Regulation of extracellular potassium in the developing hippocampus

Elevated potassium provides an ionic mechanism for deep brain stimulation in the hemiparkinsonian rat

The mechanism of high-frequency stimulation used in deep brain stimulation (DBS) for Parkinson's disease (PD) has not been completely elucidated. Previously, high-frequency stimulation of the rat entopeduncular nucleus, a basal ganglia output nucleus, elicited an increase in [K(+)](e) to 18 mm, in vitro. In this study, we assessed whether elevated K(+) can elicit DBS-like therapeutic effects in hemiparkinsonian rats by employing the limb-use asymmetry test and the self-adjusting stepping test. We then identified how these effects were meditated with in-vivo and in-vitro electrophysiology. Forelimb akinesia improved in hemiparkinsonian rats undergoing both tests after 20 mm KCl injection into the substantia nigra pars reticulata (SNr) or the subthalamic nucleus. In the SNr, neuronal spiking activity decreased from 38.2 ± 1.2 to 14.6 ± 1.6 Hz and attenuated SNr beta-frequency (12-30 Hz) oscillations after K(+) treatment. These oscillations are commonly associated with akinesia/bradykinesia in patients with PD and animal models of PD. Pressure ejection of 20 mm KCl onto SNr neurons in vitro caused a depolarisation block and sustained quiescence of SNr activity. In conclusion, our data showed that elevated K(+) injection into the hemiparkinsonian rat SNr improved forelimb akinesia, which coincided with a decrease in SNr neuronal spiking activity and desynchronised activity in SNr beta frequency, and subsequently an overall increase in ventral medial thalamic neuronal activity. Moreover, these findings also suggest that elevated K(+) may provide an ionic mechanism that can contribute to the therapeutic effects of DBS for the motor treatment of advanced PD.

Brain development and susceptibility to damage; ion levels and movements

Responses of immature brains to physiological and pathological stimuli often differ from those in the adult. Because CNS function critically depends on ion movements, this chapter evaluates ion levels and gradients during ontogeny and their alterations in response to adverse conditions. Total brain Na(+) and Cl(-) content decreases during development, but K(+) content rises, reflecting shrinkage of the extracellular and increase in the intracellular water spaces and a reduction in total brain water volume. Unexpectedly, [K(+)](i) seems to fall during the first postnatal week, which should reduce [K(+)](i)/ [K(+)](e) and result in a lower V(m), consistent with experimental observations. Neuronal [Cl(-)](i) is high during early postnatal development, hence the opening of Cl(-) conduction pathways may lead to plasma membrane depolarization. Equivalent loss of K(+)(i) into a relatively large extracellular space leads to a smaller increase in [K(+)](e) in immature animals, while the larger reservoir of Ca(2+)(e) may result in a greater [Ca(2+)](i) rise. In vivo and in vitro studies show that compared with adult, developing brains are more resistant to hypoxic/ischemic ion leakage: increases in [K(+)](e) and decreases in [Ca(2+)](e) are slower and smaller, consistent with the known low level of energy utilization and better maintenance of [ATP]. Severe hypoxia/ischemia may, however, lead to large Ca(2+)(i) overload. Rises in [K(+)](e) during epileptogenesis in vivo are smaller and take longer to manifest themselves in immature brains, although the rate of K(+) clearance is slower. By contrast, in vitro studies suggest the existence of a period of enhanced vulnerability sometime during the developmental period. This chapter concludes that there is a great need for more information on ion changes during ontogeny and poses the question whether the rat is the most appropriate model for investigation of mechanisms of pathological changes in human neonates.

Layer-specific generation and propagation of seizures in slices of developing neocortex: role of excitatory GABAergic synapses

The neonatal period is critical for seizure susceptibility, and neocortical networks are central in infantile epilepsies. We report that application of 4-aminopyridine (4-AP) to immature (P6-P9) neocortical slices generates layer-specific interictal seizures (IISs) that transform after recurrent seizures to ictal seizures (ISs). During IISs, cell-attached recordings show action potentials in interneurons and pyramidal cells in L5/6 and interneurons but not pyramidal neurons in L2/3. However, L2/3 pyramidal neurons also fire during ISs. Using single N-methyl-d-aspartate (NMDA) channel recordings for measuring the cell resting potential (Em), we show that transition from IISs to ISs is associated with a gradual Em depolarization of L2/3 and L5/6 pyramidal neurons that enhances their excitability. Bumetanide, a NKCC1 co-transporter antagonist, inhibits generation of IISs and prevents their transformation to ISs, indicating the role excitatory GABA in epilepsies. Therefore deep layer neurons are more susceptible to seizures than superficial ones. The initiating phase of seizures is characterized by IISs generated in L5/6 and supported by activation of both L5/6 interneurons and pyramidal cells. IISs propagate to L2/3 via activation of L2/3 interneurons but not pyramidal cells, which are mostly quiescent at this phase. In superficial layers, a persistent increase in excitability of pyramidal neurons caused by Em depolarization is associated with a transition from largely confined GABAergic IIS to ictal events that entrain the entire neocortex.

High frequency stimulation or elevated K+ depresses neuronal activity in the rat entopeduncular nucleus

High frequency stimulation (HFS) is applied to many brain regions to treat a variety of neurological disorders/diseases, yet the mechanism(s) underlying its effects remains unclear. While some studies showed that HFS inhibits the stimulated nucleus, others report excitation. In this in vitro study, we stimulated the rat globus pallidus interna (entopeduncular nucleus, EP), a commonly stimulated area for Parkinson's disease, to investigate the effect of HFS-induced elevation of extracellular potassium (K(+)(e)) on rat EP neuronal activity. Whole-cell patch-clamp recordings and [K(+)](e) measurements were obtained in rat EP brain slices before, during and after HFS. After HFS (150 Hz, 10 s), [K(+)](e) increased from 2.5-9.6+/-1.4 mM, the resting membrane potential of EP neurons depolarized by 11.1+/-2.5 mV, spiking activity was significantly depressed, and input resistance decreased by 25+/-6%. The GABA(A) receptor blocker, gabazine, did not prevent these effects. The bath perfusion of 6 or 10 mM K(+), with or without synaptic blockers, mimicked the HFS-mediated effects: inhibition of spike activity, a 20+/-9% decrease in input resistance and a 17.4+/-3.0 mV depolarization. This depolarization exceeded predicted values of elevated [K(+)](e) on the resting membrane potential. A depolarization block did not fully account for the K(+)-induced inhibition of EP neuronal activity. Taken together, our results show that HFS-induced elevation of [K(+)](e) decreased EP neuronal activity by the activation of an ion conductance resulting in membrane depolarization, independent of synaptic involvement. These findings could explain the inhibitory effects of HFS on neurons of the stimulated nucleus.

Do seizures affect the developing brain? Lessons from the laboratory

Laboratory models of prolonged seizures and status epilepticus in developing animals demonstrate age- and model-dependent propensity for brain injury. Even in models without overt brain injury, plasticity, which leads to epileptogenicity as well as to behavioral and cognitive effects, has been demonstrated. Brief, recurrent seizures in the neonatal period not only appear to exhibit plasticity that can be anatomically and physiologically meaningful but also seem to produce cognitive deficits. Translation of these findings into clinical practice is limited by the effects chronic therapy may have on brain development. There is little evidence that available treatments can effectively alter epileptogenesis. However, it is widely agreed that prolonged seizures and status epilepticus can carry negative consequences. Preventing epileptogenesis remains an important goal to modify the development of comorbidities, and it represents an area of research in need of much progress. For now, prevention of prolonged seizures with early intervention is important and is the most effective available option to minimize the potential short- and long-term adverse effects of prolonged seizures and optimize patient outcomes.

Inwardly rectifying K+ (Kir) channels antagonize ictal-like epileptiform activity in area CA1 of the rat hippocampus

Reactive glial cells, for example, from patients with temporal lope epilepsy have a reduced density of inward rectifying K(+) (Kir) channels and thus a reduced K(+) buffering capacity. Evidence is accumulating that this downregulation of Kir channels could be implicated in epileptogenesis. In rat hippocampal brain slices, prolonged exposure to the nonselective Kir channel antagonist, Cs(+) (5 mM), gives rise to an epileptiform field potential (Cs-FP) in area CA1 composed of an initial positive (interictal-like) phase followed by a prolonged negative (ictal-like) phase. We have previously shown that the interictal-like phase depends on synaptic activation. The present study extends these findings by showing that the ictal-like phase of the Cs-FP is (i) sensitive to osmotic expansion of the extracellular space, (ii) reversed very quickly during wash out of Cs(+), and (iii) re-established in the presence of Ba(2+) (30-200 microM) or isosmotic low extracellular concentration of Na(+) ([Na(+)](o), 51.25 mM). The interictal-like phase showed less or no sensitivity to these treatments. In the complete absence of Cs(+), the Cs-FP could be fully reconstructed by the combined application of 4-aminopyridine (0.5 mM), an isosmotic high extracellular concentration of K(+) ([K(+)](o), 7 mM), and low [Na(+)](o) (51.25 mM). These results suggest that the interictal-like phase is initiated through synaptic activation and results from an unspecific increase in neuronal excitability, whereas the ictal-like phase is entirely dependent on blockade of Kir channels in CA1. We propose that glial dysfunction-related loss of Kir channels may not alone be sufficient for starting the induction process, but will likely increase the tendency of an epileptogenic process to proceed into seizure activity.

Rhythmic neuronal discharge in the medulla and spinal cord of fetal rats in the absence of synaptic transmission

Spontaneous rhythmic neuronal activity is generated in the developing vertebrate nervous system. The patterned activity spreads diffusely throughout the fetal neuraxis. Here we demonstrate the ability of the fetal rat spinal cord and medulla to generate and transmit robust rhythmic patterns in the absence of synaptic activity. Regular rhythmic discharges were produced by fetal tissue bathed in low or zero [Ca(2+)](o) solution. The activity persisted in the presence of antagonists to neurotransmitter receptors that are known to mediate synaptic-mediated events associated with fetal rhythms. A combination of ventral root recordings and optical imaging using voltage-sensitive dyes demonstrated the extensive spread of rhythmic discharge in spinal cord and medullary neuronal populations of in vitro preparations. Whole cell recordings from medullary slices were performed to examine the ionic conductances and revealed the importance of persistent sodium conductances for generation of rhythmic activity in hypoglossal (XII) motoneurons. Rhythmic bursting in XII motoneurons persisted in the presence of gap junction blockers, although the amplitude of synchronized motor discharge recorded from nerve roots was diminished. We propose that nonsynaptically mediated conductances, potentially by extracellular ionic flux and/or ephaptic and electrotonic interactions mechanisms, act in concert with neurochemical transmission and gap junctions to promote the diffuse spread of rhythmic motor patterns in the developing nervous system.

Excitatory and inhibitory control of epileptiform discharges in combined hippocampal/entorhinal cortical slices

We examined whether epileptiform activity can be induced and prevented by mild reduction of GABA(A) receptor-mediated inhibition and non-NMDA receptor-mediated excitation, respectively, in different regions of combined hippocampal/entorhinal cortical slices from juvenile rats (P15-21). We used the receptor antagonists bicuculline (GABA(A)) and CNQX (non-NMDA) as tools to investigate the sensitivities of the CA1, the subiculum (SUB) and the medial entorhinal cortex (MEC) for generating epileptiform discharges upon extracellular stimulation. We found that low concentrations of bicuculline (<3.5 microM) were enough to induce epileptiform discharges in the three regions. These discharges were similar to those observed under high concentrations of bicuculline (>10 microM) and consisted of stereotyped population bursts, recorded both extra- and intracellularly. Interestingly, the CA1 and SUB were more susceptible to generate discharges compared to the MEC in the same slices. We also found that non-NMDA excitation was critical in controlling discharges, as they were blocked by CNQX in a concentration-dependent manner. The sensitivity of the CA1 region to CNQX was lower than that of the SUB and MEC. Based on these regional differences, we show that epileptiform activity can be pharmacologically isolated within the CA1 region in the hippocampal-entorhinal circuitry in vitro.