and time-courses of regulated exocytosis in a variety of neuronal, neuroendocrine and non-electrically excitable cells.
Section snippets
Ca2+ nano-, micro- and macrodomains in regulated exocytosis
Due to its ubiquitous action in a multitude of signaling pathways, Ca2+ acts on very different time- and length-scales, ranging from transient, highly localized sub-micrometer domains (‘Ca2+ noise’) [20], [21] to vast Ca2+ waves that slowly invade large cell populations [22]. We will use the term ‘Ca2+ domain’ to designate a sub-cellular Ca2+ transient that is associated with a Ca2+ signal on a characteristic length-scale. In general, the spatiotemporal [Ca2+] dynamics will reflect the combined
Synchronous, action-potential-triggered neurotransmitter release
Active zones of motor nerve terminals are highly specialized sites for fast release of neurotransmitter. Their architecture with the molecular co-localization between the Ca2+ channel and the Ca2+ sensor for exocytosis is exquisitely designed to facilitate the regulated tethering, docking and rapid fusion of the synaptic vesicles with the plasma membrane [109], [110]. Classical work at the frog neuromuscular junction imaged with freeze-fracture EM shows large transmembrane particles at release
Selective mobilization of vesicle sub-populations and heterogeneity of release probabilities
Whereas earlier work stressed the clear distinction between fast neurotransmission [126] on the one hand and slow ‘asynchronous’ peptide and hormone release on the other hand [51], more recent studies point toward a large diversity of geometric arrangements of Ca2+ channels and vesicles in different cell types, leading to a wide spectrum of Ca2+ signals and kinetic components of exocytosis [8], [109]. For example, at the Calyx of Held, a time-dependent model to simulate Ca2+ influx,
Bidirectional communication of astrocytes and neurons
Astrocytes represent the major cell population of the central nervous system of higher mammals. Originally, this macroglia cell type has been regarded as a sort of passive cell only providing nutrients and structural support to the neighboring neurons. Numerous studies within the past two decades demonstrated that they engage a bidirectional communication with neurons [22], [157], [158], [159], [160]. Astrocytes are equipped with a large array of transmitter and neuromodulator receptors
Stimulation of exocytosis without a Ca2+ signal
Finally, we should note that while Ca2+ is generally viewed as the principal second messenger linking stimulation to exocytosis, cyclic nucleotides, prostaglandins, and other arachidonic acid metabolites must not be neglected as putatative second messengers that mediate the effect of Ca2+ in secretion. Paradoxically, despite the important progress made in studying the complex interplay of Ca2+ influx, Ca2+ release, buffering and sequestration, we must even acknowledge the possible existence of
Note added in proof
During the typesetting of this manuscript, two more review articles were published. Garcia and co-workers offer an in-depth view of Ca2+ microdomains in chromaffin cell secretion (Garcia et al. Calcium signaling and exocytosis in adrenal chromaffin cells. Physiol Rev. 2006 Oct;86(4):1093–131.). Neher compared the exocytic control mechanisms in chromaffin cells and the Calyx synapse (E. Neher, A comparison between exocytic control mechanisms in adrenal chromaffin cells and a glutamatergic
Acknowledgements
We like to thank Bob Chow, Alain Marty, Rami Rahamimoff, Nicole Ropert and Serge Charpak for comments on the manuscript and Kerry Delaney for sharing unpublished material. Supported by a joint INSERM/Max-Planck Society AMIGO grant.
References (174)
- et al.
Presynaptic calcium and control of vesicle fusion
Curr. Opin. Neurobiol.
(2005) How does calcium trigger neurotransmitter release?
Curr. Opin. Neurobiol.
(2001)- et al.
Ca2+ and synaptic plasticity
Cell Calcium
(2005) - et al.
Time courses of calcium and calcium-bound buffers following calcium influx in a model cell
Biophys. J.
(1993) - et al.
Effects of rapid buffers on Ca2+ diffusion and Ca2+ oscillations
Biophys. J.
(1994) Vesicle pools and Ca2+ microdomains: new tools for understanding their roles in neurotranmitter release
Neuron
(1998)- et al.
Compartmentalization of the submembrane calcium activity during calcium influx and its significance in transmitter release
Biophys. J.
(1985) - et al.
Calcium domains associated with individual channels can account for anomalous voltage relations of Ca2+ dependent responses
Biophys. J.
(1984) Can a synaptic signal arise from noise?
Neuron
(2003)- et al.
Glial calcium signaling and neuron-glia communication
Cell Calcium
(2005)