Deletion of the Na-K-2Cl cotransporter NKCC1 results in a more severe epileptic phenotype in the intrahippocampal kainate mouse model of temporal lobe epilepsy

Astrocytic NKCC1 inhibits seizures by buffering Cl- and antagonizing neuronal NKCC1 at GABAergic synapses

OBJECTIVE

A pathological excitatory action of the major inhibitory neurotransmitter γ-aminobutyric acid (GABA) has been observed in epilepsy. Blocking the Cl- importer NKCC1 with bumetanide is expected to reduce the neuronal intracellular Cl- concentration ([Cl- ]i ) and thereby attenuate the excitatory GABA response. Accordingly, several clinical trials of bumetanide for epilepsy were conducted. Although NKCC1 is expressed in both neurons and glial cells, an involvement of glial NKCC1 in seizures has not yet been reported. Astrocytes maintain high [Cl- ]i with NKCC1, and this gradient promotes Cl- efflux via the astrocytic GABAA receptor (GABAA R). This Cl- efflux buffers the synaptic cleft Cl- concentration to maintain the postsynaptic Cl- gradient during intense firing of GABAergic neurons, thereby sustaining its inhibitory action during seizure. In this study, we investigated the function of astrocytic NKCC1 in modulating the postsynaptic action of GABA in acute seizure models.

METHODS

We used the astrocyte-specific conditional NKCC1 knockout (AstroNKCC1KO) mice. The seizurelike events (SLEs) in CA1 pyramidal neurons were triggered by tetanic stimulation of stratum radiatum in acute hippocampus slices. The SLE underlying GABAA R-mediated depolarization was evaluated by applying the GABAA R antagonist bicuculline. The pilocarpine-induced seizure in vivo was monitored in adult mice by the Racine scale. The SLE duration and tetanus stimulation intensity threshold and seizure behavior in AstroNKCC1KO mice and wild-type (WT) mice were compared.

RESULTS

The AstroNKCC1KO mice were prone to seizures with lower threshold and longer duration of SLEs and larger GABAA R-mediated depolarization underlying the SLEs, accompanied by higher Racine-scored seizures. Bumetanide reduced these indicators of seizure in AstroNKCC1KO mice (which still express neuronal NKCC1), but not in the WT, both in vitro and in vivo.

SIGNIFICANCE

Astrocytic NKCC1 inhibits GABA-mediated excitatory action during seizures, whereas neuronal NKCC1 has the converse effect, suggesting opposing actions of bumetanide on these cells.

Chloride transporters controlling neuronal excitability

Synaptic inhibition plays a crucial role in regulating neuronal excitability, which is the foundation of nervous system function. This inhibition is largely mediated by the neurotransmitters GABA and glycine that activate Cl^--permeable ion channels, which means that the strength of inhibition depends on the Cl^- gradient across the membrane. In neurons, the Cl^- gradient is primarily mediated by two secondarily active cation-chloride cotransporters (CCCs), NKCC1 and KCC2. CCC-mediated regulation of the neuronal Cl^- gradient is critical for healthy brain function, as dysregulation of CCCs has emerged as a key mechanism underlying neurological disorders including epilepsy, neuropathic pain, and autism spectrum disorder. This review begins with an overview of neuronal chloride transporters before explaining the dependent relationship between these CCCs, Cl^- regulation, and inhibitory synaptic transmission. We then discuss the evidence for how CCCs can be regulated, including by activity and their protein interactions, which underlie inhibitory synaptic plasticity. For readers who may be interested in conducting experiments on CCCs and neuronal excitability, we have included a section on techniques for estimating and recording intracellular Cl^-, including their advantages and limitations. Although the focus of this review is on neurons, we also examine how Cl^- is regulated in glial cells, which in turn regulate neuronal excitability through the tight relationship between this nonneuronal cell type and synapses. Finally, we discuss the relatively extensive and growing literature on how CCC-mediated neuronal excitability contributes to neurological disorders.

Spatiotemporal expression patterns of genes coding for plasmalemmal chloride transporters and channels in neurological diseases

Neuronal voltage changes which are dependent on chloride transporters and channels are involved in forming neural functions during early development and maintaining their stability until adulthood. The intracellular chloride concentration maintains a steady state, which is delicately regulated by various genes coding for chloride transporters and channels (GClTC) on the plasmalemma; however, the synergistic effect of these genes in central nervous system disorders remains unclear. In this study, we first defined 10 gene clusters with similar temporal expression patterns, and identified 41 GClTC related to brain developmental process. Then, we found 4 clusters containing 22 GClTC were enriched for the neuronal functions. The GClTC from different clusters presented distinct cell type preferences and anatomical heterogeneity. We also observed strong correlations between clustered genes and diseases, most of which were nervous system disorders. Finally, we found that one of the most well-known GClTC, SLC12A2, had a more profound effect on glial cell-related diseases than on neuron-related diseases, which was in accordance with our observation that SLC12A2 was mainly expressed in oligodendrocytes during brain development. Our findings provide a more comprehensive understanding of the temporal and spatial expression characteristics of GClTC, which can help us understand the complex roles of GClTC in the development of the healthy human brain and the etiology of brain disorders.

GABAergic circuits drive focal seizures

Epilepsy is based on abnormal neuronal activities that have historically been suggested to arise from an excess of excitation and a defect of inhibition, or in other words from an excessive glutamatergic drive not balanced by GABAergic activity. More recent data however indicate that GABAergic signaling is not defective at focal seizure onset and may even be actively involved in seizure generation by providing excitatory inputs. Recordings of interneurons revealed that they are active at seizure initiation and that their selective and time-controlled activation using optogenetics triggers seizures in a more general context of increased excitability. Moreover, GABAergic signaling appears to be mandatory at seizure onset in many models. The main pro-ictogenic effect of GABAergic signaling is the depolarizing action of GABA_A conductance which may occur when an excessive GABAergic activity causes Cl^- accumulation in neurons. This process may combine with background dysregulation of Cl^-, well described in epileptic tissues. Cl^- equilibrium is maintained by (Na⁺)/K⁺/Cl^- co-transporters, which can be defective and therefore favor the depolarizing effects of GABA. In addition, these co-transporters further contribute to this effect as they mediate K⁺ outflow together with Cl^- extrusion, a process that is responsible for K⁺ accumulation in the extracellular space and subsequent increase of local excitability. The role of GABAergic signaling in focal seizure generation is obvious but its complex dynamics and balance between GABA_A flux polarity and local excitability still remain to be established, especially in epileptic tissues where receptors and ion regulators are disrupted and in which GABAergic signaling rather plays a 2 faces Janus role.

Chloride imbalance in Fragile X syndrome

Developmental changes in ionic balance are associated with crucial hallmarks in neural circuit formation, including changes in excitation and inhibition, neurogenesis, and synaptogenesis. Neuronal excitability is largely mediated by ionic concentrations inside and outside of the cell, and chloride (Cl-) ions are highly influential in early neurodevelopmental events. For example, γ-aminobutyric acid (GABA) is the main inhibitory neurotransmitter of the mature central nervous system (CNS). However, during early development GABA can depolarize target neurons, and GABAergic depolarization is implicated in crucial neurodevelopmental processes. This developmental shift of GABAergic neurotransmission from depolarizing to hyperpolarizing output is induced by changes in Cl- gradients, which are generated by the relative expression of Cl- transporters Nkcc1 and Kcc2. Interestingly, the GABA polarity shift is delayed in Fragile X syndrome (FXS) models; FXS is one of the most common heritable neurodevelopmental disorders. The RNA binding protein FMRP, encoded by the gene Fragile X Messenger Ribonucleoprotein-1 (Fmr1) and absent in FXS, appears to regulate chloride transporter expression. This could dramatically influence FXS phenotypes, as the syndrome is hypothesized to be rooted in defects in neural circuit development and imbalanced excitatory/inhibitory (E/I) neurotransmission. In this perspective, we summarize canonical Cl- transporter expression and investigate altered gene and protein expression of Nkcc1 and Kcc2 in FXS models. We then discuss interactions between Cl- transporters and neurotransmission complexes, and how these links could cause imbalances in inhibitory neurotransmission that may alter mature circuits. Finally, we highlight current therapeutic strategies and promising new directions in targeting Cl- transporter expression in FXS patients.

NKCC1 and KCC2: Structural insights into phospho-regulation

Inhibitory neurotransmission plays a fundamental role in the central nervous system, with about 30–50% of synaptic connections being inhibitory. The action of both inhibitory neurotransmitter, gamma-aminobutyric-acid (GABA) and glycine, mainly relies on the intracellular Cl^– concentration in neurons. This is set by the interplay of the cation chloride cotransporters NKCC1 (Na⁺, K⁺, Cl^– cotransporter), a main Cl^– uptake transporter, and KCC2 (K⁺, Cl^– cotransporter), the principle Cl^– extruder in neurons. Accordingly, their dysfunction is associated with severe neurological, psychiatric, and neurodegenerative disorders. This has triggered great interest in understanding their regulation, with a strong focus on phosphorylation. Recent structural data by cryogenic electron microscopy provide the unique possibility to gain insight into the action of these phosphorylations. Interestingly, in KCC2, six out of ten (60%) known regulatory phospho-sites reside within a region of 134 amino acid residues (12% of the total residues) between helices α8 and α9 that lacks fixed or ordered three-dimensional structures. It thus represents a so-called intrinsically disordered region. Two further phospho-sites, Tyr⁹⁰³ and Thr⁹⁰⁶, are also located in a disordered region between the ß8 strand and the α8 helix. We make the case that especially the disordered region between helices α8 and α9 acts as a platform to integrate different signaling pathways and simultaneously constitute a flexible, highly dynamic linker that can survey a wide variety of distinct conformations. As each conformation can have distinct binding affinities and specificity properties, this enables regulation of [Cl^–]_i and thus the ionic driving force in a history-dependent way. This region might thus act as a molecular processor underlying the well described phenomenon of ionic plasticity that has been ascribed to inhibitory neurotransmission. Finally, it might explain the stunning long-range effects of mutations on phospho-sites in KCC2.

Molecular Mechanisms of Epilepsy: The Role of the Chloride Transporter KCC2

Epilepsy is a neurological disease characterized by abnormal or synchronous brain activity causing seizures, which may produce convulsions, minor physical signs, or a combination of symptoms. These disorders affect approximately 65 million people worldwide, from all ages and genders. Seizures apart, epileptic patients present a high risk to develop neuropsychological comorbidities such as cognitive deficits, emotional disturbance, and psychiatric disorders, which severely impair quality of life. Currently, the treatment for epilepsy includes the administration of drugs or surgery, but about 30% of the patients treated with antiepileptic drugs develop time-dependent pharmacoresistence. Therefore, further investigation about epilepsy and its causes is needed to find new pharmacological targets and innovative therapeutic strategies. Pharmacoresistance is associated to changes in neuronal plasticity and alterations of GABA_A receptor-mediated neurotransmission. The downregulation of GABA inhibitory activity may arise from a positive shift in GABA_A receptor reversal potential, due to an alteration in chloride homeostasis. In this paper, we review the contribution of K⁺-Cl^--cotransporter (KCC2) to the alterations in the Cl^- gradient observed in epileptic condition, and how these alterations are coupled to the increase in the excitability.

How Staying Negative Is Good for the (Adult) Brain: Maintaining Chloride Homeostasis and the GABA-Shift in Neurological Disorders

Excitatory-inhibitory (E-I) imbalance has been shown to contribute to the pathogenesis of a wide range of neurodevelopmental disorders including autism spectrum disorders, epilepsy, and schizophrenia. GABA neurotransmission, the principal inhibitory signal in the mature brain, is critically coupled to proper regulation of chloride homeostasis. During brain maturation, changes in the transport of chloride ions across neuronal cell membranes act to gradually change the majority of GABA signaling from excitatory to inhibitory for neuronal activation, and dysregulation of this GABA-shift likely contributes to multiple neurodevelopmental abnormalities that are associated with circuit dysfunction. Whilst traditionally viewed as a phenomenon which occurs during brain development, recent evidence suggests that this GABA-shift may also be involved in neuropsychiatric disorders due to the “dematuration” of affected neurons. In this review, we will discuss the cell signaling and regulatory mechanisms underlying the GABA-shift phenomenon in the context of the latest findings in the field, in particular the role of chloride cotransporters NKCC1 and KCC2, and furthermore how these regulatory processes are altered in neurodevelopmental and neuropsychiatric disorders. We will also explore the interactions between GABAergic interneurons and other cell types in the developing brain that may influence the GABA-shift. Finally, with a greater understanding of how the GABA-shift is altered in pathological conditions, we will briefly outline recent progress on targeting NKCC1 and KCC2 as a therapeutic strategy against neurodevelopmental and neuropsychiatric disorders associated with improper chloride homeostasis and GABA-shift abnormalities.

CNS pharmacology of NKCC1 inhibitors

The Na-K-2Cl cotransporter NKCC1 and the neuron-specific K-Cl cotransporter KCC2 are considered attractive CNS drug targets because altered neuronal chloride regulation and consequent effects on GABAergic signaling have been implicated in numerous CNS disorders. While KCC2 modulators are not yet clinically available, the loop diuretic bumetanide has been used off-label in attempts to treat brain disorders and as a tool for NKCC1 inhibition in preclinical models. Bumetanide is known to have anticonvulsant and neuroprotective effects under some pathophysiological conditions. However, as shown in several species from neonates to adults (mice, rats, dogs, and by extrapolation in humans), at the low clinical doses of bumetanide approved for diuresis, this drug has negligible access into the CNS, reaching levels that are much lower than what is needed to inhibit NKCC1 in cells within the brain parenchyma. Several drug discovery strategies have been initiated over the last ∼15 years to develop brain-permeant compounds that, ideally, should be selective for NKCC1 to eliminate the diuresis mediated by inhibition of renal NKCC2. The strategies employed to improve the pharmacokinetic and pharmacodynamic properties of NKCC1 blockers include evaluation of other clinically approved loop diuretics; development of lipophilic prodrugs of bumetanide; development of side-chain derivatives of bumetanide; and unbiased high-throughput screening approaches of drug discovery based on large chemical compound libraries. The main outcomes are that (1), non-acidic loop diuretics such as azosemide and torasemide may have advantages as NKCC1 inhibitors vs. bumetanide; (2), bumetanide prodrugs lead to significantly higher brain levels than the parent drug and have lower diuretic activity; (3), the novel bumetanide side-chain derivatives do not exhibit any functionally relevant improvement of CNS accessibility or NKCC1 selectivity vs. bumetanide; (4) novel compounds discovered by high-throughput screening may resolve some of the inherent problems of bumetanide, but as yet this has not been achieved. Thus, further research is needed to optimize the design of brain-permeant NKCC1 inhibitors. In parallel, a major challenge is to identify the mechanisms whereby various NKCC1-expressing cellular targets of these drugs within (e.g., neurons, oligodendrocytes or astrocytes) and outside the brain parenchyma (e.g., the blood-brain barrier, the choroid plexus, and the endocrine system), as well as molecular off-target effects, might contribute to their reported therapeutic and adverse effects.

Pharmacological tools to target NKCC1 in brain disorders

The chloride importer NKCC1 and the chloride exporter KCC2 are key regulators of neuronal chloride concentration. A defective NKCC1/KCC2 expression ratio is associated with several brain disorders. Preclinical/clinical studies have shown that NKCC1 inhibition by the United States FDA-approved diuretic bumetanide is a potential therapeutic strategy in preclinical/clinical studies of multiple neurological conditions. However, bumetanide has poor brain penetration and causes unwanted diuresis by inhibiting NKCC2 in the kidney. To overcome these issues, a growing number of studies have reported more brain-penetrating and/or selective bumetanide prodrugs, analogs, and new molecular entities. Here, we review the evidence for NKCC1 pharmacological inhibition as an effective strategy to manage neurological disorders. We also discuss the advantages and limitations of bumetanide repurposing and the benefits and risks of new NKCC1 inhibitors as therapeutic agents for brain disorders.

Reply to the commentary by Ben-Ari and Delpire: Bumetanide and neonatal seizures: Fiction versus reality

In this response to a commentary by Ben-Ari and Delpire on our recent study on the pharmacology of neonatal seizures in a novel, physiologically validated rat model of birth asphyxia, we wish to rectify their inaccurate descriptions of our model and data. Furthermore, because Ben-Ari and Delpire suggest that negative data on bumetanide from preclinical and clinical trials of neonatal seizures have few implications for (alleged) bumetanide actions on neurons in other brain disorders, we will discuss this topic as well. Based on the poor brain penetration of bumetanide, combined with the extremely wide cellular expression patterns of the target protein NKCC1, it is obvious that the numerous actions of systemically applied bumetanide described in the literature are not mediated by the drug's effects on central neurons.