Neuronal Excitability

POSTSYNAPTIC EVENTS

The neuronal membrane normally has a resting membrane potential around — 70 mV (inside negative in respect of outside) with Na+ and Cl— concentrated on the outside and K+ on the inside prepared to move down their concentration gradients when the appropriate ion channels are opened (Fig. 1.4). On arrival of an excitatory impulse the Na+ channels are opened and there is an increased influx of Na+ so that the resting potential moves towards the so-called equilibrium potential for Na+ (+50 mV) when Na+ influx equals Na+ outflux but at —60 to — 65 mV, the threshold potential, there is a sudden increase in Na+ influx. This depolarisation leads to the generation of a propagated action potential. The initial subthreshold change in membrane potential parallels the action of the excitatory transmitter and is graded in size according to the amount of NT released. It is known as the excitatory postsynaptic potential (EPSP) and lasts about 5 ms.

An inhibitory input increases the influx of Cl to make the inside of the neuron more negative. This hyperpolarisation, the inhibitory postsynaptic potential (IPSP), takes the membrane potential further away from threshold and firing. It is the mirror-image of the EPSP and will reduce the chance of an EPSP reaching threshold voltage.

Such clear postsynaptic potentials can be recorded intracellularly with microelec-trodes in large quiescent neurons after appropriate activation but may be somewhat artificial. In practice a neuron receives a large number of excitatory and inhibitory inputs and its bombardment by mixed inputs means that its potential is continuously changing and may only move towards the threshold for depolarisation if inhibition fails or is overcome by a sudden increase in excitatory input.

Not all influences on, or potentials recorded from, a neuron have the same time-course as the EPSP and IPSP, which follow the rapid opening of Na+ and Cl— ion channels directly linked to NT receptors. There are also slowly developing, longer lasting and smaller non-propagated (conditioning) changes in potential most of which appear to have a biochemical intermediary in the form of G-proteins linked to the activation (Gs) or

Ionic Basis Action Potential

Figure 1.4 Ionic basis for excitatory postsynaptic potentials (EPSPs) and inhibitory postsynaptic potentials (IPSPs). Resting membrane potential ( —70mV) is maintained by Na+ influx and K+ efflux. Varying degrees of depolarisation, shown by different sized EPSPs (a and b), are caused by increasing influx of Na+. When the membrane potential moves towards threshold potential (60-65 mV) an action potential is initiated (c). The iPSPs (a'b') are produced by an influx of CI—. Coincidence of an EPSP (b) and IPSP (a') reduces the size of the EPSP (d)

Figure 1.4 Ionic basis for excitatory postsynaptic potentials (EPSPs) and inhibitory postsynaptic potentials (IPSPs). Resting membrane potential ( —70mV) is maintained by Na+ influx and K+ efflux. Varying degrees of depolarisation, shown by different sized EPSPs (a and b), are caused by increasing influx of Na+. When the membrane potential moves towards threshold potential (60-65 mV) an action potential is initiated (c). The iPSPs (a'b') are produced by an influx of CI—. Coincidence of an EPSP (b) and IPSP (a') reduces the size of the EPSP (d)

inhibition (Gi) of adenylate cyclase and cyclic AMP production or IP3 breakdown (see Chapter 2). They can be excitatory (depolarising) or inhibitory (hyperpolarising) generally involving the opening or closing of K+ channels. This can be achieved directly by the G-protein or second messenger but more commonly by the latter causing membrane phosphorylation through initiating appropriate kinase activity.

Thus the activity of a neuron can be controlled in a number of ways by NTs activating appropriate receptors (Fig. 1.5). Two basic receptor mechanisms are involved:

(1) Ionotropic Those linked directly to ion channels such as those for Na+ (e.g. ACh nicotinic or some glutamate receptors) or Cl— (GABA), involving fast events with increased membrane conductance and ion flux.

(2) Metabotropic Those not directly linked to ion channels but initiating biochemical processes that mediate more long-term effects and modify the responsiveness of the neuron. With these the first messenger of synaptic transmission, the NT, activates a second messenger to effect the change in neuron excitability. They are normally associated with reduced membrance conductance and ion flux (unless secondary to

Figure 1.5 The degree of ion channel opening can be controlled (gated) either directly (ionotropic effect) or indirectly (metabotropic effect). In the former the neurotransmitter combines with a receptor that is directly linked to the opening of an ion channel (normally Na+ or Cl_) while in the latter the receptor activates a G-protein that can directly interact with the ion channel (most probably K+ or Ca2+) but is more likely to stimulate (Gs) or inhibit (Gi) enzymes controlling the levels of a second messenger (e.g. cAMP, GMP, IP3). These in turn may also directly gate the ion channel but generally control its opening through stimulating a specific protein kinase that causes phosphorylation of membrane proteins and a change in state of the ion channel. The latter (metabotropic) effects may either open or close an ion channel (often K+) and are much slower (lOOs ms to min) than the ionotropic ones (l-lOms). A variety of neuro-transmitters, receptors, second messengers and ion channels permits remarkably diverse and complex neuronal effects

Figure 1.5 The degree of ion channel opening can be controlled (gated) either directly (ionotropic effect) or indirectly (metabotropic effect). In the former the neurotransmitter combines with a receptor that is directly linked to the opening of an ion channel (normally Na+ or Cl_) while in the latter the receptor activates a G-protein that can directly interact with the ion channel (most probably K+ or Ca2+) but is more likely to stimulate (Gs) or inhibit (Gi) enzymes controlling the levels of a second messenger (e.g. cAMP, GMP, IP3). These in turn may also directly gate the ion channel but generally control its opening through stimulating a specific protein kinase that causes phosphorylation of membrane proteins and a change in state of the ion channel. The latter (metabotropic) effects may either open or close an ion channel (often K+) and are much slower (lOOs ms to min) than the ionotropic ones (l-lOms). A variety of neuro-transmitters, receptors, second messengers and ion channels permits remarkably diverse and complex neuronal effects an increased Ca2+ conductance) and may involve decreased Na+ influx (inhibitory) or K+ efflux (excitatory). Some amines (e.g. noradrenaline) may increase K+ efflux (inhibitory).

These two basic mechanisms could provide a further classification for NTs, namely fast and slow acting, although one NT can work through both mechanisms using different receptors. The slow effects can also range from many milliseconds to seconds, minutes, hours or even to include longer trophic influences. What will become clear is that while one NT can modify a number of different membrane ion currents through different mechanisms and receptors, one current can also be affected by a number of different NTs. The control of neuronal excitability is discussed in more detail in Chapter 2.

PRESYNAPTIC EVENTS

So far we have assumed that a NT can only modify neuronal activity by a postsynaptic action. Recently, interest has also turned to presynaptic events. It has been known for many years that stimulation of muscle or cutaneous afferents to one segment of the spinal cord produces a prolonged inhibition of motoneuron activity without any accompanying change in conductance of the motoneuron membrane, i.e. no IPSP. Such inhibition is probably, therefore, of presynaptic origin and is, in fact, associated with a depolarisation of the afferent nerve terminals and a reduction in release of the excitatory NT. If it is assumed that the amount of NT released from a nerve terminal depends on the amplitude of the potential change induced in it, then if that terminal is already partly depolarised when the impulse arrives there will be a smaller change in potential and it will release less transmitter (Fig. 1.6). There is no direct evidence for this concept from studies of NT release but electrophysiological experiments at the crustacean neuromuscular junction, which has separate excitatory and inhibitory inputs, show that stimulation of the inhibitory nerve, which released GABA, reduced the EPSP evoked postsynaptically by an excitatory input without directly hyperpolarising (inhibiting) the muscle fibre. Certainly when GABA is applied to various in vivo and in vitro preparations (spinal cord, cuneate nucleus, olfactory cortex) it will produce a depolarisation of afferent nerve terminals that spreads sufficiently to be recorded in their distal axons.

Such presynaptic inhibition can last much longer (50-100 ms) than the postsynaptic form (5 ms) and can be a very effective means of cutting off one particular excitatory input without directly reducing the overall response of the neuron. How GABA can produce both presynaptic depolarisation and conventional postsynaptic hyperpolarisation by the same receptor, since both effects are blocked by the same antagonist, bicuculline, is uncertain (see Chapter 2) although an increased chloride flux appears to be involved in both cases. If nerve terminals are depolarised, rather than hyper-polarised by increased chloride flux, then their resting membrane potential must be different from (greater than) that of the cell body so that when chloride enters and the potential moves towards its equilibrium potential there is a depolarisation instead of a hyperpolarisation. Alternatively, chloride efflux must be achieved in some way.

This form of presynaptic inhibition must be distinguished from another means of attenuating NT release, i.e. autoinhibition. This was first shown at peripheral noradrenergic synapses where the amount of noradrenaline released from nerve terminals is reduced by applied exogenous noradrenaline and increased by appropriate (alpha) adrenoceptor antagonists. Thus through presynaptic (alpha) adrenoreceptors, which can be distinguished from classical postsynaptic (alpha) adrenoreceptors by relatively specific agonists and antagonists, neuronal-released noradrenaline is able to inhibit its own further (excessive) release. It is a mechanism for controlling the synaptic concentration of noradrenaline. This inhibition does not necessarily involve any change in membrane potential but the receptors are believed to be linked to and inhibit adenylate cyclase. Whether autoinhibition occurs with all NTs is uncertain but there is strong evidence for it at GABA, dopamine and 5-HT terminals.

There is also the interesting possibility that presynaptic inhibition of this form, with or without potential changes, need not be restricted to the effect of the NT on the terminal from which it is released. Numerous studies in which brain slices have been loaded with a labelled NT and its release evoked by high K+ or direct stimulation show

Drug Use And Brain Function

Figure 1.6 Presynaptic inhibition of the form seen in the dorsal horn of the spinal cord. (a) The axon terminal (i) of a local neuron is shown making an axo-axonal contact with a primary afferent excitatory input (ii). (b) A schematic enlargement of the synapse. (c) Depolarisation of the afferent terminal (ii) at its normal resting potential by an arriving action potential leads to the optimal release of neurotransmitter. (d) When the afferent terminal (ii) is already partially depolarised by the neurotransmitter released onto it by (i) the arriving acting potential releases less transmitter and so the input is less effective

Figure 1.6 Presynaptic inhibition of the form seen in the dorsal horn of the spinal cord. (a) The axon terminal (i) of a local neuron is shown making an axo-axonal contact with a primary afferent excitatory input (ii). (b) A schematic enlargement of the synapse. (c) Depolarisation of the afferent terminal (ii) at its normal resting potential by an arriving action potential leads to the optimal release of neurotransmitter. (d) When the afferent terminal (ii) is already partially depolarised by the neurotransmitter released onto it by (i) the arriving acting potential releases less transmitter and so the input is less effective that such release can be inhibited by a variety of other NTs. A noradrenergic terminal has been shown to possess receptors for a wide range of substances, so-called heteroceptors (see Langer 1981, 1997) and although this may be useful for developing drugs to manipulate noradrenergic transmission it seems unlikely that in vivo all of the receptors could be innervated by appropriate specific synapses or reachable by their NT. They may be pharmacologically responsive but not always physiologically active (see Chapter 4).

CONTROL OF SYNAPTIC NT CONCENTRATION

Having briefly discussed the presynaptic control of NT release it is necessary to consider how the concentration of a NT is controlled at a synapse so that it remains localised to its site of release (assuming that to be necessary) without its effect becoming too excessive or persistent.

Although one neuron can receive hundreds of inputs releasing a number of different NTs, the correct and precise functioning of the nervous system presumably requires that a NT should only be able to act on appropriate receptors at the site of its release. This control is, of course, facilitated to some extent by having different NTs with specific receptors so that even if a NT did wander it could only work where it finds its receptors and was still present in sufficient concentration to meet their affinity requirements. Normally the majority of receptors are also restricted to the immediate synapse.

Nevertheless, from release (collection) studies we know that enough NT must diffuse (overflow) to the collecting system, be that a fine probe in vivo or the medium of a perfusion chamber in vitro, to be detected. Thus one must assume that either the concentration gradient from the collecting site back to the active synaptic release site is so steep that the NT can only reach an effective concentration at the latter, or it is not unphysiological for a NT to have an effect distal from its site of release.

Released NT, if free to do so, would diffuse away from its site of release at the synapse down its concentration gradient. The structure of the synapse and the narrow gap between pre- and postsynaptic elements reduces this possibility but this means that there must be other mechanisms for removing or destroying the NT so that it, and its effects, do not persist unduly at the synapse but are only obtained by regulated impulse controlled release. In some cases, e.g. ACh, this is achieved by localised metabolising enzymes but most nerve terminals, especially those for the amino acids and monoamines, possess very high-affinity NT uptake systems for the rapid removal of released NT. In fact these are all Na+- and Cl_-dependent, substrate-specific, high-affinity transporters and in many cases their amino-acid structure is known and they have been well studied. Transport can also occur into glia as well as neurons and this may be important for the amino acids. Of course, a further safeguard against an excessive synaptic concentration of the NT is the presence of autoreceptors to control its release.

Thus there are mechanisms to ensure that NTs neither persist uncontrollably at the synapse nor produce dramatic effects distal from it. Studies of glutamate release always show a measurable basal level (1-3 ^M), although this may not all be of NT origin, and yet it is very difficult to increase that level even by quite intense stimulation. Whether this is a safeguard against the neurotoxicity caused by the persistent intense activation of neurons by glutamate (see Chapter 9), or just to ensure that neurons remain responsive to further stimulation is unclear, as is the mechanism by which it is achieved.

Despite the above precautions, it is still possible that NT spillover and extrasynaptic action may occur and indeed could be required in some instances. Thus the diffusion of glutamate beyond the synapse could activate extrasynaptic high-affinity NMDA or metabotropic receptors (Chapter 9) to produce long-lasting effects to maintain activity in a network. This may be important in long-term potentiation and memory effects. Crosstalk between synapses could also act as a back-up to ensure that a pathway functions properly (see Barbour and Hausser 1997).

Was this article helpful?

0 0

Post a comment