Synapses

Abstract

A synapse is the specialized junction at which one neuron communicates with another cell, the elementary unit of signaling in the nervous system. At a chemical synapse, by far the most common kind, an electrical impulse arriving at the presynaptic terminal triggers calcium entry, which drives synaptic vesicles to release a neurotransmitter into a narrow cleft; the transmitter binds receptors on the postsynaptic cell and changes its electrical state. A smaller class of electrical synapses couples cells directly through gap junctions for near-instant transmission. Synapses are not fixed relays: their strength rises and falls with use, and this plasticity is widely regarded as the physical basis of learning and memory. This article covers the structure of the synapse, chemical and electrical transmission, quantal release, and synaptic plasticity, with a worked example and three interactive demonstrations.

Keywords: synapse, chemical synapse, neurotransmitter, quantal release, synaptic plasticity

What a Synapse Is

A *synapse* is the point of functional contact where a neuron passes a signal to another neuron, a muscle fiber, or a gland cell. The word was coined by Charles Scott Sherrington in 1897 to name the junction he had inferred, from the logic of reflexes, must lie between one nerve cell and the next. It is the elementary switch of the nervous system: a typical human neuron carries thousands of synapses on its dendrites and cell body, and the brain as a whole is estimated to hold on the order of a hundred trillion of them. Cognition is, at bottom, the collective activity of these junctions.

Two broad kinds exist. At a *chemical synapse* the two cells remain physically separated by a narrow gap, and the signal crosses as a puff of chemical messenger; at an *electrical synapse* the cells are joined by channels that let current flow directly from one into the other. The chemical synapse is overwhelmingly the more common in the mammalian brain and is the kind meant when the term is used without qualification. Its defining feature is that it is not a fixed wire but an adjustable contact, whose strength can be turned up or down by experience, giving the nervous system its capacity for learning and memory (Kandel, 2001).

Types of Synapses

The Medical Subject Headings vocabulary files *Synapses* (D013569) under two parent kinds at once: *Nervous System*, reflecting the structure it belongs to, and *Intercellular Junctions*, reflecting what a synapse is as a cell-biological object. It then lists the narrower descriptors shown in Table 1. This placement reflects indexing practice rather than a functional taxonomy: MeSH is a controlled vocabulary for cataloguing literature, so its subtypes group the parts and kinds of synapse that organize the research record, not a single clean classification of one thing. The entries below are glossed from their ordinary neuroscientific meaning and do not yet have their own articles on this site.

Structure of a Synapse

A chemical synapse has three parts: a *presynaptic terminal* on the transmitting cell, a *synaptic cleft* between the two cells, and a receptive specialization on the postsynaptic cell. The presynaptic terminal is a swelling at the end of an axon, packed with mitochondria and with dozens to hundreds of *synaptic vesicles*, each a tiny membrane sac holding several thousand molecules of a neurotransmitter. A patch of the terminal membrane, the *active zone*, is the actual release site, and it is precisely there that vesicles dock and fuse. Directly opposite, across the cleft, the postsynaptic membrane is thickened into the *postsynaptic density*, a protein lattice that anchors the receptors in register with the active zone.

The dimensions are small and consequential. The synaptic cleft is only about twenty nanometres wide, so a released transmitter reaches the far side in a fraction of a millisecond. Serial-section electron microscopy has reconstructed these junctions in three dimensions and shown that a synapse is a highly organized structure whose size, vesicle number, and postsynaptic density scale together, so that a physically larger synapse is also a functionally stronger one (Harris & Weinberg, 2012). More recent imaging has revealed that release sites in the active zone are aligned to within tens of nanometres with receptor clusters in the postsynaptic density, forming trans-synaptic nanocolumns that place transmitter release directly over the receptors that must catch it (Biederer, Kaeser, & Blanpied, 2017). This precise apposition is set up during development and maintenance by adhesion molecules, notably the neurexins, that span the cleft and instruct both sides to match (Südhof, 2017).

Figure 1

The Parts of a Chemical Synapse

Chemical Transmission

Chemical transmission runs as an ordered cascade, and the first demonstration steps through it. An action potential travelling down the axon invades the presynaptic terminal and depolarizes its membrane. That depolarization opens voltage-gated calcium channels concentrated at the active zone, and calcium ions rush in. Bernard Katz and Ricardo Miledi established that this calcium entry is the trigger: transmitter release fails without external calcium, and the timing of release follows the timing of calcium influx with a delay of well under a millisecond (Katz & Miledi, 1967). The tight coupling of channel to release site is itself engineered, and the properties of these presynaptic calcium channels set how much transmitter a given spike will liberate (Dolphin & Lee, 2020).

The incoming calcium is caught by a sensor on the vesicle, which drives the vesicle membrane to fuse with the terminal membrane and empty its transmitter into the cleft. Thomas Südhof and colleagues identified the core protein machinery that carries out this fast, calcium-triggered fusion and then recycles the emptied vesicle, a cycle that lets a busy terminal sustain release without exhausting its supply (Südhof, 2004). The released transmitter diffuses across the cleft and binds receptors on the postsynaptic density; those receptors open ion channels, ions flow, and the postsynaptic membrane potential shifts, either toward firing at an excitatory synapse or away from it at an inhibitory one.

Quantal Release

The single most important quantitative fact about the chemical synapse is that transmitter is not released continuously but in discrete packets. Working at the neuromuscular junction, Paul Fatt and Bernard Katz recorded tiny spontaneous voltage blips in the resting muscle, each about half a millivolt, which they interpreted as the response to a single spontaneously released packet, or *quantum*, of transmitter (Fatt & Katz, 1952). José del Castillo and Katz then showed that the evoked response to a nerve impulse is built from a whole number of these same quanta, so the *quantal content* of a response is simply how many packets it contains (del Castillo & Katz, 1954). Each quantum corresponds, we now understand, to the contents of one synaptic vesicle.

Because release is probabilistic, the number of quanta liberated by successive identical impulses varies from trial to trial, and at a low-probability synapse that count follows a Poisson distribution. This yields an elegant way to measure the quantal content without measuring anything about single quanta: if the mean content is m, the fraction of trials that release nothing at all is the failure rate P(0) = e^−m, so counting failures gives m = −ln P(0). The second demonstration plots this Poisson prediction, and the Worked Example applies the method of failures to a concrete recording. This quantal framework remains the foundation for analysing how much any synapse releases and how that changes with plasticity.

Electrical Synapses

Not every synapse works by chemistry. At an *electrical synapse* the membranes of the two cells are bridged by *gap junctions*, clusters of channels that connect the two cytoplasms directly, so that ionic current generated in one cell flows straight into the other. The first direct demonstration came from Edwin Furshpan and David Potter, who recorded at the giant motor synapse of the crayfish and found that the signal passed from one cell to the next electrically, and in that synapse in only one direction, without any chemical intermediary (Furshpan & Potter, 1959).

Electrical synapses have complementary strengths and weaknesses against chemical ones. They are extremely fast, with essentially no synaptic delay, and they are typically bidirectional, so they excel at synchronizing populations of neurons and at reflexes that must not wait. What they lack is the chemical synapse's capacity for sign inversion, amplification, and long-lasting adjustable gain. Far from being a rare curiosity, electrical synapses are now known to be widespread in the mammalian brain and to interact intimately with chemical synapses, so that many circuits use both kinds side by side (Pereda, 2014).

Synaptic Plasticity

The property that makes synapses central to cognition is that their strength is not fixed. In 1973 Timothy Bliss and Terje Lømo found that a brief burst of high-frequency stimulation in the hippocampus produced a strengthening of synaptic transmission that lasted for hours or longer, a phenomenon they named *long-term potentiation* (Bliss & Lømo, 1973). The mirror-image weakening, *long-term depression*, follows prolonged low-frequency stimulation. Together these bidirectional changes let a synapse encode the recent history of its own use, and the third demonstration illustrates how the frequency of activity determines which way, and how far, the strength moves.

At the glutamatergic synapses where they are best understood, both directions of change turn substantially on the trafficking of receptors into and out of the postsynaptic density: potentiation adds receptors and depression removes them, altering how large a response the same amount of transmitter produces (Malenka & Bear, 2004; Collingridge, Isaac, & Wang, 2004). Eric Kandel's work on simpler nervous systems showed that these functional changes are accompanied, for the longest-lasting forms, by the growth of new synaptic connections and by changes in gene expression, tying the molecular biology of the synapse directly to the storage of memory (Kandel, 2001). This is why synaptic plasticity is the dominant physical hypothesis for how the brain learns.

Worked Example: Quantal Content by the Method of Failures

Suppose an experimenter records the response of a single low-output synapse to two hundred identical presynaptic impulses, delivered one at a time. On twenty-seven of those trials the postsynaptic response is completely flat: the synapse released nothing, a failure. The question is how much transmitter the synapse releases on average, expressed as its quantal content m, the mean number of quanta per impulse.

If release is Poisson-distributed, the probability of a failure, meaning zero quanta released, is P(0) = e^−m. From the data, the observed failure rate is 27 divided by 200, which is 0.135. Setting e^−m = 0.135 and taking the natural logarithm gives m = −ln(0.135), which is about 2.00. The synapse releases, on average, two quanta per impulse.

This estimate can be cross-checked against the direct method, which divides the mean size of the evoked response by the size of a single quantum. If the average evoked potential across all two hundred trials measures 0.80 millivolts and a single spontaneous quantum measures 0.40 millivolts, then the quantal content is 0.80 divided by 0.40, again exactly 2.0. That two independent routes, one counting only failures and the other averaging amplitudes, agree on the same value is the classic evidence that transmitter genuinely comes in quantal packets. The second demonstration lets the reader vary m and confirm that a mean content of 2.0 predicts the 13.5 percent failure rate used here.

Discussion

The synapse changes how one should think about the physical basis of mind. A neuron is often pictured as a wire, but the crucial computation happens at the adjustable, probabilistic junctions between neurons rather than along the cables. Because a chemical synapse can be excitatory or inhibitory, can amplify or attenuate, and can strengthen or weaken with use, it is the site where the nervous system does its arithmetic and stores its results. A memory, on the leading account, is not a thing kept in a cell but a pattern of altered synaptic weights distributed across a network in the cerebral cortex and hippocampus (Kandel, 2001).

This has consequences throughout cognitive psychology. The probabilistic, quantal nature of release means that transmission at a single synapse is unreliable, and reliability is bought only by averaging over many synapses, which constrains how neural circuits must be built. The bidirectional plasticity of synapses supplies a concrete mechanism for learning and for the formation of the durable representations that underlie working memory and skilled behavior. And because so many disorders of cognition, from addiction to Alzheimer disease, involve synapses that are too strong, too weak, or lost, the synapse is where much of the effort to understand and treat those conditions is now concentrated (Malenka & Bear, 2004).

Current Directions

One active frontier concerns the molecular logic that specifies which cell connects to which. The neurexins and their binding partners form a combinatorial code of adhesion molecules that appears to instruct the identity and properties of a synapse, and unravelling that code is a central project, with direct relevance to the many neurodevelopmental conditions in which these genes are disrupted (Südhof, 2017). A closely related effort asks how the presynaptic active zone is assembled in the first place, and how its scaffold proteins position calcium channels the exact distance from vesicles that fast release requires (Emperador-Melero & Kaeser, 2020).

A second frontier is spatial precision. The finding that release sites and receptor clusters are aligned into trans-synaptic nanocolumns has reframed synaptic strength as a matter of nanometre-scale architecture, not merely of how much transmitter is released or how many receptors are present, and how these nanocolumns are built and remodelled during plasticity is an open question (Biederer, Kaeser, & Blanpied, 2017). Running through both frontiers is the recognition that electrical and chemical synapses are not separate worlds but interacting elements of the same circuits, so that a full account of any network must treat the two together (Pereda, 2014).

Common Misconceptions

The two neurons at a synapse touch.

At a chemical synapse the cells are separated by a synaptic cleft about twenty nanometres wide, and the signal crosses as a diffusing chemical messenger rather than by direct contact (Harris & Weinberg, 2012).

A synapse releases transmitter smoothly and reliably.

Release is quantal and probabilistic: transmitter comes in discrete packets, and an identical impulse may release two quanta, one, or none, so a single synapse is an unreliable transmitter (del Castillo & Katz, 1954).

All synapses are chemical.

Electrical synapses, in which gap junctions couple cells directly, are widespread in the mammalian brain and work alongside chemical synapses in many circuits (Pereda, 2014).

Synaptic connections are fixed once the brain matures.

Synaptic strength changes throughout life through long-term potentiation and depression, and the strongest forms of change even add or remove whole synapses (Bliss & Lømo, 1973; Kandel, 2001).

Glossary

Active zone.

The specialized patch of presynaptic membrane where synaptic vesicles dock and fuse to release transmitter, aligned opposite the postsynaptic density.

Chemical synapse.

A junction at which the signal crosses as a neurotransmitter released from the presynaptic terminal onto receptors across a cleft; the dominant type in the brain.

Electrical synapse.

A junction at which gap junctions couple two cells directly, letting ionic current pass between them with almost no delay.

Gap junction.

A cluster of channels that connects the cytoplasm of two adjacent cells, the structural basis of an electrical synapse.

Long-term depression.

A lasting weakening of synaptic strength, typically induced by prolonged low-frequency activity; the counterpart of potentiation.

Long-term potentiation.

A lasting strengthening of synaptic transmission following brief high-frequency stimulation; the leading candidate mechanism for memory.

Neurotransmitter.

The chemical messenger released from synaptic vesicles at a chemical synapse to carry the signal across the cleft to the postsynaptic cell.

Postsynaptic density.

The protein lattice thickening the postsynaptic membrane that anchors neurotransmitter receptors in register with the presynaptic active zone.

Presynaptic terminal.

The transmitting end of the axon, containing synaptic vesicles and the active zone from which transmitter is released.

Quantal content.

The mean number of quanta of transmitter released per presynaptic impulse, written m, a basic measure of synaptic strength.

Quantum.

The discrete packet of transmitter released as a unit, corresponding to the contents of a single synaptic vesicle.

Receptor.

A postsynaptic membrane protein that binds a neurotransmitter and opens an ion channel or triggers a signalling cascade in response.

Synaptic cleft.

The narrow gap, about twenty nanometres wide, separating the presynaptic and postsynaptic membranes at a chemical synapse.

Synaptic vesicle.

A small membrane-bound sac in the presynaptic terminal that stores a quantum of neurotransmitter and releases it by fusing with the membrane.

Key Researchers

Timothy V. P. Bliss

A neuroscientist at the Francis Crick Institute in London who, with Terje Lømo, discovered long-term potentiation in the hippocampus, the leading synaptic model of how memories are stored.
Wikipedia - Wikidata - Royal Society

Kristen M. Harris

A neuroscientist at the University of Texas at Austin who reconstructs the ultrastructure of synapses and dendritic spines by serial-section three-dimensional electron microscopy; elected to the National Academy of Sciences in 2024.
Faculty page - Laboratory

Pascal S. Kaeser

A neuroscientist at Harvard Medical School who studies the molecular assembly of the presynaptic active zone and how it positions release sites opposite postsynaptic receptors.
Faculty page - ORCID - Google Scholar

Eric R. Kandel

A neuroscientist at Columbia University who showed that short- and long-term memory correspond to functional and structural change at the synapse; awarded the 2000 Nobel Prize in Physiology or Medicine.
ORCID - Wikipedia - Wikidata - Nobel biography

Bernard Katz

(1911-2003). A biophysicist at University College London who established that transmitter is released in fixed multimolecular quanta and that presynaptic calcium entry triggers their release; awarded the 1970 Nobel Prize in Physiology or Medicine.
Wikipedia - Wikidata - Nobel facts

Robert C. Malenka

A neuroscientist at Stanford University who dissected the receptor-trafficking mechanisms of long-term potentiation and depression at glutamatergic synapses.
Wikipedia - Wikidata - Google Scholar - Faculty page

Charles Scott Sherrington

(1857-1952). A physiologist at the University of Oxford who coined the word synapse in 1897 to name the specialized junction he inferred between neurons; awarded the 1932 Nobel Prize in Physiology or Medicine.
Wikipedia - Wikidata

Thomas C. Südhof

A neuroscientist at Stanford University who identified the core molecular machinery of fast, calcium-triggered vesicle fusion and the neurexins that organize the synapse; awarded the 2013 Nobel Prize in Physiology or Medicine.
ORCID - Wikipedia - Wikidata - Google Scholar

Frequently Asked Questions

What is a synapse?

A synapse is the specialized junction at which one neuron transmits a signal to another cell, whether another neuron, a muscle, or a gland. It is the elementary unit of communication in the nervous system, and a single neuron may carry many thousands of them (Kandel, 2001).

What is the difference between a chemical and an electrical synapse?

At a chemical synapse the cells are separated by a cleft and the signal crosses as a released neurotransmitter, allowing amplification, sign inversion, and plasticity. At an electrical synapse gap junctions couple the cells directly, giving very fast, usually bidirectional transmission but little capacity for gain control (Pereda, 2014).

How does a chemical synapse transmit a signal?

An action potential depolarizes the presynaptic terminal, opening calcium channels; the calcium that enters triggers synaptic vesicles to fuse and release neurotransmitter into the cleft; the transmitter binds postsynaptic receptors and changes the membrane potential of the receiving cell (Katz & Miledi, 1967; Südhof, 2004).

What does it mean that transmitter release is quantal?

Transmitter is released in discrete packets called quanta, each corresponding to the contents of one synaptic vesicle. The number of quanta released by a given impulse varies from trial to trial, so release at a single synapse is inherently probabilistic (Fatt & Katz, 1952; del Castillo & Katz, 1954).

What is quantal content and how is it measured?

Quantal content is the mean number of quanta released per impulse, written m. At a low-probability synapse the failure rate equals e^−m, so counting the trials on which nothing is released gives m as the negative natural logarithm of that failure fraction (del Castillo & Katz, 1954).

What is synaptic plasticity?

Synaptic plasticity is the lasting change in the strength of a synapse with use. Brief high-frequency activity produces long-term potentiation, a strengthening, while prolonged low-frequency activity produces long-term depression, a weakening (Bliss & Lømo, 1973; Malenka & Bear, 2004).

Why are synapses important for learning and memory?

Because synaptic strength can be increased or decreased and the change persists, a pattern of activity can leave a durable trace in the pattern of synaptic weights across a network. This is the leading physical hypothesis for how the brain stores memories (Kandel, 2001).

How wide is the synaptic cleft?

The cleft at a typical chemical synapse is about twenty nanometres across, narrow enough that a released neurotransmitter reaches receptors on the far side in well under a millisecond, as shown by three-dimensional electron-microscopic reconstruction (Harris & Weinberg, 2012).

References

Biederer, T., Kaeser, P. S., & Blanpied, T. A. (2017). Transcellular nanoalignment of synaptic function. Neuron, 96(3), 680-696. https://doi.org/10.1016/j.neuron.2017.10.006

Bliss, T. V. P., & Lømo, T. (1973). Long-lasting potentiation of synaptic transmission in the dentate area of the anaesthetized rabbit following stimulation of the perforant path. The Journal of Physiology, 232(2), 331-356. https://doi.org/10.1113/jphysiol.1973.sp010273

Collingridge, G. L., Isaac, J. T. R., & Wang, Y. T. (2004). Receptor trafficking and synaptic plasticity. Nature Reviews Neuroscience, 5(12), 952-962. https://doi.org/10.1038/nrn1556

del Castillo, J., & Katz, B. (1954). Quantal components of the end-plate potential. The Journal of Physiology, 124(3), 560-573. https://doi.org/10.1113/jphysiol.1954.sp005129

Dolphin, A. C., & Lee, A. (2020). Presynaptic calcium channels: Specialized control of synaptic neurotransmitter release. Nature Reviews Neuroscience, 21(4), 213-229. https://doi.org/10.1038/s41583-020-0278-2

Emperador-Melero, J., & Kaeser, P. S. (2020). Assembly of the presynaptic active zone. Current Opinion in Neurobiology, 63, 95-103. https://doi.org/10.1016/j.conb.2020.03.008

Fatt, P., & Katz, B. (1952). Spontaneous subthreshold activity at motor nerve endings. The Journal of Physiology, 117(1), 109-128. https://doi.org/10.1113/jphysiol.1952.sp004735

Furshpan, E. J., & Potter, D. D. (1959). Transmission at the giant motor synapses of the crayfish. The Journal of Physiology, 145(2), 289-325. https://doi.org/10.1113/jphysiol.1959.sp006143

Harris, K. M., & Weinberg, R. J. (2012). Ultrastructure of synapses in the mammalian brain. Cold Spring Harbor Perspectives in Biology, 4(5), a005587. https://doi.org/10.1101/cshperspect.a005587

Kandel, E. R. (2001). The molecular biology of memory storage: A dialogue between genes and synapses. Science, 294(5544), 1030-1038. https://doi.org/10.1126/science.1067020

Katz, B., & Miledi, R. (1967). The timing of calcium action during neuromuscular transmission. The Journal of Physiology, 189(3), 535-544. https://doi.org/10.1113/jphysiol.1967.sp008183

Malenka, R. C., & Bear, M. F. (2004). LTP and LTD: An embarrassment of riches. Neuron, 44(1), 5-21. https://doi.org/10.1016/j.neuron.2004.09.012

Pereda, A. E. (2014). Electrical synapses and their functional interactions with chemical synapses. Nature Reviews Neuroscience, 15(4), 250-263. https://doi.org/10.1038/nrn3708

Südhof, T. C. (2004). The synaptic vesicle cycle. Annual Review of Neuroscience, 27, 509-547. https://doi.org/10.1146/annurev.neuro.26.041002.131412

Südhof, T. C. (2017). Synaptic neurexin complexes: A molecular code for the logic of neural circuits. Cell, 171(4), 745-769. https://doi.org/10.1016/j.cell.2017.10.024