All animals were handled and experiments were conducted in accordance with procedures approved by the Institutional Animal Care and Use Committee at the University of California, Irvine. In this study, we used a newer AlstR mouse line conditionally expressing AlstR and green fluorescent protein (GFP) following Cre recombination. This AlstR mouse line (R26-AlstR) was created by using the same targeting vector of transgene as the AlstR 192 strain (Gosgnach et al., 2006), but the vector was inserted at the Rosa26 locus to drive more stable expression. Specifically, the R26-AlstR mouse was first generated using a conditional double-stop cassette system that has been described previously (Kim et al., 2009). The AlstR gene followed by an IRES-EGFP cassette was placed downstream of the CAG promoter and FSF-LSL double-stop cassette. The Cre-dependent R26-AlstR mouse line was then produced by removing the FSF stop cassette by crossing the original double-stop AlstR mouse with the mouse line of ACTB-FLPe transgene (Rodriguez et al., 2000). To achieve Cre-directed expression of AlstRs, the R26-AlstR mouse line was cross-bred with an Emx1-Cre mouse line (Guo et al., 2000) or a Dlx5/6-Cre mouse line (Monory et al., 2006). Emx1 is a mouse homologue of the Drosophila homeobox gene empty spiracles and its expression is largely restricted to excitatory neurons in the developing and adult cerebral cortex and hippocampus (Guo et al., 2000; Gorski et al., 2002; Kuhlman and Huang, 2008); the Emx1-Cre mouse line was created with the Cre gene placed directly downstream of the Emx1 promoter (Guo et al., 2000). Dlx is a family of homeodomain transcription factors which are related to the Drosophila distal-less (Dll) gene (Panganiban and Rubenstein, 2002); in the Dlx5/6 Cre mouse, expression of Cre is directed to virtually all forebrain GABAergic neurons during embryonic development by the I56i and I56ii enhancers from the zebrafish dlx5a/dlx6a intergenic region (Monory et al., 2006).

To prepare living brain slices, the double transgenic mouse pups (postnatal day 16–21) or control pups (AlstR or Cre mouse pups) were deeply anesthetized with pentobarbital sodium (>100 mg/kg, i.p.) and rapidly decapitated, and their brains removed. For electrophysiological experiments, primary somatosensory (S1) cortical slices of 400 μm thick were cut in sucrose-containing artificial cerebrospinal fluid (ACSF) (in mM: 85 NaCl, 75 sucrose, 2.5 KCl, 25 glucose, 1.25 NaH₂PO₄, 4 MgCl₂, 0.5 CaCl₂, and 24 NaHCO₃). Slices were first incubated in sucrose-containing ACSF for 30 minutes–one hour at 32°C, and then transferred to recording ACSF (in mM: 126 NaCl, 2.5 KCl, 26 NaHCO₃, 2 CaCl₂, 2 MgCl₂, 1.25 NaH₂PO₄, and 10 glucose). For VSD imaging experiments, cortical slices were cut in ACSF with a broad-spectrum excitatory amino acid antagonist kynurenic acid (0.2 mM). The ACSF with kynurenic acid was found to better preserve slice healthiness for dye staining than the sucrose-containing ACSF. After the initial incubation at 32°C, slices were transferred for dye staining at room temperature for one hour in the ACSF and 0.1 mg/ml of the absorption VSD, NK3630 (Nippon Kankoh-Shikiso Kenkyusho Co. Ltd., Japan) and then maintained in the recording ACSF before use. Throughout the cutting, incubation, staining, and recording, the solutions were continuously supplied with 95% O₂–5% CO₂.

Our overall system of electrophysiological recordings, photostimulation, and imaging was described previously (Xu et al., 2010; Xu, 2011). The solution was fed into the slice recording chamber through a pressure driven flow system with pressurized 95% O₂–5% CO₂ with a perfusion flow rate of about 2 ml/minute. To perform whole-cell recording, individual neurons were visualized at high magnification (60x objective), and were patched with glass electrodes of 4–6 MΩ resistance that were filled with an internal solution containing (in mM) 126 K-gluconate, 4 KCl, 10 HEPES, 4 ATP-Mg, 0.3 GTP-Na, and 10 phosphocreatine (pH 7.2, 300–305 mOsm). The internal solution also contained 0.1% biocytin for cell labeling and morphological identification. Once stable whole-cell recordings were achieved with good access resistance (usually <20 MΩ), basic electrophysiological properties were examined through hyperpolarizing and depolarizing current injections. We focused on analyzing neuronal input resistance, spiking thresholds, and spike numbers at different steps of depolarizing current pulses in control ACSF, in the presence of AL, and after washout of AL. To examine AlstR-mediated inactivation, the ligand AL was added into the recording solution. Although nanomolar concentrations of AL could inactivated isolated cultured AlstR-expressing neurons (Lechner et al., 2002), 1 μM AL (unless specified otherwise) were used for facilitating the ligand penetration deep into the slices and increasing temporal speeds of the ligand infusion. The AL application for 15 minutes was estimated to produce full AlstR effects, while the washout of 20 minutes was considered to remove the added AL from the recording solution.

During laser scanning photostimulation experiments, the microscope objective was switched from 60x to 4x. Stock solution of MNI-caged-l-glutamate (Tocris Bioscience, Ellisville, MO) was added to 20 ml of ACSF for a concentration of 0.2 mM caged glutamate. The cortical slice image was acquired by a high resolution digital CCD camera, which in turn was used for guiding and registering photostimulation sites. A laser unit (DPSS Lasers, Santa Clara, CA) was used to generate 355 nm UV laser for glutamate uncaging. Short pulses of laser flashes (1 ms, 20 mW) were controlled by using an electro-optical modulator and a mechanical shutter. The laser beam formed uncaging spots, each approximating a Gaussian profile with a width of ∼100 μm laterally at the focal plane. During the experiments examining neuronal excitation profiles, photostimulation was applied to 8 × 8 patterned sites (with an inter-site space of 50 μm²) centered at the recorded neurons in a non-raster, non-random sequence to avoid revisiting the vicinity of recently stimulated sites, while current-clamp recordings measured membrane potential depolarization and spiking activity. The neuronal excitation profile was simply assessed by the number of spiking locations to the photostimulation of 8 × 8 sites. Laser scanning photostimulation permits rapid evaluation of multiple stimulation locations with no physical damage to the tissue under our experimental conditions.

During VSD imaging experiments, a 705 nm light trans-illuminated brain slices and voltage-dependent changes in the light absorbance of the dye were captured by the MiCAM02 fast imaging system (SciMedia USA Ltd, Costa Mesa, CA). The illumination was supplied by optically filtering white light produced by an Olympus tungsten-halogen lamp up to 100 W of power. Optical recording of VSD signals was performed under the 2x objective with a sampling rate of 2.2 or 4.4 ms per frame (frame resolution 88 (w) × 60 (h) pixels). The field of view covered the area of 2.56 × 2.14 mm² with a spatial resolution of 29.2 × 35.8 μm²/pixel. For the imaging of photostimulation-evoked excitatory activity with and without the AL application, VSD imaging of evoked activity was triggered and synchronized with each laser photostimulation at specified cortical sites. For each trial, the VSD imaging duration was 2000 frames including 500 baseline frames with an inter-trial interval of 12–14 seconds. To image pharmacologically induced spontaneous seizure-like activity, ACSF with 10 μM bicuculline were infused into the slices for 15 minutes before VSD imaging; the imaging experiments consisted of seven sessions each at three perfusion conditions: bicuculline only, bicuculline + AL, and bicuculline only again in ACSF. Each session had nine consecutive trials conducted with three minutes of off-illumination intervals between sessions. The number of seizure occurrences per session was measured under different conditions.

VSD signals were originally measured by the percent change in pixel light intensity [ΔI/I%; the % change in the intensity (ΔI) at each pixel relative to the initial intensity (I)]. In addition, the mean and standard deviation (SD) of the baseline activity of each pixel across the 50 frames before the spontaneous activity onset or preceding photostimulation was calculated, and VSD signal amplitudes were then expressed as SD multiples above the mean baseline signal for display and quantification. The activated pixel was empirically defined as the pixel with the amplitude ≥ 1 SD above the baseline mean of the corresponding pixel's amplitude (equivalent to the detectable signal level in the original VSD maps of ΔI/I%). VSD images were smoothed by convolution with a Gaussian spatial filter (kernel size: five pixels; SD (σ): one pixel), and a Gaussian temporal filter (kernel size: three frames; δ: one frame). In the present study, single-trial VSD signals were of sufficiently high amplitudes and could be discerned from background noise; no averaging over multiple trials was used for data presentation unless specified. To examine AlstR-mediated effects on photostimulation-evoked VSD responses, the total numbers of activated pixels and response amplitudes across the 10 frames around the peak VSD activation (e.g., at 55 ms after photostimulation) were measured for comparison across control, AL application, and washout. Images were displayed and analyzed using custom-made MATLAB programs.

For statistical comparisons across more than two groups, we used the Kruskal–Wallis test (non-parametric One-Way ANOVA) and the Mann–Whitney U-test for group comparisons. Alpha levels of p ≤ 0.05 were considered significant. All the values were presented as mean ± SE.

In the present study, we used Cre-directed expression of the AlstR transgene to excitatory and inhibitory cortical neurons by crossing the R26-AlstR mouse line conditionally expressing AlstR and GFP following Cre recombination (Gosgnach et al., 2006) with the Emx1-Cre mouse line (Guo et al., 2000) or the Dlx5/6-Cre mouse line (Monory et al., 2006). We confirmed the specificity and efficiency of these Cre mouse lines in achieving cell-type specific gene expression. As demonstrated in Figures 1A–F, through the use of a tdTomato Cre reporter line (Madisen et al., 2010), the Emx1-Cre mouse expresses Cre in a majority of excitatory pyramidal neurons of the cortex (Guo et al., 2000; Gorski et al., 2002) (Figures 1A–C), while the Dlx5/6 mouse expresses Cre specifically in inhibitory interneurons (Monory et al., 2006; Chao et al., 2010) (Figures 1D–F).

Emx1-cre: AlstR mice expressed AlstRs with GFP in cortical excitatory neurons, which was verified through GABA immunostaining (Figures 1G,H) and electrophysiological recordings (see below). Immunochemical examinations and living slice observations indicated that about 50% of excitatory cortical neurons express AlstRs, detected by their strong GFP appearance. Similarly, AlstR expression in inhibitory cortical neurons of Dlx5/6-Cre: AlstR mice was strong, as the native fluorescence enabled direct visualization and targeted electrophysiological recordings of the labeled neurons.

We started to test whether AlstR expression through the transgenic mouse approach led to functional inactivation of neuronal activity with single-cell electrophysiology. As illustrated in Figures 2A–D, consistent with GIRK activation, bath application of AL significantly decreased neuronal excitability of AlstR-expressing excitatory cells in response to somatic current injections in Emx1-Cre: AlstR mouse slices. Similar to previous findings (Lechner et al., 2002; Wehr et al., 2009), bath application of AL hyperpolarized resting membrane potentials, markedly reduced membrane input resistance of AlstR-expressing (AlstR+) neurons (Figures 2B,D), and inhibited action potential firing by the decreased spike numbers elicited across a range of depolarizing current injections (Figures 2B,C). Control neurons without AlstR expression were not affected (N = 10). Then we further characterized AlstR-mediated inhibition of evoked excitability by laser scanning photostimulation. Compared to the examination of intrinsic membrane excitability through current injections, the measurement of photostimulation-evoked neuronal excitability via glutamate uncaging is more physiologically meaningful as the evoked excitability depends on the structure and membrane characteristics of the recorded neurons, and reflects the combination of intrinsic neuronal excitability and local synaptic input. Caged glutamate was present in the bath solution, and only turned active through focal UV photolysis. As shown in Figure 2E, we recorded from individual neurons and photostimulation was applied across multiple sites. Our data established that the AlstR-mediated inactivation was sufficiently strong and it essentially abolished the photostimulation-evoked spiking activity of the recorded AlstR-expressing excitatory neurons (Figures 2F–I).

As for inhibitory cortical neurons, we showed that the AlstR system produced robust inhibition on major subtypes of inhibitory cortical neurons through examinations of AlstR-expressing inhibitory neurons in Dlx5/6-Cre: AlstR mouse slices. The electrophysiological phenotype of the inhibitory neurons examined was fast-spiking, adapting-spiking, low-threshold spiking, or late-spiking; the cells had morphologies resembling basket cells, Martinotti cells, bitufted/bipolar cells, or neurogliaform cells. Compared to excitatory neurons, bath application of AL had a much larger hyperpolarizing effect on the resting membrane potentials of AlstR-expressing inhibitory neurons (Figures 3A,B). AL application consistently inhibited all the inhibitory neurons tested by decreasing their intrinsic excitability to current injections (Figure 3C) and photostimulation-evoked excitability (Figures 3D–L). It should be noted that the level of AlstR-mediated inhibition varied with cell-types. For example, the spiking activity of AlstR+, fast-spiking cells tended to be suppressed completely while AlstR+, low-threshold spiking cells had remaining spikes in the presence of AL (Figures 3E–L). However, the overall effect of inhibition was substantial across interneuronal subtypes (Figures 3C,D).

After the examination of AlstR-mediated effects at the level of individual neurons, we went further to characterize the AlstR-mediated inhibition on neuronal population activity with our newly developed functional imaging technique. This is important because cortical circuits ultimately operate at the neuronal population level. Optical imaging with fast VSD was used to examine photostimulation-evoked population activity and to evaluate AlstR-mediated inhibition on cortical network activity. The VSD imaging can monitor neuronal activity in a large area simultaneously (Xu et al., 2010; Xu, 2011), with changes in optical absorption closely correlating with neuronal membrane depolarization. The brain slices were stained with VSD and perfused with ACSF containing caged glutamate. Then local cortical regions were stimulated with laser photostimulation, and circuit activation and output directly visualized by fast VSD imaging. For the purpose of visual display, in Figure 4 and subsequent figures, the color scale codes the strength of excitatory activity (reflecting membrane potential depolarization of neuronal ensembles) with warmer colors indicating greater excitation. In addition, the time course of VSD signal (in the percent change of pixel intensity [ΔI/I%]) from the specified region of interest (ROI) is shown with the image data.

AlstR-mediated inhibition of excitatory neuronal population activity was found to be effective across different cortical sites (Figure 4). In Emx1-Cre: AlstR mouse slices, under control conditions, photostimulation in local cortical regions initiated excitation around 5 ms after laser exposure, which quickly propagated to functionally connected cortical regions within 50 ms in a columnar manner. Stimulation in the middle cortical layer (layer 4 or layer 5a) caused excitatory activity to spread vertically to layers 2/3, 5, and 6 (Figures 4A1,A2 and 5A1,A2). Although photostimulation-evoked excitatory activity occurred in the presence of the AL peptide, evoked responses of the AlstR-expressing brain slice was clearly reduced and constrained (Figures 4B1,B2). Given that about half of excitatory cortical neurons expressed AlstRs in these slices, the AlstR-mediated inhibition of excitatory population activity was quite effective. The AL effect was reversible with the peptide washout (Figures 4C1,C2). In comparison, AL application did not have effects on photostimulation-evoked population activity in the cortical slices without AlstR expression (N = 4). The reduction of VSD activation with the AL application for Emx1-Cre: AlstR slices, compared to control, is summarized in Figure 6A. On average, the total activation size during AL application was 18% ± 2.0% of the control condition; the total response amplitude during AL application was 15.8% ± 1.8% relative to control.

Conversely, we found that AlstR-mediated inactivation of GABAergic neurons augmented excitatory population activity. Photostimulation-evoked response in the presence of AL was dramatically increased in Dlx5/6-Cre: AlstR mouse slices (Figures 5A1,A2 and B1,B2), and the effect was reversed with the AL washout (Figures 5C1,C2). In agreement with the previous finding that S1 barrels are functionally independent cortical columns (Laaris et al., 2000; Petersen and Sakmann, 2001), in normal ACSF, the lateral extent of photostimulation response in S1 layer 4 appeared to be limited to the cortical column defined structurally by the barrel (Figure 5A1). Yet, the importance of inhibitory neuronal regulation of cortical excitatory information flow is evident, as inhibition of GABAergic interneurons via the AlstR system caused the spread of excitation laterally in layer 4 as well as in wide regions of upper and deeper layers (Figure 5B1). Across the slices, the increase of VSD activation in the AL presence, compared to control, was 500.5% ± 171% and 621.2% ± 272.6% for the total activation size and response amplitude, respectively (Figure 6B).

Given that targeted expression of AlstRs in excitatory neurons can strongly reduce cortical excitability with the ligand present, we were interested in investigating whether the AlstR system could be used for controlling seizure activity that is characterized by abnormal, excessive, or synchronous excitatory neuronal activity in the brain. We performed the experiments in single Emx1-cre: AlstR slices sequentially. First, bath application of bicuculline induced spontaneous seizure-like activity in the slices. The epileptiform activity detected by VSD imaging, occurred in a small region and evolved into large propagating activity across the whole slice lasting for about 200–300 ms per event (Figures 7A1–A4 and B1,B2). After stable occurrences of the seizure activity across seven imaging sessions (126 seconds per session), AL peptide was co-applied with bicuculline and seizure occurrences continued to be monitored by VSD imaging. As illustrated in Figures 7C,D, the tendency of reduced seizure occurrences was clearly visible during the AL application. Note that the average durations or peak amplitudes of individual events during control and the AL infusion did not differ. Finally, the removal of AL from bath perfusion increased seizure occurrences, which could get back to the normal control level. For the Emx1-Cre: AlstR mouse slices tested, the average number of seizure occurrences per minute was 1.09 ± 0.14, 0.41 ± 0.12, 0.68 ± 0.38 for bicuculline-containing ACSF, co-application of bicuculline and AL, and bicuculline-containing ACSF again, respectively. The co-application of AL with bicuculline reduced the occurrence of bicuculline-induced seizure activity significantly (Figure 7E). Thus, these data suggest that the AlstR system can be used to effectively suppress seizure activity.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.