The bumetanide-sensitive Na-K-2Cl cotransporter NKCC1 as a potential target of a novel mechanism-based treatment strategy for neonatal seizures

Full access

Seizures that occur during the neonatal period do so with a greater frequency than at any other age, have profound consequences for cognitive and motor development, and are difficult to treat with the existing series of antiepileptic drugs. During development, γ-aminobutyric acid (GABA)ergic neurotransmission undergoes a switch from excitatory to inhibitory due to a reversal of neuronal chloride (Cl) gradients. The intracellular level of chloride ([Cl]i) in immature neonatal neurons, compared with mature adult neurons, is about 20–40 mM higher due to robust activity of the chloride-importing Na-K-2Cl cotransporter NKCC1, such that the binding of GABA to ligand-gated GABAA receptor-associated Cl channels triggers Cl efflux and depolarizing excitation. In adults, NKCC1 expression decreases and the expression of the genetically related chloride-extruding K-Cl cotransporter KCC2 increases, lowering [Cl]i to a level such that activation of GABAA receptors triggers Cl influx and inhibitory hyperpolarization. The excitatory action of GABA in neonates, while playing an important role in neuronal development and synaptogenesis, accounts for the decreased seizure threshold, increased seizure propensity, and poor efficacy of GABAergic anticonvulsants in this age group. Bumetanide, a furosemide-related diuretic already used to treat volume overload in neonates, is a specific inhibitor of NKCC1 at low doses, can switch the GABA equilibrium potential of immature neurons from depolarizing to hyperpolarizing, and has recently been shown to inhibit epileptic activity in vitro and in vivo in animal models of neonatal seizures. The fundamental role of NKCC1 in establishing excitatory GABAergic neurotransmission in the neonate makes it a tempting target of a novel mechanism-based anticonvulsant strategy that could utilize the well-known pharmacology of bumetanide to help treat neonatal seizures.

Abbreviations used in this paper: [Cl–]i = intracellular concentration of chloride; CNS = central nervous system; ECl = chloride equilibrium potential; EEG = electroencephalogram; EGABA = GABA equilibrium potential; GABA = γ-aminobutyric acid; Ki = inhibition constant; TLE = temporal lobe epilepsy.

Seizures that occur during the neonatal period do so with a greater frequency than at any other age, have profound consequences for cognitive and motor development, and are difficult to treat with the existing series of antiepileptic drugs. During development, γ-aminobutyric acid (GABA)ergic neurotransmission undergoes a switch from excitatory to inhibitory due to a reversal of neuronal chloride (Cl) gradients. The intracellular level of chloride ([Cl]i) in immature neonatal neurons, compared with mature adult neurons, is about 20–40 mM higher due to robust activity of the chloride-importing Na-K-2Cl cotransporter NKCC1, such that the binding of GABA to ligand-gated GABAA receptor-associated Cl channels triggers Cl efflux and depolarizing excitation. In adults, NKCC1 expression decreases and the expression of the genetically related chloride-extruding K-Cl cotransporter KCC2 increases, lowering [Cl]i to a level such that activation of GABAA receptors triggers Cl influx and inhibitory hyperpolarization. The excitatory action of GABA in neonates, while playing an important role in neuronal development and synaptogenesis, accounts for the decreased seizure threshold, increased seizure propensity, and poor efficacy of GABAergic anticonvulsants in this age group. Bumetanide, a furosemide-related diuretic already used to treat volume overload in neonates, is a specific inhibitor of NKCC1 at low doses, can switch the GABA equilibrium potential of immature neurons from depolarizing to hyperpolarizing, and has recently been shown to inhibit epileptic activity in vitro and in vivo in animal models of neonatal seizures. The fundamental role of NKCC1 in establishing excitatory GABAergic neurotransmission in the neonate makes it a tempting target of a novel mechanism-based anticonvulsant strategy that could utilize the well-known pharmacology of bumetanide to help treat neonatal seizures.

Seizures occur more often during the neonatal period than at any other age (2–3.5 seizures per 1000 live births). Neonatal seizures have profound long-term consequences, with adult survivors often developing epilepsy and significant cognitive and motor disabilities.22 Nearly any neurological insult (such as hypoxia-ischemia, metabolic derangement, hemorrhage, and infection) sustained during this period can trigger the synchronous firing of hyperexcitable neurons that underlie epileptogenesis.65 Whereas phenobarbital, benzodiazepines, and phenytoin have been the first-line treatments for neonatal seizures for years, these drugs are often ineffective and can even potentiate seizure episodes.8,10,50 A recent Cochrane Review concluded that there is little evidence to endorse the use of any of the anticonvulsants currently used in the neonatal period.6 The distinct clinical phenotype and treatment response of neonatal seizures, compared with adult seizures, is probably due to the different mechanisms of seizure generation, propagation, and termination in these 2 age groups. It is likely that an improved understanding of the molecular pathophysiology underlying neonatal seizures would lead to improved treatment by defining targets for novel mechanism-based therapeutics.

In the mammalian CNS, the [Cl]i in neurons determines the strength and polarity of GABAergic neurotransmission. The [Cl]i is largely determined by the SLC12A cation-chloride cotransporters, including the Na-K-2Cl cotransporter NKCC1 (mediating Cl entry) and the K-Cl cotransporter KCC2 (mediating Cl exit).52,54 The mature neurons of adults exhibit low [Cl]i due to minimal NKCC1 expression but robust KCC2 expression, such that GABA-induced allosteric activation of the GABAA receptor (which functions as a ligand-gated Cl channel) results in an influx of negatively charged Cl ions that triggers membrane hyperpolarization and synaptic inhibition. In adults, barbiturates and benzodiazepines decrease seizure activity by increasing the duration and/or frequency of the opening of GABAA receptor-associated Cl channels. In contrast to the mature neurons of adults, the immature neurons of neonates exhibit a much higher [Cl]i due to robust NKCC1 expression and minimal KCC2 expression, such that activation of GABAA receptors elicits an efflux of Cl ions that triggers membrane depolarization and synaptic excitation. The excitatory action of GABA in neonates plays a role in neuronal development,2,3 but also accounts for their increased seizure propensity and lowered seizure threshold.68 The high [Cl]i of immature neurons also has a tendency to render barbiturates and benzodiazepines ineffective, because these drugs (as compared with their action on mature neurons, which harbor low [Cl]i) facilitate the passive outflow of Cl down its electrochemical gradient, depolarizing and exciting neurons.19,67

Recent data have shown that the NKCC1 inhibitor bumetanide, a drug already approved by the US Food and Drug Administration as a diuretic, shows efficacy against seizures in the immature rodent brain, and thus may be clinically useful in the treatment of human neonatal seizures.19,21 This review will discuss the recent work that has provided insight into the role of NKCC1 in fostering excitatory GABAergic neurotransmission in the immature brain, how the excitatory effect of GABA in neonates makes them particularly vulnerable to the development of seizures, and how the pharmacological inhibition of NKCC1 with bumetanide might hold promise for the treatment of seizures in neonates.

Neonatal Seizures Are a Common Problem With Inadequate Treatment

Neonatal seizures, or epileptic episodes suffered by infants in the first 28 days of life, occur in 1–2% of patients in neonatal intensive care units, and are the most common manifestation of an acute neurological disorder in newborn infants.75 Most commonly caused by hypoxic ischemic encephalopathy, hemorrhage, or cerebral infarction, the presence of neonatal seizures often portends severe neurological dysfunction later in life, with high rates of adult epilepsy and long-term cognitive and motor deficits in survivors.13,75 In animal models, neonatal seizures have been shown to be injurious to the development of the brain, inducing synaptic reorganization, altering synaptic plasticity, and priming cortical neurons to increased damage from seizures sustained later in life.57 Thus, the prompt diagnosis and successful treatment of neonatal seizures is important for improving long-term neurological outcome.

Whereas seizure activity in adults is usually clinically obvious and the EEG reflects coordinated seizure activity, diagnosing seizures in neonates is difficult because seizures are often behaviorally subtle and the EEG typically demonstrates a multifocal process.65 Nonetheless, seizures in neonates are currently diagnosed using electroencephalography, with a discharge duration of 10 seconds required to diagnose an electrographic seizure (compared with 3 seconds in older age groups). However, because EEGs are not immediately available in many neonatal intensive care units, the initial diagnosis and treatment of seizures are often based on clinical assessment alone, and EEGs are obtained after the administration of antiepileptic drugs.61,65 Unfortunately, it is an all-too-common occurrence for electrographic seizures to persist in encephalopathic neonates despite anticonvulsant drugs at “therapeutic” levels.

Conventional antiepileptic drugs have limited utility in treating neonatal seizures.6,28,41 Barbiturates and benzodiazepines, which are GABAA receptor agonists that are efficacious for treating adult seizures, are currently among the first-line drugs for neonatal seizures; however, they are often ineffective, and have been shown to actually potentiate seizure activity in the immature brain.19,67 Phenytoin has been used with a similar amount of success.6,50 Barbiturates and benzodiazepines have also been known to produce a phenomenon termed “electroclinical dissociation” in neonates, whereby the overt clinical manifestations of seizures (convulsions) are inhibited, but electroencephalography-documented cortical seizure activity is either unaffected or exacerbated.16,62 This insidious effect of the GABA agonists has the potential for great harm, because it provides physicians with a false sense of security that seizures are under control, while cortical seizure activity—and its associated detrimental effects—continues.

There have been few prospective studies or randomized controlled trials of the antiepileptic drugs that are currently used to treat neonatal seizures.6,12,60 To date, the only randomized trial of antiepileptic drugs for the treatment of neonatal seizures compared the current first-line drugs (phenobarbital and phenytoin).50 In this study, the majority of neonates had asphyxia, infarction, or hemorrhage as the origin of their seizures. Complete control of electrographic seizures was achieved with either drug in only ~ 25% of neonates whose seizure frequency was increasing. Seizure control was achieved in another 15% of newborns when both agents were used concurrently. Preliminary studies of antiepileptic drugs other than phenobarbital and phenytoin have shown only modest efficacy in smaller cohorts of neonates, although sufficiently powered randomized trials are needed to conclusively demonstrate whether any of these drugs are truly effective.7,10,32,46,64 The lack of evidence-based treatment recommendations, coupled with the paucity of data regarding the underlying pathophysiology of neonatal seizures, has made their current management far from optimal.

The Strength and Polarity of GABAergic Neurotransmission Are Determined by the Intracellular Concentration of Chloride, Which Is Established by NKCC1 and KCC2

Recent research has provided information on key age-specific differences in GABAergic neurotransmission that account for the decreased seizure threshold and lack of efficacy of conventional GABA agonists in neonates. Understanding the molecular mechanisms that underlie the differential actions of GABA in immature versus mature neurons has, in turn, helped define targets for novel anticonvulsants in neonates.

In the adult cortex, GABA is the main inhibitory neuro-transmitter, in which its activity is essential for maintaining appropriate electrical activity by balancing inputs that are triggered by excitatory neurotransmitters such as glutamate. In contrast, in the neonatal cortex, GABA is an excitatory neurotransmitter, where it plays an important role in neuronal development and the activity-dependent wiring of circuits by triggering the depolarizing inputs that underlie large-scale spontaneous electrical activity.3 In immature neurons, GABA-induced membrane depolarization triggers action potentials that directly activate voltage-dependent Ca channels, and indirectly activate N-methyl-D-aspartate receptors. The subsequent GABA-induced elevations of intracellular Ca promote neuronal maturation, and are important for the genesis and maintenance of synaptic connections.3

Fig. 1.
Fig. 1.

A switch in the neuronal expression of NKCC1 and KCC2 underlies the excitatory-to-inhibitory developmental shift in GABAergic signaling, in which GABAergic neurotransmission undergoes a switch from excitatory to inhibitory due to a reversal of neuronal Cl gradients during development. The ([Cl]i) of immature neonatal neurons, compared with mature adult neurons, is much higher due to robust activity of the chloride-importing Na-K-2Cl cotransporter NKCC1, such that the binding of GABA to ligand-gated GABAA receptor-associated Cl channels triggers Cl efflux and depolarizing excitation. In adults, NKCC1 expression decreases and the expression of the genetically related chloride-extruding K-Cl cotransporter KCC2 increases, lowering [Cl]i to a level such that activation of GABAA receptors triggers Cl influx and inhibitory hyperpolarization. The excitatory action of GABA in neonates accounts for the decreased seizure threshold and poor efficacy of GABAergic anticonvulsants in this age group. Reprinted with permission from Delpire E: News Physiol Sci 15:309–312, 2000.

Because the GABAA receptor is a ligand-gated Cl channel, the resting level of [Cl]i, and the ECl, are key determinants of GABA's effect. Upon binding to GABAA receptors, GABA triggers conformational changes in receptor-associated Cl channels that permit either the passive influx or efflux of Cl ions. The directionality of Cltransport depends on the cell's [Cl]i. If the [Cl]i is below the ECl, Clenters the cell and triggers membrane hyperpolarization and inhibition; if the [Cl]i is greater than the ECl, Cl enters the cell and triggers membrane depolarization and excitation.

Recent work has shown that the differential action (excitatory vs inhibitory) of GABA in the neonate and adult CNS is due to a difference in the [Cl]i of immature versus mature neurons.3 Compared to mature adult neurons, [Cl]i is ~ 20–40 mM higher in immature neonatal neurons, a difference that is sufficient to shift the action of GABA from inhibition to excitation because the electrochemical driving force for Cl drives the flux of this anion out of the cell when GABA channels are opened. The [Cl]i is determined by the Na-K-Cl cotransporter NKCC1, which mediates Cltransport into the cell, and the K-Cl cotransporter KCC2, which mediates Cl transport out of the cell. The NKCC1 and KCC2 cotransporters are related members of the SLC12A electroneutral cation-chloride cotransporter gene family.17,25,52

The cation-chloride cotransporters are intrinsic membrane proteins that transport Cl ions, together with Na and/or K ions, across the plasma membranes of cells throughout the body. They are targets of some of the most commonly used drugs in medicine (the “loop” and thiazide diuretics), and are mutated in several inherited human diseases, including the autosomal-recessive, renal, salt-wasting Gitelman and Bartter syndromes.30 The stoichiometric coupling and directionality of the ions translocated by the cation-chloride cotransporters results in a secondarily active, nonelectrogenic (electroneutral) transport process that is energetically driven by transmembrane Na and K gradients established by Na+-K+-adenosine triphosphatase. The cation-chloride cotransporters are divided into 2 main branches, defined by the identity and stoichiometry of their transported ions, their sensitivity to pharmacological inhibitors, and genetic phylogeny.

Given the large electrochemically favorable inward gradient for Na, the Na-coupled branch comprises the Na-(K)-Cl cotransporters NCC (the target of thiazide diuretics), NKCC1, and NKCC2 (the target of Lasix, a furosemide diuretic), which load Cl ions into the cell to raise [Cl]i above ECl. Conversely, by coupling the transport of Cl to an outwardly-directed gradient for K, the 4 different K-Cl cotransporters (KCC1, KCC2, KCC3, and KCC4) primarily transport Cl ions out of the cell, lowering [Cl]i below ECl. The relative activities of the Na-(K)-Cl and K-Cl cotransporters determine [Cl]i in numerous cell types, and their activities are reciprocally regulated by the chloride-sensitive serine-threonine WNK and SPAK/OSR1 kinases.18,35,36 In neurons, NKCC1 and KCC2 are the cation-chloride cotransporters responsible for determining [Cl ]i.52

Both bumetanide and furosemide, well-known loop diuretics, are capable of inhibiting the cation-chloride co-transporters in vitro and in vivo.29,34 Bumetanide has an approximately 500-fold greater affinity for NKCC1 (Ki of ~ 0.1 μM) than for KCC2 (Ki of ~ 25–50 μM). Furosemide inhibits NKCC1 and KCC2 with equal potency (Ki of ~25–50 μM). Therefore, at low doses (2–10 μM) bumetanide is a relatively specific inhibitor of NKCC1.52 The accumulation of bumetanide in the CNS after systemic administration has not been directly measured, but the drug's high lipid:water partition coefficient and its documented anticonvulsant effects in both animals and humans in vivo suggest that the drug is able to cross the blood–brain barrier.19,21,66

A Switch in the Neuronal Expression of NKCC1 and KCC2 Underlies the Excitatory-to-Inhibitory Developmental Shift in GABAergic Signaling

In embryonic and early postnatal life, there is robust neuronal expression of NKCC1, with minimal expression of KCC2.14,53,76 This predominance of inwardly directed Cltransport via NKCC1 increases [Cl]i above ECl, such that GABAA receptor activation triggers a passive outflow of Cl ions that results in membrane depolarization and excitation. Consistent with the important role of NKCC1 for excitatory GABAergic neurotransmission, GABAA receptor-mediated depolarization is not present in NKCC1 mice,27 NKCC1 knockdown with RNA interference promotes precocious inhibitory GABAergic signaling in immature neurons,9 and excitatory GABAergic signaling can be switched to inhibitory signaling in immature neurons by transgenic overexpression of KCC2.58,69

At birth, KCC2 is expressed at very low levels,59 and the developmental negative shift in the GABA reversal potential is paralleled by a robust increase in KCC2 expression near the end of the second postnatal week in rats.76 In humans, KCC2 expression begins at the 40th week after conception.21 This increase in KCC2 expression, accompanied by the concurrent downregulation in NKCC1 expression, results in a dominance of Cl export over Cl import, decreasing [Cl]i below ECl. Consistent with the important role of KCC2 for establishing inhibitory GABAergic neuro-transmission, knockdown of KCC2 in mature hippocampal slices using RNA interference produces a positive shift in EGABA,59 cortical neurons harvested from KCC2–/– mice fail to show a developmental decrease in [Cl ]i,79 and transgenic overexpression of KCC2 in immature neurons abolishes the increases in intracellular Ca that are usually seen in early development as a result of GABA-induced membrane depolarization.40

By acting as a self-limiting trophic factor, GABA might itself trigger the developmental switch from NKCC1 to KCC2 by activating specific intracellular cascades that upregulate KCC2 gene expression.26 However, the developmental increase in KCC2 expression and the negative shift in EGABA can take place in the presence of GABAA receptor antagonists, suggesting that GABA signaling may be sufficient but not necessary for this transition.74 Recent data suggest that spontaneous cholinergic activity, by triggering Ca-dependent signaling cascades that upregulate KCC2 expression downstream of the acetylcholine nicotinic receptor, might also play a role in facilitating the excitatory-to-inhibitory GABAergic transition.42

Excitatory GABAergic Signaling in Immature Neurons Predisposes Neonates to Seizures

Increasing synaptic excitation and/or decreasing synaptic inhibition can cause neurons to become hyperactive. Because neurons harbor a multiplicity of connections with other neurons, the electrical firing of even a small population of hyperexcitable neurons, when synchronized, can progressively entrain larger neural networks until seizures ensue. In neonates, although GABA-mediated excitation plays a role in neuronal development, it also renders the developing brain particularly susceptible to seizures. In the adult cortex, excitatory glutaminergic signaling is balanced by inhibitory GABAergic signaling. However, in the neonate, the additional depolarization due to GABAA receptor activation likely adds to the excitation already initiated by glutamate neurotransmission to tip the balance of excitation/inhibition toward excessive excitation and a propensity to seizure activity. Excitation mediated by GABA has been shown to support epileptogenesis in the developing hippocampus,20 and also to decrease the seizure threshold of neonates.38 The excitatory nature of GABA signaling in immature neurons also explains why GABA agonists like barbiturates and benzodiazepines are often ineffective in reducing neonatal seizures, and can even exacerbate seizure activity.50 Clearly, new antiepileptic treatment strategies are needed for neonatal seizures.

Effect of Bumetanide, an Inhibitor of NKCC1, in Animal Models of Neonatal Seizures

Because the elevated [Cl]i of immature neurons is due to robust activity of NKCC1, this cotransporter is currently being explored as a target for novel anticonvulsant strategies for neonatal seizures. In theory, inhibition of NKCC1, by reducing [Cl]i, could reduce the GABA-mediated excitation of immature neonatal neurons, or even possibly convert the GABA response to an inhibitory one.

Dzhala et al.21 were the first to test this hypothesis in a recent seminal paper. Previous groups had established that NKCC1 expression in rats is highest in cortical neurons during the first postnatal week, begins to decrease at post-natal Day 14, and then drops to the low levels that are found in adults.53,76 Conversely, the expression of KCC2 is minimal at birth in rat cortical neurons, is low during the first postnatal week, and then attains an expression level at postnatal Day 14 that is comparable with adults.44 In rats, the higher expression levels of NKCC1 earlier in development correlate with an increased [Cl ]i49,77 and with the presence of excitatory GABAergic signaling.3,20,39,48 Dzhala and colleagues21 demonstrated that a similar expression pattern is present in the human cortex, with high NKCC1 and low KCC2 expression during the neonatal time period and before the end of the 1st year of life, a time scale that is not significantly different from the scale in the rat when adjusted for the greater length of time required for the development of the human cortex relative to the rodent brain. These data supported the hypothesis that, similar to the rat, GABA is excitatory in immature human cortical neurons, and neonates may be susceptible to seizures due to the excitatory effects of GABA during development. Because NKCC1 is known to establish the elevated levels of [Cl]i that underlie excitatory GABAergic signaling in immature neurons, these data predicted that bumetanide, an inhibitor of NKCC1, might be an effective treatment for neonatal seizures.

This hypothesis was tested by Staley67 in a series of elegant in vitro and in vivo physiological studies. These investigators found that NKCC1 blockade by bumetanide inhibited cortical seizure activity in neonatal rats both in vitro and in vivo, and this inhibition was observed at doses that have already been extensively tested in human neonates in diuresis studies.29 Specifically, they demonstrated that pharmacological inhibition of NKCC1 by bumetanide: 1) produced a negative shift in EGABA; 2) inhibited GABA-dependent synchronous excitatory activity in the immature hippocampus; 3) suppressed interictal and ictallike activity in immature hippocampal slices in vitro; and 4) attenuated kainate-induced seizure activity in vivo in neonatal rats. These anticonvulsant effects of bumetanide were shown to be specific, as they were: 1) achieved at low doses that selectively block NKCC1; 2) blocked by antagonists of the GABAA receptor, indicating that bumetanide is acting through a GABAA receptor-associated signaling pathway; 3) did not affect epileptiform activity in brain slices from NKCC1–/– mice, indicating that inhibition of NKCC1 is the mechanism by which bumetanide exerts its GABA-dependent anticonvulsant effects; and 4) did not depress epileptiform activity in mature neurons, in which the expression of NKCC1 is < 10% of that in neonatal tissue.21

Given these promising findings, it seemed reasonable to combine bumetanide, which blocks the excitatory effect of GABA in immature neurons by decreasing [Cl]i, with phenobarbital, a GABA agonist that opens GABAA receptor-associated Cl channels. Theoretically, such an increase in GABA-mediated conductance in neurons already targeted by bumetanide would serve to increase shunting inhibition and maximize the anticonvulsant power of the GABA system.

The efficacy of bumetanide, in combination with the GABA-enhancing anticonvulsant phenobarbital, was recently tested for the treatment of recurrent tonic-clonic epileptiform activity in the intact immature hippocampus in vitro.19 In this study, a low-magnesium model of neonatal seizures in the intact immature hippocampal formation in vitro was employed. Such a model has the benefits of not altering the energy gradient for NKCC1 and preserving longitudinal intrahippocampal connections. Recurrent seizures were induced in the intact hippocampal preparation by a continuous 5-hour exposure to low-magnesium solution, and the anticonvulsant efficacy of phenobarbital, bumetanide, and the combination of these drugs was then studied. Whereas phenobarbital failed to abolish or depress recurrent seizures in 70% of immature hippocampi, phenobarbital in combination with bumetanide abolished seizures in 70% of immature hippocampi, and significantly reduced the frequency, duration, and power of seizures in the remaining 30% of immature hippocampi.

Taken together, these in vitro and in vivo studies suggest that bumetanide, alone or in combination with other drugs such as phenobarbital, might be useful in the treatment of neonatal seizures in humans.19,21

Bridging the Gap From Animal Models of Neonatal Seizures to Human Subjects

At low concentrations (2–10 μM), bumetanide is a specific inhibitor of NKCC1 and has well-established pharma-cokinetic and pharmacodynamic properties in adult humans with few side effects.68 Because the expression patterns of NKCC1 during development are similar in the human and rat cortex,21 bumetanide might be useful for the treatment of seizures in human neonates. Bumetanide has been extensively used in both healthy and critically ill human full-term and preterm infants to treat fluid volume overload due to cardiac and/or pulmonary disease, so extrapolation from these studies might help guide the design of any potential pilot studies or clinical trials.43,72 However, it will be important to investigate whether the pharmacokinetics of bumetanide are altered by any of the underlying diseases that are responsible for triggering seizures in neonates, because many full-term newborns with refractory seizures have hypoxic-ischemic encephalopathy from perinatal asphyxia, which is often accompanied by multi-organ dysfunction (including hepatic and renal failure); such organ dysfunction can dramatically affect drug metabolism.12,47,73

Perhaps the best new treatment strategy for neonatal seizures and one that could be easily tested, is a combination regimen that would include bumetanide with a barbiturate like phenobarbital. In the neonatal rat brain, phenobarbital has been rendered a more effective anticonvulsant by coad-ministering it with bumetanide, which reverses the Cl gradient in immature neurons to a level such that GABAA potentiation by phenobarbital results in synaptic inhibition.19 If such a trial were to take place, neonates with persistent seizures despite an initial loading dose of phenobarbital (the current standard of care) could be offered bumetanide along with the second dose of phenobarbital. Continuous electroencephalography monitoring of patients could then be used to determine whether bumetanide reduces seizures compared with controls (those neonates treated with phenobarbital alone). Pilot studies now underway are examining the efficacy of bumetanide, administered with phenobarbital, for the treatment of neonatal seizures (FDA IND No. 101690; see http://www.cureepilepsy.org/research/current.asp).

The combination of bumetanide and a barbiturate should obviate the need to use the high doses of barbiturate or benzodiazepine that have been associated with significant side effects, such as apoptotic neurodegeneration in the developing brain and late cognitive/behavioral impairment.4,5,23,56 Moreover, because of its longstanding safe use in newborns as a diuretic, the low doses of bumetanide that are required to inhibit NKCC1 are not anticipated to produce short- or long-term side effects. However, caution must be exercised as work proceeds, and studies should be performed to determine any potential side effects of inhibiting NKCC1 in the neonatal nervous system, because GABA-mediated excitation is important for neuronal development.3 To date, bumetanide-mediated inhibition of NKCC1 in the brain, for periods of time that would far exceed the duration that would be used for the treatment of neonatal seizures, has been shown to have few developmental side effects.9,27 These modest but measurable side effects on cortical development must be weighed against the well-known detrimental effects of persistent seizures in the immature brain and the absence of knowledge regarding the developmental effects of NKCC1 inhibition in the setting of such seizures.31

Excitatory GABAergic Signaling in Adult Epilepsy Syndromes

Recent studies have also implicated excitatory GABA ergic signaling in the genesis of TLE and seizures that occur after ischemic-hypoxemic insult in adults.37,68 A feature common to these seizure syndromes is that adult neurons adopt transmembrane Cl gradients that phenotypically resemble those of neonatal neurons, rendering GABA activity excitatory instead of inhibitory.52 An increase in the relative abundance and/or activity of NKCC1 in relation to KCC2, which elevates [Cl]i, accounts for this pathological shift. For example, neurons from individuals with adult TLE accumulate Cl in hippocampal neurons to an extent that renders GABAA receptor activation excitatory,15 apparently because of increased expression of NKCC1 and decreased expression of KCC2.51 A human trial of bumetanide is currently underway in adults with medically intractable and surgically unresectable TLE associated with mesial temporal sclerosis or different cortical malformations.45

Bumetanide might also be useful for adult seizures that occur secondary to hypoxic-ischemic encephalopathy or traumatic brain injury (due to brain edema), as NKCC1 is upregulated after both of these insults, and bumetanide has been shown to decrease the excitability that accompanies the increases in neuronal [Cl]i that follow these injuries.52 For example, in rat models of cerebral hypoxic-ischemic injury, a prolonged rise in [Cl]i in hippocampal neurons renders GABAergic neurons hyperexcitable.24,33 This rise in [Cl]i, which is completely prevented by bumetanide, is associated with an increase in the expression of the phosphorylated active form of NKCC1.55 Bumetanide has also been shown to prevent ischemia-induced hyperexcitability and cell swelling.1,70 The activation of glutamate receptors and high extracellular K, factors that contribute to the pathogenesis of neuronal excitotoxicity, also have been shown to induce the phosphorylation and activation of NKCC1.63,71 These data suggest that a pathological increase in NKCC1 activity due to an increase in cotransporter phosphorylation contributes to the ischemia-induced accumulation of intracellular Cl that triggers the neuronal hyperexcitability, excitotoxicity, and injury to neurons after hypoxic-ischemic insults.11,54,78 and suggests that bumetanide might be useful for the treatment of seizures in this clinical context.

References

If the inline PDF is not rendering correctly, you can download the PDF file here.

Article Information

Address correspondence to: Kristopher T. Kahle, M.D., Ph.D., Department of Neurosurgery, Massachusetts General Hospital, Harvard Medical School, 55 Fruit Street, Boston, Massachusetts 02114. email: kkahle@partners.org.

© AANS, except where prohibited by US copyright law.

Headings

Figures

  • View in gallery

    A switch in the neuronal expression of NKCC1 and KCC2 underlies the excitatory-to-inhibitory developmental shift in GABAergic signaling, in which GABAergic neurotransmission undergoes a switch from excitatory to inhibitory due to a reversal of neuronal Cl gradients during development. The ([Cl]i) of immature neonatal neurons, compared with mature adult neurons, is much higher due to robust activity of the chloride-importing Na-K-2Cl cotransporter NKCC1, such that the binding of GABA to ligand-gated GABAA receptor-associated Cl channels triggers Cl efflux and depolarizing excitation. In adults, NKCC1 expression decreases and the expression of the genetically related chloride-extruding K-Cl cotransporter KCC2 increases, lowering [Cl]i to a level such that activation of GABAA receptors triggers Cl influx and inhibitory hyperpolarization. The excitatory action of GABA in neonates accounts for the decreased seizure threshold and poor efficacy of GABAergic anticonvulsants in this age group. Reprinted with permission from Delpire E: News Physiol Sci 15:309–312, 2000.

References

  • 1

    Beck J: Na-K-Cl cotransporter contributes to glutamate-mediated excitotoxicity. J Neurosci 23:506150682003

  • 2

    Ben-Ari Y: Developing networks play a similar melody. Trends Neurosci 24:3533602001

  • 3

    Ben-Ari Y: Excitatory actions of GABA during development: the nature of the nurture. Nat Rev Neurosci 3:7287392002

  • 4

    Bittigau PSifringer MGenz KReith EPospischil DGovindarajalu S: Antiepileptic drugs and apoptotic neurodegeneration in the developing brain. Proc Natl Acad Sci USA 99:15089150942002

  • 5

    Bittigau PSifringer MIkonomidou C: Antiepileptic drugs and apoptosis in the developing brain. Ann N Y Acad Sci 993:1031141231242003

  • 6

    Booth DEvans DJ: Anticonvulsants for neonates with seizures. Cochrane Database Syst Rev CD0042182004

  • 7

    Boylan GBRennie JMChorley GPressler RMFox GFFarrer K: Second-line anticonvulsant treatment of neonatal seizures: a video-EEG monitoring study. Neurology 62:4864882004

  • 8

    Boylan GBRennie JMPressler RMWilson GMorton MBinnie CD: Phenobarbitone, neonatal seizures, and video-EEG. Arch Dis Child Fetal Neonatal Ed 86:F165F1702002

  • 9

    Cancedda LFiumelli HChen KPoo MM: Excitatory GABA action is essential for morphological maturation of cortical neurons in vivo. J Neurosci 27:522452352007

  • 10

    Castro Conde JRHernández Borges AADoménech Martínez EGonzález Campo CPerera Soler R: Midazolam in neonatal seizures with no response to phenobarbital. Neurology 64:8768792005

  • 11

    Chen HSun D: The role of Na-K-Cl co-transporter in cerebral ischemia. Neurol Res 27:2802862005

  • 12

    Clancy RR: Summary proceedings from the neurology group on neonatal seizures. Pediatrics 117:S23S272006

  • 13

    Clancy RRLegido A: Postnatal epilepsy after EEG-confirmed neonatal seizures. Epilepsia 32:69761991

  • 14

    Clayton GHOwens GCWolff JSSmith RL: Ontogeny of cation-Cl cotransporter expression in rat neocortex. Brain Res Dev Brain Res 109:2812921998

  • 15

    Cohen INavarro VClemenceau SBaulac MMiles R: On the origin of interictal activity in human temporal lobe epilepsy in vitro. Science 298:141814212002

  • 16

    Connell JOozeer Rde Vries LDubowitz LMDubowitz V: Clinical and EEG response to anticonvulsants in neonatal seizures. Arch Dis Child 64:4594641989

  • 17

    Delpire E: Cation-chloride cotransporters in neuronal communication. News Physiol Sci 15:3093122002

  • 18

    Delpire EGagnon KB: SPAK and OSR1: STE20 kinases involved in the regulation of ion homeostasis and volume control in mammalian cells. Biochem J 409:3213312008

  • 19

    Dzhala VIBrumback ACStaley KJ: Bumetanide enhances phenobarbital efficacy in a neonatal seizure model. Ann Neurol 63:2222352008

  • 20

    Dzhala VIStaley KJ: Excitatory actions of endogenously released GABA contribute to initiation of ictal epileptiform activity in the developing hippocampus. J Neurosci 23:184018462003

  • 21

    Dzhala VITalos DMSdrulla DABrumback ACMathews GCBenke TA: NKCC1 transporter facilitates seizures in the developing brain. Nat Med 11:120512132005

  • 22

    Evans DLevene M: Neonatal seizures. Arch Dis Child Fetal Neonatal Ed 78::F70F751998

  • 23

    Farwell JRLee YJHirtz DGSulzbacher SIEllenberg JHNelson KB: Phenobarbital for febrile seizures-effects on intelligence and on seizure recurrence. N Engl J Med 322:3643691990

  • 24

    Galeffi FSah RPond BBGeorge ASchwartz-Bloom RD: Changes in intracellular chloride after oxygen-glucose deprivation of the adult hippocampal slice: effect of diazepam. J Neurosci 24:447844882004

  • 25

    Gamba G: Molecular physiology and pathophysiology of electroneutral cation-chloride cotransporters. Physiol Rev 85:4234932005

  • 26

    Ganguly KSchinder AFWong STPoo M: GABA itself promotes the developmental switch of neuronal GABAergic responses from excitation to inhibition. Cell 105:5215322001

  • 27

    Ge SGoh ELSailor KAKitabatake YMing GLSong H: GABA regulates synaptic integration of newly generated neurons in the adult brain. Nature 439:5895932006

  • 28

    Hakeem VFWallace SJ: EEG monitoring of therapy for neonatal seizures. Dev Med Child Neurol 32:8588641990

  • 29

    Hannaert PAlvarez-Guerra MPirot DNazaret CGaray RP: Rat NKCC2/NKCC1 cotransporter selectivity for loop diuretic drugs. Naunyn Schmiedebergs Arch Pharmacol 365:1931992002

  • 30

    Hebert SCMount DBGamba G: Molecular physiology of cation-coupled Cl cotransport: the SLC12 family. Pflugers Arch 447:5805932004

  • 31

    Holmes GLBen-Ari Y: Seizures in the developing brain: perhaps not so benign after all. Neuron 21:123112341998

  • 32

    Hoppen TElger CEBartmann P: Carbamazepine in phenobarbital-nonresponders: experience with ten preterm infants. Eur J Pediatr 160:4444472001

  • 33

    Inglefield JRSchwartz-Bloom RD: Activation of excitatory amino acid receptors in the rat hippocampal slice increases intra-cellular Cl and cell volume. J Neurochem 71:139614041998

  • 34

    Isenring PForbush B: Ion transport and ligand binding by the Na-K-Cl cotransporter, structure-function studies. Comp Biochem Physiol A Mol Integr Physiol 130:4874972001

  • 35

    Kahle KTRinehart Jde Los Heros PLouvi AMeade PVazquez N: WNK3 modulates transport of Cl in and out of cells: implications for control of cell volume and neuronal excitability. Proc Natl Acad Sci USA 102:16783167882005

  • 36

    Kahle KTRinehart JRing AGimenez IGamba GHebert SC: WNK protein kinases modulate cellular Cl flux by altering the phosphorylation state of the Na-K-Cl and K-Cl cotransporters. Physiology (Bethesda) 21:3263352006

  • 37

    Kahle KTStaley K: Altered neuronal chloride homeostasis and excitatory GABAergic signaling in human temporal lobe epilepsy. Epilepsy Curr 8:51532008

  • 38

    Khalilov IHolmes GLBen-Ari Y: In vitro formation of a secondary epileptogenic mirror focus by interhippocampal propagation of seizures. Nat Neurosci 6:107910852003

  • 39

    Khazipov RKhalilov ITyzio RMorozova EBen-Ari YHolmes GL: Developmental changes in GABAergic actions and seizure susceptibility in the rat hippocampus. Eur J Neurosci 19:5906002004

  • 40

    Lee HChen CXLiu YJAizenman EKandler K: KCC2 expression in immature rat cortical neurons is sufficient to switch the polarity of GABA responses. Eur J Neurosci 21:259325992005

  • 41

    Legido AClancy RRBerman PH: Neurologic outcome after electroencephalographically proven neonatal seizures. Pediatrics 88:5835961991

  • 42

    Liu ZNeff RABerg DK: Sequential interplay of nicotinic and GABAergic signaling guides neuronal development. Science 314:161016132006

  • 43

    Lopez-Samblas AMAdams JAGoldberg RNModi MW: The pharmacokinetics of bumetanide in the newborn infant. Biol Neonate 72:2652721997

  • 44

    Lu JKaradsheh MDelpire E: Developmental regulation of the neuronal-specific isoform of K-Cl cotransporter KCC2 in postnatal rat brains. J Neurobiol 39:5585681999

  • 45

    Maa EBainbridge JSpitz MCStaley KJ: Oral bumetanide add-on therapy in refractory temporal lobe epilepsy. Epilepsia 48:S62007. (Abstract)

  • 46

    Maytal JNovak GPKing KC: Lorazepam in the treatment of refractory neonatal seizures. J Child Neurol 6:3193231991

  • 47

    McBride MCLaroia NGuillet R: Electrographic seizures in neonates correlate with poor neurodevelopmental outcome. Neurology 55:5065132000

  • 48

    Owens DFBoyce LHDavis MBKriegstein AR: Excitatory GABA responses in embryonic and neonatal cortical slices demonstrated by gramicidin perforated-patch recordings and calcium imaging. J Neurosci 16:641464231996

  • 49

    Owens DFKriegstein AR: Is there more to GABA than synaptic inhibition?. Nat Rev Neurosci 3:7157272002

  • 50

    Painter MJScher MSStein ADArmatti SWang ZGardiner JC: Phenobarbital compared with phenytoin for the treatment of neonatal seizures. N Engl J Med 341:4854891999

  • 51

    Palma EAmici MSobrero FSpinelli GDi Angelantonio SRagozzino D: Anomalous levels of Cl transporters in the hippocampal subiculum from temporal lobe epilepsy patients make GABA excitatory. Proc Natl Acad Sci USA 103:846584682006

  • 52

    Payne JARivera CVoipio JKaila K: Cation-chloride co-transporters in neuronal communication, development and trauma. Trends Neurosci 26:1992062003

  • 53

    Plotkin MDSnyder EYHebert SCDelpire E: Expression of the Na-K-2Cl cotransporter is developmentally regulated in postnatal rat brains: a possible mechanism underlying GABA's excitatory role in immature brain. J Neurobiol 33:7817951997

  • 54

    Pond BBBerglund KKuner TFeng GAugustine GJSchwartz-Bloom RD: The chloride transporter Na(+)-K(+)-Cl cotrans-porter isoform-1 contributes to intracellular chloride increases after in vitro ischemia. J Neurosci 26:139614062006

  • 55

    Pond BBGaleffi FAhrens RSchwartz-Bloom RD: Chloride transport inhibitors influence recovery from oxygen-glucose deprivation-induced cellular injury in adult hippocampus. Neuropharmacology 47:2532562004

  • 56

    Reinisch JMSanders SAMortensen ELRubin DB: In utero exposure to phenobarbital and intelligence deficits in adult men. JAMA 274:151815251995

  • 57

    Rennie JMBoylan GB: Neonatal seizures and their treatment. Curr Opin Neurol 16:1771812003

  • 58

    Reynolds ABrustein ELiao MMercado ABabilonia EMount DB: Neurogenic role of the depolarizing chloride gradient revealed by global overexpression of KCC2 from the onset of development. J Neurosci 28:158815972008

  • 59

    Rivera CVoipio JPayne JARuusuvuori ELahtinen HLamsa K: The K+/Cl co-transporter KCC2 renders GABA hyper-polarizing during neuronal maturation. Nature 397:2512551999

  • 60

    Sankar RPainter MJ: Neonatal seizures: after all these years we still love what doesn't work. Neurology 64:7767772005

  • 61

    Scher MS: Seizures in the newborn infant. Diagnosis, treatment, and outcome. Clin Perinatol 24:7357721997

  • 62

    Scher MSAlvin JGaus LMinnigh BPainter MJ: Uncoupling of EEG-clinical neonatal seizures after antiepileptic drug use. Pediatr Neurol 28:2772802003

  • 63

    Schomberg SLSu GHaworth RASun D: Stimulation of Na-K-2Cl cotransporter in neurons by activation of non-NMDA ionotropic receptor and group-I mGluRs. J Neurophysiol 85:256325752001

  • 64

    Sheth RDBuckley DJGutierrez ARGingold MBodensteiner JBPenney S: Midazolam in the treatment of refractory neonatal seizures. Clin Neuropharmacol 19:1651701996

  • 65

    Silverstein FSJensen FE: Neonatal seizures. Ann Neurol 62:1121202007

  • 66

    Sipilä STSchuchmann SVoipio JYamada JKaila K: The cation-chloride cotransporter NKCC1 promotes sharp waves in the neonatal rat hippocampus. J Physiol 573:7657732006

  • 67

    Staley K: Enhancement of the excitatory actions of GABA by barbiturates and benzodiazepines. Neurosci Lett 146:1051071992

  • 68

    Staley KJ: Wrong-way chloride transport: is it a treatable cause of some intractable seizures?. Epilepsy Currents 6:1241272006

  • 69

    Stein VHermans-Borgmeyer IJentsch TJHübner CA: Expression of the KCl cotransporter KCC2 parallels neuronal maturation and the emergence of low intracellular chloride. J Comp Neurol 468:57642004

  • 70

    Su GKintner DBFlagella MShull GESun D: Astrocytes from Na(+)-K(+)-Cl(-) cotransporter-null mice exhibit absence of swelling and decrease in EAA release. Am J Physiol Cell Physiol 282:C1147C11602002

  • 71

    Su GKintner DBSun D: Contribution of Na(+)-K(+)-Cl(-) cotransporter to high-[K(+)](o)-induced swelling and EAA release in astrocytes. Am J Physiol Cell Physiol 282:C1136C11462002

  • 72

    Sullivan JEWitte MKYamashita TSMyers CMBlumer JL: Pharmacokinetics of bumetanide in critically ill infants. Clin Pharmacol Ther 60:4054131996

  • 73

    Tekgul HGauvreau KSoul JMurphy LRobertson RStewart J: The current etiologic profile and neurodevelopmental outcome of seizures in term newborn infants. Pediatrics 117:127012802006

  • 74

    Titz SHans MKelsch WLewen ASwandulla DMisgeld U: Hyperpolarizing inhibition develops without trophic support by GABA in cultured rat midbrain neurons. J Physiol 550:7197302003

  • 75

    Volpe JJ: Neurology of the Newborn PhiladelphiaWB Saunders2001. 178214

  • 76

    Wang CShimizu-Okabe CWatanabe KOkabe AMatsuzaki HOgawa T: Developmental changes in KCC1, KCC2, and NKCC1 mRNA expressions in the rat brain. Brain Res Dev Brain Res 139:59662002

  • 77

    Yamada JOkabe AToyoda HKilb WLuhmann HJFukuda A: Cl uptake promoting depolarizing GABA actions in immature rat neocortical neurones is mediated by NKCC1. J Physiol 557:8298412004

  • 78

    Yan YDempsey RJFlemmer AForbush BSun D: Inhibition of Na(+)-K(+)-Cl(-) cotransporter during focal cerebral ischemia decreases edema and neuronal damage. Brain Res 961:22312003

  • 79

    Zhu LLovinger DDelpire E: Cortical neurons lacking KCC2 expression show impaired regulation of intracellular chloride. J Neurophysiol 93:155715682005

TrendMD

Metrics

Metrics

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 174 174 34
PDF Downloads 65 65 17
EPUB Downloads 0 0 0

PubMed

Google Scholar