SLEEP DEPRIVATION IMPAIRS SPATIAL LEARNING AND MODIFIES THE HIPPOCAMPAL THETA RHYTHM IN RATS

Transcription

1 Neuroscience 173 (2011) SLEEP DEPRIVATION IMPAIRS SPATIAL LEARNING AND MODIFIES THE HIPPOCAMPAL THETA RHYTHM IN RATS R.-H. YANG, a * X.-H. HOU, a X.-N. XU, a L. ZHANG, a J.-N. SHI, a F. WANG, a S.-J. HU b * AND J.-Y. CHEN a * a Department of Nutrition and Food Hygiene, the Fourth Military Medical University, Xi an , PR China b Institute of Neuroscience, the Fourth Military Medical University, Xi an , PR China Abstract Previous work from our laboratory suggests that paradoxical sleep deprivation (PSD) decreases persistent sodium currents and hyperpolarization-activated cation currents in CA1 pyramidal neurons, and this leads to decreases in neuron excitability. Here, we further investigate the mechanisms of rhythmic theta-range activity in the hippocampus by examining the resonance characteristics of hippocampal pyramidal neurons. Sleep deprivation (SD) interfered with the rhythmic activity of theta band in the hippocampus, which may be involved in the deficit of the spatial learning ability of rats. Additionally, SD changes the voltage dependence of resonance. The effect of SD on the ion currents may contribute to the alternation of the theta resonance of neurons and further influence the physiological function. These results suggest that changes in neuron resonance lead to changes in the generation of rhythmic theta activity, and may contribute to behavioral deficits associated with theta activity during learning and memory tasks. We suggest the resonant properties of hippocampal neurons are potential targets for preventing deleterious effects of sleep deprivation IBRO. Published by Elsevier Ltd. All rights reserved. Key words: sleep deprivation, spatial learning, Morris Water Maze, resonance, pyramidal neuron, electroencephalogram. Several recent studies have shown that sleep has a key role in learning and memory. Rapid eye movement (REM) sleep is increased after learning sessions (Mandai et al., 1989; Smith and Rose, 1997), and sleep deprivation interferes with learning and memory (Guan et al., 2004; Silvestri, 2005). Animal studies indicate that the firing patterns of neurons in the hippocampus involved in the learning experience were replayed during subsequent sleep (Sutherl and McNaughton, 2000; Louie and Wilson, 2001). In a previous study, we have found that sleep deprivation (SD) impairs spatial learning ability of rats, which was associated with the decrease of neuron excitability in hippocampus (Yang et al., 2008). *Corresponding author. Tel: or or ; fax: or or addresses: fmmur.young@gmail.com (R.-H. Yang) or jy_chen@ fmmu.edu.cn (J.-Y. Chen) or sjhu@fmmu.edu.cn (S.-J. Hu). Abbreviations: ACSF, artificial cerebrospinal fluid; EEG, electroencephalogram; FFT, fast Fourier transforms; REM, rapid eye movement; SD, sleep deprivation /11 $ - see front matter 2011 IBRO. Published by Elsevier Ltd. All rights reserved. doi: /j.neuroscience Resonance is described as the ability of neurons to respond selectively to inputs at a preferred frequency, and it participates in the rhythm of a neuronal population generation (Lampl and Yarom, 1997; Hutcheon and Yarom, 2000; Wu et al., 2001). Spontaneous theta ( ) field potentials (4 10 Hz) are an important biological rhythm in many brain regions, especially in the hippocampal formation. Oscillations at theta frequency occur during REM sleep and are associated with spatial learning. It is essential to characterize the resonance of CA1 pyramidal neurons because it serves as a substrate for network activity in the hippocampus. The present study was performed to investigate the resonance characteristics of CA1 pyramidal neurons. Also, the effect of SD on the resonance and its mechanism were studied. EXPERIMENTAL PROCEDURES Animals and surgery A total of 60 adult ( g) male Sprague Dawley rats were used in the experiments. Animal care was in accord with the Principles of Medical Laboratory Animal Care issued by the National Ministry of Health. All experiments conformed to the guidelines of the National Ordinances on Experimental Animals for the ethical use of animals. Animals were housed individually and kept on a reversed light-dark h cycle, food and water were available ad libitum. Under isoflurane anaesthesia, a bipolar electrode (0.5 of diameter) was chronically implanted in the CA1 field of the right hippocampus. The coordinates of the implants were 4.0 mm posterior to bregma, 2.1 mm lateral from the midline, and 2.0 mm ventral from the skull (Paxinos and Watson, 2007). The two wires of the bipolar electrode sets were separated 1 mm. A stainless steel needle was placed on the bone over the prefrontal cortex surface as a ground electrode. Two weeks after surgery, the animals were divided into two groups of twenty: control group and SD group. Sleep deprivation Ten small platforms (8.5 cm in height and 6.0 cm in diameter) were placed (8 10 cm apart) inside a water tank made of sheet iron. The bottom of the tank was filled with 24 C water which reached up to 2 cm below the surface of the platforms. The platforms were of small diameter permitting the rat to sit, but not lie down, on the platform. The rats could easily move between the platforms but could not stretch across any two platforms to sleep, thus the animals were awoken when they experienced REM sleep-induced atonia by touching the water, and it may produce a fragmentation of NREM sleep due to repeated awakenings when the animal falls into the water surrounding the platform (Silvestri, 2005). The water in the tank was changed daily. All rats had free access to food and water.

2 R.-H. Yang et al. / Neuroscience 173 (2011) Spatial learning test Animals were divided into two groups. The sleep deprivation was performed by the multiple platform technique described above for 6 days. Control group of animals was housed in their home cage and permitted to sleep during the same period. Daily, behavioral test was performed between 15:00 h and 18:00 h, in the Morris Water Maze. The maze, 80 cm deep and 150 cm in diameter, was divided into four quadrants of equal size on the monitor screen of a computer, filled to a depth of 24 cm with water. The water in the tank was fresh each day and was maintained at C. A white, 10 cm diameter platform was placed in the center of quadrant 4 and submerged 2 cm below the water surface. Rats with red tags attached to their backs, were trained to find the hidden platform according to the spatial cues in the experimental room. Each rat was released facing the wall of the Water Maze in the four quadrants respectively. The order of quadrants was changed each day such that subjects were never exposed to a sequence of trials that they had had before. Each animal was allowed to swim for a maximum duration of 120 s in each trial to find the platform. After the animal found and got onto the platform, it was allowed to stay on the platform for 20 s. If the rat did not find the platform, the rat was guided to it and left there for 20 s. After training, the animal was dried with a fabric towel and returned to its home cage or SD tank cage. The distance swum by the rat to find the platform and time (latency) to locate the platform were recorded by an MT-200 Morris image motion system (Chengdu Technology Market Corp., PR China). Swimming speed was calculated from distance and latency. On the seventh day, the platform was removed from the Water Maze and the animals were challenged to a single trial for 120 s (probe trial). Electroencephalogram activity Before and after the behavioral training, the hippocampal electroencephalogram (EEG) were amplified (2000 ), filtered ( Hz), recorded and stored on hard disk, to be analyzed off-line. The EEG record was divided in 2.56 s windows. Each window was divided in 64 point epochs. Both ends of the epochs were smoothed with Hamming window, and the magnitude of the Fourier transform, for each epoch, was obtained using IgorPro software (WaveMetrics, Lake Oswega, OR, USA). The relative magnitude of each frequency was averaged for each window and then displayed graphically. A band-pass filter was used with the filter set up at 1 20 Hz. The proportion of theta was evaluated as a proportion of the total area under frequency-magnitude area obtained between 4 and 8 Hz (Menezes et al., 2009). In vitro electrophysiological recording The experimental rats were deeply anesthetized with pentobarbital sodium (40 mg/kg) and decapitated after 3 days SD (n 12). The control animals (n 8) were allowed to sleep during the same period. Hippocampal slices (300 m in thickness) were prepared with a vibratome (Vibroslice 752M, Campden Instruments, Loughborough, UK) and incubated with artificial cerebrospinal fluid (ACSF) containing (in mm): 124 NaCl, 2.5 KCl, 1.2 NaH 2 PO 4, 1.0 MgCl 2, 2.0 CaCl 2, 25 NaHCO 3, 10 Glucose. Slices were maintained in ACSF at 26 C for at least 1 h before being moved into the recording chamber. During the recordings, the slices were kept submerged in a chamber perfused with ACSF. In the experiments, the ACSF was saturated with 95% O 2 /5%CO 2 and the temperature was kept at 26 C. Individual neurons were visualized with a 40 water-immersion objective under a microscope (BX51WI; Olympus, Tokyo, Japan) equipped with infrared differential interference contrast optics. Whole-cell recordings were obtained from pyramidal cells using recording pipettes with a resistance of 4 7 M. Patch pipettes were filled with solution containing (in mm): 140 potassium gluconate, 10 HEPES, 10 phosphoereatine sodium salt, 2 ATP sodium salt, 0.4 GPT sodium salt and 2 MgCl 2. All Chemicals were obtained from Sigma, St. Fig. 1. Swimming routes displayed by a representative animal in the six training days and the probe trial. The line showed the fourth trial of each of the 6 d. The maze was divided into I, II, III and IV quadrants. The platform was placed in quadrant IV in the training days. Louis, MO, USA. For patch-clamp recordings, a Multiclamp 700B amplifier (Molecular Devices Corporation, Sunnyvale, CA, USA) was used. The series resistance was M. All potentials were corrected online for the junction potential by adjusting the offset of the pipette using the Multiclamp 700B commander software. Neurons were selected for further study if they had a resting membrane potential that was more negative than 50 mv and if they exhibited overshooting action potentials. Data acquisition and statistics The data were acquired with a digidata 1322A acquisition system (Molecular Devices) using pclamp 8.0 software (Molecular Devices) at a sampling rate of 10 khz for subsequent offline processing. Data values are expressed as mean SE. Student s t-tests (SigmaStat version 2.03) were used to determine the statistical significance of differences between means obtained from two different groups of neurons. ANOVA followed by post hoc pairwise comparisons (Student Newman Keuls method) (SigmaStat 2.03) were used to determine the statistical signif-

3 118 R.-H. Yang et al. / Neuroscience 173 (2011) Fig. 2. Effect of sleep deprivation on spatial learning and memory. (A) The mean escape latencies were significantly higher in the SD group compared with control group from the third day on. (B) The distance traveled during the trial. (C) The swimming speeds of SD rats were similar with that of control group. (D) The time that the animal spent in each quadrant during the probe trial. Data are means SE, * P 0.05, SD vs. Control, # P 0.001, quadrant 4 vs. the other quadrants. icance of differences between values of hippocampal theta proportion. In all experiments, a probability of 0.05 or less was considered statistically significant. For the analysis of electrical resonance, the impedance amplitude profile (ZAP) method was used to characterize the resonance behaviors of the cells (Puil et al., 1986; Pape and Driesang, 1998; Hutcheon and Yarom, 2000). During whole-cell current-clamp recording, we injected the ZAP current (from 0 to 15 Hz for 20 s) through the patch pipette and recorded the voltage response. Resonance manifested as a distinct and reproducible peak in the voltage response at a specific frequency. The ratio of the impedance at the resonance peak to the impedance at 0.5 Hz (called Q-value) (Hutcheon et al., 1996) was used to quantify the strength of the resonance. A Q-value of 1 represents no resonance. The amplitude of the ZAP current was adjusted to keep the perturbation of the membrane potential close to 10 mv, peak-to-peak, to avoid triggering action potentials. To measure the resonance frequency (Fres), we performed fast Fourier transforms (FFT) of both the membrane potential response (V) and the ZAP current (I). The impedance profile was then calculated by dividing FFT (V) by FFT(I). RESULTS Effect of SD on spatial learning Representative rat swimming routes of training are shown in Fig. 1. The animal displayed more directed and regular swimming routes as the trials elapsed. The latency traveled to find the platform was shown in Fig. 2A. Both control and SD group over the six training days exhibited significant, substantial reductions in their times to find the platform. However, compared to the control group, SD rats took significantly longer to find the platform, implying a significant impairment of spatial learning process, and this impairment occurred from training day 3 onward (P 0.05). Similarly, the distance traveled to the platform during the trial decreased along with the training, and the difference between control and SD group was significant from day 4 (Fig. 2B). There was no significant difference of swimming speeds between two groups (Fig. 2C, P 0.934). As showed in Fig. 1, searching activity was evident during the probe trial around the area where the platform had been placed earlier. Both control and SD rats spent similar percentage of time in the target quadrant ( % and %, respectively, P 0.98) (Fig. 2D), which was significantly more than that in the other quadrants (F , P 0.001). There was no significant difference between crossing times of SD and control group ( and , respectively, P 0.097).

4 R.-H. Yang et al. / Neuroscience 173 (2011) Fig. 3. Effect of sleep deprivation on the hippocampal EEG. (A) EEG traces obtained before and after the daily training throughout the six training days from one representative animal. Note the more rhythmic pattern of EEG after the training. (B) Effect of SD on the percentage of EEG in 5 8 Hz theta band before and after training. * P 0.05, SD vs. Control, # P 0.05, after training vs. before training in each day. Effect of SD on the hippocampal EEG EEG traces obtained from the representative animals are depicted in Fig. 3. The training changed the EEG pattern from an irregular mixed wave to the appearance of rhythmic activity (Fig. 3A). There was no significant difference of the proportion of theta rhythm (4 8 Hz) between control and SD group in the basal status (before training). After training, the proportion of theta was significantly increased in the control group compared with both basal status and SD group (P 0.05). However, there was only increased proportion of theta in first two training days in SD group (Fig. 3B). Resonance of the hippocampal CA1 pyramidal neurons Whole-cell patch-clamp recordings were obtained from 83 normal hippocampal CA1 pyramidal neurons in hippocam- Fig. 4. Subthreshold membrane resonance in hippocampal CA1 pyramidal neurons. (A) The voltage response was induced by depolarization and hyperpolarization in neuron. It showed that sag and rebound were induced at hyperpolarization and firings at depolarization. (B) The voltage response (upper trace) to the injection of ZAP current (lower trace; from 0 to 15 Hz, 20 s). The cell was holding at 77 mv. (C) Impedance curve (impedance magnitude, Z FFT(V)/FFT(I)) plotted as a function of input frequency, using the data in (B). It shows that the frequency of resonance is 2.10 Hz.

5 120 R.-H. Yang et al. / Neuroscience 173 (2011) pal slices. The neurons had resting potentials near 65 mv ( , n 31) and action potential thresholds close to 50 mv ( , n 31). Fig. 4A showed that the hippocampal CA1 pyramidal neuron fired with repetitive spiking and exhibited obvious sag and rebound characteristics. To test for resonance, a sinusoidal current (the ZAP current) with constant amplitude and linearly increasing frequency (0 15 Hz during 20 s; Fig. 4B, lower trace) was injected, and the voltage response was recorded (Fig. 4B, upper trace). In the example neuron of Fig. 4, the resonant hump in the voltage response corresponded to the frequency of 2.10 Hz at a membrane potential of 77 mv (Fig. 1B, C). Temperature and voltage dependence of resonance in hippocampal CA1 pyramidal neurons Most of our experiments were carried out at 26 C. This temperature does not represent the physiological state of rat hippocampal CA1 pyramidal neurons. With the membrane potential holding at 80 mv, we tested the temperature dependence of neuron resonance. Fig. 5 showed that the resonance frequency increased from 2.01 to 5.06 Hz when the temperature was raised from 26 to 35 C. The resonance frequencies were Hz, Hz, Hz and Hz at 26 C (n 15), 29 C (n 10), 32 C (n 7) and 35 C (n 6), respectively. Statistical analysis showed that the resonance frequency increased significantly with raised temperature (one-way ANOVA, F , P 0.001). These results demonstrated that the resonance of CA1 pyramidal neurons belong to the band at these sub-physiological temperatures. To determine the voltage dependence of resonance, the membrane potentials were held at different levels before applying the ZAP current. Fig. 6A shows the voltage response during ZAP current injections at different membrane potentials in one neuron. The neuron was resonant at depolarized potentials lower than 50 mv and hyperpolarized potentials beyond 70 mv. At near resting potential ( mv, n 31), however, the neuron did not exhibit obvious resonance. The average frequency of the resonance peak was close to 2.80 Hz at both depolarized ( 40 mv) and hyperpolarized ( 90 mv) potentials, but was only about 0.80 Hz near the resting potential (Fig. 6B, n 16). The average Q-value was at 40 mv and at 90 mv, and the Q-value approached 1.0 at potentials close to the resting level (Fig. 6C, n 16). These results showed that the resonance characteristics of the CA1 pyramidal neurons were temperature- and voltage-dependent, and the frequency belong to the band at sub-physiological temperatures. The resonance frequency and Q-values were higher at both hyperpolar- Fig. 5. Temperature dependence of the resonance in CA1 pyramidal neurons. (A) The voltage response to the injection of ZAP current at different chamber temperatures. Note that the frequency of resonance moved to a higher frequency band when the temperature increased from 26 to 35 C. (B) The frequency increased from 2.01 Hz at 26 C to 5.06 Hz at 35 C. (C) The frequencies at four different temperature points were compared. Note that the frequency was 4 6 Hz at near-physiological temperatures of C.

6 R.-H. Yang et al. / Neuroscience 173 (2011) Fig. 6. Voltage dependence of the resonance in CA1 pyramidal neurons. (A) The voltage response to the injection of a ZAP current at different holding membrane potentials. The different membrane potentials were obtained by different amounts of steady current injection. Stronger resonance was observed at depolarized ( 40 mv) and hyperpolarized ( 90 mv) potentials compared to potentials near the resting level ( mv, n 31). (B) The frequency was plotted at different potentials (mean SE; n 6). It can be seen that the average frequency was higher at both depolarized and hyperpolarized potentials than near the resting potential. (C) Q value (mean SE) was plotted at different potentials, obtained from the same six neurons as in (B). The Q value indicates the strength of the resonance, and is 1.0 when there is no resonance. Q value also showed the dual voltage dependence, which was closer to 1.0 at potentials near the resting level. ized and depolarized potentials than that exhibited at resting potential. Effect of SD on the voltage-dependence of resonance In the current-clamp mode, ZD7288 (14 M) and TTX (100 nm) were applied to detect the role of hyperpolarizationactivated cation current (Ih) and persistent sodium currents (I NaP ) on resonance behavior, respectively. ZAP currents were applied before and after application of ZD7288 at hyperpolarized potential ( 85 mv). Fig. 7A, B showed that the resonance was abolished and impedance magnitude was increased after ZD7288 application. At holding potential of 55 mv (depolarized potential), the resonance amplitude was reduced after TTX application (Fig. 7C, D). The impedance curve showed that the impedances were reduced and the frequency did not change after applying TTX (Fig. 7D). After 72 h SD, the dual voltage dependence of resonance was also observed in hippocampal CA1 pyramidal neurons. Compared to the control group, the curve showed a right shift as a depolarized membrane potential (Fig. 7E). There was no significant difference in the voltage dependence of the Q-value between the SD and control neurons, although the Q-values of SD neurons were numerically lower than control at both hyperpolarized and depolarized potentials (Fig. 7F, n 16). DISCUSSION The importance of the hippocampal formation for memory is well established on the basis of neuroanatomical and electrophysiological studies (Squire and Zola, 1996; Eichenbaum, 1999). In particular, spatial memory is understood to be strongly dependent on hippocampal activity in rats (Eichenbaum, 2000; Henninger et al., 2007). In a previous study, we found that SD decreased the persistent sodium currents (I NaP ) and the hyperpolarization-activated cation current (Ih) of CA1 pyramidal neurons, which contributed to the decrease of neuron excitability (Yang et al., 2010), further influencing learning and memory processing (Yang et al., 2008). In the present study, we further studied the effect of SD on the hippocampal EEG and the resonance characters of hippocampal CA1 pyramidal neurons. Our major findings are that SD disturbed the rhythmic

7 122 R.-H. Yang et al. / Neuroscience 173 (2011) groups, which indicated that SD decreased spatial learning ability. Spatial learning is understood to be strongly dependent on hippocampal processing (Squire, 1992; Eichenbaum, 1999, 2004; Henninger et al., 2007). Theta oscillation, which is regular in frequency and large amplitude in hippocampal CA1 region, has been suggested to play important roles in learning and memory (O Keefe and Recce, 1993; Buzsaki, 2002; Schall et al., 2008). Ethanol impaired memory through suppressing hippocampal theta activity (Givens, 1995). In the present study, the increase of theta rhythmic activity was observed as the training process in control group. However, the SD rats showed decreased spatial learning ability which accompanied by the lack of rhythmic activity in the EEG, the proportion of theta was significantly lower than the control group. It has been suggested that the place-learning take place concomitantly with the expression of theta rhythm in hippocampus (Olvera-Cortes et al., 2002). Thus, SD might influence the learning task of rats by disturbing the theta rhythmic activity. SD affected the resonance behavior of hyppocampal CA1 pyramidal neurons Fig. 7. Effect of SD on the voltage-dependence of resonance and its possible ion mechanism. (A) At 85 mv, the voltage response was recorded when injecting ZAP current before and after applying ZD7288 (14 M). Note that the resonance hump vanished after applying ZD7288. (B) The impedance curve before and after applying ZD7288 at a holding potential of 85 mv for the same neuron as in (A), demonstrating that the resonance was abolished and impedance magnitude increased due to the application of ZD7288. (C) The voltage response to the ZAP current at depolarized potentials ( 55 mv) when applying TTX (100 nm), demonstrating that the resonance hump decreased due to applying TTX. (D) The impedance curve before and after applying TTX at a holding potential of 55 mv in the same neuron as in (C). Note that the impedance reduced after applying TTX at the holding potentials. (E) Summary diagram of peak resonance frequency plotted as a function of membrane potential for control and SD neurons. SD induced a right shift as depolarized membrane potential. (F) Summary diagram of Q values plotted as a function of membrane potential for control and SD neurons. activity of theta band in the hippocampus, which may involve in the deficit of the spatial learning ability of rats. Additionally, SD changes the voltage dependence of resonance. The effect of SD on the ion currents may contribute to the alternation of frequency rhythms in the hippocampus and further influence its physiological functions. Theta rhythm activity contribute to the effect of SD on learning ability There is now substantial evidence that paradoxical sleep is involved with the processing of learning and memory. A main observation in the present study is that SD severely impaired spatial learning. With the Morris Water Maze spatial learning set protocol, the escape latencies of SD rats were significantly more delayed than in the control The results of the present study indicated that the resonance of the CA1 neuron was temperature and voltage dependent, and was abolished by ZD7288 at hyperpolarized potential, which is consistent with the results of other studies about resonance in hippocampus (Hu et al., 2002; Wang et al., 2006). When we applied TTX (100 nm) to block I NaP, the resonance hump was still observed at depolarized potential but with reduced amplitude. Hu et al. (Hu et al., 2002) reported that resonance was absent after higher concentration of TTX (1 M) application, and the resonance reappeared at further depolarizing potential in the presence of TTX. After SD, the voltage dependence of resonance shifted to a depolarized membrane potential. Under voltage-clamp, SD decreased the amplitude of I NaP in CA1 pyramidal neurons (Yang et al., 2010). We suggested that I NaP is involved in the effect of SD on the resonance of CA1 pyramidal neurons. There may be other ion channels contribute to the resonance behavior and play a role in the effect of SD. Further studies are needed to explore the mechanisms of resonance changes. Theta oscillation is referred to as network phenomena. Although the role of single neuron oscillations in the generation of neuronal networks remains unclear, it is reported that a single neuron can modulate its own rhythmicity and that local factors may regulate neuronal recruitment and synchronization, which may be important during hippocampal functions (Leung and Yu, 1998). Staba et al. reported that burst discharges in the hippocampus participate in memory consolidation (Staba et al., 2002). After SD, the neuron excitability was decreased; on the other side, the voltage-dependence shift of resonance suggested a more depolarizing potential requirement for resonance generation and maintenance. These changes of a single neuron may influence theta oscillation in hippocampus.

8 R.-H. Yang et al. / Neuroscience 173 (2011) CONCLUSION In summary, SD disturbed the hippocampal theta rhythmic activity in learning process, which may involved in the deficit of the spatial learning ability of rats. SD induced a depolarization shift of the voltage-dependence of neuron resonance. I NaP and Ih reduction may contribute to the effect of SD on resonance of neurons, which may participate in the generation of theta oscillation in the hippocampus. Our data have led us to conclude that the changes of neuron resonance affected the generation of theta rhythmic activity, and by extension contributed to the deficits of performing certain learning and memory tasks after SD. Acknowledgments This work was supported by a National Natural Science Foundation of China [ ] and a grant from the Fourth Military Medical University [XJ200502]. REFERENCES Buzsaki G (2002) Theta oscillations in the hippocampus. Neuron 33: Eichenbaum H (1999) The hippocampus and mechanisms of declarative memory. Behav Brain Res 103: Eichenbaum H (2000) A cortical-hippocampal system for declarative memory. Nat Rev Neurosci 1: Eichenbaum H (2004) Hippocampus: cognitive processes and neural representations that underlie declarative memory. Neuron 44: Givens B (1995) Low dose of ethanol impair spatial working memory and reduce hippocampal theta activity. Alcohol Clin Exp Res 19: Guan Z, Peng X, Fang J (2004) Sleep deprivation impairs spatial memory and decreases extracellular signal-regulated kinase phosphorylation in the hippocampus. Brain Res 1018: Henninger N, Feldmann RE Jr, Futterer CD, Schrempp C, Maurer MH, Waschke KF, Kuschinsky W, Schwab S (2007) Spatial learning induces predominant downregulation of cytosolic proteins in the rat hippocampus. Genes Brain Behav 6: Hu H, Vervaeke K, Storm JF (2002) Two forms of electrical resonance at theta frequencies, generated by M-current, h-current and persistent Na current in rat hippocampal pyramidal cells. J Physiol 545: Hutcheon B, Miura RM, Puil E (1996) Subthreshold membrane resonance in neocortical neurons. J Neurophysiol 76: Hutcheon B, Yarom Y (2000) Resonance, oscillation and the intrinsic frequency preferences of neurons. Trends Neurosci 23: Lampl I, Yarom Y (1997) Subthreshold oscillations and resonant behavior: two manifestations of the same mechanism. Neuroscience 78: Leung LS, Yu HW (1998) Theta-frequency resonance in hippocampal CA1 neurons in vitro demonstrated by sinusoidal current injection. J Neurophysiol 79: Louie K, Wilson MA (2001) Temporally structured replay of awake hippocampal ensemble activity during rapid eye movement sleep. Neuron 29: Mandai O, Guerrien A, Sockeel P, Dujardin K, Leconte P (1989) REM sleep modifications following a Morse code learning session in humans. Physiol Behav 46: Menezes RCA, Ootsuka Y, Blessing WW (2009) Sympathetic cutaneous vasomotor alerting responses (SCVARs) are associated with hippocampal theta rhythm in non-moving conscious rats. Brain Res 1298: O Keefe J, Recce ML (1993) Phase relationship between hippocampal place units and the EEG theta rhythm. Hippocampus 3: Olvera-Cortes E, Cervantes M, González-Burgos I (2002) Place-learning, but not cue-learning training, modifies the hippocampal theta rhythm in rats. Brain Res Bull 58: Pape HC, Driesang RB (1998) Ionic mechanisms of intrinsic oscillations in neurons of the basolateral amygdaloid complex. J Neurophysiol 79: Paxinos G, Watson C (2007) The rat brain in stereotaxic coordinates, Vol. 6. San Diego, CA: Elsevier Academic Press. Puil E, Gimbarzevsky B, Miura RM (1986) Quantification of membrane properties of trigeminal root ganglion neurons in guinea pigs. J Neurophysiol 55: Schall KP, Kerber J, Dickson CT (2008) Rhythmic constraints on hippocampal processing: state and phase-related fluctuations of synaptic excitability during theta and the slow oscillation. J Neurophysiol 99: Silvestri AJ (2005) REM sleep deprivation affects extinction of cued but not contextual fear conditioning. Physiol Behav 84: Smith C, Rose GM (1997) Post training paradoxical sleep in rats is increased after spatial learning in the Morris Water Maze. Behav Neurosci 111: Squire LR (1992) Memory and the hippocampus: a synthesis from findings with rats, monkeys, and humans. Psychol Rev 99: Squire LR, Zola SM (1996) Structure and function of declarative and nondeclarative memory systems. Proc Natl Acad Sci U S A 93: Staba RJ, Wilson CL, Fried I, Engel J Jr (2002) Single neuron burst firing in the human hippocampus during sleep. Hippocampus 12: Sutherl GR, McNaughton B (2000) Memory trace reactivation in hippocampal and neocortical neuronal ensembles. Curr Opin Neurobiol 10: Wang WT, Wan YH, Zhu JL, Lei GS, Wang YY, Zhang P, Hu SJ (2006) Theta-frequency membrane resonance and its ionic mechanisms in rat subicular pyramidal neurons. Neuroscience 140: Wu N, Hsiao CF, Chandler SH (2001) Membrane resonance and subthreshold membrane oscillations in mesencephalic V neurons: participants in burst generation. J Neurosci 21: Yang RH, Hu SJ, Wang Y, Zhang WB, Luo WJ, Chen JY (2008) Paradoxical sleep deprivation impairs spatial learning and affects membrane excitability and mitochondrial protein in the hippocampus. Brain Res 1230: Yang RH, Wang WT, Hou XH, Hu SJ, Chen JY (2010) Ionic mechanisms of the effects of sleep deprivation on excitability in hippocampal pyramidal neurons. Brain Res 1343: (Accepted 2 November 2010) (Available online 10 November 2010)