(A) VSDI set-up. A stained slice was placed in an interface chamber. Liquid level of the perfusion source (Pool1) was stabilized by refilling solution through a pump from another ACSF pool (Pool2) at the same rate as bath perfusion. THT microscope and MiCAM02 camera system were used to record fluorescent images using an excitation filter (ex), a dichroic mirror (dic), an emission filter (em), an objective lens (obj) and a projection lens (pro). Drugs desired for bath application were added to both pools simultaneously. In some experiments, a stimulation electrode (stim) and a field (field) recording electrode were also introduced. (B) Stimulation-evoked responses are shown in slices stained with Di-4-ANEPPS (left) or FM 1–43 (right), confirming the voltage sensitivity of Di-4-ANEPPS. Colored pixels represent fluorescent changes at levels indicated in the legend on the right. Single 0.1 ms 400 μA stimulation was delivered to Schaffer-collateral fibers through a bipolar electrode (arrow) to evoke neuronal depolarization responses. Field recordings and VSDI signals (traces below each image) at a spot in CA1 stratum radiatum (white circle) demonstrated that VSDI signals corresponded well with field potentials in Di-4-ANEPPS stained slice. The arrow heads above traces indicate time of stimulation and the vertical blue broken line indicate the time when the displayed images were taken. Images and traces are averages of 10 stimulation trials. (C) Drifting of VSDI signals at a spot in deep EC area (3×3 pixel area) during 40 min of imaging in slices perfused with ACSF. After correction of drifting using a 2-dimentional function in BV-Analyze software, imaging signals showed a stable baseline at a low noise level (RMS or root mean square, 0.1%). (D) A VSDI image of a TTX-treated slice after 5 min of KCl (4 mM) bath application. The trace below the image shows VSDI signal values over time at indicated spot (3×3 pixels). VSDI baseline signal was stable during 20 min of perfusion and significant fluorescent responses were induced by bath addition of KCl (4 mM).

(A) Approximate locations of individual neurons studied for nicotine-induced (bath application) depolarizations with whole-cell current clamp in the presence of 1 μM TTX. (B) Upper traces show averaged membrane potential recordings obtained from neurons in selected regions as indicated in A. Due to the irresponsiveness of some neurons in the CA1SR region, traces from 4 out of 7 CA1SR neurons that responded to nicotine with > 1 mV depolarization were averaged and shown; this was done so that the time course of the nicotine effect was more evident. Depolarizations by nicotine (10 μM) in all neurons were mostly blocked by DHßE (1 μM), and ECVI neurons had the largest amplitude of responses among the regions studied. The lower traces are averaged membrane potential recordings from 5 neurons in each specified region during 5 min of bath-application with a lower concentration of nicotine (0.1 μM). Note that this lower concentration of nicotine (0.1 μM) induced significantly depolarization in neurons of ECVI (n = 6, p<0.05, student t test on peak values when comparing to the control recordings without nicotine application), but not in those of either SbSO or ECV (n = 5 p>0.05). Scale bars: vertical, 10 mV for upper traces, 1 mV for lower traces; horizontal, 5 min. (C) Peaks of membrane depolarization induced by nicotine (10 μM) in neurons shown in A. Each data point represents data from one neuron, and the black error bars are medians with interquartile ranges. ECVI neurons had significantly more depolarization when compared to the other regions (p<0.05, student t test to compare ECVI and SbSO values since they are normally distributed, nonparametric Kruskal-Wallis test followed by Dunn's multiple comparison test were used to compare the rest). (D) Dose-response curves demonstrate that neurons in ECVI were the most sensitive to nicotine. Data points represent averages of peak responses in at least 3 neurons.

An α7-only neuron responded to the local application (5-s duration) of ACh (4 mM) with a fast desensitizing current (upper left). The bath-application of nicotine (Nic, 10 μM, upper right) to this neuron failed to induce a membrane depolarization. A fast-desensitizing current response in another α7-only neuron (black trace in lower left) was significantly prolonged by PNU-120596 (PNU, 10 μM, red traces in lower left). The bath application of the same concentration of nicotine to this PNU-treated neuron induced a significant depolarization. Experiments were performed in the presence of TTX (1 μM), atropine (10 μM) and DHßE (1 μM).

(A) In the presence of atropine (10 μM), NBQX (20 μM) and APV (25 μM), sIPSCs were recorded from an ECV neuron when voltage-clamped at −80 mV with a KCl-based internal solution containing QX314. There was no significant change in sIPSC frequency with bath-applied nicotine (Nic, 10 μM) (p>0.05, KS test in pooled data of 3 neurons, cumulative probability data not shown). The GABA_A receptor blocker gabazine (1 μM) was applied at the end of the experiment to verify that the events were mediated by GABA_A receptors. (Scale bar: horizontal, 2 min; vertical, 50 pA). Note that this ECV neuron had a much less evident nicotine-induced current response than that of Figure 6A, which is a reflection of variability in nicotine responses among ECV neurons demonstrated in Table 1. (B) Increases in sIPSC frequency were observed in an ECV neuron during longer duration (1 min) local applications of ACh (4 mM) to the ECVI region (2nd trace), but not to either ECIII, ECV or control regions in the presence of antagonists to muscarinic AChRs (10 μM atropine), α7 nAChRs (10 nM MLA), and ionotropic glutamate receptors (10 μM NBQX, 25 μM APV). Locations of the local ACh applications were the same as in Figure 6B. The recordings were performed using a CsGluc-based internal solution containing QX314, and the neuron was voltage-clamped at 0 mV. (Scale bar: horizontal, 1 min; vertical, 50 pA, which also applies to C and D). (C) Increase in sIPSC events was recorded in an ECV neuron during local ACh application to the ECVI region (same condition as B). This increase was blocked by 1 μM DHßE, which was partially reversed by the removal of DHßE for 20 min. At the end of experiments, all sIPSCs were eliminated by gabazine (1 μM). (D) Increases in sIPSC events recorded in ECV neurons during local ACh application to the ECVI region were eliminated by 1 μM TTX, demonstrating that ECVI neurons influence towards ECV neurons require action potential firing. (E) Pooled results from 5 neurons demonstrated that intervals of sIPSC events (Int) in ECV neurons were decreased by the activation of non-α7 nAChR-containing neurons in ECVI, but not in ECV or ECIII.

(A) Images acquired at indicated time interval in the presence of 1μM TTX, either without nicotine bath perfusion (1st row) or with 10 μM nicotine (the rest of the rows). The nicotine-induced VSDI signals were not blocked by pre-perfusion of the α7 nAChR antagonist MLA (10 nM, 3rd row), but by the non-α7 nAChR antagonist DHßE (1μM, 4th row). Further 4mM KCl application induced depolarization in slices when nicotine effects were blocked by DHßE (4th row), confirming the viability of the slices. The last row shows a slice stained with FM 1–43 (a non-voltage-sensitive dye), which failed to induce detectable fluorescence changes after nicotine application. The figure legend (top right) indicates values of fluorescence change corresponding to color of pixels. Any pixel with a fluorescence change below 0.25% was left transparent so that a gray-scale background image of the slice could be shown to indicate the location of voltage changes. (B) Activation map of nicotine effect in TTX treated slices show sequence of nicotine-induced depolarization (images derived from the same slice as the 3rd row in A. Colored pixels indicate areas depolarized at 2 (green), 3 (brown) and 4 (red) min after bath-application of nicotine (10 μM). The map was derived after smoothing with a 9×9 spatial filter. The nicotine-induced depolarization started in ECVI and part of SbSO at 2 min, and then extended to neighboring regions at 3 and 4 min. Colored areas are regions above a threshold of 0.25% fluorescent change. (C) Averages of fluorescent changes in ECVI induced by bath-application of nicotine (10 μM) in the presence of TTX (1 μM, red line, n = 8), TTX and MLA (10 nM, blue line, n = 5), or TTX and DHßE (1 μM, green line, n = 6). Error bars: s.e.m.. Abbreviations: PRh, pararhinal cortex; Sb, subiculum; so, stratum oriens; sr, stratum radiatum; slm, striatum lacunosum moleculare; Den, dentate; PaSb, parasubiculum; PrSb, presubiculum.

(A) An ECVI neuron demonstrated a rapidly activating and desensitizing α7 nAChR-mediated current in response to the local application of choline (upper trace), and a mixture of fast (α7 nAChR-mediated) and slow desensitizing current (non-α7 nAChR-mediated) evoked by the local application of ACh (lower trace). (B) Scatter plots of functional α7 (top) and non-α7 (bottom) nAChR-mediated currents evoked by local application to neurons in the various regions. Significantly larger non-α7 nAChR-mediated currents were found in ECVI and SbSO neurons than CA1 neurons. Each data point represents the value from an individual neuron, and the error bars are median and interquartile ranges (see Table 1 for statistical analysis; see methods for isolation of α7 and non-α7 nAChR-mediated currents). Please note that the y-axis scales of the top segment are different from the bottom segment in B; this was done in order to better demonstrate the differences among regions with smaller responses.

(A) sEPSCs were recorded from an ECV neuron when voltage-clamped at −80 mV with a CsGluc-based internal solution containing QX314. Increase of sEPSC frequency by bath-applied nicotine (Nic, 10 μM) was reversed by DHßE (10 μM). The recorded ECV neuron also responded to nicotine with a slower inward current response, which was evidenced by a baseline shift after nicotine application. The ionotropic glutamatergic receptor blockers NBQX (20 μM) and APV (25 μM) were applied at the end of experiment to verify that the recorded events were mediated by glutamatergic receptors. (Scale bar: horizontal, 2 min; vertical, 50 pA). (B) Pooled data from 3 ECV neurons showed that the cumulative probability of sEPSC intervals significantly decreased between 1–5 min after nicotine application (red line, p<0.05, KS test), and recovered to control levels (black line) between 1–5 min after DHßE application (Blue line, p>0.05, KS test). (C) To confirm that ECVI neurons were the major source of nicotine-inducible glutamatergic inputs to ECV neurons, ACh (4 mM) was locally applied to neighboring regions of the EC through sharp glass pipettes (tip diameter ~1 μm) while an ECV neuron was recorded under whole-cell voltage-clamp. The figure shows the various locations of ACh application (long arrow heads). Besides the control location which was set away from the slice, ACh applications were ~180–220 μM away from the recorded neurons. (D) In the presence of atropine (10 μM) and MLA (10 nM), sEPSC events in ECV neurons were significantly increased after brief (350 ms) local application of ACh to ECVI neurons, but not after the application of ACh to either ECIII or neighboring ECV neurons. A glutamatergic receptor inhibitor CNQX (20 μM) completely blocked the sEPSC events. The slow inward current in the recordings was consistent with the direct activation of nAChRs on the recorded ECV neuron.

(A) Both bath (black circles) and local (red circles) application of nicotine (0.1 μM) for 10 min increased the amplitude of EPSCs recorded from ECVI neurons evoked by single stimuli. Data points and error bars represent mean and s.e.m. of eEPSCs recorded before (open circles), during (closed circles), and after (open circles afterwards) the bath (n = 5) or local (n = 6) nicotine application. Representative traces on top were derived from bath application experiments at indicated time points. The scale bar represents 50 pA/20 ms, which is the same for traces in B and C. (B) Tetanic stimulation (100 Hz, 100ms, black upward arrow) in Sb area induced STP in ECVI neurons (n = 5). Traces on top represent eEPSCs at indicated time points. (C) Local application of nicotine (0.1 μM) for 10 min to the recorded ECVI neuron boosted the STP to LTP (black circles, n = 7). When applied together with DHßE (1 μM), nicotine neither increased eEPSC amplitude, nor boosted STP into LTP (red circles, n = 5). Representative traces on top were derived from eEPSCs of nicotine experiment at indicated time points. (D) MLA (10 nM) blocked the increase of eEPSC amplitude by nicotine, but failed to stop generation of LTP (black circles, n = 6). However, after dialyzing cells with the calcium chelator BAPTA (10 mM, red circles, n = 4), nicotine failed to boost STP into LTP. (E) Paired-pulse ratio before (open bars) and after (closed bars) various treatments. Nic, nicotine alone (n = 6). STP, short-term potentiation after tetanic stimulation (n = 3). Nic+STP, STP preceded with nicotine treatment (n = 3). MLA+Nic, MLA co-applied with nicotine (n = 4). MLA+Nic+STP, STP preceded with MLA and nicotine application. *: significantly different from control (p<0.01students t-test).