Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2017 Jun 28.
Published in final edited form as: Epilepsia. 2015 Dec 11;57(2):222–232. doi: 10.1111/epi.13270

Bumetanide reduces seizure progression and the development of pharmacoresistant status epilepticus

Sudhir Sivakumaran *,†, Jamie Maguire †
PMCID: PMC5487491  NIHMSID: NIHMS866021  PMID: 26659482

Summary

Objective

We investigated the role of chloride homeostasis in seizure progression and development of pharmacoresistant status epilepticus (SE) by pharmacologically targeting the Na-K-Cl cotransporter (NKCC1) with bumetanide. We also investigated the ability of bumetanide to restore the efficacy of diazepam following SE.

Methods

Kainic acid (KA)–induced SE in vivo and 0-Mg2+-induced seizure-like events (SLEs) in vitro were monitored using electroencephalography (EEG) recordings in freely moving adult male mice and extracellular field potential recordings in acute entorhinal cortex-hippocampus slices, respectively. The ability of bumetanide to decrease epileptiform activity and prevent the development of pharmacoresistance to diazepam following SE was evaluated.

Results

Bumetanide treatment significantly reduced KA-induced ictal activity in vivo and SLEs in vitro. In addition, bumetanide restored the efficacy of diazepam in decreasing ictal activity following SE in both the in vivo and in vitro models.

Significance

Our data demonstrate an anticonvulsant effect of bumetanide on KA-induced seizures in adult mice, suggesting a role for chloride plasticity in seizure progression. These data also demonstrate that the erosion of inhibition during seizure progression could underlie the development of pharmacoresistant SE and implicate a role for chloride plasticity in this process.

Keywords: GABA, NKCC1, Pharmacoresistance, Epilepsy, Chloride, Status epilepticus, Epilepsy

Status epilepticus (SE), clinically defined as continuous epileptiform activity or two or more seizures lasting longer than 5 min, is a life-threatening, medical emergency characterized by unremitting, persistent seizures, with an associated mortality rate of approximately 20%.1 Although the first-line treatment for SE are benzodiazepines (BDZs), prolonged seizures result in the development of pharmacoresistance in approximately 60% of patients.1

One popular mechanism proposed to underlie the development of pharmacoresistance is a reduction in γ-aminobutyric acid (GABA)ergic inhibition mediated by BDZ-sensitive, GABAAR-γ2-containing receptors following SE.2 Recently, compromised GABAergic inhibition due to pathologic changes in chloride homeostasis has been proposed to contribute to pharmacoresistant SE.3 Effective GABAergic inhibition and modulation by BDZs require maintenance of low intracellular chloride levels, which is primarily accomplished by KCC2 in the adult.

During development when intracellular chloride levels are high due to high levels of NKCC1 expression and low levels of KCC2, GABA is depolarizing and neonatal seizures are pharmacoresistant.4,5 Status epilepticus in the intact neonatal hippocampus results in the accumulation of intracellular chloride, involving a NKCC1-dependent mechanism,6 and blocking NKCC1 with bumetanide has been demonstrated to exhibit anticonvulsant actions both in vitro and in vivo.4 Furthermore, bumetanide has also been shown to restore the anticonvulsant effects of phenobarbital in the immature brain.5 These anticonvulsant actions of bumetanide are mediated by an NKCC1-dependent mechanism, since similar effects are not observed in NKCC1-knockout mice,4,5 and are thought to be unique to the immature brain due to the excitatory actions of GABA.4,5,7 However, alterations in NKCC1 and KCC2 have also been demonstrated following SE in adult animals. Seizure progression results in intracellular chloride loading in neuronal cultures and a loss of KCC2 and a shift in EGABA following status epilepticus in vivo.8 Of interest, loss of KCC2 or pharmacologic blockade of KCC2 reduces the efficacy of diazepam, implicating KCC2 in the development of pharmacoresistance. 3,10

The aim of this study is to investigate whether pathologic changes in chloride homeostasis contribute to GABAergic dysfunction associated with seizure progression and the development of pharmacoresistant SE in the adult. To test this hypothesis, we utilized the NKCC1 inhibitor, bumetanide, to investigate whether alterations in chloride homeostasis and the capacity for GABAergic inhibition impacts seizure progression and the development of pharmacoresistance both in vivo and in vitro. Our data demonstrate anticonvulsant effects of bumetanide in the adult and its ability to restore the seizure-suppressing effects of diazepam. Together, these data support previous studies demonstrating compromised GABAergic inhibition during seizure progression,11 which may contribute directly to the generation of epileptiform activity,12,13 leading to the development of pharmacoresistant SE. This study implicates a role for chloride plasticity in the development of pharmacoresistant SE and provides convincing experimental evidence for the anticonvulsant effects of bumetanide in the adult.

Materials and Methods

Animal handling

Adult male C57BL/6 mice were obtained from Jackson Lab and housed at the Tufts University School of Medicine, Division of Laboratory Animal Medicine. All mice were handled according to protocols approved by the Institutional Animal Care and Use Committee (IACUC).

Electroencephalography (EEG) recordings

We performed in vivo EEG recordings in awake, behaving animals to assess seizure susceptibility following administration of the chemoconvulsant KA. KA is an established model of temporal lobe epilepsy in rodents in which seizure activity progresses to SE and development of pharmacoresistance (reviewed in Reddy and Kuruba14). EEG recordings were acquired and analyzed using PowerLab and LabChart Pro (AD Instruments), as described previously.15–17 Electrographic epileptiform activity was measured for 2 h following either an intraperitoneal (i.p.) injection with 20 mg/kg KA, pilocarpine (340 mg/kg), or intrahippocampal administration of 18.8 mM KA (500 nl), similar to previous studies.18 A single dose of bumetanide was used at a final concentration of 0.2 mg/kg or 2.0 mg/kg, i.p., and 54.8 μM (500 nl) for intrahippocampal administration. As a control, mice were administered chlorothiazide (CTZ, 10 mg/Kg, i.p.),19 which is an effective diuretic but lacks effects on NKCC1 and KCC2,20 and is predicted to lack central nervous system (CNS) permeability.21 Diazepam was administered at a final concentration of 20 mg/kg, i.p., 1 h following KA administration (Fig. 5) or 5 μm for in vitro experiments. In the KA model, animals enter SE, a period characterized by an unremitting, persistent epileptiform activity, approximately 1 h after KA administration. Therefore, this time point was chosen to administer diazepam or diazepam + bumetanide in order to examine the development of pharmacoresistance in SE. Abnormal periods of EEG activity (Figs. 1e and 5Ab) that cannot be defined as a ictal activity, including periods of rhythmic spiking lasting >30 s, along with ictal events, were collectively defined as “epileptiform activity.” These criteria have been used successfully previously by our group15–17 as well as by experts in the field.22

Figure 5.

Figure 5

Open in a new tab

Bumetanide prevents the development of pharmacoresistance in vivo. (A), Representative traces of electrographic epileptiform activity showing the full duration of the recording in from mice treated with diazepam, in the absence (top, grey trace) or presence (bottom, blue trace) of bumetanide (0.2 mg/kg). The time of administration of drugs are labeled accordingly. A higher time resolution of brief episodes of the EEG traces are shown below indicating electrographic activity during baseline (a), 1 h after KA(b), and 45 min (c) and 235–240 min (d) after diazepam treatment in the absence (top, gray trace) or presence (bottom, blue trace) of bumetanide. Note the visible reduction in epileptiform activity in the presence of bumetanide compared to diazepam alone (inset d). (B), The average histograms demonstrate that bumetanide treatment enhanced the ability of diazepam to decrease the percent time exhibiting epileptiform activity. Values ± SEM; n = 3–7 mice per experimental group; ANOVA, ***p < 0.001, *p < 0.05. Epilepsia © ILAE

Figure 1.

Figure 1

Open in a new tab

Bumetanide decreases KA-induced epileptiform activity in vivo. (A) Representative traces showing KA-induced epileptiform activity for the full 2 h recording period in vivo in vehicle- or bumetanide-treated mice, top (black) and bottom (red) traces, respectively. EEG traces are labeled indicating drug administration times (vehicle, bumetanide (0.2 mg/kg) and KA) accordingly. A higher time resolution of brief episodes of the EEG traces are shown below indicating electrographic activity during baseline (a), following vehicle or bumetanide treatment (b) and at 30, 60,90, and 120 min, respectively, following KA administration (c-f). (B) Histograms showing the latency (min), average epileptiform duration (min), and epileptiform activity (%) of KA-induced epileptiform activity in vehicle- (black) and bumetanide-(red) treated mice. Values ± standard error of the mean (SEM); n = 5–9 mice per experimental group; ANOVA, **P < 0.05. Epilepsia © ILAE

Electrophysiologic recordings

Mice were anesthetized with isoflurane and decapitated, and the brain was quickly removed and immersed in ice-cold sucrose-based cutting solution containing (in mM): 87 NaCl, 2.5 KCl, 0.5 CaCl2, 25 NaHCO3, 1.25 NaH2PO4, 7 MgCl2, 50 sucrose, and 25 d-glucose, equilibrated with 95% O2 and 5% CO2. Acute 400 μm transverse hippocampal slices were prepared using a Leica Vibratome and were allowed to recover for 1 h in normal artificial cerebrospinal fluid (nACSF) containing (in mM) 126 NaCl, 26 NaHCO3, 1.5 NaH2PO4, 2.5 KCl, 2 CaCl2, 2 MgCl2, and 10 dextrose (300—310 mOsm) at 34°C.

Field potential recordings were obtained with glass micro-electrodes filled with nACSF (resistance 1–5 mOhm) and positioned in layer III of medial entorhinal cortex (mEC). Epileptiform activity was recorded using the 0-Mg2+ model with ACSF containing (in mM) 126 NaCl, 26 NaHCO3, 1.5 NaH2PO4, 5 KCl, 2 CaCl2, and 10 dextrose (300— 310 mOsm) at 34°C. In this model, slices develop a progressive, characteristic pattern of epileptiform activity, including seizure-like events (SLEs),11 and SLEs have been shown to be resistant to treatment with conventional AEDs.23

Data acquisition was carried out using a PowerLab hardware and software (AD Instruments), and data analysis was performed using LabChart Pro (AD Instruments) and pClamp10 (Molecular Devices LLC, U.S.A.).

Western blot analysis

Western blot analysis was carried out as described previously.24,25 Blots were probed with a polyclonal antibody specific for KCC2, P-KCC2 (Ser940), NKCC1, or b-tubulin. See Data S1 for additional experimental details.

Results

Bumetanide decreases KA-induced epileptiform activity in adult mice in vivo

To determine whether bumetanide can suppress seizures in adult mice, we measured epileptiform activity following administration of KA (20 mg/kg) using in vivo EEG recording in vehicle-treated (0.09% injection saline) or bumetanide-treated (0.2 or 2 mg/kg) mice. Intraperitoneal injection of both 0.2 and 2 mg/kg bumetanide 30 min prior to KA administration effectively reduced epileptiform activity (Fig. 1A). In vehicle-treated controls, KA administration results in sustained electrographic and behavioral seizures that persist throughout the recording period (Fig. 1A, gray trace and inset c-f, top traces). Bumetanide-treated animals exhibit a decrease in epileptiform activity, with a marked reduction in high-frequency discharges and increase in persistent low-frequency activity (Fig. 1A, black trace and insets c-f, bottom traces). Bumetanide-treated mice exhibit an increased latency to the onset of epileptiform activity (0.2 mg/kg: 15.4 ± 0.7 min, ANOVA, P = 0.014, F1,12 = 8.27; 2 mg/kg: 13.3 ± 0.7 min, ANOVA, P = 0.012, F1,9 = 9.63) compared to vehicle-treated mice (7.9 ± 1.4 min, n = 5– 9 mice per experimental group).

Bumetanide treatment also decreased the average duration of epileptiform events (0.2 mg/kg: 3.9 ± 0.3 min, ANOVA, P = 0.03, F1,12 = 5.8; 2 mg/kg: 3.6 ± 0.6 min, ANOVA, P = 0.04, F1,9 = 5.92) and percent time exhibiting epileptiform activity during the 2 h post-KA administration (0.2 mg/kg: 64.1 ± 2.9%, ANOVA, P = 0.037, F1,12 = 5.53; 2 mg/kg: 58 ± 0.2%, ANOVA, P = 0.003, F1,9 = 21.98) compared to vehicle-treated mice (duration: 6.02 ± 1.1 min; percent time: 76.9 ± 2.9%, n = 5–9 mice per experimental group). Quantitative comparisons in the power of epileptiform activity over time were made between vehicle- (Fig. S1A) and bumetanide-treated mice (Fig. S1B) by determining the total electrical power in EEG studies of vehicle and bumetanide-treated mice. KA-induced epileptiform activity in vehicle-treated mice was correlated with a progressive increase in EEG power, which remained high throughout the recording period (Fig. S1A). In contrast, bumetanide-treated mice demonstrate a decrease in the total power, which remains lower throughout the recording period compared to vehicle-treated mice (Fig. S1B). In addition, even though bumetanide treatment reduced the total power to near-baseline levels (Fig. S1B, top), spectral analysis demonstrated a decrease in the power of high frequency ictal activity and the presence of persistent low frequency activity over the 2 h recording period, post-KA administration (Fig. S1B, bottom; see also Fig. 2). To take a closer look at the changes in the frequency of epileptiform activity in vehicle- and bumetanide-treated mice, a detailed analysis of the electrographic epileptiform activity (see Materials and Methods and Data S1) was performed that demonstrates a progressive increase in the duration of high-frequency, ictal activity (duration: 1.5 ± 0.2 min; Fig. 2A) in vehicle-treated mice, with a decrease in low-frequency epileptiform activity, 2 h post-KA administration (6.6 ± 2.4 per minute, Fig. 2B). Bumetanide treatment decreased the progression of ictal activity, as evident by a decrease in the duration of high-frequency, ictal events (0.2 mg/kg: 0.59 ± 0.05 min, ANOVA, P < 0.001, F1,13 = 34.94; 2 mg/kg: 0.65 ± 0.05 min, ANOVA, P = 0.007, F1,12 = 10.42) and a decrease in the percent time exhibiting ictal-like activity (0.2 mg/kg: 8.58 ± 0.95%, ANOVA, P < 0.001, F1,13 = 20.37; 2 mg/kg: 10.7 ± 2.2%, ANOVA, P = 0.015, F1,12 = 7.87) compared to vehicle-treated mice (duration: 1.5 ± 0.2 min; percent time: 27.9 ± 5.6%, n = 5–9 mice per experimental group, Fig. 2A). Conversely, bumetanide treatment prevents the progressive loss of low-frequency activity, 2 h post-KA administration (0.2 mg/kg: 19.2 ± 3.14 events per minute; 2 mg/kg: 22.3 ± 3.75 events per minute) compared to vehicle-treated mice (6.6 ± 2.4 events per minute), which can also be appreciated in the power spectrum analysis (Fig. 2B). These data suggest that bumetanide treatment may alter the course of seizure progression. The anticonvulsant actions of bumetanide are specific and cannot be replicated using another diuretic that lacks central actions. Mice treated with chlorothiazide (CTZ; 10 mg/kg, i.p.), a diuretic that does not exhibit central effects, did not show significant difference in percent time exhibiting epileptiform activity (72.7 ± 4.1%, n = 6) post-KA administration, compared to vehicle-treated mice (76.9 ± 2.9%, n = 9, P = 0.8).

Figure 2.

Figure 2

Open in a new tab

Bumetanide preserves the low-frequency epileptiform activity and attenuates ictal discharges in vivo. In depth analysis of electrographic seizures recorded in vivo (see Materials and Methods), revealed changes in ictal (A) and lowfrequency epileptiform activity (B) in bumetanide-treated mice compared to vehicle controls. A, Quantification of ictal events in 60 s bins shows that bumetanide-treated mice exhibit significantly shorter ictal activity compared to vehicle-treated mice, and an overall reduction in the frequency of ictal events following KA treatment. A (inset), Bumetanide treatment (0.2 and 2 mg/kg, i.p.) significantly reduced the duration of ictal events and percent time exhibiting ictal activity compared to vehicle-treated controls. Representative power spectral analysis demonstrates a reduction in high-frequency, ictal activity in the presence of bumetanide (bottom) compared to controls. B, Vehicle-treated mice show a progressive loss of low-frequency epileptiform activity (black circles); whereas, bumetanide (0.2 and 2 mg/Kg, i.p., red and gray circles, respectively) treatment prevents this loss of low frequency epileptiform activity. Representative power spectral analysis reflects the progressive loss of low-frequency epileptiform activity in vehicle controls (top), which is prevented in animals treated with bumetanide (bottom). Values ± SEM; n = 5–9 mice per experimental group; ANOVA, *** P < 0.001, ** P < 0.01 and *P < 0.05. Epilepsia © ILAE

Bumetanide decreases intrahippocampal KA-induced seizures

To further investigate whether the anticonvulsant actions of systemically administered bumetanide are due to central actions, we investigated the anticonvulsant potential of intrahippocampal administration of bumetanide. Vehicle or bumetanide (54.8 μm, 500 nl) was injected directly into the hippocampus 30 min prior to intrahippocampal administration of KA (18.8 mM, 500 nl), while electrographic activity was simultaneously recorded in the ipsilateral hippocampus of adult mice. Intrahippocampal bumetanide treatment reduced epileptiform activity compared to vehicle-treated mice (Fig. 3A,B). Similar to earlier observations with systemic administration of bumetanide (Fig. 1), intrahippocampal bumetanide resulted in a significant increase in the latency to the onset of epileptiform activity (vehicle: 1.9 ± 0.3 min; bumetanide: 21.9 ± 11.4 min, P = 0.03), a reduction in duration of epileptiform events (vehicle: 5.1 ± 0.7 min; bumetanide: 2.8 ± 0.6 min), and a decrease in the total time exhibiting epileptiform activity (vehicle: 86.9 ± 3.6%; bumetanide: 43.8 ± 10.3%, n = 5 mice per experimental group, P = 0.009). These data provide evidence that direct bumetanide application to the hippocampus can effectively decrease seizures similar to systemic (i.p.) administration in adult mice.

Figure 3.

Figure 3

Open in a new tab

Intrahippocampal bumetanide effectively decreases KA-induced epileptiform activity in vivo. (A) Electrographic epileptiform activity recorded from the ipsilateral hippocampus of vehicle- (left, black traces) or bumetanide- (right, red traces) treated mice, following direct intrahippocampal administration of KA. The traces represent continuous EEG recordings from intrahippocampal-treated vehicle or bumetanide mice. (B) Histograms showing the latency (min), average epileptiform duration (min), and epileptiform activity (%) of KA-induced epileptiform activity in vehicle (black) and bumetanide (red) treated mice. Values ± SEM;n = 5 mice per experimental group; *p < 0.05, **p < 0.01. Epilepsia © ILAE

Bumetanide decreases pilocarpine-induced epileptiform activity in adult mice in vivo

To determine whether the anticonvulsant effects of bumetanide are model specific, we assessed the effect of bumetanide on pilocarpine-induced SE in adult mice using in vivo EEG recording. Bumetanide treatment (0.2 mg/kg, i.p.) 30 min prior to pilocarpine administration decreased the duration of epileptiform activity (39.1 ± 9.5 min) and percent time exhibiting epileptiform activity (20.7 ± 5.2%) following pilocarpine treatment compared to vehicle-treated mice (duration: 80.5 ± 8.3 min; percent time: 42.8 ± 4.9%, Fig. S3, n = 4 mice per experimental group, P = 0.008 & P = 0.011, respectively). These data suggest that the anticonvulsant effects of bumetanide are not model specific.

Bumetanide decreases SLEs in acute slices in vitro

We used the 0-Mg2+ model to evaluate whether the ability of bumetanide to decrease seizures in vivo can be reproduced in vitro. Bumetanide (10 μm) decreased the duration of seizure-like events (SLEs) (17.1 ± 5.1 s) and increased the inter-SLE interval (ISI) (447.2 ± 176.1 s) compared to vehicle-treated, 0-Mg 2+controls (SLE duration: 30.4 ± 5.9 s, ISI: 286.9 ± 98.4 s, n = 9 slices, six mice per experimental group, P = 0.03, Fig. 4A,B). Similar to the results obtained in vivo, these data demonstrate the ability of bumetanide to reduce the progression of epileptiform activity in vitro and support central actions of bumetanide.

Figure 4.

Figure 4

Open in a new tab

Seizure-like events (SLEs) recorded using the in vitro 0-Mg2+ model are suppressed with bumetanide. (A) Representative field potential recording of 0-Mg2+-induced SLEs (top trace) recorded from layer III of the medial entorhinal cortex in acute slices, before (0-Mg2+) and after treatment with bumetanide (Bume 10 μm). The drug administration time is indicated accordingly. The parts of the field potential recordings with brief episodes of SLEs labeled with a and b, corresponding to before (0-Mg2+, a) and after treatment with bumetanide (10 μm, b), respectively, are shown on an expanded time scale (bottom traces). (B) Scatter plot of SLE duration (left) and inter-SLE-interval (ISI, right) in slices treated with 0-Mg2+ alone (black circles) or 0-Mg2++bumetanide (red circles). Individual slices are represented with dotted lines and the averaged changes are represented with solid lines. Values ± SEM; *p < 0.05. Epilepsia © ILAE

Bumetanide prevents development of pharmacoresistance in vivo and in vitro

Within an hour following KA administration (20 mg/kg, i.p.), vehicle-treated animals develop SE, characterized as persistent, uninterrupted electrographic seizures that lasts on average 58.7 ± 4.6 min until the recording was terminated. Diazepam (20 mg/kg) was administered 1 h after KA treatment. Following entry into SE, we observe decreased efficacy of diazepam (20 mg/kg) to suppress seizures, as evident from persistence of ictal activity toward the end of the recording period (Fig. 5A, gray trace and inset d, top trace) similar to what we observed in vehicle-treated mice. Diazepam does not significantly alter the duration (45.9 ± 31.6 min) or percent time exhibiting ictal activity during the final 150–300 min of the recording (41.9 ± 3.1%) compared to mice that received only vehicle (duration: 30.4 ± 2.6 min and % ictal activity: 55.5 ± 6.6%), demonstrating the pharmacoresistance in this model. Of interest, when bumetanide (0.2 mg/kg, i.p.) is administered in combination with diazepam 1 h following KA treatment, the duration (4.0 ± 2.9 min) and percent time exhibiting ictal activity (2.0 ± 1.2%) are decreased compared to vehicle (see also Fig. S4, duration: 30.4 ± 2.6 min and % ictal activity: 55.5 ± 6.6%), bumetanide alone (see also Fig. S4, duration: 32.3 ± 12.2 min; % ictal activity: 9.8 ± 3.1%), or diazepam alone (duration: 45.9 ± 31.6 min; % ictal activity: 41.9 ± 3.1%) (n = 3–7 mice per experimental group, ANOVA, p < 0.001 and p < 0.05, F18 = 96.2, Fig. 5B, see also, Fig. S4). Furthermore, bumetanide treatment together with diazepam preserved low frequency activity compared to diazepam alone. Mice treated with diazepam alone show a progressive loss of low-frequency activity (9.5 ± 9.4 events per minute, Fig. S5) concomitant with the development of increased ictal activity (Fig. 5B); whereas, bumetanide (0.2 mg/kg, i.p.) treatment in combination with diazepam maintains the low-frequency activity (37.6 ± 12.3 events per minute, Fig. S5), and prevents the development of ictal activity (Fig. 5B). These data demonstrate the effectiveness of bumetanide in decreasing seizure activity when administered 1 h after seizures have been initiated with KA (Fig. 5B), although bumetanide in combination with diazepam is even more effective (Fig. 5B).

Similarly, we found that bumetanide (10 μm) enhanced the efficacy of diazepam (5 μm) to suppress SLEs in vitro (Fig. 6A-C). We confirmed that diazepam alone did not prevent the progressive increase in the duration of SLEs (28.4 ± 10.3 s) or decrease in the inter-SLE interval (ISI) (168.3 ± 75.6 s) of SLEs compared to controls (SLE duration: 24.4 ± 8.3 s, ISI: 325.3 ± 148.2 s), indicating progression of epileptiform activity despite diazepam treatment (n = 7 slices, six mice per experimental group, p = 0.1, Fig. 6A,C). Bumetanide treatment in combination with diazepam decreased the duration (31.7 ± 7.7 s vs. 0-Mg2+ controls: 44.1 ± 7.8 s, p = 0.015) and increased the ISI (698.1 ± 147.1 s vs. 0-Mg2+ controls: 469.8 ± 113.1 s, n = 14 slices, nine mice per experimental group, p = 0.013, Fig. 6C) of SLEs. These data suggest that bumetanide in combination with diazepam decreases seizure progression in vitro, which may reach a floor effect due to the robust reduction of SLEs in both bumetanide and bumetanide + diazepam treatment groups.

Figure 6.

Figure 6

Open in a new tab

Bumetanide restores the efficacy of diazepam in vitro. (A, B) Representative field potential recordings of 0-Mg2+-induced SLEs recorded from layer III of the medial entorhinal cortex in slices treated with diazepam (DZP, 5 μm) in the absence (A) or presence (B) of bumetanide (Bume, 10 μm). Traces are labeled indicating drug administration times accordingly. The parts of the field potential recordings with brief episodes of SLEs (a-d), corresponding to before (0-Mg2+, a and c) and after treatment with diazepam alone (DZP 5 μm, b) or with diazepam in the presence of bumetanide (DZP 5 μm + Bume, 10 μm, d), are shown on an expanded time scale (bottom traces). (C) Scatter plot of 0-Mg2+-inducedSLE duration (left) and inter-SLE-interval (ISI, right) in slices treated with diazepam (DZP) alone (gray circles) or diazepam (DZP) + bumetanide (blue circles). Individual slices are represented with dotted lines and the averaged changes are represented with solid lines. Values ± SEM; *P < 0.05. Epilepsia © ILAE

Altered chloride cotransporter expression following KA-induced seizures

To determine whether seizure progression and the development of status epilepticus is associated with alterations in chloride homeostasis due to alterations in the expression of NKCC1 or KCC2, we performed Western blot analysis on the total protein isolated from the hippocampus 2 h following seizures induced with KA in vehicle or bumetanide (0.2 mg/kg, i.p.) treated mice. We observe a decrease in total KCC2 protein levels in both vehicle (104.3 ± 13.5 O.D. units/50 μg total protein) and bumetanide-treated mice (115.4 ± 7.4 O.D. units/50 μg total protein) 2 h following KA-induced seizures compared to control, seizure-free mice (132.0 ± 3.7 O.D. units/50 μg total protein) (Fig. S2, n = 9–12 mice per experimental group; p < 0.001; F2,28 = 11.19). In contrast, there is a significant increase in NKCC1 expression following KA-induced seizures in the hippocampus of both vehicle (112.5 ± 15.6 O.D. units/50 μg total protein) and bumetanide-treated mice (118.9 ± 21.4 O.D. units/50 μg total protein) compared to seizure-free controls (65.3 ± 10.7 O.D. units/50 μg total protein) (Fig. S2, n = 10–12 mice per experimental group; p = 0.01;F2,30 = 5.29).

The activity and surface expression of KCC2 is regulated by phosphorylation at residue Ser940.26 Western Blot analysis of total protein 2 h following seizures induced with KA reveals a decrease in P-KCC2 in both vehicle (87.3 ± 8.8 O.D. units/50 μg total protein) and bumetanide-treated mice (71.7 ± 6.4 O.D. units/50 μg total protein) compared to control, seizure-free mice (128.9 ± 14.1 O.D. units/50 μg total protein) (Fig. S2, n = 5 mice per experimental group; p < 0.01).

Discussion

The data presented in this manuscript are consistent with a role for compromised GABAergic inhibition during seizure progression in the development of pharmacoresistant SE, which has traditionally been thought to be caused by alterations in GABAergic signaling as a result of decreased expression of benzodiazepine-sensitive GABAARs.23,27,28 Although the anticonvulsant effects of bumetanide were previously thought to be unique to the immature brain,4,5 herein we demonstrate that bumetanide exerts anticonvulsant actions even in the mature brain and prevents the development of pharmacoresistant SE.

The anticonvulsant effects of bumetanide in the adult have been controversial. Recently the efficacy of bumetanide in suppressing seizures in vivo has been questioned (reviewed in Loscher et al. 29). The limited ability of bumetanide to cross the blood-brain barrier and its short half-life have cast doubts30 on the seizure suppressing effects of bumetanide.4,31 Despite the limitations in the bioavailability of bumetanide, anticonvulsant effects have been demonstrated in numerous seizure models, including maximal electroshock seizures (MES), pentylenetetrazole (PTZ), KA, and kindling, in mice, rats, and nonhuman primates (for review see Loscher et al.29 and Puskarjov et al.32). In addition, bumetanide has been shown to exert anticonvulsant actions against neonatal seizures in humans,7 enhance the anticonvulsant actions of phenobarbital in animal models, and to exert “disease-modifying effects” against pilocarpine-induced seizures.34 However, recent studies have suggested that bumetanide alone is incapable of blocking SE,35 which may be due to differences in methodology including species differences, inconsistent timing of bumetanide administration, the use of the prodrug (BUM5), the use of anesthesia in some experiments, and methods used to measure epileptiform activity. In this study, we relied solely on electrographically recorded seizures in vivo in the KA model, an accepted model for studying SE,14 whereas, previous studies relied on behavioral observations, which is not a reliable method for evaluating the efficacy of anticonvulsant treatments or for studying pharmacoresistance.36 Despite the differences in the experimental approaches, the majority of studies find anticonvulsant effects of bumetanide (for review see Loscher et al.29 and Puskarjov et al.32), and the present study is an important confirmation of these findings, demonstrating reproducibility in multiple models of epilepsy in numerous laboratories, which is essential to validate these studies.

In light of the limited brain penetration of bumetanide, it is curious to note that numerous labs have observed anticonvulsant effects of bumetanide. One explanation may be that seizures are known to cause a breakdown in the blood–brain barrier,37,38 which may facilitate the central actions of bumetanide. Here we demonstrate equivalent anticonvulsant actions of both 0.2 and 2 mg/kg bumetanide in vivo. This lack of dose response may be due to the fact that the concentration of bumetanide gaining access to the brain falls on the shallow part of the dose–response curve, presenting little difference in efficacy. The limited brain penetration of bumetanide has also spurred the idea that the anticonvulsant actions of bumetanide may be mediated by effects in the periphery or in brain regions with a fenestrated blood–brain barrier, such as the hypothalamus.32 However, our data suggest that the anticonvulsant effects are mediated by central actions of bumetanide. Bumetanide is a known diuretic and seizures can cause brain swelling (i.e., uptake of fluid from the blood). In theory, it is plausible that diuresis could reduce swelling and alleviate pressure on the brain. However, systemic (i.p.) administration of a specific inhibitor of renal Na+-Cl- cotransporter, chlorothiazide (CTZ), a diuretic which acts in periphery but lacks central effects, did not exert seizure suppressing effects, arguing against a role for diuresis in the anticonvulsant effects of bumetanide. Furthermore, we entertained the idea that bumetanide may alter seizure induction in response to intraperitoneal administration of KA by altering the excretion and/or metabolism of KA. However, we demonstrate the anticonvulsant effects of locally applied (intrahippocampal) bumetanide on seizures induced by intrahippocampal administration of KA. Furthermore, the ability of bumetanide to decrease epileptiform activity in vitro also suggests central actions of bumetanide in seizure reduction.

Given that our data points to central actions of bumetanide, the limited ability of bumetanide to cross the blood– brain barrier greatly diminishes its therapeutic potential. To overcome this limitation, there have been efforts to produce prodrugs of bumetanide with better brain penetration and decreased diuretic effects,30 or to increase brain concentrations of bumetanide using an anion transport inhibitor.39 One of the prodrugs has shown promise in blunting seizure susceptibility in chronically epileptic mice and rats and enhances the anticonvulsant actions of phenobarbital.30 The data presented here demonstrate a modest effect of bumetanide on epileptiform activity and suggest that a prodrug of bumetanide, with better bioavailability, would likely have more robust anticonvulsant effects. Although the exact mechanism mediating the central anticonvulsant actions of bumetanide or the prodrug of bumetanide is unknown, it is likely due to their effect on chloride homeostasis and GABAergic inhibition.

KA-induced seizures result in an initial interictal barrage resulting in GABA spillover,40 which could induce Cl- loading during or after the ictal events.41 In fact, chloride loading associated with epileptiform activity has been demonstrated in vitro.6,9 Recently, it was demonstrated that a collapse in chloride gradient and excitatory actions of GABA drive epileptiform afterdischarges both in vitro and in vivo.12 Similarly, driving interneurons using optogenetic stimulation induced ictal discharges in the presence of 4-aminopyridine (4-AP) in vitro.13 In cultured primary neurons, a glutamate challenge results in an intracellular chloride load and collapse of hyperpolarizing GABAergic responses42 and in reduced efficacy of diazepam.3 Together these data suggest that intracellular chloride accumulation during seizures reduce the strength of GABAergic inhibition and the efficacy of GABAA receptor modulators, which is consistent with a lack of diazepam sensitivity in mice lacking KCC2.10 Of interest, the spread of epileptiform activity occurs simultaneously with the erosion of inhibition in vitro,11 suggesting that the loss of GABAergic inhibition plays a critical role in the progression of epileptiform activity. However, the mechanisms underlying the loss of inhibition associated with epileptiform activity remain unclear. Our data suggest that the loss of efficacy of GABAergic agonists during seizure progression is associated with a decrease in KCC2 expression and an increase in NKCC1 expression in the hippocampus, similar to changes reported in the pilocarpine model of SE,43 although the methods employed in this study cannot discriminate between changes in NKCC1 expression on neurons versus glial cells. These data suggest that chloride loading during seizure progression may underlie compromised GABAergic inhibition. Given that our in vivo data corroborate the in vitro findings in this study, our observations are consistent with the erosion of inhibition associated with epileptiform activity, which can be blocked, or at least delayed, with bumetanide treatment.

Herein we demonstrate that even in the mature brain, regulating chloride homeostasis by targeting NKCC1 can exert anticonvulsant effects. Bumetanide likely mediates its anticonvulsant action by pharmacologically blocking the increased NKCC1 expression associated with seizure progression, thereby preventing intracellular chloride accumulation and the erosion of GABAergic inhibition. The data provide evidence for controlling chloride homeostasis for the treatment of epilepsy in combination with benzodiazepines to combat pharmacoresistant SE. This study takes the first steps in determining the therapeutic potential of bumetanide for combinatorial treatment of SE, which will undoubtedly require additional study.

Key Points.

  • Bumetanide exerts anticonvulsant effects in the adult

  • Central actions of bumetanide mediate the anticonvulsant effects

  • Bumetanide restores the anticonvulsant efficacy of diazepam

Acknowledgments

The authors would like to thank Dr. Steve Moss and members of his laboratory, in particular Dr. Tarek Deeb, for many thoughtful discussions relevant to this manuscript. J.M. was supported by NS073574 and a Research Grant from the Epilepsy Foundation. S.S. was supported by the Academy of Finland.

Biography

graphic file with name nihms866021b1.gif

Sudhir Sivakumaran is a postdoctoral fellow in Dr. Jamie Maguire's lab at Tufts University School of Medicine.

Footnotes

Disclosure: The authors declare that they have no conflict of interest. We confirm that we have read the Journal's position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.

Supporting Information: Additional Supporting Information may be found in the online version of this article:

Figure S1. Bumetanide reverses KA-induced increases in electroencephalographic (EEG) power.

Figure S2. Alterations in the expression of chloride cotransporters following KA-induced seizures.

Figure S3. Bumetanide reduces pilocarpine-induced seizures.

Figure S4. Bumetanide is effective in reducing seizure activity after the development of KA-induced seizures.

Figure S5. Bumetanide enhances the ability of diazepam in maintaining low-frequency activity compared to diazepam alone.

Data S1. Supplementary methods.

References

  • 1.Lowenstein DH. Status epilepticus: an overview of the clinical problem. Epilepsia. 1999;40(Suppl 1):S3–S8. doi: 10.1111/j.1528-1157.1999.tb00872.x. [DOI] [PubMed] [Google Scholar]
  • 2.Goodkin HP, Kapur J. The impact of diazepam's discovery on the treatment and understanding of status epilepticus. Epilepsia. 2009;50:2011–2018. doi: 10.1111/j.1528-1167.2009.02257.x. [DOI] [PubMed] [Google Scholar]
  • 3.Deeb TZ, Nakamura Y, Frost GD, et al. Disrupted Cl(—) homeostasis contributes to reductions in the inhibitory efficacy of diazepam during hyperexcited states. Eur J Neurosci. 2013;38:2453–2467. doi: 10.1111/ejn.12241. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Dzhala VI, Talos DM, Sdrulla DA, et al. NKCC1 transporter facilitates seizures in the developing brain. Nat Med. 2005;11:1205–1213. doi: 10.1038/nm1301. [DOI] [PubMed] [Google Scholar]
  • 5.Dzhala VI, Brumback AC, Staley KJ. Bumetanide enhances phenobarbital efficacy in a neonatal seizure model. Ann Neurol. 2008;63:222–235. doi: 10.1002/ana.21229. [DOI] [PubMed] [Google Scholar]
  • 6.Dzhala VI, Kuchibhotla KV, Glykys JC, et al. Progressive NKCC1-dependent neuronal chloride accumulation during neonatal seizures. J Neurosci. 2010;30:11745–11761. doi: 10.1523/JNEUROSCI.1769-10.2010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Kahle KT, Staley KJ. Neonatal seizures and neuronal transmembrane ion transport. In: Noebels JL, Avoli M, Rogawski MA, Olsen RW, Delgado-Escueta AV, editors. Jasper's basic mechanisms of the epilepsies. 4th. 2012. p. 1066. [PubMed] [Google Scholar]
  • 8.Li X, Zhou J, Chen Z, et al. Long-term expressional changes of Na+ -K + -Cl— co-transporter 1 (NKCC1) and K + -Cl— co-transporter 2 (KCC2) in CA1 region of hippocampus following lithium-pilocarpine induced status epilepticus (PISE) Brain Res. 2008;1221:141–146. doi: 10.1016/j.brainres.2008.04.047. [DOI] [PubMed] [Google Scholar]
  • 9.Kapur J, Coulter DA. Experimental status epilepticus alters gamma-aminobutyric acid type A receptor function in CA1 pyramidal neurons. Ann Neurol. 1995;38:893–900. doi: 10.1002/ana.410380609. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Tornberg J, Segerstrale M, Kulesskaya N, et al. KCC2-deficient mice show reduced sensitivity to diazepam, but normal alcohol-induced motor impairment, gaboxadol-induced sedation, and neurosteroidinduced hypnosis. Neuropsychopharmacology. 2007;32:911–918. doi: 10.1038/sj.npp.1301195. [DOI] [PubMed] [Google Scholar]
  • 11.Trevelyan AJ, Sussillo D, Yuste R. Feedforward inhibition contributes to the control of epileptiform propagation speed. J Neurosci. 2007;27:3383–3387. doi: 10.1523/JNEUROSCI.0145-07.2007. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Ellender TJ, Raimondo JV, Irkle A, et al. Excitatory effects of parvalbumin-expressing interneurons maintain hippocampal epileptiform activity via synchronous afterdischarges. J Neurosci. 2014;34:15208–15222. doi: 10.1523/JNEUROSCI.1747-14.2014. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Yekhlef L, Breschi GL, Lagostena L, et al. Selective activation of parvalbumin- or somatostatin-expressing interneurons triggers epileptic seizurelike activity in mouse medial entorhinal cortex. J Neurophysiol. 2015;113:1616–1630. doi: 10.1152/jn.00841.2014. [DOI] [PubMed] [Google Scholar]
  • 14.Reddy DS, Kuruba R. Experimental models of status epilepticus and neuronal injury for evaluation of therapeutic interventions. Int J Mol Sci. 2013;14:18284–18318. doi: 10.3390/ijms140918284. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.O'Toole KK, Hooper A, Wakefield S, et al. Seizure-induced disinhibition of the HPA axis increases seizure susceptibility. Epilepsy Res. 2014;108:29–43. doi: 10.1016/j.eplepsyres.2013.10.013. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Lee V, Maguire J. Impact of inhibitory constraint of interneurons on neuronal excitability. J Neurophysiol. 2013;110:2520–2535. doi: 10.1152/jn.00047.2013. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Maguire JL, Stell BM, Rafizadeh M, et al. Ovarian cycle-linked changes in GABA(A) receptors mediating tonic inhibition alter seizure susceptibility and anxiety. Nat Neurosci. 2005;8:797–804. doi: 10.1038/nn1469. [DOI] [PubMed] [Google Scholar]
  • 18.Krook-Magnuson E, Armstrong C, Oijala M, et al. On-demand optogenetic control of spontaneous seizures in temporal lobe epilepsy. Nat Commun. 2013;4:1376. doi: 10.1038/ncomms2376. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Rieg T, Bundey RA, Chen Y, et al. Mice lacking P2Y2 receptors have salt-resistant hypertension and facilitated renal Na+ and water reabsorption. FASEB J. 2007;21:3717–3726. doi: 10.1096/fj.07-8807com. [DOI] [PubMed] [Google Scholar]
  • 20.Gamba G, Miyanoshita A, Lombardi M, et al. Molecular cloning, primary structure, and characterization of two members of the mammalian electroneutral sodium-(potassium)-chloride cotransporter family expressed in kidney. J Biol Chem. 1994;269:17713–17722. [PubMed] [Google Scholar]
  • 21.Doniger S, Hofmann T, Yeh J. Predicting CNS permeability of drug molecules: comparison of neural network and support vector machine algorithms. J Comput Biol. 2002;9:849–864. doi: 10.1089/10665270260518317. [DOI] [PubMed] [Google Scholar]
  • 22.Castro OW, Santos VR, Pun RYK, et al. Impact of corticosterone treatment on spontaneous seizure frequency and epileptiform activity in mice with chronic epilepsy. PLoS ONE. 2012;7:e46044. doi: 10.1371/journal.pone.0046044. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Wahab A, Albus K, Gabriel S, et al. In search of models of pharmacoresistant epilepsy. Epilepsia. 2010;51:154–159. doi: 10.1111/j.1528-1167.2010.02632.x. [DOI] [PubMed] [Google Scholar]
  • 24.Maguire J, Ferando I, Simonsen C, et al. Excitability changes related to GABAA receptor plasticity during pregnancy. J Neurosci. 2009;29:9592–9601. doi: 10.1523/JNEUROSCI.2162-09.2009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Sarkar J, Wakefield S, Mackenzie G, et al. Neurosteroidogenesis is required for the physiological response to stress: role of neurosteroid-sensitive GABAA receptors. J Neurosci. 2011;31:18198–18210. doi: 10.1523/JNEUROSCI.2560-11.2011. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Lee HH, Walker JA, Williams JR, et al. Direct protein kinase C-dependent phosphorylation regulates the cell surface stability and activity of the potassium chloride cotransporter KCC2. J Biol Chem. 2007;282:29777–29784. doi: 10.1074/jbc.M705053200. [DOI] [PubMed] [Google Scholar]
  • 27.Naylor DE, Liu H, Wasterlain CG. Trafficking of GABA(A) receptors, loss of inhibition, and a mechanism for pharmacoresistance in status epilepticus. J Neurosci. 2005;25:7724–7733. doi: 10.1523/JNEUROSCI.4944-04.2005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Deeb TZ, Maguire J, Moss SJ. Possible alterations in GABAA receptor signaling that underlie benzodiazepine-resistant seizures. Epilepsia. 2012;53:79–88. doi: 10.1111/epi.12037. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Loscher W, Puskarjov M, Kaila K. Cation-chloride cotransporters NKCC1 and KCC2 as potential targets for novel antiepileptic and antiepileptogenic treatments. Neuropharmacology. 2013;69:62–74. doi: 10.1016/j.neuropharm.2012.05.045. [DOI] [PubMed] [Google Scholar]
  • 30.Tollner K, Brandt C, Topfer M, et al. A novel prodrug-based strategy to increase effects of bumetanide in epilepsy. Ann Neurol. 2014;75:550–562. doi: 10.1002/ana.24124. [DOI] [PubMed] [Google Scholar]
  • 31.Hochman DW. The extracellular space and epileptic activity in the adult brain: explaining the antiepileptic effects of furosemide and bumetanide. Epilepsia. 2012;53:18–25. doi: 10.1111/j.1528-1167.2012.03471.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Puskarjov M, Kahle KT, Ruusuvuori E, et al. Pharmacotherapeutic targeting of cation-chloride cotransporters in neonatal seizures. Epilepsia. 2014;55:806–818. doi: 10.1111/epi.12620. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Cleary RT, Sun H, Huynh T, et al. Bumetanide enhances phenobarbital efficacy in a rat model of hypoxic neonatal seizures. PLoS ONE. 2013;8:e57148. doi: 10.1371/journal.pone.0057148. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Brandt C, Nozadze M, Heuchert N, et al. Disease-modifying effects of phenobarbital and the NKCC1 inhibitor bumetanide in the pilocarpine model of temporal lobe epilepsy. J Neurosci. 2010;30:8602–8612. doi: 10.1523/JNEUROSCI.0633-10.2010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Tollner K, Brandt C, Erker T, et al. Bumetanide is not capable of terminating status epilepticus but enhances phenobarbital efficacy in different rat models. Eur J Pharmacol. 2015;746:78–88. doi: 10.1016/j.ejphar.2014.10.056. [DOI] [PubMed] [Google Scholar]
  • 36.Vermoesen K, Smolders I, Massie A, et al. The control of kainic acid-induced status epilepticus. Epilepsy Res. 2010;90:164–166. doi: 10.1016/j.eplepsyres.2010.04.001. [DOI] [PubMed] [Google Scholar]
  • 37.Salar S, Maslarova A, Lippmann K, et al. Blood–brain barrier dysfunction can contribute to pharmacoresistance of seizures. Epilepsia. 2014;55:1255–1263. doi: 10.1111/epi.12713. [DOI] [PubMed] [Google Scholar]
  • 38.Pont F, Collet A, Lallement G. Early and transient increase of rat hippocampal blood–brain barrier permeability to amino acids during kainic acid-induced seizures. Neurosci Lett. 1995;184:52–54. doi: 10.1016/0304-3940(94)11166-g. [DOI] [PubMed] [Google Scholar]
  • 39.Tollner K, Brandt C, Romermann K, et al. The organic anion transport inhibitor probenecid increases brain concentrations of the NKCC1 inhibitor bumetanide. Eur J Pharmacol. 2015;746:167–173. doi: 10.1016/j.ejphar.2014.11.019. [DOI] [PubMed] [Google Scholar]
  • 40.Huberfeld G, Wittner L, Clemenceau S, et al. Perturbed chloride homeostasis and GABAergic signaling in human temporal lobe epilepsy. J Neurosci. 2007;27:9866–9873. doi: 10.1523/JNEUROSCI.2761-07.2007. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Miles R, Blaesse P, Huberfeld G, et al. Chloride homeostasis and GABA signaling in temporal lobe epilepsy. In: Noebels JL, Avoli M, Rogawski MA, Olsen RW, Delgado-Escueta AV, editors. Jasper's basic mechanisms of the epilepsies. 4th. 2012. pp. 581–590. [PubMed] [Google Scholar]
  • 42.Kitamura A, Ishibashi H, Watanabe M, et al. Sustained depolarizing shift of the GABA reversal potential by glutamate receptor activation in hippocampal neurons. Neurosci Res. 2008;62:270–277. doi: 10.1016/j.neures.2008.09.002. [DOI] [PubMed] [Google Scholar]
  • 43.Eftekhari S, Mehrabi S, Soleimani M, et al. BDNF modifies hippocampal KCC2 and NKCC1 expression in a temporal lobe epilepsy model. Acta Neurobiol Exp (Wars) 2014;74:276–287. doi: 10.55782/ane-2014-1993. [DOI] [PubMed] [Google Scholar]

RESOURCES