Ion channel

A protein containing a regulated, selective pore, which mediates the flow of particular ions down their electrochemical gradient from one side of a biological membrane to the other.

Ion channels belong to families of proteins that translocate substances across biological membranes. These families include channels, pumps, and transporters (Aidley 1998). Channels open to permit the flow of ions down their electrochemical gradient. They are fundamental to all types of excitable cells. Pumps drive ions against their electrochemical gradient. Transporters bind specific ligands and undergo a conformational change that releases the ligand on the other side of the membrane (Reith 1997;Amara 1998). In some cases the distinction between a channel and a transporter is blurred (Valverde et al. 1992).

Ion channels are classified on the basis of their ion selectivity and the cellular signal that opens or closes ('gates') them (Hille 1992; Rudy and Iverson 1992; Johnston and Wu 1995; Aidley 1998; Conn 1998). "Taxonomy on the basis of the preferred translocated ion distinguishes among sodium (Na+), potassium (K+), "calcium (Ca2+), and chloride (Cl-) channels. Each of these types is further classified on the basis of some phenotypic or functional characteristic, such as kinetics of activation and inactivation, modulation by ligands, or assumed physiological role. Taxonomy on the basis of gating distinguishes between 'voltage'-, 'second messenger'-, and 'agonist'-gated channels. 'Voltage-gated' channels are regulated by membrane potential, 'second messenger-gated' channels by second messengers such as cyclic nucleotides or Ca2+ ("intracellular signal transduction cascade), and 'agonist-' or 'ligand-gated' channels by "neurotransmitters (see "receptor). Voltage-gated and second messenger-gated channels are considered as members of the same super-family. With the advances in molecular biology and the identification of hundreds of genes that encode channel proteins, additional taxonomies are now possible, which take into account previously unknown 'evolutionary kinship' (e.g. Jan and Jan 1992).

The history of research on ion channels is an intriguing chapter in biophysics and neurobiology, now spanning over a century (reviewed in Hille 1992; Armstrong and Hille 1998; Colquhoun and Sakmann 1998). It began with pioneering observations on the phenomenology of excitable membranes (Bernstein 1902, cited in Hille 1992). It then evolved through a set of methodological and conceptual breakthroughs to yield robust, quantitative "models of excitability (e.g. Hodgkin and Huxley 1952). In recent years, a combination of cellular physiology and molecular biology started to unveil the detailed structure-function relationship of individual channels and their interaction with other cellular components. One of the hallmarks of the field is an attempt to comprehend the computations, temporal dynamics, and use-dependent modification of multiple types of channels in neuronal compartments (e.g. Koester and Sakmann 1998). The molecular and cellular biology of ion channels is no doubt one of the most booming subdisciplines of neuroscience. At the time of writing, 5000 papers a year are published on voltage-gated channels alone. Our interest here lies in the role of channels in learning and memory, which narrows the field a bit, but still not enough to do minimal justice to its rich spectrum of themes in a few sentences. Hence, only a few prominent examples will be noted.

It is useful to consider the role of channels in learning and memory in the context of the operation of hypothetical cellular 'learning machines'. Such 'biological machines' are expected to include several types of molecular devices that embody "acquisition, association, storage, and readout of information (Dudai 1993). Cellular acquisition devices are receptors capable of responding to incoming stimuli. Cellular conjuncture devices are "coincidence detectors, associating multiple stimuli. Cellular storage devices are capable of retaining a change in cellular function over time. Readout devices are macromolecules whose activity in "retrieval expresses the experience-dependent alteration in cellular activity. Several of the above roles can be realized in a single molecule.

Ligand-gated channels fit to serve as cellular acquisition devices, although they can also subserve association, retention, and retrieval of information. The

Fig. 40 An example of an agonist-gated (alias ligand-gated) ion channel. This is a highly simplified scheme of the a-amino-3-hydroxy-5-methyl-4-lsoxazole propionic acid (AMPA)-type *glutamate receptor, which mediates fast excitatory transmission in the vertebrate brain, and is considered to subserve neuronal plasticity, e.g. *LTP, and possibly memory. For example, it is proposed that AMPA receptor channel availability increases upon use-dependent facilitation of central 'synapses (e.g. Nayak et al. 1998). It is also a target for experimental memory-enhancing drugs ('nootropics). Glutamate binds to a site on the extracellular domain (not indicated in the drawing), and activates the channel. The channel itself is a central aqueous pore engulfed by multiple types of subunits (GluR) that transverse the neuronal membrane.The pore mediates primarily sodium (Na+) influx, potassium efflux (not shown), and, depending on the composition of the subunits that form the channel 'wall', *calcium (Ca2+) influx. (a) The activated channel allows only Na+ in. (b) The activated channel allows calcium influx as well. Q is the amino acid glutamine, R is arginine, which is positively charged and can block calcium entry. Some naturally occurring poisons, e.g. certain spider toxins, can also block calcium permeability. Out, In—outer and inner faces of the neuronal membrane, respectively. (Adapted from Pellegrini-Giampietro et al. 1997.)

Fig. 40 An example of an agonist-gated (alias ligand-gated) ion channel. This is a highly simplified scheme of the a-amino-3-hydroxy-5-methyl-4-lsoxazole propionic acid (AMPA)-type *glutamate receptor, which mediates fast excitatory transmission in the vertebrate brain, and is considered to subserve neuronal plasticity, e.g. *LTP, and possibly memory. For example, it is proposed that AMPA receptor channel availability increases upon use-dependent facilitation of central 'synapses (e.g. Nayak et al. 1998). It is also a target for experimental memory-enhancing drugs ('nootropics). Glutamate binds to a site on the extracellular domain (not indicated in the drawing), and activates the channel. The channel itself is a central aqueous pore engulfed by multiple types of subunits (GluR) that transverse the neuronal membrane.The pore mediates primarily sodium (Na+) influx, potassium efflux (not shown), and, depending on the composition of the subunits that form the channel 'wall', *calcium (Ca2+) influx. (a) The activated channel allows only Na+ in. (b) The activated channel allows calcium influx as well. Q is the amino acid glutamine, R is arginine, which is positively charged and can block calcium entry. Some naturally occurring poisons, e.g. certain spider toxins, can also block calcium permeability. Out, In—outer and inner faces of the neuronal membrane, respectively. (Adapted from Pellegrini-Giampietro et al. 1997.)

N-methyl-D-aspartate receptor channel (NMDAR) is but one example. This macromolecular complex is composed of two types of subunits, NR1 and NR2. The NR2 subunit contains the recognition site for the "neurotransmitter "glutamate, and NR1, for glycine (Anson et al. 1998).1 The channel is basically the inner pore formed by the aggregation of four subunits (e.g. Rosenmund et al. 1998), always containing NR1 and at least one of multiple NR2 subtypes. It is a Ca2+ channel that is blocked by magnesium ions under resting conditions. Binding of glutamate activates the receptor but does not remove the magnesium block. To release the latter, the membrane must become depolarized. The NMDA receptor channel is hence a coincident detector, gated by both agonist and voltage (e.g. Seeburg et al. 1995). It is assumed to play a decisive part in acquisition of certain forms of "long-term potentiation and learning. The NMDA receptor channel is probably more than a cellular acquisition device; it is known to undergo lasting changes in response to neuronal activation (Rosenblum et al. 1996). Some of these changes may not relate to the properties of the current but rather to the interfacing of the complex with intracellu-lar cascades. These post-translational modifications may turn the complex into a device that stores information (i.e. functional change in the nerve cell) over the first few hours after training.

Another example for the role of channels in learning relates to "synaptic facilitation in the circuit that subserves "sensitization of defensive reflexes in "Aplysia (Byrne and Kandel 1996). In this system a major contribution to synaptic facilitation, a cellular analogue of "sensitization, is made by use-dependent enhancement of excitability and neurotransmitter release in the sensory-to-motor synapses in the circuit. It involves state- and time-dependent modulation of voltage-dependent as well as voltage-independent potassium (K+) conductances. For our purpose, suffice it to note that: (a) K+ channels are critical for the synaptic change that contributes to the behavioural change in the reflex; (b) these channels play a part in storage (part of the memory at the cellular "level is the lasting modification in K+ conductances, although the memory-keeping step may not be in the channel itself, but rather in a "protein kinase that keeps modifying the re-modified as well as the newly synthesized copies of channel molecules); and (c) K+ channels are also readout devices, at least in the short-term (in retrieval, the action potential, which encodes the test stimulus in the sensory neuron, encounters a presynaptic membrane with modified channel(s), and therefore triggers a modified sensory-to-motor signal). Whether the same or similar channels also play a part in storage and retrieval of long-term memory is still unclear ("late response genes).

The roles of NMDAR in NMDA-dependent LTP and of K+ channels in learning in Aplysia are only selected examples (for additional proven or postulated roles of channels in neuronal and behavioural "plasticity, see Changeux et al. 1998; Blackwell and Alkon 1999; MacDermott et al. 1999). Computations performed by batteries of channels will surely occupy prominent positions in future "algorithms of biological learning, and in "models that interrelate events at the molecular, cellular, circuit, and behaviour levels (Dudai 1997b; "reduction). And, finally, on the more pragmatic side: ion channels also provide promising targets for cognitive enhancers (Eid and Rose 1999, "nootropics).

Selected associations: Coincidence detector, Neurotransmitter, Receptor, Reduction

1The reader may wonder why, if this is the case, the receptor channel is named after glutamate rather than glycine. Well, this has partially to do with the history of the field, but is utterly justified: under physiological conditions the concentration of glycine is usually sufficient to occupy the glycine site, whereas the concentration of glutamate is a sensitive function of incoming stimuli and hence critical in determining the activity state of the complex.

Was this article helpful?

0 0

Post a comment