The integrity of the hippocampal network depends on the coordination of excitatory and inhibitory signaling, which are under dynamic control by various regulatory influences such as the cholinergic systems. ZSET1446 (ST101; spiro[imidazo[1,2-a]pyridine-3,2-indan]-2(3H)-one) is a newly synthesized azaindolizinone derivative that significantly improves learning deficits in various types of Alzheimer disease (AD) models in rats. We examined the effect of ZSET1446 on the nicotinic acetylcholine (ACh) receptor (nAChR)-mediated regulation of synaptic transmission in hippocampal slices of rats. ZSET1446 significantly potentiated the facilitatory effect of nicotine and ACh on the frequency of spontaneous postsynaptic currents (sPSCs) recorded in CA1 pyramidal neurons with a maximum effect at 100 pM (tested range, 10 pM–1000 pM). The basal sPSC frequency without ACh was not affected. Such potentiation by ZSET1446 was observed in both the pharmacologic isolations of inhibitory and excitatory sPSCs and markedly reduced by blockade of either α7 or α4β2 nAChRs. ZSET1446 did not affect ACh-activated inward currents or depolarization of interneurons in the stratum radiatum and the lacunosum moleculare. These results indicate that ZSET1446 potentiates the nicotine-mediated enhancement of synaptic transmission in the hippocampal neurons without affecting nAChRs themselves, providing a novel possible mechanism of procognitive action that might improve learning deficits in clinical therapy.
Alzheimer disease (AD) is a neurodegenerative disorder characterized by progressive loss of memory and cognition. Although the effective therapeutic approaches against the AD remain highly limited (Holtzman et al., 2011), one of the most promising strategies is believed to be the potentiation of acetylcholine (ACh)-mediated signaling, especially that by nicotinic ACh receptors (nAChRs) (Levin and Rezvani, 2002). This strategy stems from various observations supporting the link between cognitive function and nAChRs in AD patients and animal models, including the decreased nAChRs in AD patients (Rinne et al., 1991; Perry et al., 1995) and direct inhibition of nAChR function by β-amyloid peptide (Aβ) (Wang et al., 2000a,b; Pettit et al., 2001). In fact, cholinesterase inhibitors such as donepezil and rivastigmine improve cognitive functions in animal models of AD (Dong et al., 2005; Van Dam et al., 2005) and also in human AD patients (Hansen et al., 2008; Farlow et al., 2010).
ZSET1446 (ST101; spiro[imidazo[1,2-a]pyridine-3,2-indan]-2(3H)-one) is a newly synthesized azaindolizinone derivative that significantly improves the learning deficits in the rat AD models induced by an i.c.v. injection of Aβ, by an i.p. injection of scopolamine, and in mice with genetically accelerated senescence (Yamaguchi et al., 2006, 2012). It also improves memory deficits in other rodent dementia models, including those induced by methamphetamine administration (Ito et al., 2007) and olfactory bulbectomy (OBX) (Han et al., 2008; Yamamoto et al., 2013). In addition, ZSET1446 prevents a reduction in nicotine-induced increase in extracellular ACh level by intracerebral microinjection of Aβ in the hippocampus in freely moving rats (Yamaguchi et al., 2006). These lines of evidence accumulated in the behaving animals suggest that ZSET1446 exert procognitive effects in animal models of AD through affecting cholinergic systems. A possible mechanism that has been proposed to underlie this effect is a stimulation of T-type voltage-dependent Ca2+ channels (T-VDCCs), that would lead to enhanced long-term potentiation (LTP) in the somatosensory cortex (Moriguchi et al., 2012).
It remains still equivocal, however, how such an effect is linked to the modification of synaptic activities in the hippocampus. For example, such enhancement of LTP by ZSET1446 does not occur in the hippocampus (Moriguchi et al., 2012), despite manifest increase in extracellular ACh level, both that occurring spontaneously and that stimulated by nicotine, by systemic injection of ZSET1446 (Yamamoto et al., 2013). Interestingly, such an enhancing effect of ZSET1446 on extracellular ACh level, in a manner sensitive to a T-VDCC inhibitor, could be observed both in naïve and OBX rodents (Yamamoto et al., 2013), which is in clear contrast to the LTP enhancement that was observed only in OBX animals (Han et al., 2008). Therefore, how ZSET1446 affects the synaptic transmission in a manner related to changes in the extracellular ACh level in the hippocampal network remains to be clarified. In the present study, aiming to elucidate the procognitive mechanism of ZSET1446, we analyzed effects of ZSET1446 on hippocampal synaptic transmissions, particularly those activated by exogenous application of choline receptor ligands, in acute slices prepared from young naïve rats using whole-cell patch clamp technique.
Materials and Methods
All experimental procedures were approved by the Jikei University Animal Care and Use Committee and performed in accordance with the Guidelines for Animal Experiments of the Jikei University School of Medicine, which conform to the Guiding Principles for the Care and Use of Laboratory Animals of the Science Council of Japan (for the experiments performed at Jikei University School of Medicine) and approved by the Animal Care and Use Committee of the Central Research Laboratory of Zenyaku Kogyo Co., Ltd., and performed in accordance with the Guideline for the Care and Use of Laboratory Animals established at the Central Research Laboratory, Zenyaku Kogyo Co., Ltd. (for the experiments performed at the Central Research Laboratory of Zenyaku Kogyo Co., Ltd). All efforts were made to reduce the unnecessary suffering of the animals and the number of animals used and to refine the experimental design. Wistar/ST rats (Japan SLC, Shizuoka, Japan) aged 15–22 days of either sex were anesthetized with isoflurane (5% in 100% O2) and decapitated. Two to three coronal hippocampal slices 400 µm thick were made with a linear slicer (PRO 7; Dosaka, Kyoto, Japan) in ice-cold cutting artificial cerebrospinal fluid (aCSF) composed of (in mM) 87 NaCl, 2.5 KCl, 0.5 CaCl2, 7 MgCl2, 1.25 NaH2PO4, 10 d-glucose, 25 NaHCO3, and 75 sucrose (pH 7.4 bubbled with 95% O2 + 5% CO2; osmolarity, ∼315 mOsm/kg). The slices were incubated in standard aCSF (in mM: 125 NaCl, 3 KCl, 2 CaCl2, 1.3 MgCl2, 1.25 NaH2PO4, 10 d-glucose, 0.4 l-ascorbic acid, and 25 NaHCO3) saturated with 95% O2 + 5% CO2 for 25–30 minutes at 37°C and then kept at room temperature until the recording. The slice was fixed in a recording chamber (∼0.4-ml volume) under nylon strings attached to a U-shaped platinum frame, then submerged in standard aCSF, and continuously perfused with aCSF at a flow rate of 2–2.5 ml/min. The CA1 pyramidal cells and interneurons were visually identified under an upright microscope (BX-51WI, Olympus, Tokyo, Japan) with infrared differential interference contrast (IR-DIC) optics. The IR-DIC images were captured using an IR CCD camera (IR-1000, Dage-MTI, Michigan City, IN) and stored digitally on a computer. For the experiments in which the slices were pretreated by scopolamine, the brain slices were incubated for 25–30 minutes at 37°C in the standard aCSF as described already but with scopolamine (10 µM) added, and the slices are left in the same solution at room temperature 30–90 minutes before being transferred to the recording chamber. In these experiments, the slices in the recording chamber were submerged in and continuously perfused with aCSF also containing scopolamine (10 µM) and the test solution for ZSET1446 (10 nM) also contained the same concentration of scopolamine.
Whole-Cell Patch-Clamp Recordings
Patch electrodes were fabricated from borosilicate glass capillaries of 1.2-mm outer diameter (1B120F-4; World Precision Instruments, Sarasota, FL) with a programmable puller (P-97; Sutter Instruments, Novato, CA). The pipettes were filled with one of the following internal solutions: 1) “K-gluconate-based” internal solution containing (in mM) 120 potassium gluconate, 6 NaCl, 1 CaCl2, 2 MgCl2, 2 ATPMg, 0.5 GTPNa, 12 phosphocreatine Na2, 5 EGTA and 10 HEPES hemisodium (pH 7.2 as adjusted with KOH; osmolarity, ∼310 mOsm/kg). This internal solution was used to observe changes in membrane potential, membrane current and stimulation-evoked excitatory postsynaptic currents. 2) “CsCl-based” internal solution containing (in mM) 136 CsCl, 1 CaCl2, 2 ATPMg, 1/2 phosphocreatine Na2, 5 EGTA and 10 HEPES hemisodium (pH 7.3 as adjusted with CsOH; osmolarity, ∼310 mOsm/kg). This internal solution was used for recordings of spontaneously occurring EPSCs and inhibitory postsynaptic currents (IPSCs). The tip resistance of the electrodes filled with these internal solutions was 3–6 MΩ.
Under visualization of neurons with IR-DIC optics, whole-cell recording was made from visually identified pyramidal neurons in the CA1 (for postsynaptic current recordings) or interneurons in the stratum radiatum and lacunosum-moleculare (for ACh-activated inward current and ACh-induced depolarization of interneurons). Whole-cell capacitance was compensated. Changes in series resistance was monitored by observing the membrane current responses to fixed command pulse and verified after the experiments by superimposing the response current curves. The transmembrane current and potential were recorded using an Axopatch 200B amplifier (Axon Instruments), low-pass filtered at 2 kHz, and sampled at 4 kHz with a PowerLab interface (AD Instruments, Sydney, Australia). For the voltage-clamp recordings, the membrane potential was clamped at a –70 mV (junction potential compensated). All recordings were made at room temperature (20–25°C).
To observe evoked excitatory postsynaptic currents, the membrane currents were recorded with the “K-gluconate-based” internal solution and the Schaffer collateral/commissural pathway was stimulated using a bipolar stimulating electrode (World Precision Instruments) with a constant-current stimulus (0.043–0.22 mA; 100 µs; 0.05 Hz). The holding potential was −70 mV. To observe the spontaneous postsynaptic currents (sPSCs) in CA1 pyramidal neurons, the pipettes were filled with “CsCl-based” internal solution, and holding potential was set at −70 mV. Spontaneous IPSCs and EPSCs (sIPSCs and sEPSCs, respectively) were separately recorded in the presence of kynurenic acid (3 mM) and bicuculline (10 µM), respectively.
To observe the currents activated by ACh and changes in membrane potentials in the interneurons, the recordings were made with “K-gluconate-based” internal solution.
Localized Application to Neurons Being Recorded.
Two types of agonists, ACh and nicotine, were applied locally to the slice for 30 seconds and 2 minutes, respectively, with a glass pipette (inner diameter, 0.7 mm), the tip of which was placed 2–5 mm upstream of the recording electrode. Thus, a faster rise of the agonist and faster exchange of the solution around the cells recorded than those achieved by bath application were ensured. When ACh was applied locally in the presence of atropine and neostigmine, the local application solution of ACh also contained the same concentration of these antagonists to ensure that the antagonist concentration did not change by application. Nicotine was applied alone.
ACh was dissolved at a concentration of 1 mM with a HEPES-buffered aCSF composed of (in mM) 135 NaCl, 3 KCl, 1.25 NaH2PO4, 10 glucose, 10 HEPES hemisodium, 0.4 l-ascorbic acid, 2 CaCl2 and 1.3 MgCl2, (pH 7.4, adjusted with NaOH). A pipette with a size equivalent to that used for patch-clamp recordings was filled with this solution and connected to a Pneumatic Picopump (PV830; World Precision Instruments). ACh was applied to neurons with a pressure pulse (8 psi, 10–25 ms) from a pipette of which the tip was located in the vicinity of the neuron being recoded. The bath superfusion was continued during the puffer application.
ZSET1446, atropine, neostigmine, bicuculline, kynurenic acid, methyllycaconitine (MLA), dihydro-β-erythroidine (DHβE), mibefradil, and tetrodotoxin (TTX) were dissolved with aCSF at each final concentration and bath-applied. These agents were perfused at least for 10 minutes before the application of agonists being tested.
Ex Vivo Application of Scopolamine.
In this series of experiments, the brain slices were incubated at 37°C in the same aCSF solution as used for other experiments but containing scopolamine (10 µM). In addition, recordings including the cell exploring phases before establishment of whole-cell patch configuration were made in the constant presence of scopolamine (10 µM).
ACh chloride, neostigmine bromide, (-)-nicotine, kynurenic acid, bicuculline methiodide, DHβE hydrobromide, and mibefradil dihydrochloride were obtained from Sigma Chemical Co. (St. Louis, MO). Atropine sulfate was obtained from Wako (Osaka, Japan). TTX was purchased from Nacalai tesque (Kyoto, Japan). MLA perchlorate was purchased from Calbiochem (La Jolla, CA).
Data and Statistical Analysis
The recorded membrane current and potential were analyzed off-line by an Igor Pro 5 (WaveMetrics, Portland, OR) program with the Igor Pro procedures written by F.K. To compare the event frequency before and after agonist application, the mean event frequency over a 30-second period before the application was defined as “before agonist” value, and the mean event frequency over a 30-second period starting from the maximal effect was defined as the “after agonist” value. The effects of antagonists on the effect of agonist on the event frequency were examined by comparing the event frequency within the same 30-second period defined according to the control response to the agonist in the absence of the antagonist. The mean amplitude of sIPSCs was calculated by averaging the amplitude of consecutive 100-second IPSC events appearing before and after drug applications.
The values are expressed as mean ± S.E.M. Differences were compared with the nonparametric Mann-Whitney’s U test for normalized values and paired or unpaired t test for raw measurement values. Differences with P < 0.05 were considered significant.
Limited Effects of ZSET1446 on Basic Neuronal Properties of CA1 Neurons
On the basis of the previous in vivo studies showing possible involvements of the hippocampus in the procognitive effects of ZSET1446 (Yamaguchi et al., 2006), we targeted neurons in the CA1 area in the hippocampus of naïve rats. Recordings of evoked EPSCs of CA1 pyramidal neurons indicated that ZSET1446 at concentrations from 100 pM to 10 nM in the slices prepared from rats aged P12–P20 and recorded in the absence and presence of scopolamine (10 µM) did not exert significant effects during observation for 10–90 minutes (with ZSET1446 at 10 nM, 60 minutes; 103.3% ± 7.8% of before application, n = 5, P = 0.15 in the absence of scopolamine in rats at P12–13; 126.0% ± 12.9% of before application, n = 5 in the presence of scopolamine in rats at P15-20, P = 0.15; Wilcoxon test).
In addition, we did not observe detectable and consistent changes in the holding current in any of these recordings (shift in holding current at 60–90 minutes of recording in the absence of scopolamine, −5.3 pA − 3.3 pA; n = 5, than in the presence of scopolamine, +0.86 − 14.6 pA; n = 5). Furthermore, ZSET1446 did not affect the basal sPSC frequency recorded in the absence of glutamate and GABA receptor blockers (Fig. 1E; ZSET1446, 100 pM; n = 6) and basal sIPSC frequency from pyramidal neurons recorded in the presence of glutamate receptor blocker (Fig. 2I; ZSET1446, 10 pM − 1 nM; n = 3–5). These data suggest that ZSET1446 exerts limited effects on the basal neuronal excitability and synaptic transmission.
ZSET1446 Potentiated the Facilitatory Effect of nAChR Stimulation on sPSC Frequency
Because an involvement of cholinergic systems, in particular, the nAChR systems, has been suggested for the behavioral procognitive effects of ZSET1446 (Yamaguchi et al., 2006), we then analyzed the effects of this drug on nAChR activation-triggered increase in the frequency of sPSCs. These were recorded with CsCl-based internal solution at a holding potential of −70 mV in the absence of glutamate and GABA receptor blockers. Under such conditions, both EPSCs and IPSCs were recorded as inward sEPSC events.
To selectively activate nAChRs, the recordings in this series of experiments were made in the presence of atropine (5 µM). In addition, to reduce the influence of extracellular degradation of ACh through cholinesterase activity, neostigmine (10 µM) was present in the extracellular solution. ACh (5 µM; 30 seconds) was locally applied to CA1 regions repeatedly for four times at an interval of 10 minutes except for that between the second and third application, which was 13 minutes (Fig. 1, A and B). This 13-minute interval was composed of a 3-minute period for ACh effect observation and a 10-minute application of ZSET1446 before the third ACh application. This same application protocol was carried out in 12 neurons, in six of which the third and fourth applications were done after the 10-minute and 20-minute applications of ZSET1446, respectively.
Local application of ACh increased the sPSC frequency from 4.2 ± 0.3 events/s to 7.4 ± 0.7 events/s (n = 12; P < 0.001, paired t test) at the first ACh application. This was termed “ACh-evoked PSC facilitation.” This ACh-evoked PSC facilitation in response to the first ACh application was used as a “control increment with ACh” and was used for normalizing the changes in ACh-induced sPSC frequency in response to the second, third, and fourth applications in each neuron.
In six neurons onto which ZSET1446 was not applied, ACh-evoked PSC facilitation was observed in response to repeated ACh application. The time course of this increase was almost unchanged, even with four repeated applications (Fig. 1A); however, the magnitude of this ACh-evoked PSC facilitation tended to be slightly attenuated in the course of repeated application (Fig. 1, A2 and open circles in C). Whereas the first ACh application significantly increased the PSC frequency (P < 0.01; n = 6; Fig. 1D, open circles), the ACh-evoked PSC facilitations were no more significant after the second application (Fig. 1D, open circles).
In six neurons, ZSET1446 was added to the aCSF at 10 minutes before the third ACh application (3 minutes after the second ACh application). In contrast to that without ZSET1446 (Fig. 1, A and open circles in D), ACh significantly increased sPSC frequency at the third and fourth application in the presence of ZSET1446 (Fig. 1, B and filled circles in D). The normalized increase in sPSC frequency by ACh at the third and fourth ACh application was significantly larger than that observed in neurons without ZSET1446 (Fig. 1C). No significant difference was seen in the basal frequency of sPSCs between the ZSET1446-treated and untreated groups at any of time points immediately before the ACh applications (P > 0.36; unpaired t test; n = 6; Fig. 1E). These results indicate that ZSET1446 potentiates ACh-evoked PSC facilitation. One possible mechanism would be preventing the gradual desensitization of this ACh effect during repeated application.
ACh-Evoked PSC Facilitation Resulted Mostly from Enhanced Inhibitory Transmission
The experimental conditions used in the previous section to record sPSCs do not allow distinguishing EPSCs and IPSCs without detailed analyses of kinetics because both these EPSCs and IPSCs appear as inward currents with a reversal potential of approximately 0 mV. To determine whether this effect of ZSET1446 resulted from potentiation of ACh effect on EPSC or IPSC, we made ACh application in the presence of bicuculline (10 µM) to pharmacologically isolate EPSCs. The addition of bicuculline markedly decreased the basal sPSC frequency by 87.3% ± 3.0% (from 3.1 ± 0.3 events/s to 0.4 ± 0.1 events/s; n = 6; Supplemental Fig. 1, A–D). This result indicates that most of the spontaneously occurring PSCs were mediated by GABAA receptor–mediated transmission in this recording condition. In the presence of bicuculline, the peak ACh response became 26.2% ± 12.0% that recorded in the absence of bicuculline (the ACh-induced increment of sPSC frequency was 2.6 ± 0.7 events/s in the absence of bicuculline and was 0.5 ± 0.2 events/s in the presence of bicuculline; n = 6; Supplemental Fig. 1D), indicating that the most of facilitatory effect of ACh was due to increasing sIPSC frequency. The ACh-evoked EPSC facilitation recorded in the presence of bicuculline was only 0.8 ± 0.7 events/s and 1.2 ± 1.0 events/s before and after ZSET1446, respectively. We failed to detect significant difference between these values (n = 4; P = 0.28, paired t test). In addition, these ACh-evoked increments in sEPSC frequency were much smaller than that in sIPSC frequency both in the presence and absence of ZSET1446. These results suggest that most (∼80%) of the increase in sPSC with ACh was due to increased frequency of sIPSC resulting from GABA-mediated transmission and the increase in frequency with ACh by ZSET1446 as described herein are due mostly to the facilitated GABA release.
Nicotine-Evoked sIPSC Facilitation Was Potentiated by ZSET1446
We then asked which kind of upstream mechanisms underlie such potentiation of ACh-mediated inhibitory transmission by ZSET1446. Based on the preceding results, we analyzed the influence of ZSET1446 with those of a nicotine-induced increase in sIPSC frequency, which was recorded under pharmacologic isolation by adding kynurenic acid (3 mM) to the aCSF. The sIPSCs were recorded with CsCl-based internal solution as inward currents. In addition, we performed local applications of nicotine at 5 µM, a “threshold” concentration without too manifest effect on sIPSC frequency compared with larger concentrations, as confirmed by preliminary application (see Supplemental Fig. 2; increase by 19% of pre-nicotine with 5 µM nicotine, P = 0.062, paired t test; n = 6), and examined the effect of ZSET1446 thereon. Unlike the experiments shown in Fig. 1, nicotine was applied only once and its effect was analyzed in each slice that had been perfused with ZSET1446 and specific blockers to avoid strong influence of desensitization-mediated attenuation of the small effect of low-concentration nicotine.
As expected based on the preceding results, ZSET1446 potentiated the increase in sIPSC frequency by local application of nicotine (5 µM; 2 minutes) without affecting the basal sIPSC frequency at the range of 10 pM to 1 nM (Fig. 2I, left bars). The concentration-response relation of this potentiating effect of ZSET1446 on nicotine-induced facilitation was bell-shaped with a maximum effect at 100 pM in the tested range from 10 pM to 1000 pM (Fig. 2, A–C, H). The nicotine-induced increase in sIPSC frequency was not accompanied by a significant increase in sIPSC amplitude both in the absence and presence of ZSET1446 (Fig. 2J). It is therefore likely that effects on postsynaptic GABAA receptors would play limited roles in the effects of nicotine and ZSET1446.
As both α7- and α4β2-type nicotine receptors are expressed and operational in the GABAergic interneurons in the hippocampus (Alkondon et al., 1999), we examined whether blockade of one of these subtypes affects the effect of ZSET1446. In the presence of selective blockers MLA (against α7-type, D) and DHβE (against α4β2-type, E), either of which was added to the aCSF at least 10 minutes before the nicotine application, the potentiating effect of ZSET1446 was significantly smaller than that in the absence of these nicotinic receptor blockers, and the changes in the nicotine effect was no more significant (Fig. 2, D, E, and H). MLA and DHβE did not significantly affect the basal sIPSC frequency measured in the presence of ZSET1446 (Fig. 2I; P > 0.89; unpaired t test). These results suggest that both types of nicotinic receptors are involved and required in the full expression of potentiating effect of ZSET1446.
It has been reported that ZSET1446 affects T-VDCCs (Moriguchi et al., 2012). We examined whether blockade of T-VDCC affects the effect of ZSET1446. Under blockade of T-VDCCs with mibefradil (10 µM), 5 µM nicotine failed to increase sIPSC frequency in the presence of ZSET1446 (Fig. 2, F and H). Mibefradil did not significantly affect the basal sIPSC frequency measured in the presence of ZSET1446 (Fig. 2I; P = 0.46; unpaired t test). These results indicate that a T-VDCC plays a role in the potentiation of nicotine-triggered GABA release facilitation by ZSET1446.
We next examined whether the potentiating effect of ZSET1446 on nicotine-induced facilitation depended on the generation of action potential. Under blockade of action potential generation with TTX, 5 µM nicotine failed to increase sIPSC frequency both in the absence (Supplemental Fig. 2C) and presence of ZSET1446 (Fig. 2, G and H).
The basal sIPSC frequency observed without nicotine was markedly and significantly smaller in the presence of TTX than in both the absence and presence of ZSET1446 (100 pM, Fig. 2I), suggesting that a large portion of sIPSC events depend on action potential generation in the GABAergic interneurons. This might suggest that facilitatory effect of ZSET1446 on nicotine-induced increase in sIPSC frequency is due to action potential-dependent GABA release and subsequent inhibitory synaptic transmission.
Effect of ZSET1446 on Interneuron Excitation by ACh
A possible mechanism that might underlie this potentiating effect of ZSET1446 is an enhancement of nicotinic receptor functions themselves; however, this possibility seems unlikely because of the following two lines of evidence. First, as described already, this effect of ZSET1446 does not depend on any specific type of nicotinic receptors involved in the release facilitation, suggesting that ZSET1446 rather affects the processes that do not depend on a specific type of nAChRs. Second, ZSET1446, at the same concentration (100 pM) for the sIPSC frequency increase, did not apparently affect the amplitude of the ACh-induced inward currents (Fig. 3, A and D) recorded in the presence of atropine and neostigmine. For this purpose, we recorded transmembrane currents from eight interneurons located in the stratum radiatum, and ACh (1 mM) was puffer-applied from a pipette placed in the vicinity of the cell being recorded to analyze the kinetics of the ACh-induced current in response to the brief application (10–25 ms; Fig. 3, A and C). Of these, six neurons showed “fast” MLA-sensitive response, characterized by faster rise and decay, to puffer application of ACh and 2 neurons showed “slow” DHβE-sensitive response, characterized by slower rise and decay kinetics (Fig. 3, B and D; Alkondon et al., 1999). Interestingly, both fast (Fig. 3, B, E, and F) and slow (Fig. 3, D–F) inward currents by ACh in distinct interneurons were insensitive to ZSET1446, suggesting again that the effect of ZSET1446 is not mediated by a specific subtype of nAChRs. In addition, ZSET1446 application did not exert any detectable significant effect on the holding current (after a continuous application for 10–30 minutes, mean, −0.08 pA; range, −0.56 pA − 0.60 pA; n = 8 neurons. P = 0.53, paired t test). This finding suggests that global change in the resting membrane potential by ZSET1446 does not underlie its facilitatory effect on ACh-induced GABA release. These results argue against the interpretation that ZSET1446 increased sIPSC frequency through evoking larger depolarization of inhibitory interneurons in response to ACh application.
To confirm this directly, we measured the membrane potential of interneurons and analyzed the effect of ZSET1446 on ACh-induced depolarization. ACh (20 µM) was bath-applied at 10- or 13-minute intervals. The depolarization by ACh was not significantly affected by ZSET1446 (100 pM; 10 minutes and 20 minutes; Fig. 3, G–I). We also confirmed that the number of action potentials appearing during this ACh-induced depolarization was not significantly different between before (Fig. 3I, “2nd”) and after (Fig. 3I, “3rd” and “4th”) ZSET1446 application.
Effect of ZSET1446 on the Inhibitory Synaptic Transmission in the Hippocampus
This study provides the in vitro evidence for the synaptic mechanism underlying the effect of ZSET1446, a potent procognitive drug that significantly improves the learning and memory deficits in various rodent AD models (Yamaguchi et al., 2006; Han et al., 2008; Yamamoto et al., 2013). The main conclusion of this study is that whereas ZSET1446 does not affect basal synaptic activities, excitatory or inhibitory, in the hippocampal CA1, it enhances the facilitatory effect of nicotinic receptor activation on the frequency of spontaneous transmission, especially the inhibitory ones. It is therefore conceivable that, in the in vivo brain, cholinergic influence to the hippocampal interneurons would be potentiated by ZSET1446, which would ameliorate and rectify the whole hippocampal network activity.
The mechanism with which ZSET1446 potentiates ACh-evoked facilitation of inhibitory transmission remains to be identified. The present results indicate that 1) basic membrane excitability was not affected by ZSET1446; 2) GABA and GABAA receptor-mediated inhibitory transmission is the most predominant target of this potentiation; 3) both types of nAChRs (i.e., MLA-sensitive α7 and DHβE-sensitive α4β2 nAChRs) play significant roles and blockade of any of these significantly attenuate the ZSET1446 effect; 4) the potentiation by ZSET1446 of nicotine-induced GABA release facilitation was attenuated under T-VDCC inhibition; 5) inward currents mediated by these classes of nAChRs are not affected by ZSET1446, ruling out possible direct or indirect effects on the receptor channels, unlike PNU-120596 (Hurst et al., 2005); 6) depolarization after the nAChR activation was not affected, ruling out possible changes in the membrane excitability; and 7) the release facilitation required action potential generation. Therefore, plausible mechanisms underlying the potentiating effect of ZSET1446 on ACh-induced GABA release facilitation observed in this study would involve potentiation of excitation-release coupling that is more specific in GABAergic interneurons expressing any of α7 and α4β2 nAChRs (Léna et al., 1993; Tang et al., 2011).
Targets of ZSET1446
The molecular target of ZSET1446 in improving cognitive functions remains only poorly identified. Recently, Moriguchi et al. (2012) reported that mibefradil completely blocks ZSET1446-induced enhancement of long-term potentiation in the cortex and that ZSET1446 stimulates the CaV3.1-mediated Ca2+ current (i.e., T-VDCC-mediate current) in neuro2A cells overexpressing recombinant human CaV3.1. This report is of particular importance because an ACh-induced facilitation of asynchronous GABA release in the hippocampus is blocked by nAChR antagonists and also by T-VDCC blockers (Tang et al., 2011). They suggest that T-VDCCs play accessary role in modulating the excitation-release coupling in the hippocampal interneuron terminals. Our results that mibefradil prevented significant potentiation by ZSET1446 of nicotine-evoked IPSC facilitation (Fig. 2, F and H) suggest an involvement of the effects on T-VDCCs in this potentiation. It is therefore likely that, together with the increase in the extracellular ACh level after administration of ZSET1446, both in naïve and AD models (Yamaguchi et al., 2006; Yamamoto et al., 2013), modulation of GABA release through interaction between nAChR and T-VDCC at the GABAergic terminal is one of the principal targets of ZSET1446.
Similarly, it has been reported that nicotine-mediated enhancement of CA1 synaptic plasticity requires GABAergic transmission in AD model transgenic mice (Rosato-Siri et al., 2006). Interestingly, they have shown that this effect is mediated by both α7- and α4β2-type nAChR, in a similar manner to the present report. Absalom et al. (2013) reported that MLA can also inhibit α4β2 nAChR-mediated currents at an IC50 of 2.0 × 10-7 M, a concentration four times larger than the concentration used in this study. In addition, according to the concentration-response curve in Fig. 1 of their article, the inhibition of α4β2 nAChR-mediated current by MLA was almost negligible at the concentration we used in this study (50 nM). Thus, it is unlikely that ZSET1446 affects only the α4β2 nAChR-mediated currents; rather, it exerts its effects on nicotine-induced release facilitation by activating both α4β2 and α7 nAChRs.
Recently, accumulated lines of evidence point out essential roles of GABAergic interneurons in specific types of learning and cognitive loss in AD models and suggest that amelioration or enhancement of GABAergic transmission would be one of the promising strategies of AD therapy (Klausberger et al., 2003; Nitz and McNaughton, 2003; Rosato-Siri et al., 2006; Cui et al., 2008; Andrews-Zwilling et al., 2010; Murray et al., 2011). Together, one of the plausible mechanisms of the effects of ZSET1446 in improving the cognitive function would result, at least partly, from modulation of the link among nAChRs, T-VDCCs, and GABA release in the CA1 interneurons. Whether the effects on these factors are mediated by the same molecular mechanism as the ACh release facilitation remains to be elucidated in the future study.
An interesting feature of the effect of ZSET1446 is its irregular dose dependency (i.e., its bell-shaped dose-response relationship) observed in different types of preparations. At the molecular level, Moriguchi et al. (2012) found that 1 nM ZSET1446 caused weaker phosphorylation of CaMKII in the hippocampal CA1 slices than with 100 pM ZSET1446. Likewise, in the same article, Moriguchi et al. (2012) reported a steep bell-shaped dose-response curve of the direct effect of ZSET1446 on the intracellular Ca2+ concetration of cells heterologously expressing T-type VDCCs. At the behavior level, Yamaguchi et al. (2006) indicated that step-through latency at the passive-avoidance task peaked at 0.01 mg/kg, which exerted stronger effect than 0.1 mg/kg. Likewise, in this study, the potentiation of nicotine-evoked facilitation of IPSC was more manifest with 100 pM ZSET1446 than with 1 nM (Fig. 2H). The mechanism underlying these bell-shaped dose-response relationships remains unidentified. For example, Moriguchi et al. (2012) attributed this property to distinct facilitatory effects of ZSET1446 on both glutamatergic and GABAergic transmissions; however, this interpretation cannot be applied to the present finding of bell-shaped response of IPSC frequency recorded under blockade of excitatory transmission. Rather, as shown in the heterologously expression systems (Moriguchi et al., 2012), it is likely that ZSET1446 would have two opposite effects on the target molecule, such as the T-type VDCCs. Identification of these distinct effects are subjects for future studies. In this regard, appropriate determination of the clinical dosage would be critical in applying this drug to patients in the future.
One of the most plausible interpretation is therefore the potentiating effect of ZSET1446 involves effects on specific VDCC and/or Ca2+-permeable receptor channels that are tightly linked with nACh receptors. The release facilitation by nAChR activation would involve the following three mechanisms, depending on the structure and preparation: 1) direct exocytosis triggered by the Ca2+ entry through nAChR, in particular with the α7-type (Gray et al., 1996; Zappettini et al., 2011); 2) local depolarization by nAChR current, which then activates nearby VDCCs that trigger exocytosis (Léna and Changeux, 1997); and 3) depolarization by nAChRs giving rise to action potential generation at the “preterminal” axonal regions resulting in canonic action potential-triggered exocytosis in a manner sensitive to TTX (Léna et al., 1993; Alkondon and Albuquerque, 2001; Kanno et al., 2005). As potentiation with ZSET1446 could not be observed in the presence of TTX, the first and the second mechanisms are less likely to underlie the potentiation with ZSET1446 and, therefore, enhancement of ACh-triggered preterminal depolarization, which cannot be detected by somatic recordings, by ZSET1446 seems a plausible possibility.
It has been demonstrated in various central structures that agonists for α7-containing and α4β2-containing nAChRs could facilitate transmitter release (Alkondon et al., 1997, 1999; Ji and Dani, 2000; Kanno et al., 2005). Such effects are attributed to mixed expression of these nAChR subunits in single hippocampal neuron. Indeed, in this study, interneurons, even those with similar somatic morphology and location, showed distinct types of AChR-mediated inward currents with distinct pharmacologic properties, and ZSET1446 failed to affect any of these distinct currents (Fig. 3). This result rules out the possibility that a specific nAChRs is the target of ZSET1446. Rather, it is more likely that some intracellular processes linking nAChR-triggered depolarization and GABA release is affected by ZSET1446. In addition, the potentiation by ZSET1446 of ACh-, or nicotine-induced GABA release facilitation was significantly attenuated in the presence of either of these blockers for distinct nAChR subtypes. A plausible interpretation of this result is that distinct types of nAChRs located in the vicinity of release machinery are concurrently activated by ACh or nicotine and function in a synergistic manner to exert release facilitation and that ZSET1446 potentiates this process. In support of this idea, despite strong additive effects in response to simultaneous full activation of α7 and α4β2 nAChRs with large concentration of agonists, they show synergistic nonadditive effects when these receptors are weakly stimulated (Zappettini et al., 2011). In the present study, we have used a near-threshold concentration of nicotine (5 µM; see also Supplemental Fig. 2B) to clearly observe the potentiating effect of ZSET1446 (Fig. 2). Therefore, it might be possible that each of these receptor subtypes had been not fully activated by this concentration but, when activated concurrently, caused a synergistic effect and underwent modulation by ZSET1446. This might explain why blockade of any of α7 and α4β2 nAChRs strongly attenuated its potentiating effect.
Currently, no clear framework has been established in understanding the mechanism of action of the drugs against cognitive impairment. The most popular drugs, such as donepezil, galantamine, and rivastigmine, are cholinesterase inhibitors or AChR modulators that are imagined to improve cognitive functions through potentiating the cholinergic modulation of hippocampal function. In this context, the effects of ZSET1446 in potentiating the ACh-induced GABA release facilitation, as demonstrated in this study, would be expected to exert similar effects to those cholinesterase inhibitors that would also potentiate such mechanisms by increasing synaptic ACh concentration and/or by potentiating AChR function (Santos, et al., 2002). Another important aspect of the effects of ZSET1446 was general absence of the effect on excitatory transmission and spontaneous transmission as confirmed by recording various activities in the slice as shown in this study. One of the speculative interpretations is that because its effect was highly selective to a specific mechanism underlying the processes downstream to nAChR activation in the interneurons, ZSET1446 selectively potentiate this specific regulatory system of inhibitory transmission and leads to effective, but not excessive, amelioration of the cognitive function without affecting other central nervous system functions.
The authors thank Kazuhiro Hashimoto, chairman and executive director of Zenyaku Kogyo Co., Ltd., for encouragement and constant support and Dr. Eiji Shigetomi (University of Yamanashi) and Dr. Robert A. Shiurba for fruitful discussions.
Participated in research design: Kato, Yamaguchi, Hino, Takeda.
Conducted experiences: Takeda.
Data analyses: Takeda, Kato.
Wrote or contributed to the manuscript writing: Takeda, Yamaguchi, Hino, Kato.
- Received August 26, 2015.
- Accepted November 16, 2015.
Kentaro Takada, Yoshimasa Yamaguchi, and Masataka Hino are employees of Zenyaku Kogyo Co., Ltd. The Department of Neuroscience of the Jikei University of School of Medicine, of which Fusao Kato is the director, received support for consumables and equipment for unspecified research purposes from Zenyaku Kogyo Co., Ltd.
- artificial cerebrospinal fluid
- Alzheimer disease
- infrared differential interference contrast
- long-term potentiation
- nicotinic acetylcholine receptor
- olfactory bulbectomy
- spontaneous excitatory postsynaptic currents
- spontaneous inhibitory postsynaptic currents
- spontaneous postsynaptic currents
- T-type voltage-dependent Ca2+ channels
- Copyright © 2016 by The American Society for Pharmacology and Experimental Therapeutics