An Examination of Intrinsic Property
Differences Between Hippocampal CA1 and
CA3 Sharp-Wave Ripples
Date: 16th of October 2015
Author: Josien Elizabeth Visser Student number: 10213791
Study: Brain and Cognitive Sciences Research Master, Universiteit van Amsterdam
Daily supervisor: M.Sc. Ivana Milojevic Overall supervisor: Prof. Dr. C.M.A. Pennartz
Swammerdam Institute for Life Sciences, Cognitive and Systems Neuroscience group Universiteit van Amsterdam
Abstract
The hippocampus is a major component in the vertebrate brain, which plays a key role in episodic and spatial memory. Sharp-Wave Ripples (SPWRs) are local and fast network oscillations in the hippocampus which occur during awake immobility and slow-wave sleep. SPWRs are thought to promote synchronization of neuronal firing. This synchronization gain could strengthen synapses and promote memory consolidation. SPWRs can be observed in hippocampal areas CA1 and CA3. These areas are often analysed together due to firing pattern similarities. However, differences in cellular properties would suggest some major dissimilarities between those areas. In this research, differences in SPWR properties between area CA1 and CA3 were examined. Extracellular tetrode recordings were obtained when rats performed a navigational task and during slow-wave sleep. During these events, physiological data from CA1 and CA3 was gathered and the intrinsic properties of SPWRs were studied. It was observed that ripple frequency is higher in CA1 compared to CA3. On the contrary, this study revealed that ripple amplitude is higher in CA3 compared to CA1. In addition, in one rat a significant small difference in duration between ripples in CA1 and CA3 was found, but this effect was not observed in the other rat.
Introduction
The hippocampus (Greek: hippos = ‘horse’, campus = ‘monster’; often referred to as ‘seahorse’) is a major component located in the medial temporal lobe of the vertebrate brain. This brain area plays a key role in navigating, as well as episodic and spatial memory (Dragoi & Buzsáki, 2006; O’Keefe & Dostrovsky, 1971; Squire, 1992). The ‘hippocampal formation’ is a definition that is assigned to the Cornu Ammonis (CA) areas as well as the dentate gyrus (DG) and subiculum. Furthermore, the ‘hippocampus proper’ refers to the CA areas only, which are numbered from CA1 until CA3 and consist of three layers. These layers are called: stratum oriens, stratum pyramidale and stratum radiatum/ lacunosum
moleculare. Moreover, the stratum pyramidale contains cell bodies of pyramidal neurons (Dokter & von Bohlen und Halbach, 2012; Green, 1964). As can be observed in figure 1, the hippocampal formation contains a loop that is associated with synaptic transmission, called the trisynaptic circuit. In this circuit, signals from the cortex proceed via the entorhinal cortex (EC) to the DG. The DG comprises granule cells, which project to the pyramidal cells in CA3 along the mossy fibres. CA3 is connected with CA1 through the Schaffer Collaterals. Lastly, CA1 projects back to EC (Andersen, Bliss & Skrede, 1971). These findings show that the regions in the hippocampus are closely connected, which indicates that these areas all interplay in order to establish navigation or memory consolidation.
Figure1. Schematic representation of hippocampal areas and their connections in the rodent brain. The first projection in the trisynaptic circuit connects the entorhinal cortex with the dentate gyrus (1). The dentate gyrus is in close connection with CA3 via the Mossy fibres (2). CA3 projects to CA1 trough the Schaffer Collaterals (3). Fibres from CA1 link back to the entorhinal cortex (4). (Adapted from Dokter & von Bohlen und Halbach, 2012). Two major types of oscillatory patterns are present in the hippocampus and these fluctuations are associated with different behavioural states. Firstly, theta rhythms (4-12Hz) can be observed during active exploration (Wishaw & Vanderwolf, 1973). Secondly, large amplitude-irregular activity (LIA) occurs during slow-wave sleep (SWS) and awake immobility (e.g. consummatory behaviour and grooming). LIA is characterised by large-amplitude oscillations of a wide-range of frequencies. Moreover, sharp-waves and sharp-wave ripples
(SPWRs) can be observed during this state (Vanderwolf, 1969). Sharp waves are epochs with a large negative, amplitude that are often accompanied by SPWRs. In addition, SPWRs are local and fast hippocampal oscillations (110-200Hz) (Buzsáki et al., 1992).
It is thought that SPWRs are important for memory consolidation. Consolidation is the process in which memory traces are gradually stabilized (McGaugh, 2000). Long-term potentiation (LTP) of synaptic transmission is thought to be the dominant cellular mechanism for this phenomenon. LTP is the continual chemical strengthening of synapses due to previous neuronal firing patterns. This occurrence was first observed in rabbits. In these animals, a high-frequency stimulation resulted in stronger and prolonged excitatory postsynaptic potentials (EPSPs) after one pulse. It was concluded that recent neural firing can enhance synaptic transmission effectiveness (Bliss & Lømo, 1973).
It is believed that during memory consolidation, besides LTP, also a reorganizational process occurs in which episodic or spatial memory is initially stored in the hippocampus, and gradually gives rise to neocortical substrates. This theory is supported by evidence that damage to the hippocampus disrupts recent memories, but old memories remain intact (Corkin, S. 1984; Sutherland et al., 2001). Moreover, research indicates that the hippocampus is in close connection with the cortex. By using anterograde and retrograde tracing methods in rhesus monkeys, it was revealed that the cortical-hippocampal pathway is established from CA1 via the subiculum towards the prefrontal cortex (Goldman-Rakic, Selemon & Schwartz; 1984). Furthermore, the hippocampus has a reciprocal connection with the entorhinal cortex. The entorhinal cortex is convergently connected to the perirhinal and parahippocampal cortices, which project to various areas in the neocortex (Lavenex & Amaral, 2000). These findings suggest that consolidation is not only the strengthening of a memory trace, but also a relocation process due to close inter-area communication between the hippocampus and the neocortex. However, there are different theories about the remaining role of the hippocampus in long-term episodic and spatial memory, after this consolidation process (McClelland, 1995; Nadel & Moscovitch, 1997; Squire & Alvarez, 1995). There are multiple arguments for the theory that SPWRs are involved in the memory consolidation process. Firstly, SPWRs are abundant in different areas of the hippocampus during awake immobility (Buzsáki et al., 1992), which is, as mentioned before, a structure that is associated with spatial and episodic memory (Squire, 1992). Secondly, SPWRs occur during SWS. SWS takes place during the deepest stage of sleep called non-rapid eye movement (NREM). It is believed that memory consolidation particularly occurs during NREM, since research indicates that during SWS, neurons that were active during a spatial task are re-activated during this sleep stage (Kudrimoti et al., 1999). Neuronal reactivation could promote LTP, which is important for synaptic stabilization. Furthermore, inhibitory cells increase their firing frequency during ripples in a phase-locked manner, which diminishes pyramidal cell firing when these interneurons are active. However, when interneurons are simultaneously inactive during SPWRs, pyramidal cells in various hippocampal areas fire in synchrony (Stark et al., 2014). This synchrony gain could promote LTP. Lastly, there is effective
hippocampal-neocortical temporal linkage during ripples that presumably provides a temporal window for information transmission between these areas (Sirota et al., 2003). A growing body of experimental evidence shows that SPWRs indeed are important for memory consolidation. Eschenko et al. (2008) discovered that after learning an odour-discrimination task, frequency and duration of SPWRs in post task SWS increases. Furthermore, ripple occurrence is highest in the first 30 minutes after the discrimination task and decreases over time. In addition, rats that fail to learn this task do not show increased ripple frequency and duration, which suggests that SPWRs are crucial for this type of learning. Further evidence for the relevance of SPWRs in consolidation was provided by place cell research (Lee and Wilson (2002)). Place cells are spatially tuned cells localized in hippocampal CA1 and CA3, which become active when the animal crosses a specific place in space, referred to as a ‘place field’ (O’Keefe & Dostrovsky, 1971). Lee and Wilson (2002) observed that when rats repeatedly walked a trajectory, the same place cells that were associated with this activity would fire in the identical temporal order during SPWRs, while the rat was in post-task SWS. This observation implies the presence of a replay mechanism. Replay is the reactivation of a sequence of place cells, which occurred during activity, outside their place field. Hippocampal replay could be important for memory consolidations, since pyramidal cells repeatedly fire during this phenomenon, which could promote synaptic enhancement (Bi & Poo, 1998). Furthermore, during SWS ripples, hippocampal-neocortical synchronization is established, which means that relocation of the memory could also be influenced by this replay mechanism (Sirota et al., 2003). In addition, disrupting SPWRs, by hippocampal afferent stimulation, during SWS can impair spatial learning due to altered neuronal firing in this area (Ego-Stengel & Wilson, 2010; Girardeau et al., 2009). Altogether, these results suggest that SPWRs, during SWS, play a crucial role in learning and memory. Not only during sleep, but also during immobile wakefulness, sharp- wave ripples arise in the brain (aSPWRs). Similarly to neuronal firing patterns during SWS SPWRs, research by Foster and Wilson (2006) indicates that during periods of immobility, replay of place cells sequences occurs during aSPWR epochs. However, during wakeful immobility, replay of place cells mainly takes place in reversed direction compared to the place cell sequence regarding exploration of the environment. Moreover, this event was detected more frequently when the animal was exploring a novel maze. It was suggested that this might be a mechanism for learning about recent experiences (Foster & Wilson, 2006). Likewise, during SPWR events, place field sequences that represent immediate future paths can also be observed (Pfeiffer & Foster, 2013). In this study electrophysiological recordings were performed in the rat’s hippocampus while the animal performed a spatial navigation task. The arena was a 2x2m open field with 36 foraging sites. The animal had to find a food reward in either an unknown or a predictable location. Once the rat received liquid chocolate at the unknown location, the animal had to go to the predictable location in order to obtain a new reward. It was observed that during SPWRs, place cell sequences closely encoded the rat’s immediate future trajectory. In conclusion, during wakefulness SPWRs probably have a role in consolidation of recent memories and are likely to be important for future routes planning.
SPWRs are the most synchronous oscillation in the brain. Since SPWRs promote hippocampal pyramidal cells to fire in a tight high-frequency synchrony, synapse strength can be altered (Buzsáki et al., 1992). SPWRs are induced due to a population burst of CA3 cells. Synchronous discharge of neurons in CA3 activates CA1 pyramidal cells and interneurons via the recurrent Schaffer collaterals (Buzsáki & Lopez da Silva, 2012; Ylinen et al., 1995). Many models that try to explain the underlying cellular mechanisms ripple generation in CA1 have been proposed (Maier et al., 2003; Taxidis et al., 2012; Ylinen et al., 1995). In order to provide consensus about this issue, Stark and his colleagues (2012) combined optogenetic manipulations with extracellular electrophysiology in area CA1 in rodents. With this technique, pyramidal cells and interneurons could be artificially depolarized and high frequency oscillations (HFOs) that arose after this procedure were examined. Firstly, pyramidal cells were optigenetically depolarized. This resulted in origination of HFOs, which properties resembled that of SPWRs. On the contrary, activation of interneurons did not result in the development of ripple events. However, it was observed that these inhibitory cells are necessary for the coordination of neuronal spiking into organized fluctuations. Furthermore, the inhibitory potentials generated by interneurons are presumably the emitters of a high network synchronization, because inhibitory cells can alter the discharge probability of pyramidal cells in a short temporal window (Buzsáki, 1997). This indicates that pyramidal cell firing is necessary for HFO generation, but inhibitory cells play a role in the organisation of high frequency oscillations. Furthermore, if pyramidal cell illumination was pared with a GABAA
receptor blocker HFOs would not arise. This indicates that fast inhibition is also a crucial mechanism for SPWR generation. Subsequently, the influence of pyramidal cells and interneurons on spontaneous ripples were studied. In order to investigate this, spontaneous ripples were detected and closed-loop optogenetics was performed on excitatory and inhibitory cells. On the one hand, pyramidal cell stimulation resulted in a duration increase of HFOs. On the other hand, HFOs were eliminated after direct excitatory cell silencing, or indirect suppression, via interneuron activation. These results suggest a model in which feedback and reciprocal inhibition between pyramidal cells and interneurons is combined (fig 2). According to this model, HFOs are induced due to pyramidal cell activation caused by extrinsic stimulation. Reciprocally interconnected interneurons organise this excitatory firing pattern into high frequency oscillations, also known as ripples (Stark et al., 2012). In addition, during these SPWR events approximately 10-18% (50,000 – 100,000) of all neurons in the hippocampus produce one or more action potentials (Csicsvari et al., 1999). This large synchrony gain could increase synaptic plasticity. In summary, due to the cellular properties of the hippocampus, SPWRs can increases synchronous discharge of pyramidal cells. This is a mechanism that promotes plasticity, which could be important for memory consolidation.
Figure2. Schematic representation cellular model ripple generation in CA1. External input activates pyramidal cells (pyr). The
reciprocally connected
interneurons (int) will fire, which will simultaneously inhibit both pyramidal cells and induce high frequency oscillations.
In place cells research, CA1 and CA3 are often analysed together due to firing pattern similarities (Muller & Kubie, 1989). However, there seems to be differences in place field encoding between these areas (Leutgeb et al., 2004). The proportion of active place cells in CA3 is significantly lower compared to CA1. Moreover, place cells show more overlap in between areas in CA1, which means that there is an increased probability that a CA1 place cell is active in two distinct spatial locations. This assumes that place cells in CA3 could give more information about the exact location of the rat. Moreover, the same study indicated that while exploring a novel environment, CA3 place cell representations developed slower than in CA1 (Leutgeb et al., 2004). In summary, these data suggest that there could be a distinct role for CA1 and CA3 in place encoding.
CA1 and CA3 also differ with regard to SWPR properties. In vitro research found that the intra-burst frequency is increased in CA1 (210Hz ± 16Hz) compared to CA3 (193 ± 14Hz) and the degree of pyramidal population synchrony is higher in CA1 (Csicsvari et al., 2000; Maier
et al., 2003). However, the prevalence of ripples is similar in CA1 and CA3 (Decker et al.,
2009; Ramadan et al., 2009). Decker et al., (2009) also investigated CA1-CA3 ripple property differences in vitro and observed that ripple amplitude is higher in CA3 (2.6±0.15mV) compared to CA1 (1.5±0.17mV). Furthermore, ripple duration was slightly longer in CA3 (61.0±1.2ms) than in CA1 (54.6±1.1ms). However, this difference was not statistically verified. Similar to in vitro research, differences in ripple properties between CA1 and CA3 have also been observed in in vivo studies. The study by Sullivan et al., (2011) also discovered that SPWRs in CA3 have a higher oscillation frequency compared to CA1. However, this difference in oscillatory frequency between the areas was much larger than the fluctuation rate observed in vitro. It was revealed that SPWRs in the CA3 area often oscillate at a frequency below 110 Hz and hippocampal region CA1 ripples resonate at approximately 170 Hz.
In order to gain more knowledge about the differences and similarities in CA3 and CA1 SPWRs in vivo, this study examined ripple properties (duration, intra-burst frequency, amplitude and spectral power) in both areas. In these experiments ripple data was obtained by performing electrophysiological tetrode recordings in rats, during sleep and while the animals were involved in a spatial navigation task. Data was gathered while tetrodes were
positioned in either hippocampal area CA1 or CA3. It will be hypothesised that the data will follow the results from previous literature. The duration of SPWRs are hypothesized to be slightly longer in CA3 compared to CA1. It is expected that CA1 ripples will resonate around 170Hz. In CA3 the preferred frequency of approximately 110Hz will be envisioned. Lastly, it will be hypothesized that SPWR amplitude is higher in CA3 compared to CA1.
Material & Methods
Animal and Surgical procedures
Data was collected from two male Lister Hooded rat. Upon arrival the first animal was 7 weeks old and weighed 185.5 grams. The second rat was 8 weeks old and weighed 225.5 grams. Animals were housed in a group of three under a reversed 12h light/ 12h dark cycle with light turned on at 08:00am. In the first two weeks of arrival, the animals were handled by the experimenters and habituated on the maze. Subsequently, the rats were trained to learn the behavioural spatial navigation task (see paragraph ‘behavioural procedures’). The rats were trained and tested in their active period. The animals were food-restricted up to 85% of the ad libitum individual growth curve. After pretraining to proficiency, at least 70% correct trials on the spatial navigational task (see paragraph ‘behavioural procedures’ for specifications of a correct trial), recording electrodes were implanted. Surgery was
performed when the first rat was 20 weeks old and weighed 386 grams. The second animal was implanted at an age of 18 weeks and weighed 350 grams. The rats were anesthetized with isoflurane and head fixed in a stereotaxic apparatus. Buprenorphine (0.01-0.01 mg.kg) and meloxicam (2mg/kg) were administered to provide prolonged analgesia. In order to prevent infections, the animals were given with Baytril (5mg/kg). After the skull was exposed, lidocainebase (10-20mg) was applied as a local anaesthetic on the periost. As figure 3
presents, two craniotomies were performed, which were above the hippocampus (AP = -2.8mm ML = -2.78mm) and above the prelimbic cortex (AP = +3.13mm, ML=-1.17mm). The rats were implanted with a custom built dual-drive loaded with 36 tetrodes (18 per brain area). This drive design was based on a split drive model, used for recordings from two brain areas simultaneously (Lansink et al., 2007). During this study, hippocampal area CA3 and the prelimbic cortex were concurrently recorded. In this paper only the hippocampal data will be considered. As can be seen in figure 4, 18 tetrodes (California Fine Wire Co., size 0.0005 (12.5μm), Tungsten 99.95%, HFV insulation) were lowered towards the hippocampus with anterior angle of 9.11° and posterior 10.81° from the midline. The target areas were reached in approximately 7 to 10 days after surgery. Targeting CA1 and CA3 was confirmed by
monitoring the local field potentials (LFPs) and cell density. Furthermore, the depth of the tetrode was closely tracked by calculating the approximate distance the tetrode had travelled. When the tetrodes were positioned in hippocampal area CA1, two recording sessions, on separate days, were performed. Subsequently, the tetrodes were lowered towards area CA3 and the other recording sessions were conducted. Reference electrodes for the hippocampus were placed in the corpus callosum. All experimental procedures were conducted in accordance with the Dutch National Animal Experiments regulations (de Wet op Dierproeven) and with approval of the animal ethics committee of the Universiteit van Amsterdam.
Behavioural Procedures
To provide neurobiological data that can give more information about the intrinsic property differences of SPWRs observed in CA1 and CA3 extracellular tetrode recordings were
performed while the rat was involved in a spatial navigation task. As can be seen in fig 5, the maze consisted of two elevated side arms with a start chamber in between. A modifiable trajectory of the maze was located in front of the start chamber, called spatial exploration trajectory (SET). Maze alterations could be made in order to investigate physiological changes when the rats would explore a novel trajectory. However, in this paper only the maze
configuration that is presented in figure 5 is considered. On this altered T-maze a food well was located at the end of each arm and the start chamber and SET were separated by a sliding door. In SET, each arm contained an oblong, green light-emitting diode (LED).
Figure 4. Locations of the hippocampal recordings. Tetrodes were lowered with an angle (Anterior 9.11° and posterior 10.81°). During the first two sessions, CA1 was recorded. In the following recordings, data from CA3 was gathered.
The rats began the task by placement of the animal in the start chamber with a closed door. A small reward (50 µl 15% liquid sucrose) could be obtained by poking one of the reward ports on the side arms. Due to this nose poke, one of the green LEDs switched on that indicated at which arm in SET the following, large reward could be collected. Since the side arms were elevated with respect to the SET, the rats were able to observe the light and trajectory, which provided a possibility for the animals to plan future movement in order to obtain the next reward. The rats had to return to the start chamber and to wait until the door slid down, in order to enter SET. Moreover, when the door opened, the green LED switched off. If the animal went to the reward port at the arm were the LED light was illuminated before the door opened, the choice was counted as correct. However, if the other reward port in SET was poked, the choice was classified as incorrect. If a correct choice was executed, a large reward (120 µl 15% liquid sucrose) was provided. Thereafter, the rats had to go back to the start chamber and new trial started.
Figure5. Maze configuration. First, the rat was placed in the start chamber and had to obtain a small reward at
one of the side arms. Subsequently, one of the lights turned on. Thereafter, the animal had to wait for the door until it opened. Once the door slid down the light would turn off. The rat had to poke the reward port on the end of the arm in Spatial Exploration Trajectory (SET) that was illuminated in order to earn a big reward.
During recording sessions, awake SPWRs could be observed during consummatory behaviour (at the rewards ports), while the rats was waiting in front of the door, when the rat showed grooming behaviour and during other periods of immobility. Furthermore, before the behavioural task, the rats were placed on a flowerpot for 30 to 60minutes, covered with a cloth in order to record SPWRs during pre-task sleep.
Data collection
After sufficient recovery from surgery, the recording sessions would start. Physiological signals were recorded during pre-task sleep and while the rats were freely moving on the maze, performing the navigational task. The maze task was conducted under red light. Behaviour of the rat was recorded with a camera (JAI/CV-S3200P, 6mm camera lens, PAL-752x582 pixel resolution, 8.6x8.3 pixel size, 0.09 Lux sensitivity on sensor)that was mounted on the ceiling above the maze. In addition, animal location and head direction were captured by video-tracking LEDs, mounted on the sides of the dual-drive. Neurophysiological signals from the tetrodes were acquired by a 144 channel recording system (Neuralynx Cheetah Software, Bozeman, United States). LFPS were sampled at 2 kHz and band-pass filtered between 1-500Hz on all tetrodes. Furthermore, ripples were band-pass filtered between 100-300Hz. Extracellular action potentials were sampled at 32 kHz and band-pass filtered between 6-6000Hz. Events of the navigational task were acquired and time-stamped by the recording system.
Histology
After the last recording session, the animals were injected with a lethal dose of 20% Euthasol (2.0 mL). In order to visualize the tetrode end points, a small electrical current of 12µA was passed for 10s through all leads, one lead per tetrode, which caused small lesions in the brain. Thereafter, a perfusion with formaldehyde was performed (4% paraformaldehyde). When the fixation was completed, the head and drive were stored in the fridge for at least 24 hours to allow post fixation. Subsequently, the brains were extracted and cut in coronal sections of 40µm using a vibratome (Leica, VT1000S, Germany). Lastly, the sections were Nissl stained in order to make the tetrode tracks distinguishable.
Ripple detection
The procedures for ripple detection were based on those described previously (Csicsvari et
al., 1999) with additional refinements in order to minimize false-positive detections and
optimize ripple detection. LFP was filtered between 100 and 250Hz (Notch filter). It was decided to filter the signal above 100Hz instead of 80Hz in order to prevent gamma waves to be detected and influence the results. SPWRs were identified as periods in which the LFP power exceeded 3SD above baseline noise. The beginning and end of the ripple was
classified as the time point in which the LFP power fell below 1SD. The minimal ripple length was 25ms and ripple events separated by less than 100ms were merged.
Data analysis
For all ripples duration, frequency, peak amplitude and power were determined.
Subsequently, one lead during the recording sessions with tetrodes in CA1 and one during the session in CA3 with the highest number detected SPWRs was selected for analysis. For each variable, normality was tested with a Kolmogorov-Smirnov test and determined based on a histogram and q-q plot. If the data met the normality conditions, an independent samples t-test was conducted to compare this ripple property of CA1 and CA3 SPWRs. However, if the normality requirement were not fulfilled an alternative test, the Wilcoxon rank sum test, was carried out. Since 5 different variables were compared a Bonferroni correction was performed in order to counteract the problem of multiple comparisons. Therefore, the hypotheses were tested at p = 0.01. Moreover, the mean power for every frequency that could be observed in CA1 and CA3 was calculated and plotted.
Another purpose of this ripple examination study was to determine whether a tetrode was located in hippocampal layer CA1 or CA3. Therefore, a ripple categorization script was programmed which could classify leads from all tetrodes on various ripple parameters. These
classification paratmeters were determined by selecting two tetrodes that were highly likely to be in either CA1 or CA3. This selection process was executed by looking at the turning sheets, where the depth of the tetrodes were kept. Also, amount of cells and the amplitude and shape of SPWRs were examined. Subsequently, it was determined which ripple
properties differed between CA1 and CA3 and between tetrodes that were not in one of these layers. It was observed that the best classification ripple properties were frequency and amplitude. Based on these parameter values the tetrode classification script was programmed (for example see Supplementary Table I and II).
Results
Histology
To investigate the differences between hippocampal CA1 and CA3 SWPR intrinsic characteristics, electrophysiological recordings in these areas were carried out in two rats. Firstly, data was gathered while the tetrodes were placed in CA1 and thereafter a recording session with tetrodes in CA3 was performed. In order to verify the tetrode location in CA3, Nissl stained coronal section were visually examined. Figure 6 provides an example of a stained section from the first rat. During the last recording session, hippocampal cells were not observed on any of the leads anymore. This is an indication for drive instabilities, thus the recordings were ceased and the rat was sacrificed.
Figure 7 depicts a Nissl stained section with tetrode marks of the second animal. Since the drive fell off after finishing one of the recording sessions, no lesions could be performed in this animal. However, the tetrode tracks are still visible, which indicates that the leads went into the correct direction (fig 7).
Figure6. Nissl stained coronal sections of the hippocampus in the right hemisphere of rat 1. The tetrodes were designated to be placed in hippocampal area CA3.
Figure7. Nissl stained coronal sections of the hippocampus in the right hemisphere of the second rat. The tetrodes were targeted to be placed in hippocampal area CA3. Due to complications no lesions could be made. However, tetrode tracks are visible.
SPWR amplitude
While doing the analysis, firstly, it was observed that in both animals the amount of detected ripples was much higher in CA1 (rat 1, n=976; rat 2, n = 531) compared to CA3 (rat 1, n = 364; rat 2, n = 97). Figure 8 illustrates examples of detected ripples in hippocampal area CA1 and CA3 extracted from band-pass filtered data between 100 and 250Hz. Subsequently, in order to test the differences between CA1 and CA3 ripple amplitude, an independent samples t-test was conducted. In the first rat a significant difference could be observed between area CA1 (M = 40.8µV, SD = 11.79) and CA3 (M = 55.80µV, SD = 12.11) in the hippocampus; t(1338) = 70.01, p < 0.0001 (fig 8A). The effect size for this analysis (d = 0.43) was found to be close to the Cohen’s conventions for a medium effect (d = .50). In the second rat the t-test also yielded a significant result (t (626) = 4.60, p = 0.0055). However, the difference in amplitude between hippocampal area CA1 (M = 46.90µV, SD = 14.12) and CA3 (M = 48.42µV, SD = 7.97) was much smaller in this animal (fig 9B). The effect size (d = 0.13) was found to be small according to Cohen’s values (d = .2).
Figure 8. Example ripples in the CA1 and CA3 area of the hippocampus. Extracted from filtered data (Band-pass filter 100-250Hz).
SPWR Duration
In order to test the hypothesis that ripple duration is the same in the two areas, again an independent samples t-test was performed. In rat 1 a significant difference was detected between CA1 (M = 0.055s, SD = 0.018) and CA3 (M = 0.060s, SD = 0.02285); t (1338) = 1.11,
p = 0.00045. Cohen’s effect size value (d = 0.24) suggested a small practical significance (fig
10A). On the contrary, the variability in ripple duration between CA1 (M = 0.055, SD = 0.021) and CA3 (M = 0.059, SD = 0.021) was not significant in the second rat (t (626) = 0.24,
p=0.0651) (fig 10B). The effect size for the ripple duration analysis in the second rat (d =
0.19), indicates a small practical significance.
SPWR frequency
After the ripple duration analysis, frequency data was examined. Because the data was skewed for one variable in both rats, a Wilcoxon rank sum test was conducted. As figure 11A indicates, in the first rat frequency was significantly higher in CA1 (M = 166.05Hz, SD = 23.86, Mdn = 170.9Hz) compared to CA3 (M = 118.38Hz, SD = 6.55, Mdn = 117.19Hz); Z = 27.73, p < 0.0001. Correspondingly, the difference between CA1 and CA3 ripple frequency was also significant for the second rat (fig 11B); Z = 14.5, p < 0.0001. The analysis of the second animal revealed that the frequency was also higher in CA1 (M = 165.57Hz, SD = 24.79, Mdn = 167.97Hz) compared to CA3 (M = 116.36Hz, SD = 4.23, Mdn = 155.23Hz). The effect sizes for rat 1 (r = 0.76) met the conventions for a large effect. Additionally, the effect size for the frequency data of rat 2 (r = 0.58) exceeded the conditions for a medium effect.
A B
Figure9. Averaged Sharp-Wave Ripple (SPWR) amplitude of data recorded from either CA1 or CA3 in rat 1 (9A) and rat 2 (9B), with 95% confidence intervals.
Ns = P > 0.01; ** = P ≤ 0.01; *** = P ≤ 0.001; **** = P ≤ 0.0001
A B
Figure10. Averaged Sharp-Wave Ripple (SPWR) duration of data recorded from either CA1 or CA3 in rat 1 (10 A) and rat 2 (10B), with 95% confidence intervals.
Ns = P > 0.01; ** = P ≤ 0.01; *** = P ≤ 0.001; **** = P ≤ 0.0001
A B
Figure11. Averaged Sharp-Wave Ripple (SPWR) frequency of data recorded from either CA1 or CA3 in rat 1 (11A) and rat 2 (11B), with 95% confidence intervals.
Ns = P > 0.01; ** = P ≤ 0.01; *** = P ≤ 0.001; **** = P ≤ 0.0001
SPWR power
Figure 12 presents the mean power for every oscillatory frequency that was observed in area CA1 and CA3. From this graph it can be inferred that the mean power in CA1 is increased around 170Hz. In rat 1 there is also a power increase at 200Hz, but in rat 2 this was not observed. In CA3 of the first rat, the peak power can be found at 110Hz and this decreases with increasing power. In the second rat the highest power can be observed at 120Hz.
Figure 12. Group mean power spectra. Power across frequencies for both rats in CA1 (above) and CA3 (below) with the standard deviations.
Discussion
In vivo electrophysiology recordings enabled us to characterize the differences between
hippocampal area CA1 and CA3 intrinsic ripple properties. Because area CA1 and CA3 have comparable firing patterns (O’Keefe & Dostrovski, 1971; Leutgeb et al., 2004) and there is a close connection between these regions via the Schaffer collaterals (Andersen, Bliss & Skrede, 1971), one might expect that oscillatory properties are similar in these sections. However, this study revealed that some SPWR properties can be similar, but most ripple characteristics are highly different between those areas.
This research indicated that ripple frequency and amplitude are properties that are dissimilar between area CA1 and CA3 in the hippocampus. Moreover, the power of the frequency was highest around 170Hz in CA1 and in CA3 the largest amplitude could be found around 110Hz and 120Hz in the first and second rat respectively. In contrast, duration differences between hippocampal area CA1 and CA3 was non-significant in one rat, but significantly different in the other.
Histology
From the Nissl-stained sections it could be observed that there were leads which were placed (near) CA3. In the second rat this could not be concluded, since the lesions could not be carried out. However, the tetrode tracks reveal that the tetrodes went in the brain in the direction of CA3.
SPWR amplitude
In vitro experiments indicate that SPWR amplitude is higher in CA3 (2.6±0.15mV) compared
to CA1 (1.5±0.17mV) (Decker et al., 2009). The results of this study support and augment the findings by showing that in vivo it was also observed that in hippocampal area CA3 (Rat1: A = 55.80±12.11µV; Rat2: A = 48.42±7.97µV) compared to CA1 (Rat1: A = 40.8 ±11.79µV; Rat2: A = 46.90±14.12µV. However, in this study, the observed amplitude is smaller compared to the study in vitro by Decker and his colleagues (2009). This discrepancy can be caused by a variation
in cell properties in a living animal compared to cell slices as well as a difference in recording technique and ripple detection parameters. In the study by Decker et al., (2009) recording pipettes were used in order to record the field potentials. Moreover, in order to distinguish the SPWRs from the raw-data, Decker and his colleagues band-pass filtered the signal from 100 to 350Hz and the detection threshold was set to 4-6SD above baseline noise. In this study the data was band-pass filtered from 100 to 250 Hz and the power had to exceed 3SD in order to detect SPWRs. This variation in threshold could have influences the discrepancy in SPWR amplitude.
SPWR duration
Previous in vitro research revealed that SPWR duration in hippocampal area CA3 is 61.0±1.2ms and in CA1, 54.6±1.1ms (Decker et al., 2009). However, in the study by Decker and his colleague, these values were not statistically compared. In this research, similar duration times were observed. In the first rat, the average CA3 ripple duration was
60±22.9ms and in CA1, 55±18ms, which was a significant difference. On the contrary, the results of the second rat were not significantly different (CA3, 59±21ms; CA1, 55±21ms). Consequently, this study cannot give any consensus about whether there is a difference in ripple duration between hippocampal area CA1 and CA3. A possible explanation for this discrepancy in significance between the first and second rat could be that there was a difference in SPWR sample size between the first rat (CA1, n = 976; CA3, n = 364) and the second animal (CA1, n = 531; CA3, n = 97). In conclusion, there seems to be a trend with a longer CA3 SPWR duration compared to CA1, but future research is necessary to investigate whether there is a significant duration discrepancy between these areas.
SPWR frequency and power
Recently, in vivo electrophysiological experiments have shown that the preferred oscillation frequency in CA3 is below 110Hz. Furthermore, a frequency of 170Hz is mostly observed hippocampal CA1 (Sullivan et al., 2011). It is thought that the CA1 frequency is higher due to the presence of interconnected interneurons in this area, which discharge at a high frequency (Buzsáki et al., 1997). According to the present study, the resonance frequency is also higher in CA1 compared to CA3. It was found that the average frequency in CA1 for the first rat was 166.05Hz and for the second animal 165.57Hz. This value is negligible lower compared to the ripple frequency found in CA1 by Sullivan et al., (2011). Furthermore, the average resonance frequencies for CA3 was 118.38Hz and 116.36Hz for rat1 and rat2 respectively, which is a bit higher than was observed in previous literature.
The highest mean power in CA1 was observed around 170Hz. In contrast, in rat 1 in this study, the highest mean peak power could also be observed around 200Hz. However, this could have occurred due to the low amount of detected SPWRs with higher frequency that biased the data. The highest mean peak power in CA3 was observed at 110Hz in rat1 and the power decreased with increasing frequency. Ripples detected in CA3 of rat 2 had the highest power occurred around 120Hz. However, this could be biased due to the low amount of detected CA3 SPWRs in the second rat. In addition, Sullivan and his colleagues showed that ripples can oscillate at a frequency range from 50-300Hz. In this study it was decided to filter from 100 till 250Hz, since it was observed that when the signal was filtered from 80Hz and above, gamma was often detected as a ripple. However, this could have led to a failure of detection of ripples with a lower oscillation frequency. Moreover, in order to examine the interaction and phase-consistency between CA1 and CA3 SPWRs, a simultaneous recording in these two regions should be performed.
In summary, from this study it can be concluded that SPWR properties in CA1 and CA3 are significantly different. Due to firing pattern similarities, CA1 and CA3 are often analyses together. However, this research implicates that due to these major differences these two areas should be considered separately.
Acknowledgements
I would like to offer my special thanks to Prof. Dr. Pennartz and Dr. Lansink for giving me this internship opportunity. Moreover, I would like to express my very great appreciation to Ivana Milojevic, M.Sc. for her guidance, enthusiastic encouragement and useful critiques on my work. My grateful thanks are also extended to Drs. Jan Lankelma and Silviu Rusu, Ph.D. for their help in improving my programming skills. I also would like to thank Drs. Jan Lankelma for his help in doing the statistical analysis. Furthermore, assistance provided by Jeroen Bos, M.Sc. was greatly appreciated. Finally, I would like to thank the students (from my side of the wall) for the discussions and the great time!
References
1. Andersen, P., Bliss, T.V.P. & Skrede, K.K. (1971). Lamellar organization of hippocampal excitatory pathways. Experimental Brain Research, 13, 222-238.
2. Buzsáki, G. (1997). Functions for interneuronal nets in the hippocampus. Canadian Journal of Physiology and Pharmacology, 75, 508-515.
3. Buzsáki, G., Horváth, Z., Urioste, R., Hetke, J. & Wise, K. (1992). High-frequency network oscillations in the hippocampus. Science, 256, 1025-1027.
4. Buzsáki, G. & Lopes da Silva, F. (2012). High frequency oscillations in the intact brain. Progress in Neurobiology, 98, 241-249.
5. Bliss, T.V.P. & Lømo, T. (1973). Long-lasting potentiation of synaptic transmission in the dentate area of the anaesthetized rabbit following stimulation of the performant path. The Journal of Physiology, 232, 331-356.
6. Corkin, S. (1984). Lasting consequences of bilateral medial temporal lobectomy: clinical course and experimental findings H.M. Seminars in Neurology, 4, 249-259.
7. Csicsvari, J., Hirase, H., Czurkó, A., Mamiya, A. & Buzsáki, G. (1999). Oscillatory coupling of hippocampal pyramidal cells and interneurons in the behaving rat. The Journal of Neuroscience, 19, 274-287.
8. Csicsvari, J., Hirase, H., Mamlya, A. & Buzsáki, G. (2000). Ensemble patterns of hippocampal CA3-CA1 neurons during sharp wave associated population events. Neuron, 28, 585-594.
9. Decker, J.M., Wójtowicz, A.M., Ul Haq, R., Braunewell, K.H., Heinemann, U. & Behrens, C.J. (2009). C-type natriuretic peptide decreases hippocampal network oscillations in adult rats in vitro. Neuroscience, 164, 1764-1775.
10. Dokter, M. & von Bohlen und Halback, O. (2012). Neurogenesis within the adult hippocampus under physiological conditions and in depression. Neural Regeneration Research, 7, 552-559.
11. Dragoi, G. & Buzsáki, G. (2006). Temporal encoding of place sequences by hippocampal cell assemblies. Neuron, 50, 145-157.
12. Draguhn, A., Traub, R., Schmitz, D., & Jefferys, J. (1998). Electrical coupling underlies high-frequency oscillations in the hippocampus in vitro. Nature, 394, 189–192.
13. Eschenko, O., Ramadan, W., Mölle, M., Born J. & Sara, S.J. (2008). Sustained increase in hippocampal sharp-wave ripple activity during slow-wave sleep after learning. Learning and Memory, 15, 222-228.
14. Ego-Stengel, V. & Wilson, M.A. (2009). Disruption of ripple-associated hippocampal activity during rest impairs spatial learning in the rat. Hippocampus, 20, 1-10.
15. Foster, D.J. & Wilson, M.A. (2006). Reverse replay of behavioural secquences in hippocampal place cells during the awake state. Nature, 440, 680-682.
16. Giap, B.T., Jong, C.N., Ricker, J.H., Cullen, N.K. & Zafonte, R.D. (2000). The hippocampus: Anatomy, pathophysiology and regenerative capacity. Journal of Head Trauma Rehabilitation, 15, 875-894.
17. Goldman-Rakic, P.S., Selemon, L.D. & Schwartz, M.L. (1984). Dual pathways connecting the dorsolateral prefrontal cortex with the hippocampal formation and parahippocampal cortex in the rhesus monkey. Neuroscience, 12, 719-743.
18. Green, J.D. (1964). The Hippocampus. The American Physiological Society, 44, 561-608.
19. Kudrimoti, H.S., Barnes, C.A. & McNaughton, B.L. (1999). Reactivation of hippocampal cell assemblies: effects of behavioural state, experience and EEG dynamics. The Journal of Neuroscience, 19, 4090-4101.
20. Lavenex, P. & Amaral, D. G. (2000). Hippocampal-neocortical interaction: a hierarch of associativity. Hippocampus, 10, 420-430.
21. Lansink, C.S., Bakker, M., Buster, W., Lankelma, J., van der Blom, R., Westdorp, R., Joosten, R.N.J.M.A., McNaughton, B.L. & Pennartz, C.M.A. (2007). A split microdrive for simultaneous multi-electrode recordings from two brain areas in awake small animals. Journal of Neuroscience Methods, 162, 129-138.
22. Lee, A.K. & Wilson, M.A. (2002). Memory of sequential experience in the hippocampus during slow wave sleep. Neuron, 36, 1183-1194.
23. Leutgeb, S., Leutgeb, J.K., Treves, A., Moser, M. & Moser, E.I. (2004). Distinct ensemble codes in hippocampal areas CA3 and CA1. Science, 305, 1295-1298.
24. Maier, N., Nimmrich, V. & Draguhn, A. (2003). Cellular and network mechanisms underlying spontaneous sharp wave-ripple complexes in mouse hippocampal slices. The Journal of Physiology, 550, 873-887.
25. McClelland, J.L., McNaughton, B.L. & O’Reilly, R.C. (1995). Why there are complementary learning systems in the hippocampus and neocortex: insights from the successes and failures of connectionist models of learning and memory. Psychological Review, 102, 419-457.
26. McGaugh, J.L. (2000). Memory- a century of consolidation. Science, 287, 248-251.
27. Montogomery, S.M. & Buzsáki, G. (2007). Gamma oscillations dynamically couple hippocampal CA3 and CA1 regions during memory task performance. Proceedings of the National Academy of Sciences, 104, 14495-14500.
28. Muller, R.U. & Kubie, J.L. (1989). The firing of hippocampal place cells predicts the future position of freely moving rats. The Journal of Neuroscience, 9, 4101-4110.
29. Nadel, L. & Moscovitch, M. (1997). Memory consolidation, retrograde amnesia and the hippocampal complex. Current Opinion in Neurobiology, 7, 217-227.
30. O’Keefe, J. & Dostrovsky, J. (1971). The hippocampus as a spatial map. Preliminary evidence from unit activity in the freely-moving rat. Brain Research, 34, 171-175.
31. Paxinos, G. & Watson, C. (1998). The rat brain in stereotaxic coordinates. New York, Academic Press.
32. Pfeiffer, B.E. & Foster, D.J. (2013). Hippocampal place cell sequences depict future paths to remembered goals. Nature, 497, 74-79.
33. Ramadan, W., Eschenko, O., Sara, S.J. (2009) Hippocampal sharp wave/ ripples during sleep for consolidation of associative memory. PLOS one, 1-9.
34. Sirota, A., Csicsvari, J., Buhl, D. & Buzsáki, G (2003). Communication between neocortex and hippocampus during sleep in rodents. Proceedings of the National Academy of Sciences, 100, 2065-2069.
35. Stark, E., Roux, L., Eichler, R., Senzai, Y., Royer, S. & Buzsáki, G. (2014). Pyramidal cell-interneuron interactions underlie hippocampal ripple oscillations. Neuron, 83, 467-480.
36. Sullivan, D., Csicsvari, J., Mizuseki, K., Montgomery, S., Diba, K. & Buzsáki, G. (2011). Relationships between hippocampal sharp waves, ripples and fast gamma oscillations: influence of dentate and entorhinal cortical activity. The Journal of Neuroscience, 31, 8605-8616.
37. Sutherland, R.J., Weisend, M.P., Mumby, D., Astur, R.S., Hanlon, F.M., Koerner, A., Thomas, M.J., Wu, Y., Moses, S.N., Cole, C., Hamilton, D.A. & Hoesing, J.M. (2001). Retrograde amnesia after hippocampal damage: recent vs remote memories in two tasks. Hippocampus, 27-42.
38. Squire, L.R. (1992). Memory and the hippocampus: A synthesis from findings with rats, monkeys, and humans. Psychological review, 99, 195-231.
39. Squire, L.R. & Alvarez, P. (1995). Retrograde amnesia and memory consolidation: a neurobiological perspective. Current Opinion in Neurobiology, 5, 169-177.
40. Taxidis, J., Coombes, S., Mason, R. & Owen, M.R. (2011). Modeling Sharp Wave-Ripple complexes through a CA3-CA1 network model with chemical synapses, Hippocampus, 22, 995-1017
41. Vanderwolf, C.H. (1969). Hippocampal electrical activity and voluntary movement in the rat. Electroencephalography and Clinical Neurophysiology, 26, 407-418.
42. Wishaw , I.Q. & Vanderwolf, C.H. (1973). Hippocampal EEG and behavior: change in amplitude and frequency of RSE (Theta rhytm) associated with spontaneous and learned movement patterns in rats and cats. Behavioural Biology, 8, 461-484.
43. Ylinen, A., Bragin, A., Nádasdy, Z., Jandó, G., Szabó, I., Sik, A. & Buzsáki, G. (1995). Sharp Wave- associated high-frequency oscillation (200Hz) in the intact hippocampus: Network and intracellular mechanisms. The Journal of Neuroscience, 15, 30-46.
Supplementary files
CA1 Freq Amp CA3 Freq Amp No Layer Freq Amp
HPPC1_1 165.99 58.93 HPPC1_2 166.00 57.10 HPPC1_3 166.62 59.59 HPPC1_4 166.05 40.88 HPPC2_1 150.41 37.88 HPPC2_2 132.79 14.27 HPPC2_3 157.83 26.07 HPPC2_4 160.72 28.13 HPPC3_1 168.20 41.14 HPPC3_2 167.01 29.39 HPPC3_3 167.80 30.40 HPPC3_4 167.16 30.08 HPPC4_1 127.49 33.34 HPPC4_2 125.16 34.16 HPPC4_3 127.86 33.34 HPPC4_4 130.28 23.71 HPPC5_1 149.68 39.46 HPPC5_2 143.35 37.71 HPPC5_3 151.42 37.83 HPPC5_4 149.05 27.56 HPPC6_1 149.52 37.59 HPPC6_2 157.08 40.29 HPPC6_3 142.28 36.04 HPPC6_4 141.52 23.41 HPPC7_1 116.42 0.92 HPPC7_2 116.67 36.59 HPPC7_3 117.35 36.52 HPPC7_4 117.78 26.56 HPPC8_1 119.90 31.23 HPPC8_2 139.43 42.82 HPPC8_3 117.66 33.07 HPPC8_4 119.05 24.09 HPPC9_1 147.08 42.70 HPPC9_2 144.21 40.36 HPPC9_3 129.28 45.56 HPPC9_4 151.21 33.62 HPPC10_1 163.02 75.20 HPPC10_2 163.54 84.81 HPPC10_3 164.20 58.43 HPPC10_4 163.57 62.11 HPPC11_1 162.68 89.09 HPPC11_2 159.63 88.94 HPPC11_3 162.06 86.56 HPPC11_4 160.94 65.75 HPPC12_1 168.35 38.52 HPPC12_2 161.10 36.93 HPPC12_3 165.84 27.13 HPPC12_4 154.03 25.25
HPPC13_1 162.97 50.23 HPPC13_2 162.54 52.62 HPPC13_3 164.79 45.55 HPPC13_4 166.28 33.60 HPPC14_1 122.46 33.60 HPPC14_2 122.46 27.66 HPPC14_3 117.81 26.26 HPPC14_4 131.79 22.26 HPPC15_1 161.66 57.98 HPPC15_2 163.16 72.26 HPPC15_3 161.71 68.70 HPPC15_4 164.58 45.24 HPPC16_1 152.05 72.20 HPPC16_2 152.00 71.96 HPPC16_3 152.60 50.85 HPPC16_4 152.71 50.81
Supplementary Table I. Ripple categorization with tetrodes designated to be in hippocampal area CA1 of the first rat. Based on the tetrodes that were highly likely to be in this layer the classification conditions were set to ripple amplitude >= 35mV and ripple frequency >= 135Hz.
CA1 Freq Amp CA3 Freq Amp No Layer Freq Amp
HPPC1_1 115.71 74.47 HPPC1_2 115.87 72.29 HPPC1_3 115.39 76.48 HPPC1_4 166.05 40.88 HPPC2_1 116.92 97.27 HPPC2_2 118.08 22.28 HPPC2_3 117.93 59.18 HPPC2_4 117.50 55.97 HPPC3_1 115.03 63.30 HPPC3_2 115.92 44.34 HPPC3_3 116.46 44.95 HPPC3_4 116.03 45.28 HPPC4_1 117.18 67.97 HPPC4_2 117.75 69.19 HPPC4_3 117.20 67.99 HPPC4_4 117.51 46.76 HPPC5_1 116.73 61.40 HPPC5_2 116.45 58.25 HPPC5_3 116.97 54.99 HPPC5_4 116.94 41.98 HPPC6_1 116.37 99.20 HPPC6_2 116.84 99.62 HPPC6_3 117.06 103.95 HPPC6_4 117.96 66.98 HPPC7_1 118.39 82.44 HPPC7_2 118.22 84.19 HPPC7_3 118.10 83.84 HPPC7_4 118.62 56.36
HPPC8_1 118.50 41.44 HPPC8_2 125.56 77.82 HPPC8_3 117.76 46.90 HPPC8_4 119.16 33.32 HPPC9_1 155.32 57.95 HPPC9_2 156.36 60.19 HPPC9_3 131.47 77.88 HPPC9_4 156.48 40.10 HPPC10_1 161.21 65.68 HPPC10_2 157.29 77.43 HPPC10_3 157.22 47.98 HPPC10_4 154.66 53.39 HPPC11_1 151.59 67.06 HPPC11_2 148.88 58.64 HPPC11_3 157.00 76.79 HPPC11_4 153.26 34.43 HPPC12_1 117.06 63.82 HPPC12_2 116.55 68.99 HPPC12_3 117.20 47.78 HPPC12_4 116.59 49.95 HPPC13_1 120.17 45.47 HPPC13_2 120.42 43.70 HPPC13_3 119.59 36.54 HPPC13_4 122.20 30.35 HPPC14_1 117.06 62.23 HPPC14_2 117.02 70.61 HPPC14_3 117.14 59.34 HPPC14_4 117.76 46.79 HPPC15_1 118.44 41.91 HPPC15_2 118.94 47.90 HPPC15_3 118.85 50.60 HPPC15_4 119.40 34.28 HPPC16_1 129.49 38.21 HPPC16_2 127.99 37.19 HPPC16_3 127.48 27.07 HPPC16_4 132.70 24.41 Supplementary Table II. Ripple categorization with tetrodes designated to be in hippocampal area CA3 of the first rat. Based on the tetrodes that were highly likely to be in this layer the classification conditions were set to ripple amplitude >= 35mV and ripple frequency <= 120Hz.
