Changes in Extracellular Ionic Composition q

Changes in Extracellular Ionic Composition q JL Stringer, Baylor College of Medicine, Houston, TX, United States Ó 2017 Elsevier Inc. All rights reserved. Introduction 1 Background 1 Methods 2 Recent Results 2 Future Goals 4 Further Reading 5 Relevant Website 5 Introduction All neurons and glia maintain an electrical potential across the cell membrane, which is negative relative to the extracellular environment. This negative potential provides a driving force for charged particles to move across the membrane. Every nerve impulse is dependent on this movement of ions. Most of these ions move through ion channels, along concentration gradients generated by electrogenic pumps, but some ions move across membranes on transporters. Abnormalities in the control of ionic movement across membranes can result in abnormalities in neuronal excitability. With neuronal activity, action potential firing results in movement of potassium out of the cell and sodium into it. A single action potential will result in little change in the ion concentrations in the extracellular space. During seizures (ie, intense neuronal activity), however, the amount of potassium moving out of neurons exceeds the ability of the brain to regulate precisely the ion concentrations, and the extracellular potassium will increase from a resting level of around 3 mm to 10 12 mm. At the same time calcium, magnesium, sodium and chloride enter cells, and the extracellular concentrations of these ions will decrease. After any activity that alters the ionic gradients, there needs to be correction of the concentrations back to the baseline, to resting state. Potassium, which is elevated, can diffuse away, can be taken up into neurons or glia by active pumps, or can be taken up by passive processes into astrocytes. Ions that are low in the extracellular space (Na þ,ca þþ,mg þþ,cl ) can be adjusted by diffusion back into the area of low concentration, or ions can be pumped out of cells back into the extracellular space. Astrocytes are the most numerous type of glial cell. They are involved in the maintenance of the blood brain barrier, in the regulation of water and ion homeostasis, and in metabolism of amino acid transmitters. Astrocytes are proposed to play a major role in the maintenance of the extracellular space, including the concentrations of the various ions. In particular, astrocytes are thought to contribute a significantly to the removal of excess potassium through glial spatial buffering, through activation of the glial Na K pump, and via activation of inwardly rectifying potassium channels. Astrocytes also have sodium, chloride, and bicarbonate transporters and exchangers that contribute to the regulation of these ions in the extracellular space. And astrocytes express multiple calcium channels, suggesting that they also play a role in regulation of calcium levels. Background Since ion-selective electrodes have been available, a number of investigators have measured changes in extracellular potassium in a several brain regions during neuronal activity. During seizures or intense neuronal activity, the extracellular potassium increases to 10 12 mm. In the hippocampus, the peak potassium levels measured during seizures are slightly higher in the dentate gyrus than in CA1, but overall the changes are similar. At the same time as the potassium is increasing, calcium, magnesium, sodium and chloride enter cells, thus lowering the extracellular concentrations of these ions. As a result of these ionic shifts, the extracellular space shrinks and becomes acidic. Increases in extracellular potassium ([K þ ] 0 ) are known to increase neuronal excitability. The small increases in [K þ ] 0 measured with interictal bursts have been postulated to contribute to the transition between interictal and ictal activity. The increases in [K þ ] 0 have also been shown to facilitate spread of seizure activity to nearby and synaptically connected regions. Interestingly, during seizure activity, the [K þ ] 0 increases to a ceiling of about 10 mm it does not continue to rise with ongoing neuronal activity. This observation suggests that the mechanisms that are clearing the [K þ ] 0 are keeping pace with the amount of potassium moving out of the neurons with each action potential. Understanding the mechanism(s) that clear the potassium from the extracellular space would lead to an understanding of how the brain can regulate neuronal excitability. Intra- and extracellular ph changes may have fundamental importance in regulating neuronal excitability. Small changes in intracellular and extracellular ph have been shown to alter the function of ligand- and voltage-gated channels. Since transient q Change History: October 2015. Stringer updated the section References. Reference Module in Neuroscience and Biobehavioral Psychology http://dx.doi.org/10.1016/b978-0-12-809324-5.00160-7 1

2 Changes in Extracellular Ionic Composition ph changes may be confined to localized regions of the extracellular space, these ph transients may play a role in signaling by altering neuronal excitability locally. Movement of hydrogen ions across membranes in the central nervous system is controlled largely by two families of acid-base transporters. The Na/H (NHE) and the bicarbonate-dependent transport families are a diverse group of transport proteins. The sodium/hco 3 cotransporter (NBC) is found in glia throughout the hippocampus and in the granule cells of the dentate gyrus, supporting a role of glia in the regulation of ph in the hippocampus and providing a possible explanation for the differential ph regulation in the dentate gyrus compared to CA1. In one model of ph regulation, neuronal activity causes an increase in extracellular potassium and depolarization of astrocytes, which stimulates the movement of H þ out of the astrocytes; as a result, the intracellular space within the glia becomes more alkaline while the extracellular space becomes more acidic. The combination of the regional diversity in ph regulation and the sensitivity of neuronal excitability to small changes in extracellular ph suggest that ph regulation may play a part in spread of epileptiform activity. Intra- and extracellular calcium also contribute to neuronal excitability. Normally intracellular calcium is maintained at very low levels, but measurable increases occur in neurons with action potential firing. Changes in extracellular calcium are somewhat more complex, in that neuronal activity alone does not necessarily result in significant and sustained decreases in calcium. However, any decrease in extracellular calcium would increase neuronal excitability, in part due to a loss of surface charge screen. Decreasing calcium in the perfusion solution results in multiple population spikes in CA1 in response to a single stimulation, indicating multiple action potential firing instead of the normal single action potential. If the extracellular calcium decreases below 0.5 mm, spontaneous epileptiform activity has been reported. In the presence of small elevations in extracellular potassium, neurons in the hippocampus are more sensitive to small decreases in calcium. Changes in extracellular magnesium, sodium and chloride have not been studied to the same extent as potassium, calcium and ph. However, during epileptiform activity, decreases in extracellular sodium and biphasic changes in extracellular chloride have been measured. Lowering of the extracellular magnesium would be predicted to increase transmitter release from presynaptic terminals, thus increasing neuronal excitability. Lowering of magnesium would also decreases the surface charge screen, and if the decrease is sufficient, NMDA receptors may be affected. Since chloride concentrations affect function of GABA receptors, alterations in the extracellular chloride could produce changes in inhibitory function. Methods There have been two major approaches to studying changes in extracellular ion concentrations and their roles in epileptogenesis. One is to monitor ion concentrations (in either anesthetized rats or in in vitro brain slice preparations) with ion-sensitive electrodes before, during and after seizure activity induced by any of a number of methods. The second approach is to alter the extracellular ion concentration to determine the effect this manipulation has on epileptiform activity. The latter procedure is most commonly done in in vitro systems of brain slices or organotypic cultures. Recording of extracellular ion concentrations is carried out using double-barrel ion-sensitive electrodes. Double-barrel electrodes are prepared by silanizing one barrel of the electrode to make it hydrophobic and then filling just the tip of that side with an ion exchange resin for the ion of interest. The remainder of the electrode is then filled with a solution containing the ion to be measured. The other side of the electrode is filled with a saline solution for recording of the extracellular field potentials. The electrode is calibrated in a series of standard solutions around the expected concentrations of that ion in the brain. The electrode is placed into the brain region of interest, and the ion signal is determined by a differential amplifier. Both barrels of the electrode measure the field potential changes, and the side with the ion-exchange resin also measures the concentration of that particular ion. The differential amplifier then subtracts the field potential from the additive signal, providing an output that reflects the changes in ion concentration. These double-barrel electrodes have been used in both in vivo and in vitro preparations. A newer method involves insertion of one barrel (of a double electrode ) inside a second barrel. This modification shortens the effective column length of the ion exchange resin, which improves the response time of the electrode. However, because of the physical properties of this electrode, it works well only in in vitro preparations. Using either method, one can record changes in extracellular ions before, during and after epileptiform activity. The effects of changes in extracellular ions can be determined in in vitro systems by altering the perfusion solution to levels of ions recorded in vivo. For example, when recording extracellular calcium in vivo, it was determined that the calcium falls to various levels during different types of epileptiform activity. These levels of extracellular calcium can then be reproduced (in isolation) in vitro to determine the significance of these changes. One can ask, for example, if this lower level of calcium is sufficiently low to block synaptic transmission, or only low enough to increase excitability. Such information is, in turn, used to address the question of whether/how the changes in physiological function due to the change in ionic concentration contribute to epileptiform activity. Of course, changing levels of only one ion does not mimic the situation in vivo, where many ions will change simultaneously during epileptiform activity. However, combinations of ion manipulations can be reproduced in in vitro preparations. Recent Results It is well known that extracellular potassium concentrations ([K þ ] 0 ) increase during intense neuronal activity, including seizures (Fig. 1). Glial cells are believed to play a major role in the regulation of the [K þ ] 0 increase. Using ion-selective electrodes, we

Changes in Extracellular Ionic Composition 3 Figure 1 Field recording in CA1 of the extracellular field potential (top) and the extracellular potassium concentration (bottom) before, during and after a 20 Hz stimulation to the contralateral CA3 region in vivo. examined the [K þ ] 0 changes in the dentate gyrus of urethane-anesthetized adult rats, comparing the changes in the presence of reactive astrocytes and after reduction of glial function. The regulation of [K þ ] 0 in the dentate gyrus was determined by measuring the ceiling level of [K þ ] 0 and the half-time of recovery of [K þ ] 0, during and after seizures produced by 20 Hz trains of stimulation to the angular bundle. Reactive astrocytes were induced by repeated seizures and their presence was confirmed by a qualitative increase in glial fibrillary acidic protein (GFAP) and appearance of vimentin immunoreactivity. In the presence of reactive astrocytes, there was no significant change in the peak [K þ ] 0 during seizures, but there did appear to be a trend toward a faster recovery of the [K þ ] 0 toward baseline compared to control animals. To inhibit glial function, fluorocitrate (FC), a reversible metabolic inhibitor, was injected into the dentate gyrus region and the regulation of [K þ ] 0 was monitored for 8 h. Fluorocitrate significantly slowed the rate of recovery of [K þ ] 0. Overall, the results suggested that normal glial function is required for the recovery of elevated [K þ ] 0 after seizures in vivo. It was somewhat surprising that the ceiling level of [K þ ] 0 was not elevated after loss of astrocytic function. At this ceiling level, it is presumed that the potassium is in a state of active equilibrium (or steady-state). Active neurons release potassium, which is then taken up by neurons or glia at the same rate. Based on this hypothesis, it would be predicted that changing glial cell function (either increase or decrease) would result in a change in the ceiling level of [K þ ] 0 during seizures. The results of these experiments did not support this hypothesis. In addition, although there was a statistically significant change in the rate of recovery of [K þ ] 0 after seizures when glia function was altered, the change was relatively small. One must wonder what physiological significance this small change would have on neuronal function or excitability. In order to investigate further the mechanisms that might contribute to regulation of activity-dependent variations in [K þ ] 0,we examined spontaneous recurrent epileptiform activity induced in the dentate gyrus of hippocampal slices from adult rats. Seizures were induced by perfusion with 8 mm potassium and 0-added calcium medium. Local application of TTX blocked local [K þ ] 0 changes, suggesting that potassium is released and taken up locally. Perfusion with barium (0.2 0.3 mm) or cesium (1 3 mm), blockers of the inward rectifying potassium channel that is postulated to mediate glial uptake of potassium, did not alter any of the measures of [K þ ] 0 regulation. If gap junctions between glial cells are critical for the spatial buffering of the [K þ ] 0, then reduction of gap junction conductance should alter [K þ ] 0 regulation. Decreasing gap junctional conductance only decreased the ceiling level of [K þ ] 0. Perfusion with furosemide (which blocks cation/chloride cotransporters) or perfusion with low chloride increased the ceiling level of [K þ ] 0. Bath or local application of ouabain, a Na þ /K þ -ATPase inhibitor, increased the baseline [K þ ] 0, slowed the rate of [K þ ] 0 recovery, and induced spreading depression. Together, these findings suggest that potassium redistribution by glia plays only a minor role in the regulation of [K þ ] 0 in this model. The major regulator of [K þ ] 0 in this model appears to be uptake via ana þ /K þ -ATPase, most likely neuronal. Studies of changes in hydrogen ion concentration (ph) have proved to be quite interesting, in part because the changes during seizure activity are not the same in all brain regions. It is clear that neuronal activity results in alterations in both intracellular and extracellular ph. The measured changes in extracellular ph are within the range of ph values that have been shown to alter neuronal activity. In vivo and in vitro, in the CA1 region, recurrent epileptiform activity induced by stimulus trains, bicuculline and kainic acid results in biphasic ph shifts, consisting of an initial extracellular alkalinization followed by a slower acidification (Fig. 2). In CA1, the acidification consistently peaks after termination of the afterdischarge, and the peak level of acidification correlates with the duration of the afterdischarge, suggesting a contribution of neuronal activity to the level of acidification. Both GABAergic and glutamatergic neurotransmission have been reported to cause alkalinization. In the CA1 region, this alkalinization was decreased by inhibition of carbonic anhydrase and increased by addition of GABA agonists (diazepam and phenobarbital), which is consistent with a key contribution to ph changes by activation of GABA A receptors (rather than through activation of glutamatergic receptors). The observation that ph fluctuation is linked to synaptic transmission is supported by the observation (in slices) that, in nonsynaptic conditions, there is no initial alkalinization. In the dentate gyrus in vivo and in vitro, seizure activity induced by stimulus trains to the angular bundle, by injection of either bicuculline or kainic acid, or in in vitro non-synaptic conditions was only associated with extracellular acidification.

4 Changes in Extracellular Ionic Composition CA1 recording 5 mv stimulus 1 0.1 ph 2 10 sec 3 Figure 2 Field recording (top) and extracellular ph (bottom) are shown before, during and after a stimulus train to the contralateral CA3 region in the hippocampus in vivo. Early on (1) synaptic activity results in alkalinization. The peak acidification (2) is dependent on the afterdischarge duration. The recovery phase (3) is the least well understood. The role of astrocytes in these extracellular ph changes during neuronal activity was examined using local injection of fluorocitrate and fluoroacetate into the CA1 cell layer, where there is both alkalinization and acidification. Both glial toxins reduced the peak level of acidification, without changing the lengthening of the afterdischarge. The peak level of acidification was still correlated with the total discharge duration, but the levels of acidification were consistently lower than in control animals. Administration of either glial toxin had no effect on the peak alkalinization during the stimulus train, on the rate of recovery from peak level of acidification, or on the amplitude or frequency of the neuronal discharge during the afterdischarge. These results suggest that, in normal conditions, astrocytes contribute to the acidification of the extracellular space that occurs in response to intense neuronal activity. Such acidification may contribute to feedback regulation of neuronal excitability. Changes in extracellular calcium ([Ca 2þ ] 0 ) have also been studied in the hippocampus in vivo. In urethane-anesthetized Sprague- Dawley rats, changes in [Ca 2þ ] 0 were measured in response to stimulus frequencies of 10 and 20 Hz. Maximal changes in [Ca 2þ ] 0 were recorded in the cell body layers of CA1 and the dentate gyrus. In the dentate gyrus there was a secondary drop in [Ca 2þ ] 0 with the appearance of maximal dentate activation (appearance of the large amplitude population spikes, Fig. 3); that is, the largest changes in [Ca 2þ ] 0 in the dentate gyrus were recorded during maximal dentate activation. To investigate the role of astrocytes in regulation of in [Ca 2þ ] 0, local injections of fluorocitrate, a metabolic toxin selectively taken up by astrocytes, were used. FC, but not vehicle, caused a small but significant decrease in the maximal changes in [Ca 2þ ] 0 in CA1, but an increase in the dentate gyrus. The results suggest that astrocytes play a minimal role in the regulation of [Ca 2þ ] 0 during epileptiform activity. Future Goals Because concentrations of the various ions in the extracellular space clearly contribute to neuronal excitability, understanding the mechanisms by which these ions are regulated will further our understanding of the manner in which neuronal excitability can be controlled. This insight is especially important for understanding the conditions in which seizures begin and how seizures are terminated. Action potentials open channels, allowing ions to flow down concentration gradients, resulting in changes in the extracellular ion concentrations. Much of the work in the past has focused on the significance of these changes in ionic concentrations on Figure 3 Extracellular DC field recording (top) and extracellular calcium concentration (bottom) in the dentate gyrus (DG) before, during and after a 20 Hz stimulus train to the contralateral CA3 region in vivo.

Changes in Extracellular Ionic Composition 5 neuronal function. But it is also important to determine the mechanisms by which the ion concentrations are returned to the baseline levels, the recovery of the ion concentrations, especially after intense neuronal activity such as a seizure. Largely because they do not fire action potentials, the study of glial cells and their role in brain function is an area of investigation that has lagged behind the study of neurons. However, it is now clear that glia contribute to neuronal function. One way in which they can control neuronal excitability is in the maintenance of the extracellular environment. There are many unanswered questions about the relationship between glial function in controlling extracellular ion concentrations and neuronal activity. For example: Does the glial spatial buffer for potassium slow the recovery of the potassium gradient that is necessary for normal neuronal function? Can the astrocytes regulate extracellular calcium concentrations and if so, how does that affect neurotransmitter release? Are there changes in the mechanisms that regulate the extracellular environment in a chronically epileptic brain (compared to normal brain), and do these changes contribute to the epileptic state? It is known that a number of ion channel mutations are present in families with epilepsies. Do these channel mutations alter the ionic composition in the extracellular space? Channel biophysics would suggest that a mutation that results in a channel staying open longer would allow more ions to flow through that channel. But would this difference in ion flow contribute to the seizures in this instance, or are the seizures arising from another mechanism? It is known that a number of pumps, transporters or co-transporters are responsible for the maintenance of ion gradients. One could postulate that a single nucleotide change in a transporter protein could result in a shift in the efficiency of the transport, resulting in an increased (or decreased) seizure threshold. Finally, are there methods by which one could intervene to influence the regulation of these ion gradients, and thus to reduce the likelihood of a seizure starting? or to augment the ability of the brain to stop a seizure once it has started? For example, more rapid or efficient transport activity could shorten the post-ictal state and improve recovery after a seizure. Further Reading Chesler, M., 1990. The regulation and modulation of ph in the nervous system. Prog. Neurobiol. 34, 401 427. Fedirko, N., Svichar, N., Chesler, M., 2006. Fabrication and use of high-speed, concentric H þ - and Ca 2þ -selective microelectrodes suitable for in vitro extracellular recording. J. Neurophysiol. 96, 919 924. Heinemann, U., Stabel, J., Rausche, G., 1990. Activity-dependent ionic changes and neuronal plasticity in rat hippocampus. Prog. Brain Res. 83, 197 214. Jeffreys, J.G.R., 1995. Nonsynaptic modulation of neuronal activity in the brain: electric currents and extracellular ions. Physiol. Rev. 75, 689 713. Kaila, K., Chesler, M., 1998. Activity-evoked changes in extracellular ph. In: Kaila, K., Ransom, B.R. (Eds.), ph and Brain Function. Wiley-Liss, New York, pp. 309 337. Krishnan, G.P., Bazhenov, M., 2011. Ionic dynamics mediate spontaneous termination of seizures and post-ictal depression state. J. Neurosci. 31, 8870 8882. Owen, J.A., Barreto, E., Cressman, J.R., 2013. Controlling seizure-like events by perturbing ion concentration dynamics with periodic stimulation. PLoS One 8, e73820. http://dx.doi.org/10.1371/journal.pone.0073820. Ransom, B.R., 2000. Glial modulation of neural excitability mediated by extracellular ph: a hypothesis revisited. Prog. Brain Res. 125, 217 228. Somjen, G.G., 1979. Extracellular potassium in the mammalian nervous system. Ann. Rev. Physiol. 41, 159 177. Somjen, G.G., 2002. Ion regulation in the brain. Neuroscientist 8, 254 267. Stringer, J.L., Aribi, A.M., 2003. Effects of glial toxins on extracellular acidification in the hippocampal CA1 region in vivo. Epilepsy Res. 54, 163 170. Sykova, E., 2004. Extrasynaptic volume transmission and diffusion parameters of the extracellular space. Neuroscience 129, 861 876. Xiong, Z.-Q., Stringer, J.L., 1999. Astrocytic regulation of the recovery of extracellular potassium after seizures in vivo. Eur. J. Neurosci. 11, 1677 1684. Xiong, Z.-Q., Stringer, J.L., 2000. Sodium pump activity, not glial spatial buffering, clears potassium after epileptiform activity induced with 0-added calcium and 8 mm potassium in the dentate gyrus in vitro. J. Neurophysiol. 83, 1443 1451. Xiong, Z.-Q., Stringer, J.L., 2000. Extracellular ph responses differ between CA1 and dentate gyrus during seizure activity in vivo and in vitro. J. Neurophysiol. 83, 3519 3524. Relevant Website http://www.brain-map.org/welcome.do Allen Institute for Brain Science Includes the Allen Brain Atlas, Which Is an Interactive, Genome-Wide Image Database of Gene Expression in the Mouse Brain. A Combination of RNA In Situ Hybridization Data, Detailed Reference Atlases and Informatics Analysis Tools Are Integrated to Provide a Searchable Digital Atlas of Gene Expression.