Serotonin dependent masking of hippocampal sharp wave ripples

Neuropharmacology 101 (2016) 188e203

Contents lists available at ScienceDirect

Neuropharmacology journal homepage: www.elsevier.com/locate/neuropharm

Serotonin dependent masking of hippocampal sharp wave ripples Rizwan ul Haq a, c, Marlene L. Anderson a, Jan-Oliver Hollnagel a, Franziska Worschech a, Muhammad Azahr Sherkheli c, Christoph J. Behrens a, Uwe Heinemann a, b, * €tsmedizin, Berlin, Germany Inst. Neurophysiology, Charit
e Universita €tsmedizin, Berlin, Germany Neuroscience Research Center, Charit
e Universita c Department of Pharmacy, Hazara University, Havelian Campus, 22500, Pakistan a

b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 17 April 2015 Received in revised form 4 August 2015 Accepted 21 September 2015 Available online 26 September 2015

Sharp wave ripples (SPW-Rs) are thought to play an important role in memory consolidation. By rapid replay of previously stored information during slow wave sleep and consummatory behavior, they result from the formation of neural ensembles during a learning period. Serotonin (5-HT), suggested to be able to modify SPW-Rs, can affect many neurons simultaneously by volume transmission and alter network functions in an orchestrated fashion. In acute slices from dorsal hippocampus, SPW-Rs can be induced by repeated high frequency stimulation that induces long-lasting LTP. We used this model to study SPW-R appearance and modulation by 5-HT. Although stimulation in presence of 5-HT permitted LTP induction, SPW-Rs were “masked” e but appeared after 5-HT wash-out. This SPW-R masking was dose dependent with 100 nM 5-HT being sufficient e if the 5-HT re-uptake inhibitor citalopram was present. Fenfluramine, a serotonin releaser, could also mask SPW-Rs. Masking was due to 5-HT1A and 5-HT2A/C receptor activation. Neither membrane potential nor membrane conductance changes in pyramidal cells caused SPW-R blockade since both remained unaffected by combining 5-HT and citalopram. Moreover, 10 and 30 mM 5-HT mediated SPW-R masking preceded neuronal hyperpolarization and involved reduced presynaptic transmitter release. 5-HT, as well as a 5-HT1A agonist, augmented paired pulse facilitation and affected the coefficient of variance. Spontaneous SPW-Rs in mice hippocampal slices were also masked by 5-HT and fenfluramine. While neuronal ensembles can acquire long lasting LTP during higher 5-HT levels, lower 5-HT levels enable neural ensembles to replay previously stored information and thereby permit memory consolidation memory. © 2015 Elsevier Ltd. All rights reserved.

Keywords: 5-HT 5-HT1A receptors LTP SPW-R Fenfluramine Citalopram NAN-190 8-OH-DPAT

1. Introduction In freely moving rodents, recordings of field potentials or local EEG activity in the hippocampus indicate theta activity with superimposed gamma oscillations during explorative behavior (Buzsaki, 1986; Vanderwolf, 1969; O'Keefe and Nadel, 1978). In contrast, during consummatory behavior, behavioral immobility, and slow wave sleep, hippocampal field potential activity is dominated by sharp wave ripples (SPW-Rs) (Buzsaki, 1986), during which stored information is thought to be replayed in form of sequentially activated pyramidal cells in a temporally compressed manner, thus allowing memory consolidation. GABAergic neurons play important roles in selecting those cells that are involved in the

* Corresponding author. Neuroscience Research Center & Institute of Neuro
Universit€ physiology, Charite atsmedizin, Berlin, Germany. E-mail address: [email protected] (U. Heinemann). http://dx.doi.org/10.1016/j.neuropharm.2015.09.026 0028-3908/© 2015 Elsevier Ltd. All rights reserved.

formation of a neuronal ensemble. Switching from theta rhythm activity to SPW-Rs and vice versa in vivo is rather abrupt. Variations in cholinergic tone (Hasselmo and McGaughy, 2004; Vandecasteele et al., 2014) are involved in these transitions, however, other neuromodulators such as serotonin (5-HT, 5-hydroxytryptamine) may also contribute. 5-HT, released from fibers which originate in the medial and € m and Fuxe, 1964; Aghajanian et al., lateral raphe nuclei (Dahlstro 1967), can modify the behavior of many neurons simultaneously by volume transmission and thereby orchestrate neuronal interactions among many regions of the brain. Serotonergic mechanisms are activated during intracranial self-stimulation (Jacques, 1979) and are important in mood regulation (Heinz et al., 2001; Roth, 1994). They involve the activation of a wide variety of receptors, most of which are G protein coupled. One exception to the rule are ionotropic 5-HT3 receptors (Barnes and Sharp, 1999; Kroeze et al., 2002), strongly expressed on interneurons in the

R. ul Haq et al. / Neuropharmacology 101 (2016) 188e203

hippocampus (McMahon and Kauer, 1997; Sudweeks et al., 2002). 5-HT affects postsynaptic excitability by activation of GIRK channels, thereby hyperpolarizing pyramidal cells (Andrade and Nicoll, 1987; Behr et al., 1997; Schmitz et al., 1998c; Segal, 1980). In addition, 5-HT reduces glutamate release in the entorhinal cortex, hippocampus and subiculum via activation of 5-HT1A receptors at least in part due to reduced presynaptic Ca2þ influx (Schmitz et al., 1998b). Interestingly, 5-HT affects memory functions with a prominent role of 5-HT1A receptors (Meneses and Liy-Salmeron, 2012; Cowen and Sherwood, 2013). It can be released from serotonergic fibers independent of neuronal activity by fenfluramine (Richter-Levin and Segal, 1990). In rat hippocampal slices, 5-HT and fenfluramine modulate both pharmacologically and stimulus induced gamma oscillations (Wojtowicz et al., 2009) and decrease SPW activity in the dentate gyrus (Richter-Levin and Segal, 1990; Kubota et al., 2003), potentially interfering with memory formation. Interestingly it was recently demonstrated that activation of serotonergic neurons in the raphe caused disappearance of ripple activity and interfered with memory consolidation (Wang et al., 2015). This, however, stands in contrast to another in vivo study (Ponomarenko et al., 2003) where blocking 5HT1A receptors led to reduced ripple activity, perhaps pointing towards an augmenting effect of 5-HT on ripple oscillations in vivo. The molecular mechanisms behind these effects remain unknown. Their importance though lies not only in our understanding of the brains ability to quickly switch from one brain state to another. Patients who suffer from memory related problems as well as patients suffering from depression who are treated with serotonin re-uptake inhibitors might directly be influenced by this serotonergic modulation. In ventral rodent hippocampal slices SPW-Rs occur spontaneously (Maier et al., 2003; Papatheodoropoulos and Kostopoulos, 2002). In addition, SPW-Rs can also be induced with stimulation protocols able to induce late LTP (Behrens et al., 2005). In this study we report that the appearance of SPW-Rs, evoked though high frequency stimulation, is masked by 5-HT in a dose dependent manner, presumably via activation of 5-HT1A and 5-HT2A/C

189

receptors. Neuromodulators may thereby regulate the activation probability of previously formed neural assemblies. 2. Methods 2.1. Slice preparation and solutions Animal procedures were performed in accordance with the guidelines of the European Communities Council and approved by the regional Berlin animal ethics committee (LaGeSO Berlin: T0068/02). We decapitated adult Wistar rats (aged 6e8 weeks, >200 g, Charles River Laboratories, Sulzfeld, Germany) under deep ether anesthesia or under anesthesia induced by isoflurane and laughing gas (1% isoflurane in 70% N2O and 30% O2) and male C57Bl/ 6N mice aged 12e14 weeks. We prepared horizontal hippocampal slices (400 mm/at bregma - 4.7 to 7.3 mm) at an angle of ~12 in the fronto-occipital direction (with the frontal portion up) using a vibratome (752 M Vibroslice, Campden Instruments, Loughborough, England, or Leica VT1200 vibratome, Wetzlar, Germany). Slices were prepared in cooled (~4  C) artificial cerebrospinal fluid (aCSF) containing (in mM): NaCl 129, NaHCO3 21, KCl 3, CaCl2 1.6, MgSO4 1.8, NaH2PO4 1.25, glucose 10, saturated with 95% O2 e 5% CO2. They were immediately transferred to an interface chamber perfused with aCSF at 36 ± 0.2  C (flow rate: ~1.8 ml/min, pH 7.4, osmolarity: 297e303 mosmol/kg) and allowed to recover for 2e3 h before experiments were started. For each experimental condition, only one slice per animal was used with recording durations between 2 and 3 h. However, different protocols were often employed in 2e3 slices from the same animal in order to minimize the number of animals. 2.2. Recordings Extracellular field potentials (FPs) were recorded under interface conditions from stratum pyramidale of area CA3 and CA1 with a custom-made amplifier using microelectrodes filled with 154 mM NaCl (5e10 MU). For intracellular recordings, sharp

Fig. 1. Induction of SPW-Rs in rat hippocampal slices. A: Position of stimulation electrode in SR of area CA1 and recording electrodes in stratum pyramidale of area CA3 and CA1. B. Sample recording of stimulus-induced SPW-Rs in a condensed form. Note the appearance of SPW-Rs after 3 repetitions of the stimulation protocol. C. Average properties of induced SPW-Rs in area CA3 and CA1. Note that amplitude and duration of events in area CA3 are larger than in CA1.

190

R. ul Haq et al. / Neuropharmacology 101 (2016) 188e203

Fig. 2. Laminar profile and current source density (CSD) analysis of stimulus-induced SPW-Rs in area CA3 of rat hippocampus. A: Schematic of horizontal slice. Note the placement of the 16-channel recording probe with the outer most electrode in CA3 stratum oriens (SO), electrode 5 in stratum pyramidale (SP) and the following electrodes in stratum radiatum (SR) and stratum lacunosum moleculare (SLM). B: Sample trace of one stimulus-induced SPW-R. Only every second recorded trace is shown, with SP on the third depicted trace. C: CSD analysis of stimulus-induced SPWs from the same slice after low-pass filtering with 80 Hz. Shown is the averaged field potential of 29 SPW-Rs on top of the corresponding heat map with current source in red and sinks in blue. Note the strong current source in the SP, followed by a sink in SR and SO and another smaller source in SLM D: Four examples of CSDs and field potentials of randomly chosen stimulus-induced single SPW-R events. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.).

microelectrodes (70e90 MU) were pulled from borosilicate glass (o.d. 1.2 mm) and filled with 2.5 M potassium acetate. Signals were amplified using a SEC 05L amplifier (NPI Instruments, Tamm, Germany). Sharp microelectrodes were preferred over patch clamp recordings as the required measurements often took considerable time. All data were low-pass filtered at 3 kHz, digitized at 10 kHz and stored on computer disk using a CED 1401 interface (Cambridge Electronic Design, Cambridge, UK). Intracellular recordings were accepted when membrane potentials were negative to e 62 mV, action potential (AP) amplitudes exceeded 75 mV and input resistance was >25 MU. Laminar profiles were recorded using 16 channel A1X16 silicon probes (NeuroNexus, Ann Arbor, MI, United States of America) with an inter electrode distance of 50 mm. These were placed vertically to the pyramidal cell layer in CA3. The signals were amplified 200 times; high pass filtered at 1 Hz and low-pass

filtered at 3 KHz using an EXT-16DX extracellular amplifier (NPI Instruments, Tamm, Germany). 2.3. Drugs All drugs were purchased from SigmaeAldrich (Taufkirchen, Germany), if otherwise, it is noted. Apart from NAN-190 and cisapride, which were dissolved in DMSO (final concentration 0.01%), drugs were dissolved in aCSF and applied by continuous bath perfusion. Serotonin (5-HT) was applied in concentrations from 100 nM to 50 mM. Relatively high concentrations of 5-HT and other pharmacological agents were chosen as they were usually applied for a short time period (~30 min) and equilibration kinetics within the tissue are known to be slow in interface slice chambers (Müller et al., 1988; Gloveli et al., 1995). In addition, many of these agents

R. ul Haq et al. / Neuropharmacology 101 (2016) 188e203

191

Fig. 3. Laminar profile and CSD analysis of spontaneous SPW-Rs in area CA3 of murine hippocampus. A: Scheme of a representative murine hippocampal slice. The 16-channel recording probe is placed as in Fig. 2A. SP in CA3 and the granule cell layer of the dentate gyurs are schematically depicted. B: Sample trace of one spontaneous SPW-R. For clear arrangement, only every second recorded trace is shown, with SP on the third depicted trace. C: CSD analysis of spontaneous SPWs from the same slice after low-pass filtering at 80 Hz. Shown is the averaged field potential of 306 SPW-Rs on top of the corresponding heat map. Note that current source and sink follow the same pattern as in the stimulusinduced SPW-R CSD analysis shown in Fig. 2C with a strong current source in SP, followed by a weaker sink in SR. D: Four examples of CSDs and field potentials from randomly chosen spontaneous single SPW-R events.

are prone to oxidation in the highly oxygenated aCSF. To test more physiological serotonin effects, it was either co-applied (100 nM) with 10 mM citalopram, a selective serotonin re-uptake inhibitor (Hyttel, 1994), or by inducing the release of endogenous serotonin from serotonergic fibers using fenfluramine (10e20 mM) without a previous serotonin application to eventually fill 5-HT stores (Wojtowicz et al., 2009). 2.4. Stimulation protocols For every stimulation protocol, we used a bipolar platinum electrode (25 mm, tip separation: 100e150 mm) placed in the stratum radiatum (SR) of CA1. Population spike responses were evoked in CA1 or CA3 using submaximal (70%) stimulus intensity (1.5e3 V). LTP in area CA3 was induced by applying three consecutive tetani (400 ms; 100 Hz; 40-s interval) to SR in CA1.

SPW-Rs were induced with a high frequency stimulation (HFS) protocol consisting of three trains of 40 pulses with a 10 ms interval, 100 ms duration, and a 40 s intertrain interval which were repeated every 5 min up to 7 times. Paired pulses of intracellular recorded EPSPs were evoked with two consecutive pulses applied to the proximal SR of CA1 (inter-pulse interval: 50 ms, repeated every 40 or 50 s) with stimulation intensities between 1 - 2 V which was subthreshold for action potential generation in area CA1. In experiments where paired pulse stimulation was applied in the presence of bicuculline (5 mM), CA3 was removed to prevent the generation of epileptiform discharges. 2.5. Data analysis We analyzed different components of SPW-Rs by filtering the raw data using different digital filter functions of the Spike2

192

R. ul Haq et al. / Neuropharmacology 101 (2016) 188e203

Fig. 4. Effects of 5-HT on induction of LTP and SPW-Rs. Aa: HFS induces prominent LTP with nearly 250% in control experiments (n ¼ 7). The inset above indicates that a first antidromic population spike in area CA3 induced by stimulation of SR in area CA1 was not affected by the HFS stimulation protocol, while field EPSPs and secondary population spikes were strongly enhanced, indicative for associational LTP. Ab: A sample FP trace, showing the induction of SPW-Rs by HFS protocol. Note that SPW-Rs appeared after application of the 3rd HFS train. Ba&b: Similar experiment as in Aa&b in presence of serotonin. Note that LTP was slightly impaired and induction of SPW-Rs by HFS was prevented in 5-HT treated slices. Note that SPW-Rs appeared after washout of serotonin and LTP amplitude was also increased. C: Analog to A & B, the effect of 5-HT on stimulus induction of LTP and SPW-Rs in presence of the 5-HT1A receptor antagonist NAN-190 (10 mM). Note that NAN-190 antagonized the 5-HT mediated effects on induction of LTP and SPW-Rs. D: Comparison of properties of SPW-Rs which appeared after washout of 5-HT with those induced in absence of 5-HT or in presence of NAN-190 and 5-HT. Note that there are no significant differences among each condition with respect to incidence, duration, amplitude, or ripple frequency. Data are based on comparison of 7 experiments with 5-HT, 8 experiments without 5-HT, and 5 experiments in presence of NAN-190 and 5-HT.

software (Cambridge Electronic Design, Cambridge, UK). For ripple detection, we used a band pass filter of 95e400 Hz. Ripple frequency was determined from intervals between ripple maxima using custom-made software (Maier et al., 2003). For sharp wave detection, we low pass filtered the recordings at 80 Hz. For the analysis of SPW-R amplitude and duration, 15 consecutive events were analyzed for each condition from each slice. Current source density analysis was based on simultaneous recordings of field potentials at 16 positions (50 mm inter electrode distance) vertically arranged to stratum pyramidale. Recordings with the employed electrode ensemble were accepted when the electrode trips remained separated during the recordings as recognized by visual inspection and by analyzing the voltage differences between adjacent electrode positions. Current source density (CSD) was calculated with a custom written script that calculates the second spatial derivative using procedures described previously (Nicholson and Freeman, 1975; Dietzel et al., 1989; Gabriel et al., 1998). In order to test for a possible presynaptic mechanism underlying serotonin-mediated SPW-R modulation, we investigated the paired pulse ratio (PPR) and the coefficient of variance (CV) of evoked EPSPs (Faber and Korn, 1991). We calculated the PPR by dividing the amplitude of the second evoked EPSP by that of the first evoked

EPSP. For comparison of the PPR before and after drug application, we averaged 20 values for each condition per cell. In order to analyze the coefficients of variance, we analyzed data of evoked EPSPs before and after drug application. In summary, the ratio of the coefficient of variance squared, ‘r’, was plotted against the modification factor ‘p’ where r ¼ CV2 before drug application/CV2 after drug application, and p ¼ M after drug application/M before drug application. CV ¼ standard deviation of the amplitude of evoked EPSPs/mean amplitude of evoked EPSPs, ‘M’ ¼ mean amplitude of evoked EPSPs (Faber and Korn, 1991). All data were reported as mean ± standard error of mean (SEM). Statistical significance was determined by using paired Student's t-test, one way Anova, and Wilcoxon signed rank test. p < 0.05 (*) was considered to indicate significant difference. 3. Results 3.1. Induction of SPW-Rs One of the mechanisms underlying memory formation is long lasting LTP in select synapses, which will affect the interaction between a set of neurons and thereby form a neuronal ensemble which when activated can replay stored information during sharp

R. ul Haq et al. / Neuropharmacology 101 (2016) 188e203

193

Fig. 5. Effects of serotonin on stimulus induced SPW-Rs. Aa: Example of field potential (FP) recording obtained from area CA3 showing the effect of 30 mM 5-HT. Note the rather abrupt suppression of SPW-Rs shortly after 5-HT application. Ab and Ac: SPW-R activity on different time scales. Note that SPW-R activity was not replaced by theta or gamma activity. B, C and D: Dose-dependent effects of 5-HT. Note that co-application of the serotonin selective re-uptake inhibitor citalopram with 5-HT resulted in a complete blockade of network activity already with 100 nM. E: Sample recordings of SPW-Rs in CA3 showing an increase in intrinsic 5-HT release in hippocampal slices by fenfluramine (10 mM) causes sudden suppression of SPW-R activity. F: Summary plot showing dose-dependence and reversibility of 5-HT effects, and effects on SPW-R incidence of 5-HT combined with citalopram.

Table 1 Property of stimulus induced SPW-Rs. Comparison of different properties of SPW-Rs under control condition, during application of different drugs and their wash out in area CA3 and CA1. Amplitude of SPW-Rs (mV) Control 5-HT 30 mM; n ¼ 11 5-HT 10 mM; n ¼ 11 5-HT 3 mM; n ¼ 6 Citalopram 10 mM þ 5-HT 100 nM; n ¼ 7 Fenfluramine 10 mM; n¼8 5-HT 10 mM þ NAN-190 10 mM; n ¼ 7 8-OH-DPAT; n ¼ 8

a Me-5-HT 20 mM; n ¼ 7 SR 57227A 1 mM; n ¼ 5 Cisapride 30 mM; n ¼ 4

CA3 CA1 CA3 CA1 CA3 CA1 CA3 CA1 CA3 CA1 CA3 CA1 CA3 CA1 CA3 CA1 CA3 CA1 CA3 CA1

3.7 1.9 3.81 1.73 3.16 1.45 2.7 1.56 3.11 1.6 3.12 1.39 3.05 1.55 3.65 1.97 3.64 1.94 3.29 1.59

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

0.11 0.09 0.11 0.11 0.18 0.07 0.16 0.15 0.12 0.1 0.11 0.13 0.07 0.07 0.10 0.09 0.07 0.07 0.1 0.13

Duration of SPW-Rs (ms)

Wash in

Washout

e e e e 2.22 0.86 e e e e 3.27 1.37 e e e e 3.51 1.91 3.35 1.53

3.4 1.8 3.85 1.69 3.39 1.37 2.8 1.55 e e 3.39 1.35 e e e e e e e e

± 0.1* ± 0.1*

± 0.12 ± 0.16

± ± ± ±

0.08 0.07 0.11 0.11

± ± ± ± ± ± ± ±

0.13 0.06 0.1 0.09 0.21 0.07 0.15 0.19

± 0.08 ± 0.16

Control 53.9 42.9 55.3 43.3 54.9 41.7 53.1 43.6 56.6 43.9 54.7 42.7 52.0 44.5 54.6 44.6 53.4 44.13 59.82 44.5

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

1.8 1.2 1.6 1.5 1.6 1.6 1.8 1.0 1.7 1.2 1.6 1.5 2.0 1.6 1.3 1.0 1.4 1.7 1.6 1.4

Ripple frequency (Hz)

Wash in

Washout

e e e e 51.9 40.3 e e e e 55.8 38.8 e e e e 52.97 43.3 58.1 45.5

53.7 43.5 53.2 40.4 51.6 39.7 51.9 41.6 e e 53.8 40.3 e e e e e e e e

± 1.6 ± 1.8

± 2.0 ± 1.4

± ± ± ±

1.9 1.3 1.6 1.5

± ± ± ± ± ± ± ±

1.6 1.1 1.6 1.6 2.0 1.5 1.7 1.5

± 1.9 ± 1.4

Control 176.2 172.5 168.8 164.5 176.4 170.7 173.4 167.9 174.3 170.9 183.7 172.3 171.8 168.6 170.4 160.0 175.5 170.5 174.4 169.9

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

3.4 4.1 2.3 2.6 3.0 3.8 4.5 2.8 4.7 4.1 4.4 4.1 5.1 4.3 5.0 4.4 3.8 3.9 3.9 2.6

Wash in

Washout

e e e e 175.1 170.0 e e e e 179.2 172.9 e e e e 176.7 170.6 173.2 170.8

175.4 171.6 166.3 162.5 174.9 169.7 168.4 166.2 e e 178.6 172.0 e e e e e e e e

± 3.9 ± 4.0

± 3.6 ± 3.8

± ± ± ±

5.3 4.8 3.1 2.7

± ± ± ± ± ± ± ±

3.1 3.9 3.4 2.6 4.8 3.6 4.7 2.6

± 4.1 ± 3.8

*: p < 0.05.

wave ripples (SPW-Rs). High frequency stimulation (HFS) of the stratum radiatum (SR) of area CA1 induces stable associational LTP in area CA3 (Huang and Kandel, 1995; Behrens et al., 2005; Ul Haq et al., 2012) of rat hippocampal slices. Interestingly, repetition of this stimulation protocol can induce the appearance of SPW-Rs in CA3 from where they propagate into CA1 (Fig. 1). SPW-R incidence was identical in both regions, representing an average of 9.3 ± 0.5 SPW-Rs per min (n ¼ 66 slices, Fig. 1Ca). In area CA3, SPW-Rs lasted on average 53.7 ± 1.9 ms, and in CA1 43.4 ± 1.3 ms

(n ¼ 66 slices, p < 0.05, Fig. 1Cb). Mean SPW-R amplitude in CA3 was 3.3 ± 0.13 mV, which was significantly higher than in area CA1 where SPW-Rs showed an average amplitude of 1.7 ± 0.2 mV (n ¼ 66 slices, p < 0.05). Ripple oscillations, which are superimposed on SPWs, showed a similar mean frequency in both areas, 174.5 ± 2.8 Hz in CA3, and 172.8 ± 2.4 Hz in CA1 (Fig. 1Cd). These values compare well to previous reports (Behrens et al., 2005; Ul Haq et al., 2012). Fig. 2 shows the laminar profile of stimulusinduced SPW-R complexes with a 16 channel probe placed as

194

R. ul Haq et al. / Neuropharmacology 101 (2016) 188e203

indicated in Fig. 2A. Fig. 2B depicts the original field potentials (FPs) of every second electrode of a typical SPW-R complex. Lowpass filtering with 80 Hz and averaging the obtained sharp waves results in the map depicted in Fig. 2C, where the averaged FPs are laid over a color coded heat map of current source density (CSD). The CSD heat map indicates a current sink in stratum radiatum (SR) and a large source in stratum pyramidale (SP) accompanied by a second smaller source in stratum laconosum moleculare (SLM), similar to previous observations on FP profiles in vivo (Buzsaki et al., 1983). Importantly, the current source in stratum pyramidale can precede the current sink in stratum radiatum by 1e2 ms. This presumably points towards neuronal inhibition which precedes synaptic excitation. Fig. 2D depicts four arbitrarily chosen recordings of individual sharp waves with their respective CSD profiles. These 4 sample recordings indicate that the weak source in SLM is not constant and that sometimes the sink in stratum radiatum is segregated into two sinks which are spatially and temporally segregated. It must be noted that the source in stratum pyramidale in these recordings does not always precede the sink in stratum radiatum. In murine hippocampal slices, SPW-Rs appear spontaneously without electrical stimulation. In order to validate the stimulus induced SPW-R model, we compared the FPs and CSDs between stimulus-induced rat and spontaneous murine SPW-Rs. Depicted in Fig. 3 are spontaneous murine SPW-Rs, recorded in hippocampal area CA3. They present a large variability in amplitude and duration and appear with a much higher incidence than the stimulus-induced ones in rat hippocampus. Spontaneous SPW-Rs occurred in CA3 with a frequency of 2.8 ± 0.06 per second. Their amplitude was on average 0.25 ± 0.07 mV and the mean ripple frequency 199.3 ± 1.7 Hz. Summarized over 7 slices from 3 mice, we noted a mean SPW amplitude in stratum pyramidale of 0.15 ± 0.1 mV, an average incidence of 3.1 ± 1.36 events per second with an average ripple frequency of 208.8 ± 15.7 Hz (n ¼ 7 slices). As in rat, sharp waves are positive in stratum pyramidale and

negative in stratum radiatum (Fig. 3B and C). Murine hippocampal slices are smaller than rat hippocampal slices. Differences in ripple frequency may result from differences in the distances axons are required to span in order to integrate other neurons into the ensemble activity or from differences in signal noise ratio. The averaged laminar profile and calculated CSD profile are depicted in Fig. 3C. As is the case with rat hippocampus, also in these murine slice recordings there is a current source in pyramidal cell layer preceding the current sink in stratum radiatum. Fig. 3D illustrates 4 arbitrarily chosen laminar profiles from the same experiment indicating a strong current source in stratum pyramidale usually preceding the current sinks in stratum radiatum by 1e2 ms. Fig. 3D further indicates a strong variability in the strength of the current sink for different events with sometimes segregated current sinks in stratum radiatum. 3.2. Serotonin prevents the expression of SPW-Rs but not of LTP We assumed that serotonin would not interfere with the induction of long lasting LTP in CA3 but might prevent the appearance of SPW-Rs, since serotonin levels during wakefulness are much higher than during sleep. In order to test this, we applied high levels of serotonin (30 mM) before the HFS protocol. In these experiments, SPW-Rs did not appear even when stimulus intensity was supra-maximal for the induction of population spikes in area CA3 (Fig. 4; n ¼ 7). Also, SPW-Rs would not appear if the number of HFS repetitions was increased to more than seven repetitions, thus extending beyond the about three to four repetitions sufficient to induce stable SPW-Rs. The induction of long lasting LTP could be impaired and therefore we compared LTP induction in absence and in presence of 5-HT. We used three stimulus trains of the kind which we also used for SPW-R induction and which causes almost saturating LTP under control conditions. We found that LTP could readily be induced in the presence of 5-HT, although the augmentation of stimulus-induced responses typical for induced LTP was

Fig. 6. 5-HT mediated suppression of SPW-Rs in murine hippocampal slices. Aa: Representative FP recording of spontaneous SPW-R activity from murine hippocampal slices in the area CA1 showing that 5-HT (50 mM) reversibly blocked SPW-Rs. Ab: SPW-R sample traces shown on an expanded time scale under control condition (left), in presence of serotonin (middle) and after 30 min wash out (right). Ac: Traces of a single SPW-R event (black trace) with filtered data (gray trace) before, during, and after serotonin application. B: Analog to A, showing the effects of fenfluramine (20 mM) on SPW-Rs in area CA1 of murine hippocampal slices.

R. ul Haq et al. / Neuropharmacology 101 (2016) 188e203

195

Fig. 7. Suppression of SPW-R activity is mediated by 5-HT1A & 2A/C receptors. Aa: A representative FP recording of SPW-R activity in area CA3 and CA1 shows 5-HT1A antagonist NAN-190 (10 mM) prevented the blockade of SPW-Rs by serotonin. Ab: Sample traces depicting SPW-R activity on an expanded time scale under control condition (left) and in presence of NAN-190 co-applied with 10 mM 5-HT (right). Ba: 5-HT1A receptor agonist 8-OH-DPAT (1 mM) blocked stimulus-induced SPW-Rs. Bb: Sample recordings of SPW-Rs taken from Ba on an expanded time scale under control conditions (left) and in the presence of 8-OH-DPAT (right). Ca & b: Analog to B, 5-HT2A/C receptor agonist a methyl serotonin (20 mM) suppressed SPW-R activity. Da: Representative FP recordings of SPW-R activity in area CA3 and CA1 showing that the 5-HT3 agonist SR 57227A (1 mM) had no significant effects on SPW-Rs. Db: Sample traces depicting SPW-R activity on an expanded time scale under control condition (left) and in the presence of SR 57227A (right). Ea: Representative FP recordings of SPW-R activity in area CA3 showing that NAN-190 (10 mM) did not affect sharp wave ripple activity but prevented the blocking effect of fenfluramine (10 mM). Eb: Sample recordings of SPW-Rs before and after application of the cocktail.

196

R. ul Haq et al. / Neuropharmacology 101 (2016) 188e203

R. ul Haq et al. / Neuropharmacology 101 (2016) 188e203

somewhat smaller (194.6 ± 0.8%; n ¼ 6, Fig. 4Ab) than that induced by recurrent stimulation in absence of 5-HT (249.1 ± 1.4%; n ¼ 7 Fig. 4Aa). In fact, near threshold stimulation for LTP induction in area CA3 was previously shown to be prevented by serotonin if the stimulation was weak, but could be induced by applying stronger stimulation (Villani and Johnston, 1993). Interestingly, when 5-HT was washed out, the amplitude of evoked responses increased further, suggesting that LTP was at least partially masked by inhibitory effects of 5-HT (Fig. 4B). Repetitive HFS did not induce SPW-Rs in the presence of serotonin, but SPW-Rs appeared after washout of 5-HT e both in area CA3 and CA1. These SPW-Rs had similar characteristics compared to those induced under control conditions without any prior 5-HT application (Fig. 4D; n ¼ 7; p > 0.05). As a control, we repeated the experiment in the presence of the 5-HT1A receptor antagonist NAN-190. When 5-HT (30 mM) was applied in the presence of NAN190 (10 mM), we could not only induce LTP (Fig. 4Ca), but also SPWRs in the presence of 5-HT (Fig. 4Cb). The properties of these SPWRs were well comparable to those induced under control conditions as indicated in Fig. 4D.

197

alterations in afterhyperpolarization and also in neuronal input resistance comparable to effects of serotonin itself (Segal, 1990), which also compares well to studies on gamma oscillations with 50 mM serotonin (Wojtowicz et al., 2009). In our experiments, fenfluramine (10 mM) completely masked stimulus-induced SPWRs in area CA3 and CA1 (Fig. 5E; n ¼ 8) beginning 1002.7 ± 132.8 s after application of fenfluramine. Next, we tested whether serotonin also masks spontaneous SPW-Rs in murine hippocampal slices. As SPW-Rs are often very small in area CA3, we focused on recordings in area CA1. With 30 mM 5-HT we observed a suppression of SPW-Rs in 4 out of 9 slices, while we observed complete suppression with 50 mM (Fig. 6). Application of the 5-HT releasing agent fenfluramine (20 mM) suppressed SPW-R activity completely (Fig. 6). The difference in sensitivity to 5-HT may indicate a difference between mice and rats regarding uptake and metabolism of 5-HT and the density of 5-HT containing fibers as well as expression of 5-HT receptors.

3.4. SPW-R suppression is mediated by 5-HT1A and 5-HT2A/C receptors

3.3. Serotonin mediated masking of SPW-Rs How serotonin can mask the appearance of SPW-Rs, we elaborated by testing the effects of 5-HT on stimulus-induced SPW-Rs as indicated in Fig. 5. Serotonin (30 mM) abruptly and reversibly masked stimulus-induced SPW-Rs in rat hippocampal slices, both in area CA3 and CA1 (n ¼ 11). Complete SPW-R suppression took 380 ± 33.9 s (Fig. 5A). When only 10 mM serotonin were applied, suppression was also reversible (Fig. 5B; n ¼ 11) and its latency was not significantly different from 30 mM 5-HT, suggesting that 10 mM may be the saturating concentration. In another set of experiments, serotonin (5 mM) also concealed SPW-Rs (n ¼ 3, data not shown). 5HT application in a relatively low concentration (3 mM, n ¼ 6) did not completely mask SPW-R activity except for one experiment (Fig. 5C). However, it significantly attenuated network activity as the SPW-R incidence was reduced from 10.3 ± 0.49 to 3.8 ± 0.42 SPW-Rs per min (Fig. 5C and F; n ¼ 6; p ¼ 0.00001). Similarly, with the same concentration, SPW-R amplitude was also reduced significantly in both CA3 and CA1 (Table 1; p ¼ 0.0009 and 0.00002). It thus seems that the applied concentrations of serotonin for SPW-R suppression are rather high, perhaps due to reuptake and degradation of the drug within a slice. We therefore determined the effects of 5-HT in presence of the selective serotonin re-uptake inhibitor citalopram. Importantly, in the presence of 10 mM citalopram, 5-HT reversibly masked SPW-Rs already with 100 nM (Fig. 5D and F; n ¼ 7). After serotonin washout, SPW-Rs reappeared in a similar abrupt manner. The incidence of SPW-Rs was similar before and after the blockade. SPW-R properties regarding amplitude, duration and ripple frequency, were not significantly different before and after serotonin application (for each condition p > 0.05; Table 1). We next tested whether intrinsic serotonin can lead to SPW-R masking. In the past, fenfluramine has been shown to release serotonin from presynaptic terminals of serotonergic fibers causing hyperpolarization of neurons,

Serotonin interacts with various receptors and we investigated which receptors are involved in 5-HT-mediated suppression of SPW-Rs. When serotonin was applied in presence of the antagonist of the 5-HT1A receptor, NAN-190 (10 mM), serotonin no longer concealed SPW-Rs. This was seen in 7 experiments (Fig. 7A). Application of the 5-HT1A agonist 8-OH-DPAT ((±)-8-Hydroxy-2(dipropylamino) tetralin hydrobromide) in concentrations of 1 or 2 mM lead to suppression of SPW-Rs, suggesting that the effect of serotonin is mediated by 5-HT1A receptors (Fig. 7B, n ¼ 8). 8-0HDPAT is known to hyperpolarize pyramidal cells and certain interneurons (Andrade and Nicoll, 1987; Schmitz et al., 1995b; Johnston et al., 2014). The 5-HT2A/C agonist alpha-methylserotonin (aMe 5-HT 20 mM) was also capable of suppressing SPW-Rs (Fig. 7C, n ¼ 7), possibly by hyperpolarizing pyramidal cells (Uneyama et al., 1993). The 5-HT3 agonist SR-57227A is known to activate interneurons (Sudweeks et al., 2002). SR-57227A (1 mM) did not result in any major changes of SPW-R properties such as incidence, ripple frequency, duration, or amplitude (n ¼ 5; P ¼ 0.13, 0.86, 0.82 and 0.19 respectively; Fig. 7D; Table 1). Furthermore, application of the 5HT4 agonist cisapride (30 mM) caused a small but insignificant increase in SPW-R incidence from 9.1 ± 0.43 to 10.2 ± 1.3 SPW-Rs/min (n ¼ 4; p ¼ 0.097). The involvement of 5-HT1A receptors is further supported by the finding that the suppressive effects of fenfluramine (10 or 20 mM, n ¼ 6) were prevented by prior application of NAN-190 (Fig. 7E). The effect was somewhat weaker in murine hippocampal slices expressing spontaneous SPW-Rs, where higher concentrations of fenfluramine were needed. In 2 slices however, the effect of fenfluramine was strongly reduced by NAN-190 as well (data not shown). These findings suggest that serotonin-mediated SPW-Rs suppression is mediated via activation of 5-HT1A and/or 5-HT2A/C receptors.

Fig. 8. 5-HTemediated suppression of SPW-R activity precedes hyperpolarization of CA3 pyramidal cells. A: Simultaneous extra- and intracellular recordings of SPW-R activity before and after application of 5-HT. Note that, as indicated in inset, postsynaptic membrane hyperpolarization occurred after cessation of SPW-R activity. B: Blocking effect of 5-HT on extracellular SPW-Rs and on complex activity during SPW-Rs in a participating cell. Note that the participating cell activated 2 action potentials (APs) before and after SPW-R blockade on top of the complex series of EPSPs during SPW-R activity. C: Recordings from another cell which did not fire APs during SPW-R activity. Note the initial hyperpolarization during the SPW-R complex, followed by a compound EPSP. Insets represent filtered activity on an expanded time scale as indicated by squares. D: Graphical representation of changes in resting membrane potential (Da) and input resistance (Db) of CA3 pyramidal neurons by different doses of 5-HT. Note that 5-HT significantly affected resting membrane potential and input resistance only with higher concentrations, and that 100 nM 5-HT co-applied with citalopram had no effect on resting membrane potential and input resistance of CA3 pyramidal cells.

198

R. ul Haq et al. / Neuropharmacology 101 (2016) 188e203

R. ul Haq et al. / Neuropharmacology 101 (2016) 188e203

199

Fig. 10. Serotonin-mediated augmentation of paired-pulse ratio via activation of 5-HT1A receptors. A: Responses induced by activation of Schaffer collaterals and commissural fibers with a 50 m stimulus interval, pulse width 100 ms. Aa: Top: Sample recording of paired evoked EPSPs obtained from a CA1 pyramidal cell under control condition (left) and in presence of 5-HT (right); bottom: time course of serotonin effects on paired-pulse ratio (PPR) under control condition and in presence of 5-HT, averaged over 8 neurons. Note that 5HT application significantly increases PPR. Ab: Coefficient of variance analysis of single evoked EPSPs in CA1 neurons pointing to a presynaptic action of 5-HT. Ba & b: Same analyses based on recordings in 7 CA1 pyramidal cells of the effects of 5-OH-DPAT. C: Effects of 5-HT on stimulus evoked EPSPs in presence of the GABAA receptor antagonist bicuculline (5 mM). Note that analyses of paired pulse behavior and coefficient of variance yield similar results as in case of preserved inhibition. Experiments in C were performed after removal of CA3 in order to prevent spontaneous epileptiform discharges.

3.5. Stability of SPW-R activity Although SPW-Rs can be masked by 5-HT, their properties recover after 5-HT wash out. The recordings from rat hippocampal slices presented in Fig. 8B and C shows that the activity of a given neuron was similar before and after serotonin application. As demonstrated in Fig. 8B, some cells displayed compound EPSPs with evoked action potentials before and after 5-HT application. On the other hand, some cells did not fire any action potentials and synaptic activity in these cells was characterized by an initial hyperpolarization followed by a depolarization subthreshold for action potential generation (Fig. 8C). The experiment illustrated in Fig. 9A shows two simultaneously recorded cells in area CA3 which

participate in the generation of SPW-Rs. Both cells fired action potentials, however, at different times during the course of SPW-Rs and thus can be identified as members of one neuronal ensemble formed by previous induction of associative LTP in area CA3. This supports the notion that the composed activity underlying and accompanying SPW-Rs is due to sequential activation of neurons responsible for the generation of SPW-Rs. After recovery from SPWR masking, both cells still presented the same number of action potentials during SPW-Rs as before application of serotonin, indicating their role in the generation of network activity remained mostly unchanged. In the particular experiment of Fig. 9B, both cells were inactive during SPW-R activity. However, one cell showed subthreshold compound EPSPs while the other showed a

Fig. 9. Simultaneous recording of SPW-Rs and paired recordings of CA3 pyramidal cells before and after 5-HT application. Aa: Comparison of paired recordings from CA3 pyramidal cells before and after SPW-R 5-HT mediated suppression (top: FP recording of SPW-Rs, middle: neuron one; bottom: neuron two). Ab: Example of responses during single SPW-Rs taken from Aa. Note that the simultaneous recording from two cells indicates that action potentials come at different time points during the complex EPSP, suggesting sequential activation of neurons during a SPW-R complex. Ba: Simultaneous recordings of two cells during SPW-R activity which did not participate in SPW-Rs but received mixed inhibitory and excitatory input (cell 1), and excitatory input below SPW-R induction threshold (cell 2). Bb: Analog to Ab. Note the similarity of responses before and after washout of 5-HT.

200

R. ul Haq et al. / Neuropharmacology 101 (2016) 188e203

compound EPSP that was preceded by an inhibitory potential. Following serotonin suppression, both cells showed the same type of activity as before serotonin application. Similar observations were made in all other 26 recorded cells. Of those, 11 cells were actively involved in SPW-R generation before serotonin application as indicated by action potential firing during SPW-Rs, and displayed similar activity after serotonin wash out (Figs. 8B and 9A). Similarly, the remaining 15 cells which were not actively participating in the generation of SPW-Rs, remained silent after serotonin wash out (Figs. 8C and 9B). 3.6. SPW-R masking does not primarily depend on hyperpolarization and/or changes in input resistance in pyramidal cells Simultaneous intracellular recordings revealed that 10 and 30 mM serotonin caused a transient and moderate hyperpolarization of the resting membrane potential (Vrmp) in CA3 pyramidal cells. 10 mM serotonin changed the Vrmp from 64.1 ± 0.5 mV to 70.2 ± 1.7 mV, and 30 mM from 64.9 ± 0.5 mV to 71.5 ± 1.1 mV (Fig. 8A and D; n ¼ 9 for each concentration; p ¼ 0.0057 and 0.00043 respectively). These serotonin concentrations similarly caused a significant reduction in the input resistance (Ri) of CA3 pyramidal neurons from 41.3 ± 1.7 to 29.7 ± 1.7 MU, and from 46.7 ± 1.9 to 33.9 ± 1.6 MU, respectively (Fig. 8D; n ¼ 9 for each concentration; p ¼ 0.000044 and 0.000007 respectively). 10 & 30 mM serotonin mediated effects on Vrmp and input resistance were reversible after washout (Fig. 8D; n ¼ 6 for each condition; p > 0.05 control vs. 5-HT washout). This compares well to previous publications (Segal, 1980; Sokolova et al., 1998; Schmitz et al., 1995b) and suggests that potential changes in postsynaptic membrane properties might underlie SPW-R masking. Therefore, we compared the time required for 5-HT mediated SPW-R masking with the time required for membrane potential and resistance changes. Importantly, we found that the time required for SPW-R disappearance is significantly shorter than the time needed for the onset of Vrmp hyperpolarization, which is true for all recorded CA3 neurons including those in which we did simultaneous recordings of two cells. This is illustrated in the inset of Fig. 8A. SPW-Rs were terminated 399 ± 32.4 s and 380 ± 33.5 s after 10 and 30 mM 5-HT application, respectively, while membrane potentials started to hyperpolarize significantly after 553.9 ± 32.8 s and 507.1 ± 37.8 s, respectively (Fig. 8; n ¼ 9 for each concentration; p ¼ 0.001 for 10 mM and 0.025 for 30 mM 5-HT). This suggests that SPW-R disappearance was not primarily depending on pyramidal cell hyperpolarization. Combined application of 5-HT (100 nM) with the serotonin re-uptake inhibitor citalopram revealed no significant changes in membrane potential or membrane resistance, suggesting that SPW-R masking is not caused by hyperpolarization and increased membrane conductance (Fig. 8D; n ¼ 5; p ¼ 0.22 and 0.77 for Vrmp and Ri respectively). 3.7. Serotonin affects presynaptic function The present findings suggest that SPW-R suppression was not necessarily linked to postsynaptic hyperpolarization of pyramidal cells. Instead, it may be due to reduced transmitter release and therefore reduced neuronal synchronization. Indeed, it was previously shown that agents which reduce presynaptic Ca2þ entry into CA3 pyramidal cell terminals block SPW-Rs. This has been shown for adenosine (Maier et al., 2011a) and for a adrenergic receptor agonists (Ul Haq et al., 2012) as well as for the GABAB receptor agonist baclofen (Hollnagel et al., 2014). In addition, 5-HT reduces presynaptic Ca2þ entry through 5-HT1A receptors (Schmitz et al., 1995a). Therefore, serotonin might reduce excitatory coupling

within a neuronal ensemble interconnected through axon collaterals and could provide moderate disconnection of neurons within an activated ensemble, consequently suppressing the appearance of SPW-Rs. Alterations of presynaptic Ca2þ entry usually lead to modified paired pulse behavior. We evoked responses from stratum radiatum in area CA1 by activating Schaffer collaterals and commissural fibers which originate from CA3 pyramidal cells. Their branched axonal fibers interconnect neurons in area CA3. We thus avoided activation of other inputs onto CA3 cells, as would be the case if stimulation would have been applied to stratum radiatum in CA3. Additionally, one set of paired pulse experiments required the blockade of GABAergic transmission which would have induced epileptiform discharges in area CA3. A typical experiment is shown in Fig. 10A. It indicates that 5-HT reduces the first stimulus evoked intracellular response which increases paired pulse index (Fig. 10A; n ¼ 8; p ¼ 0.00038). The effect was identical for 8-OH-DPAT (Fig. 10B; n ¼ 7; p ¼ 0.021). In addition, we calculated the coefficient of variance of evoked synaptic potentials and found that both 5-HT and 8-OH-DPAT affected evoked potentials indicative of a presynaptic effect. In the initial experiments we did not block GABAergic potentials. We therefore performed a series of experiments where we applied 5 mM bicuculline after CA3 removal which was previously shown to block phasic inhibition under our experimental conditions (Liotta et al., 2011). These results are depicted in Fig. 10C and support our conclusion of a presynaptic 5-HT effect. As shown with intact inhibition, the first evoked response was reduced and the paired pulse index was increased (n ¼ 9; p ¼ 0.0011). Also, the coefficient of variance was similarly affected as under control conditions without bicuculline application. 4. Discussion and conclusion Our study shows that serotonin can mask SPW-Rs both in murine and rat acute hippocampal slices. SPW-Rs are neuronal events important for memory formation and consolidation. It is widely believed that during SPW-Rs, previously activated pyramidal cells are sequentially reactivated in a temporally condensed manner (Diba and Buzsaki, 2008; Foster and Wilson, 2006). In this sense, ripples seem to reflect the sequential activation of neurons. Recent optogenetic experiments in area CA1 in vivo suggest ripple oscillations depend on nearby pyramidal cell interactions which disappear when they are hyperpolarized, or when parvalbumin positive interneurons are activated (Stark et al., 2014). In area CA1, focal activation of interneurons seems to be important for the pacing of pyramidal cell firing (Stark et al., 2014), while recent optogenetic and pharmacological intervention in area CA3 of murine hippocampal slices suggested activation of parvalbumin containing interneurons might be more crucial for SPW-R generation (Schlingloff et al., 2014). An important role for GABAergic cells would be expected as most neurons should be inhibited when the cells involved in memory formation are reactivated during consummatory behavior and slow wave sleep. The interplay between inhibitory and excitatory synaptic transmission in SPW-R generation is further elucidated by focal layer-specific blocking (Schonberger et al., 2014). Interestingly, Serotonin is capable of reducing both excitatory and inhibitory synaptic coupling (Schmitz et al., 1995a, 1995b). If explicit memories are generated by strengthening individual synaptic contacts between hippocampal neurons, and if their activity is replayed during SPW-Rs in a temporarily compressed manner, the induction of LTP would be a prerequisite for memory formation. With our model of stimulus induced SPW-Rs, we confirm that late LTP induction protocols (Reymann et al., 1985; Huang and Kandel, 1994) can also induce SPW-Rs in area CA3.

R. ul Haq et al. / Neuropharmacology 101 (2016) 188e203

The properties of these stimulus induced SPW-Rs compare well with spontaneous SPW-Rs recorded in murine hippocampal slices. They are blocked by CNQX, suggesting excitatory interaction between neurons (Behrens et al., 2005), and can also be blocked by GABAA receptor antagonists (Behrens et al., 2007). Laminar profiles and CSD distributions are very similar between murine and rat SPW-Rs and can be also well compared to activity recorded in vivo (Buzsaki et al., 1983). The current source in stratum pyramidale often precedes the sharp wave associated sink in stratum radiatum. This is in line with an important function during SPW-R generation of perisomatic inhibitory cells (Stark et al., 2014; Schlingloff et al., 2014; Maier et al., 2011b). While LTP, strong stimulation provided, can be induced in presence of serotonin (Villani and Johnston, 1993; Kemp and Manahan-Vaughan, 2005), SPW-Rs appear only after serotonin washout. This indicates a modification of the network during LTP induction by serotonin which permits the network to generate SPW-Rs after washout. The delayed SPW-R appearance is due to their masking by serotonin in a dose dependent manner, which compares well to effects on sharp waves in in vivo hippocampal EEG recordings (Richter-Levin and Segal, 1990). The effect was dependent on 5-HT1A or 5-HT2A/C receptor activation but insensitive to activation of excitatory 5-HT3 receptors, which are expressed on subtypes of interneurons (Turner et al., 2004; Morales et al., 1996, 2007; Chittajallu et al., 2013). During SPW-Rs, many neurons are silent. This is expected if specific event related information is transmitted through the hippocampus. The rapid sequential activation of neurons during sharp wave ripples will ensure stabilization of a memory trace and thereby support memory consolidation. The larger amplitudes of SPW-Rs in rat hippocampal slices compared to those in murine slices (Maier et al., 2003) suggest that during stimulus-induced SPW-Rs more neurons are activated. This observation has been interpreted as pro-epileptic. In favor of this interpretation stands, that similar repeated high frequency stimulations in vivo can induce epileptic seizures. But there, stimulus intensities usually are much higher because they are adjusted to intensities above seizure induction threshold. When repeated, such stimulations induce secondarily generalized seizures (Goddard et al., 1969) where many neurons in wide parts of the brain fire synchronously. Also, in contrast to epileptiform discharges induced by GABA receptor or potassium channel blockade typically lasting between 80 and some 100 ms, pyramidal cells during stimulus induced SPW-Rs do not fire multiple action potentials but rather 1 or 2 if at all. The associated rises in [Kþ]o further differentiates stimulus induced SWP-Rs from epileptiform discharges (Behrens et al., 2007). Oxygen consumption is also much higher during epileptiform discharges than during SPW-Rs (Jarosch et al., 2015). One model of seizure induction relies on increasing extracellular potassium concentrations. But rather than being pro-epileptic, the threshold for seizure induction is higher in slices in which SPW-Rs have been induced compared to naïve slices (Liotta et al., 2011). SPW-R masking required rather high 5-HT concentrations. Active concentrations may be much lower than 5 mM, since slice equilibration with a given neurotransmitter or neuromodulator is a very slow process e particularly in interface chambers e governed by diffusional equilibration, uptake, and metabolism (Müller et al., 1988; Gloveli et al., 1995; Sokolova et al., 1998). In order to avoid 5HT uptake and degradation, we repeated the experiments in presence of 10 mM citalopram which permitted complete and reversible SPW-R masking already with 100 nM serotonin. Moreover, the serotonin releaser fenfluramine also masked SPW-Rs, indicating that it is likely mediated though physiological serotonin concentrations. Serotonergic masking of SPW-Rs seemed to be dependent on 5-

201

HT1A and 5-HT2A/C receptor activation as suggested by appropriate pharmacological tests (Rydelek-Fitzgerald et al., 1990; Paluchowska et al., 1999) but was little affected by a 5-HT4 receptor agonist. This is odd, considering experience from our lab regarding effects of such agents on hippocampal output synapses in the subiculum (Wawra et al., 2014). This may point to regional differences regarding synaptic modulation by serotonin. In accord with previous reports (Andrade and Nicoll, 1987; Beck and Choi, 1991), we noted strong and lasting hyperpolarization when 10 or 30 mM 5-HT were applied. However, hyperpolarization only began after SPW-R masking and may therefore not solely account for SPW-R suppression. Instead, co-application of 100 nM 5-HT with 10 mM citalopram masked SPW-Rs without affecting input resistance or membrane potential, suggesting SPW-R masking is not directly linked to an increase in pyramidal cell membrane conductance and/ or general hyperpolarization, but rather due to reduced synaptic coupling e perhaps mediated by a presynaptic effect. Previous studies indicate that 5-HT1A receptors mediate reduced presynaptic Ca2þ entry in area CA1 (Schmitz et al., 1995a). This is in line with the finding that 5-HT can interact with high threshold voltage activated Ca2þ channels via 5-HT1A receptor activation (Penington and Kelly, 1990). Paired pulse stimulation resulted in typical frequency potentiation which was augmented by 5-HT and by 8-OHDPAT. The effect persisted when paired pulse stimulation was studied in presence of 5 mM bicuculline. We additionally recorded paired pulses in area CA1 to study Schaffer collateral mediated synaptic transmission. Schaffer collaterals are the axon collaterals of CA3 pyramidal cells and likely possess the same properties as those fibers involved in associational interaction among CA3 pyramidal cells, as well as between CA3 pyramidal cells and inhibitory interneurons. In area CA1, 5-HT still augmented the paired pulse index, pointing to a presynaptic effect. Our conclusion is further supported by the analysis of the coefficient of variance of evoked EPSPs of CA1 pyramidal neurons which were similarly affected by serotonin in normal and in disinhibited slices, as well as by 8-0HDPAT. Norepinephrine (Ul Haq et al., 2012), adenosine, cannabinoids (Maier et al., 2011a), and the GABAB receptor agonist baclofen (Hollnagel et al., 2014) suppress SPW-Rs by impairing synaptic efficacy presumably due to reduced presynaptic Ca2þ entry. Thus it appears that all known agents able to reduce presynaptic transmitter release would be able to weaken the synchronization required for concerted neuronal activity necessary for SPW-R generation. This suggests that the strength of presynaptic Ca2þ influx, modulated by local and systemic neuromodulators, sets the likelihood of SPW-R ensemble activity. Interestingly, acetylcholine also suppresses SPW-Rs (Norimoto et al., 2012) while it depolarizes hippocampal pyramidal cells. Also in this case it is possible that suppression is mediated through alteration of presynaptic Ca2þ entry (Egorov et al., 1996). Certainly, serotonin, norepinephrine, and acetylcholine have different functions in animal behavior. They share however, that their release indicates a novel situation to the animal with the need for formation of new interactions between cells in the CNS We can not completely rule out the possibility that 5-HT suppresses SPW-Rs by affecting interneurons. Previous studies have indicated that 5-HT does not affect postsynaptic GABAergic currents. Thus, 5-HT mediated reduced inhibition was often due to a decrease of excitatory input to interneurons (Segal, 1990; Behr et al., 1997; Schmitz et al., 1998c, 1998a) with a particular action on cholecystokinin expressing interneurons (Winterer et al., 2011). However, 5HT also hyperpolarized some interneurons (Schmitz et al., 1995b) and could thereby interfere with the formation of neuronal ensemble activity between interneurons and pyramidal cells (Maier et al., 2011c; Pangalos et al., 2013; Stark et al., 2014; Schlingloff et al., 2014). Previous reports showed that intraventricularly applied 5-HT1A

202

R. ul Haq et al. / Neuropharmacology 101 (2016) 188e203

receptor antagonists reduced ripple activity while 5-HT3 receptor antagonists increase the number of ripples (Ponomarenko et al., 2003). The applied concentrations were very high, reaching 80e100 mM, permitting unspecific effects. The intraventricularly injected antagonists spread through the ventricles and also affect other brain regions including respiratory centers in the brain stem, thereby changing levels of other neuromodulators, oxygen tension etc. Control experiments with fenfluramine were not carried out. Moreover, the Ponomarenko study depended solely on ripple analysis which can occur independently of SPWs (Draguhn et al., 1998; Schmitz et al., 2001). Serotonin acts on different receptors involved in different signaling cascades. Hence, there are different ways how learning and memory can be modulated by 5-HT (Asin and Fibiger, 1984; Fletcher et al., 1999). In early experiments it was shown that lesions of the raphe nuclei have profound effects on learning, particularly in tasks which require spatial orientation learning. This defect could be reverted by grafting serotonergic neurons into the hippocampus (Richter-Levin and Segal, 1989). Here we can show that at least in hippocampal areas CA3 and CA1 serotonin transiently masks SPW-Rs. 5-HT1A receptors seem to play an important role in these processes as indicated previously (Sarnyai et al., 2000; Meneses and Liy-Salmeron, 2012). Our findings do not only concern the modulation of memory formation. They are also relevant for the treatment of major depression since serotonin re-uptake inhibitors are used as treatment. During slow wave sleep, where serotonin levels are very low, the effects of a re-uptake inhibitor may wane, permitting SPW-R generation and the consolidation of new memories. This might be an underlying mechanism to alterations in cognitive functions in depressed patients. Conflict of interest The authors declare no conflict of interest. Acknowledgments This research was supported by DFG grant He 1128-16-2 and 17.1 and by the Bernstein Focus on Learning grant from the BMBF. RUH was supported by a grant from the HEC in Pakistan. We are grateful for helpful comments to Dr. N Maier (Berlin) as well as to Prof. I Mody (UCLA, Los Angeles). References Aghajanian, G.K., Rosecrans, J.A., Sheard, M.H., 1967. Serotonin: release in the forebrain by stimulation of midbrain raphe. Science 156, 402e403. Andrade, R., Nicoll, R.A., 1987. Pharmacologically distinct actions of serotonin on single pyramidal neurones of the rat hippocampus recorded in vitro. J. Physiol. (Lond) 394, 99e124. Asin, K.E., Fibiger, H.C., 1984. Spontaneous and delayed spatial alternation following damage to specific neuronal elements within the nucleus medianus raphe. Behav. Brain Res. 13, 241e250. Barnes, N.M., Sharp, T., 1999. A review of central 5-HT receptors and their function. Neuropharmacology 38, 1083e1152. Beck, S.G., Choi, K.C., 1991. 5-Hydroxytryptamine hyperpolarizes CA3 hippocampal pyramidal cells through an increase in potassium conductance. Neurosci. Lett. 133, 93e96. Behr, J., Empson, R.M., Schmitz, D., Gloveli, T., Heinemann, U., 1997. Effects of serotonin on synaptic and intrinsic properties of rat subicular neurons in vitro. Brain Res. 773, 217e222. Behrens, C.J., van den Boom, L.P., de, H.L., Friedman, A., Heinemann, U., 2005. Induction of sharp wave-ripple complexes in vitro and reorganization of hippocampal networks. Nat. Neurosci. 8, 1560e1567. Behrens, C.J., van den Boom, L.P., Heinemann, U., 2007. Effects of the GABA(A) receptor antagonists bicuculline and gabazine on stimulus-induced sharp waveripple complexes in adult rat hippocampus in vitro. Eur. J. Neurosci. 25, 2170e2181. Buzsaki, G., 1986. Hippocampal sharp waves: their origin and significance. Brain Res. 398, 242e252. Buzsaki, G., Leung, L.W., Vanderwolf, C.H., 1983. Cellular bases of hippocampal EEG

in the behaving rat. Brain Res. 287, 139e171. Chittajallu, R., Craig, M.T., McFarland, A., Yuan, X., Gerfen, S., Tricoire, L., Erkkila, B., Barron, S.C., Lopez, C.M., Liang, B.J., Jeffries, B.W., Pelkey, K.A., McBain, C.J., 2013. Dual origins of functionally distinct O-LM interneurons revealed by differential 5-HT(3A)R expression. Nat. Neurosci. 16, 1598e1607. Cowen, P., Sherwood, A.C., 2013. The role of serotonin in cognitive function: evidence from recent studies and implications for understanding depression. J. Psychopharmacol. 27 (7), 575e583. €m, A., Fuxe, K., 1964. Evidence for the existence of monoamine-containing Dahlstro neurons in the central nervous system. Acta Physiol. Scand. Suppl. 62 (Suppl. 232). Diba, K., Buzsaki, G., 2008. Hippocampal network dynamics constrain the time lag between pyramidal cells across modified environments. J. Neurosci. 28, 13448e13456. Dietzel, I., Heinemann, U., Lux, H.D., 1989. Relations between slow extracellular potential changes, glial potassium buffering, and electrolyte and cellular volume changes during neuronal hyperactivity in cat brain. Glia 2, 25e44. Draguhn, A., Traub, R.D., Schmitz, D., Jefferys, J.G.R., 1998. Electrical coupling underlies high-frequency oscillations in the hippocampus in vitro. Nature 394, 189e192. Egorov, A.V., Heinemann, U., Müller, W., 1996. Muscarinic activation reduces changes in [Ca2þ]o evoked by stimulation of Schaffer collaterals during blocked synaptic transmission in rat hippocampal slices. Neurosci. Lett. 214, 187e190. Faber, D.S., Korn, H., 1991. Applicability of the coefficient of variation method for analyzing synaptic plasticity. Biophys. J. 60, 1288e1294. Fletcher, P.J., Korth, K.M., Chambers, J.W., 1999. Depletion of brain serotonin following intra-raphe injections of 5,7-dihydroxytryptamine does not alter damphetamine self-administration across different schedule and access conditions. Psychopharmacology (Berl) 146, 185e193. Foster, D.J., Wilson, M.A., 2006. Reverse replay of behavioural sequences in hippocampal place cells during the awake state. Nature 440, 680e683. Gabriel, S., Kivi, A., Eilers, A., Kov
acs, R., Heinemann, U., 1998. Effects of barium on stimulus-induced rises in [Kþ]0 in juvenile rat hippocampal area CA1. NeuroReport 9, 2583e2587. Gloveli, T., Albrecht, D., Heinemann, U., 1995. Properties of low Mg2þ induced epileptiform activity in rat hippocampal and entorhinal cortex slices during adolescence. Dev. Brain Res. 87, 145e152. Goddard, G.V., McIntyre, D.C., Leech, C.K., 1969. A permanent change in brain function resulting from daily electical stimulation. Exp. Neurol. 25, 295e330. Hasselmo, M.E., McGaughy, J., 2004. High acetylcholine levels set circuit dynamics for attention and encoding and low acetylcholine levels set dynamics for consolidation. Prog. Brain Res. 145, 207e231. Heinz, A., Mann, K., Weinberger, D.R., Goldman, D., 2001. Serotonergic dysfunction, negative mood states, and response to alcohol. Alcohol Clin. Exp. Res. 25, 487e495. Hollnagel, J.O., Maslarova, A., Haq, R.U., Heinemann, U., 2014. GABAB receptor dependent modulation of sharp wave-ripple complexes in the rat hippocampus in vitro. Neurosci. Lett. 574, 15e20. Huang, Y.Y., Kandel, E.R., 1994. Recruitment of long-lasting and protein kinase Adependent long-term potentiation in the CA1 region of hippocampus requires repeated tetanization. Learn Mem. 1, 74e82. Huang, Y.Y., Kandel, E.R., 1995. D1/D5 receptor agonists induce a protein synthesisdependent late potentiation in the Ca1 region of the Hippocampus. Proc. Natl. Acad. Sci. U. S. A. 92, 2446e2450. Hyttel, J., 1994. Pharmacological characterization of selective serotonin reuptake inhibitors (SSRIs). Int. Clin. Psychopharmacol. 9 (Suppl. 1), 19e26. Jacques, S., 1979. Brain stimulation and reward: “pleasure centers” after twenty-five years. Neurosurgery 5, 277e283. Jarosch, M.S., Gebhardt, C., Fano, S., Huchzermeyer, C., Ul Haq, R., Behrens, C.J., Heinemann, U., 2015. Early adenosine release contributes to hypoxia-induced disruption of stimulus-induced sharp wave-ripple complexes in rat hippocampal area CA3. Eur. J. Neurosci. 42, 1808e1817. Johnston, A., McBain, C.J., Fisahn, A., 2014. 5eHT1A receptor-activation hyperpolarizes pyramidal cells and suppresses hippocampal gamma oscillations via Kir3 channel-activation. J. Physiol. 592 (pt19), 4187e4199. Kemp, A., Manahan-Vaughan, D., 2005. The 5-hydroxytryptamine4 receptor exhibits frequency-dependent properties in synaptic plasticity and behavioural metaplasticity in the hippocampal CA1 region in vivo. Cereb. Cortex 15, 1037e1043. Kroeze, W.K., Kristiansen, K., Roth, B.L., 2002. Molecular biology of serotonin receptors structure and function at the molecular level. Curr. Top. Med. Chem. 2, 507e528. Kubota, D., Colgin, L.L., Casale, M., Brucher, F.A., Lynch, G., 2003. Endogenous waves in hippocampal slices. J. Neurophysiol. 89, 81e89. Liotta, A., Caliskan, G., Ul Haq, R., Hollnagel, J.O., Rosler, A., Heinemann, U., Behrens, C.J., 2011. Partial disinhibition is required for transition of stimulusinduced sharp wave-ripple complexes into recurrent epileptiform discharges in rat hippocampal slices. J. Neurophysiol. 105, 172e187. Maier, N., Morris, G., Schuchmann, S., Korotkova, T., Ponomarenko, A., Bohm, C., Wozny, C., Schmitz, D., 2011a. Cannabinoids disrupt hippocampal sharp waveripples via inhibition of glutamate release. Hippocampus. Maier, N., Nimmrich, V., Draguhn, A., 2003. Cellular and network mechanisms underlying spontaneous sharp wave-ripple complexes in mouse hippocampal slices. J. Physiol. 550, 873e887. Maier, N., Tejero-Cantero, A., Dorrn, A.L., Winterer, J., Beed, P.S., Morris, G.,

R. ul Haq et al. / Neuropharmacology 101 (2016) 188e203 Kempter, R., Poulet, J.F., Leibold, C., Schmitz, D., 2011b. Coherent phasic excitation during hippocampal ripples. Neuron 72, 137e152. Maier, N., Tejero-Cantero, A., Dorrn, A.L., Winterer, J., Beed, P.S., Morris, G., Kempter, R., Poulet, J.F., Leibold, C., Schmitz, D., 2011c. Coherent phasic excitation during hippocampal ripples. Neuron 72, 137e152. McMahon, L.L., Kauer, J.A., 1997. Hippocampal interneurons are excited via serotonin-gated ion channels. J. Neurophysiol. 78, 2493e2502. Meneses, A., Liy-Salmeron, G., 2012. Serotonin and emotion, learning and memory. Rev. Neurosci. 23, 543e553. Morales, M., Battenberg, E., De Lecea, L., Bloom, F.E., 1996. The type 3 serotonin receptor is expressed in a subpopulation of GABAergic neurons in the rat neocortex and hippocampus. Brain Res. 731, 199e202. Morales, M., Hein, K., Vogel, Z., 2007. Hippocampal interneurons co-express transcripts encoding the alpha7 nicotinic receptor subunit and the cannabinoid receptor 1. Neuroscience. Müller, W., Misgeld, U., Heinemann, U., 1988. Carbachol effects on hippocampal neurons in vitro: dependence on the rate of rise of carbachol tissue concentration. Exp. Brain Res. 72, 287e298. Nicholson, C., Freeman, J.A., 1975. Theory of current source-density analysis and determination of conductivity tensor for anuran cerebellum. J. Neurophysiol. 38, 356e368. Norimoto, H., Mizunuma, M., Ishikawa, D., Matsuki, N., Ikegaya, Y., 2012. Muscarinic receptor activation disrupts hippocampal sharp wave-ripples. Brain Res. 1461, 1e9. O'Keefe, J., Nadel, L., 1978. The Hippocampus as a Cognitive Map. Oxford University Press, Oxford. Paluchowska, M.H., Mokrosz, M.J., Bojarski, A., Wesolowska, A., Borycz, J., Charakchieva-Minol, S., Chojnacka-Wojcik, E., 1999. On the bioactive conformation of NAN-190(1) and MP3022 (2), 5-HT(1A) receptor antagonists. J. Med. Chem. 42, 4952e4960. Pangalos, M., Donoso, J.R., Winterer, J., Zivkovic, A.R., Kempter, R., Maier, N., Schmitz, D., 2013. Recruitment of oriens-lacunosum-moleculare interneurons during hippocampal ripples. Proc. Natl. Acad. Sci. U. S. A. 110, 4398e4403. Papatheodoropoulos, C., Kostopoulos, G., 2002. Spontaneous, low frequency (approximately 2-3 Hz) field activity generated in rat ventral hippocampal slices perfused with normal medium. Brain Res. Bull. 57, 187e193. Penington, N.J., Kelly, J.S., 1990. Serotonin receptor activation reduces calcium current in an acutely dissociated adult central neuron. Neuron 4, 751e758. Ponomarenko, A.A., Knoche, A., Korotkova, T.M., Haas, H.L., 2003. Aminergic control of high-frequency (similar to 200 Hz) network oscillations in the hippocampus of the behaving rat. Neurosci. Lett. 348, 101e104. €demann, R., Ott, T., Matthies, H., 1985. Reymann, K.G., Malisch, R., Schulzeck, K., Bro The duration of long-term potentiation in the CA1 region of the hippocampal slice preparation. Brain Res. Bull. 15, 249e255. Richter-Levin, G., Segal, M., 1989. Raphe cells grafted into the hippocampus can ameliorate spatial memory deficits in rats with combined serotonergic/ cholinergic deficiencies. Brain Res. 478, 184e186. Richter-Levin, G., Segal, M., 1990. Effects of serotonin releasers on dentate granule cell excitability in the rat. Exp. Brain Res. 82, 199e207. Roth, B.L., 1994. Multiple serotonin receptors: clinical and experimental aspects. Ann. Clin. Psychiatry 6, 67e78. Rydelek-Fitzgerald, L., Teitler, M., Fletcher, P.W., Ismaiel, A.M., Glennon, R.A., 1990. NAN-190: agonist and antagonist interactions with brain 5-HT1A receptors. Brain Res. 532, 191e196. Sarnyai, Z., Sibille, E.L., Pavlides, C., Fenster, R.J., McEwen, B.S., Toth, M., 2000. Impaired hippocampal-dependent learning and functional abnormalities in the hippocampus in mice lacking serotonin(1A) receptors. Proc. Natl. Acad. Sci. U. S. A. 97, 14731e14736. Schlingloff, D., Kali, S., Freund, T.F., Hajos, N., Gulyas, A.I., 2014. Mechanisms of sharp wave initiation and ripple generation. J. Neurosci. 34, 11385e11398. Schmitz, D., Empson, R.M., Heinemann, U., 1995a. Serotonin and 8-OH-DPAT reduce excitatory transmission in rat hippocampal area CA1 via reduction in presumed presynaptic Ca2þ entry. Brain Res. 701, 249e254.

203

Schmitz, D., Empson, R.M., Heinemann, U., 1995b. Serotonin reduces inhibition via 5-HT1A receptors in area CA1 of rat ventral hippocampal slices in vitro. J Neurosci 15, 7217e7225. Schmitz, D., Gloveli, T., Empson, R.M., Draguhn, A., Heinemann, U., 1998a. Serotonin reduces synaptic excitation in the superficial medial entorhinal cortex of the rat via a presynaptic mechanism. J. Physiol. (Lond) 508, 119e129. Schmitz, D., Gloveli, T., Empson, R.M., Heinemann, U., 1998b. Comparison of the effects of serotonin in the hippocampus and the entorhinal cortex. Mol. Neurobiol. 17, 59e72. Schmitz, D., Gloveli, T., Empson, R.M., Heinemann, U., 1998c. Serotonin reduces polysynaptic inhibition via 5-HT1A receptors in the superficial entorhinal cortex. J. Neurophysiol. 80, 1116e1121. Schmitz, D., Schuchmann, S., Fisahn, A., Draguhn, A., Buhl, E.H., Petrasch-Parwez, E., Dermietzel, R., Heinemann, U., Traub, R.D., 2001. Axo-axonal coupling. a novel mechanism for ultrafast neuronal communication. Neuron 31, 831e840. Schonberger, J., Draguhn, A., Both, M., 2014. Lamina-specific contribution of glutamatergic and GABAergic potentials to hippocampal sharp wave-ripple complexes. Front. Neural Circuits 8, 103. Segal, M., 1980. The action of serotonin in the rat hippocampal slice preparation. J. Physiol. (Lond ) 303, 423e439. Segal, M., 1990. Serotonin attenuates a slow inhibitory postsynaptic potential in rat hippocampal neurons. Neuroscience 36, 631e641. € scher, W., Heinemann, U., 1998. Comparison Sokolova, S., Schmitz, D., Zhang, C.L., Lo of effects of valproate and trans-2-en-valproate on different forms of epileptiform activity in rat hippocampal and temporal cortex slices. Epilepsia 39, 251e258. Stark, E., Roux, L., Eichler, R., Senzai, Y., Royer, S., Buzsaki, G., 2014. Pyramidal cellinterneuron interactions underlie hippocampal ripple oscillations. Neuron 83, 467e480. Sudweeks, S.N., Hooft, J.A., Yakel, J.L., 2002. Serotonin 5-HT(3) receptors in rat CA1 hippocampal interneurons: functional and molecular characterization. J. Physiol. 544, 715e726. Turner, T.J., Mokler, D.J., Luebke, J.I., 2004. Calcium influx through presynaptic 5HT3 receptors facilitates GABA release in the hippocampus: in vitro slice and synaptosome studies. Neuroscience 129, 703e718. Ul Haq, R., Liotta, A., Kovacs, R., Rosler, A., Jarosch, M.J., Heinemann, U., Behrens, C.J., 2012. Adrenergic modulation of sharp wave-ripple activity in rat hippocampal slices. Hippocampus 22, 516e533. Uneyama, H., Ueno, S., Akaike, N., 1993. Serotonin-operated potassium current in CA1 neurons dissociated from rat hippocampus. J. Neurophysiol. 69, 1044e1052. Vandecasteele, M., Varga, V., Berenyi, A., Papp, E., Bartho, P., Venance, L., Freund, T.F., Buzsaki, G., 2014. Optogenetic activation of septal cholinergic neurons suppresses sharp wave ripples and enhances theta oscillations in the hippocampus. Proc. Natl. Acad. Sci. U. S. A. 111, 13535e13540. Vanderwolf, C.H., 1969. Hippocampal electrical activity and voluntary movement in the rat. Electroencephalogr. Clin. Neurophysiol. 26, 407e418. Villani, F., Johnston, D., 1993. Serotonin inhibits induction of long-term potentiation at commissural synapses in hippocampus. Brain Res. 606, 304e308. Wang, D.V., Yau, H.J., Broker, C.J., Tsou, J.H., Bonci, A., Ikemoto, S., 2015. Mesopontine median raphe regulates hippocampal ripple oscillation and memory consolidation. Nat. Neurosci. 18, 728e735. Wawra, M., Fidzinski, P., Heinemann, U., Mody, I., Behr, J., 2014. 5-HT4-receptors modulate induction of long-term depression but not potentiation at hippocampal output synapses in acute rat brain slices. PLoS One 9, e88085. Winterer, J., Stempel, A.V., Dugladze, T., Foldy, C., Maziashvili, N., Zivkovic, A.R., Priller, J., Soltesz, I., Gloveli, T., Schmitz, D., 2011. Cell-type-specific modulation of feedback inhibition by serotonin in the hippocampus. J. Neurosci. 31, 8464e8475. Wojtowicz, A.M., van den Boom, L., Chakrabarty, A., Maggio, N., Haq, R.U., Behrens, C.J., Heinemann, U., 2009. Monoamines block kainate- and carbacholinduced gamma-oscillations but augment stimulus-induced gamma-oscillations in rat hippocampus in vitro. Hippocampus 19, 273e288.