Neuronal carbonic anhydrase VII provides GABAergic excitatory drive to exacerbate febrile seizures

Brain carbonic anhydrases (CAs) are known to modulate neuronal signalling. Using a novel CA VII (Car7) knockout (KO) mouse as well as a CA II (Car2) KO and a CA II/VII double KO, we show that mature hippocampal pyramidal neurons are endowed with two cytosolic isoforms. CA VII is predominantly expressed by neurons starting around postnatal day 10 (P10). The ubiquitous isoform II is expressed in neurons at P20. Both isoforms enhance bicarbonate-driven GABAergic excitation during intense GABAA-receptor activation. P13–14 CA VII KO mice show behavioural manifestations atypical of experimental febrile seizures (eFS) and a complete absence of electrographic seizures. A low dose of diazepam promotes eFS in P13–P14 rat pups, whereas seizures are blocked at higher concentrations that suppress breathing. Thus, the respiratory alkalosis-dependent eFS are exacerbated by GABAergic excitation. We found that CA VII mRNA is expressed in the human cerebral cortex before the age when febrile seizures (FS) occur in children. Our data indicate that CA VII is a key molecule in age-dependent neuronal pH regulation with consequent effects on generation of FS.

Western analysis of muscle and liver tissue. For Western blotting, 10 µg protein lysates from hippocampus, skeletal muscle and liver were separated on reducing 15 % SDSpolyacrylamide gels and blotted onto a nitrocellulose membrane. Blots were probed with our rabbit CA VII antibody at a dilution of 1:250. Detection was done with Lasuser 3000 (Fujifilm) using chemoluminescence ECL-Kit (Pierce).
Detection of intrapyramidal carbonic anhydrase activity. In order to detect cytosolic CA activity, we exposed slices to nominally CO 2 /HCO 3 free HEPES solution for 100-200 s in the absence and presence of the membrane-permeant CA inhibitor acetazolamide (AZ 100 M, Sigma-Aldrich, St.Louis, MO) and quantified the AZ induced suppression of the maximum rate of intracellular alkalinization (dpH i /dt max ). Since AZ is known to inhibit both extra-and intracellular CA activity (Chesler, 2003), the control responses were done in the presence of the poorly permeant CA inhibitor benzolamide (BA, 10 µM) applied for 5 minutes before withdrawal of CO 2 /HCO 3 to exclude effects mediated by inhibition of extracellular CAs. BA was a kind gift from Dr. E.R. Swenson (University of Washington, Seattle, WA).
pH i was measured using the H + -sensitive indicator BCECF. For the dye loading, brain slices were transferred to the recording chamber and perfused with standard solution.
BCECF-AM (Invitrogen) was pressure-injected into the extracellular space of CA1 pyramidal layer using a conventional patch pipette, filled with 10 M BCECF-AM in standard solution.
In each slice, somatic pH i was measured from 3-5 CA1 pyramidal neurons (see Inset in Figure   2A). Both loading and pH i measurements were done in the continuous presence of TTX (1 µM).
Live-cell imaging was performed with an upright Zeiss Axio Scope II (60x water immersion objective n.a. 0.9). BCECF was excited with an OptoLED double wavelength light source (436 ± 10 nm and 500 ± 10 nm, Cairn, UK). Emitted light passed through a 535 ± 12.5 nm band pass filter and was registered with a PCO SensiCam CCD camera. The diameter of the region of interest (ROI) was about 3x5 µm, and the ROI was positioned at the site of the highest local signal. Ratiometric images were captured (at 0.1-0.2 Hz) and analyzed with WinFluor software (courtesy of Dr. John Dempster, University of Strathclyde, Glasgow, UK).
For analysis the original pH-trace was smoothed with the Savitzky-Golay algorithm (20 data points frame, 2 nd order polynomial) (Savitzky and Golay, 1964)  where j indicates the number of sample.
Electrophysiological recordings. Whole-cell patch-clamp recordings were performed from visually identified CA1 pyramidal cells. The glass micropipettes had a resistance of 4.5-6.0 when filled with intracellular solution consisting of (in mM) 124 K-gluconate, 5 KCl, 2 NaOH, 10 HEPES, 0.5 CaCl 2 , 5 EGTA, 2 MgATP (pH adjusted to 7.3 with KOH). Recordings were made using an EPC 10 amplifier (HEKA Elektronik, Lambrecht/Pfalz, Germany) in the voltage-clamp or current-clamp mode and corrected for a calculated 15.0 mV liquid junction potential for whole-cell recordings (Barry, 1994). For field recordings, 4.5-6.0 M glass microelectrodes were filled with the standard solution and placed in the stratum pyramidale. The signal was high-pass filtered at 10 Hz. GABA A receptor-mediated synaptic responses were evoked in CA1 pyramidal neurons using a monopolar glass microelectrode filled with standard solution (resistance 4.5-6.0 M , see Fig. 3) or, for a more intense GABA A R activation, with a bipolar metal electrode, placed in the border of stratum radiatum (sr) and stratum lacunosum-moleculare (slm) (Alger and Nicoll, 1982;Staley et al., 1995;Kaila et al., 1997). The intensity of high-frequency stimulation (HFS; 40 pulses at 100 Hz, given every 5 minutes) applied via the bipolar electrode (Kaila et al., 1997)

EEG measurements and FS induction.
Surgery and EEG recordings. On the day of experiments, typically two P13-14 animals from the same litter were studied. At this age CA VII KO mice were of comparable weight as their WT littermates (6.9 ± 0.2 g n = 11 vs. 6.4 ± 0.2 g, n = 9, respectively, P = 0.14, Student's ttest). For cortical EEG recordings ball-tipped Teflon-insulated silver wire electrodes (75 µm wire diameter, ball tip diameter ca. 1 mm; Advent Research Materials Ltd, Oxford, UK) were connected to microconnectors (Microtech Inc., Boothwyn, PA, USA) and fixed with dental acrylic. Electrodes were implanted under isoflurane anesthesia (maintained at 2-2.5 %).
Small craniotomies were made using a drill equipped with a 0.6 mm diameter carbide burr and electrodes were placed epidurally above right parietal cortex (AP 2.5 mm; ML 1.3 mm).
A reference electrode was placed above cerebellum. The incision was sutured using 6-0 nylon monofilament. After surgery, mice recovered for two hours in a warm environment Seizure induction. Seizures were induced using a heated chamber as described previously for neonatal rats (Schuchmann et al., 2006;2008). In parallel with the EEG and video recordings we monitored rectal temperature with a mouse rectal probe (BAT-10; Physitemp, Science Products GmbH, Hofheim, Germany) and breath rate with a piezo crystal sensor placed on the abdomen (Pico Movement Sensor; Temec, Kerkrade, Netherlands) (Schuchmann et al., 2008). After baseline recording at 30-34 o C for 10 minutes, pups were moved to a preheated chamber kept at 43 ± 1 o C for mice and at 48 ± 1 o C when rat pups were used. Animals were exposed to hyperthermia until their body temperature reached a critical high value of 42.5-43 o C (Dubé and Baram, 2006), after which the experiment was terminated and animals decapitated.
To test the effect of CO 2 on seizure generation, the heated chamber was supplied with a constant flow of pre-heated (

Detection of CA VII in the developing human neorcortex and hippocampus.
Exon array analysis was performed as previously described (Kang et al., 2011). Briefly, total RNA was isolated from specific brain regions of neocortex and hippocampus using a nonphenolic procedure (RNeasy Plus Mini Kit, Qiagen), followed by DNase treatment (TURBO DNase, Ambion). Optical density values at 260/280 were consistently above 1.9 (NanoDrop, Thermo Scientific), and the values of RNA integrity were selected above 5 (RIN>5, Agilent Bioanalyzer). Synthesized cDNA (5.5 µg) using WT Expression kit (Ambion) was labeled and loaded onto individual Affymetrix Human Exon 1.0 ST arrays. Microarrays were hybridized at 45 °C for 16-24 hours, washed and stained using an Affymetrix FS450 fluidics station, according to manufacturer recommendations. Microarrays were scanned on a GeneChip Scanner 3000 and visually inspected for hybridization artifacts. The raw image files (.DAT files) were analysed using Affymetrix GeneChip Operating Software to generate .CEL files.
Droplet digital PCR. An aliquot of the total RNA that was previously extracted from each brain region was used for secondary validation using droplet digital PCR analysis.
One ug of total RNA was used for cDNA synthesis using oligo dT primers and SuperScript III First-strand synthesis Supermix (Invitrogen), and subsequently diluted with nuclease-free water to 1 ng/ul cDNA. Gene-specific high-melt temperature primers and FAM-probes for