Abstract
Absence epilepsy is an idiopathic generalized non-convulsive type of epilepsy associated with aberrant activity in the thalamocortical network. The common neuronal network mechanism of spike–wave discharges (a manifestation of absence epilepsy) and sleep spindles suggests a close relationship between them. This paper overviews electroencephalographic (EEG) properties of anterior sleep spindles in a genetic Wistar Albino Glaxo from Rijswijk (WAG/Rij) rat model of absence epilepsy. Epileptic discharges in WAG/Rij rats appear spontaneously, and their incidence increases with age. In epileptic rats, time–frequency profile of sleep spindles distinguished from that in non-epileptic subjects by shorter duration, lower intraspindle frequency, and contained less slow-wave components, etc. Some pro-epileptic modifications of spindle activity can also be observed in EEG in epileptic WAG/Rij rats.
Introduction
Sleep spindles are the hallmark of non-rapid eye movement (NREM) sleep in humans and animals (reviewed in De Gennaro and Ferrara, 2003; De Gennaro et al., 2005). The name “spindle” refers to its characteristic waxing and waning envelope. Sleep spindles were first described by Loomis et al. (1935). The first commonly accepted definition was given by Rechtschaffen and Kales (1968): waxing and waning oscillations of 12–14 Hz and of at least 0.5 s duration. In animals, the frequency of sleep spindles tends to be slightly lower, 7–14 Hz, than in humans (Steriade, 2003). Sleep spindles have met great interest of neurobiologists and clinicians, because they are linked to synaptic plasticity in neuronal networks and processes of memory consolidation (reviewed in Fogel and Smith, 2011; Bodizs et al., 2014; Ujma et al., 2015) and dreaming processes (Nielsen et al., 2017).
Similar to humans, there are two topographically specific types of sleep spindles in rats: anterior and posterior (Terrier and Gottesmann, 1978). Anterior sleep spindles in rats only partially resemble anterior spindles in humans, but posterior spindles seem to be rather specific. In the rat (Wistar strain), posterior sleep spindles are distinguished from anterior spindles by significantly lower amplitude (two–three times lower as measured in the cortex), higher frequency, and shorter duration (almost twice shorter as measured in the cortex; Gandolfo et al., 1985). The origin of posterior sleep spindles in rats includes “septo-hippocampal region involved in the theta genesis and spreading then to the above occipital cortex” (p. 161; Gandolfo et al., 1985). Anterior spindles are predominantly appear in the frontal cortex and apparently have thalamic origin (Gandolfo et al., 1985). In rats with genetic predisposition to absence epilepsy, anterior spindles are likely to be functionally related to epileptic spike–wave discharges, SWD, e.g., electroencaphalographic manifestation of absence epilepsy (Meeren et al., 2009; van Luijtelaar, 1997; van Luijtelaar and Bikbaev, 2007; reviewed in Coenen and van Luijtelaar, 2003).
The idea that both sleep spindles and epileptic SWD [electroencephalographic (EEG) manifestation of absence epilepsy] are produced by the thalamus has been proposed a long time ago. Initially, both spontaneous spindle waves and SWD were found to be “recruiting” response evoked by repetitive stimulation of intralaminar thalamic nuclei (Morison and Dempsey, 1942; Jasper and Drooglever-Fortuyn, 1946). Later on, spindle waves appeared to be more similar to the “augmenting” response, e.g., a pattern evoked by repetitive simulation of sensorimotor thalamic nuclei (Spencer and Brookhart, 1961; Morin and Steriade, 1981).
In 1968, Pierre Gloor introduced a corticoreticular theory of primary generalized absence epilepsy assuming that sleep waves could be transformed into epileptic spike–wave activity in the neocortex due to cortical hyperexcitability (Gloor, 1968). This theory was developed in the model of feline-generalized penicillin and acknowledged that systemic or intracortical injections of penicillin [gamma-aminobutyric acid-A (GABA-A) antagonist] caused transformation of cortical field potential of sleep spindles. In particular, two or three consequent spindle waves were fused into a “spike” component of spike–wave complexes, whereas the next spindle waves were eliminated and replaced by a slow “wave” (reviewed in Kostopoulos, 2000). A variety of experimental and theoretical studies have further supported the idea that sleep spindles are functionally related to SWD (Avanzini et al., 1992; Avoli and Gloor, 1982; van Luijtelaar, 1997; references in Destexhe and Sejnowski, 2001). First, sleep spindles typically appear during 2 sleep stage and they are more numerous at sleep onset similar to SWD that could usually be recorded during drowsiness and light slow-wave sleep (van Luijtelaar and Coenen, 1988; Drinkenburg et al., 1991). Second, sleep spindles and SWD are generated in the same thalamocortical circuit (Fig. 1).
Thalamocortical network mechanisms of sleep spindle and spike–wave discharges as established in rat models of absence epilepsy. (A) Examples of the frontal EEG recorded in 8-month-old WAG/Rij rat. (B) The neocortical part of the network includes epileptic focus in the somatosensory cortex (in orange); the thalamic part includes reticular, relay, and posterior thalamic nuclei. The posterior thalamic nucleus (orange) is a high-order nucleus that specifically involved in the generation of spike–wave discharges, but its role in sleep spindles has not been investigated yet. VPm/VPl: ventroposterior medial and lateral nuclei; MGN: medial geniculate nucleus; LGN: lateral geniculate nucleus
Citation: Sleep Spindles & Cortical Up States 2, 1; 10.1556/2053.01.2017.004
Over the last few decades, a considerable number of experimental and computational studies have been focused on synaptic, cellular, and network mechanisms of epileptic spike–wave complexes in EEG (Avanzini et al., 1992, 1993; Kandel and Buzsaki, 1997; Destexhe and Sejnowski, 2001; Steriade, 2003; Meeren et al., 2002, 2009; Polack et al., 2007; van Luijtelaar and Sitnikova, 2006). Sleep spindles and SWD share a common thalamocortical mechanism, but originate from different neuronal sources (Fig. 1; reviewed in Leresche et al., 2012; Sitnikova, 2010). Sleep spindles are triggered by thalamic neurons, whereas SWD are initiated locally in the neocortex as in human patients (Westmijse et al., 2009), as well as in genetic rat models, e.g., in facial projection area of the somatosensory cortex, layers 5/6 (Meeren et al., 2002; Polack et al., 2007). Recently, it was found that the posterior thalamic nucleus (Fig. 1) was involved in the initiation of spontaneous SWD in Wistar Albino Glaxo from Rijswijk (WAG/Rij) rats (Lüttjohann et al., 2013; Lüttjohann and van Luijtelaar, 2015). The posterior thalamic nucleus is a higher-order thalamic nucleus of the somatosensory system, which receives its main driving input from epileptic source in the neocortex and sends widespread projections to the cortex and to the reticular thalamic nucleus (reviewed in Lüttjohann and van Luijtelaar, 2015). The role of higher-order thalamic nucleus, including the posterior nucleus, in generation of sleep spindles has not been explored yet.
A common thalamocortical network mechanism of sleep spindles and SWD imply that some features of sleep spindle activity might change due to epileptogenic processes. This theory was proved in vitro and in vivo (in animal models, cited above). Only a few reports described sleep spindles in patients with absence epilepsy. Myatchin and Lagae (2007) found fewer sleep spindles in stage 2 sleep in patients with childhood absence epilepsy. Similarly, Kellaway et al. (2011) demonstrated that the average rate of sleep spindles was lower and their duration was shorter in patients with generalized absence seizures (e.g., patients with 3 Hz SWD in EEG) as compared to the control group. Sleep spindles appeared to be altered even after the successful treatment of childhood absence epilepsy. As it was found in 11-year-old boy with a history of childhood absence epilepsy, sleep spindles were distorted 3 years after cessation of his treatment (valproate monotherapy; Kokkinos et al., 2011). In this boy, sleep spindles were completely absent throughout the NREM period of the first sleep cycle; during the following sleep cycles, the mean rate of occurrence and mean amplitude were below the normal expected values for the boy’s age; and only during the brief ascending branches of NREM stage 2 sleep spindles were found to be normal. Considering ethical and methodological limitations in human EEG research, studies in genetic animal models with spontaneous absence seizures are highly beneficial.
During the last decade, sleep spindles have been intensively explored in vivo in WAG/Rij rats with genetic predisposition to absence epilepsy. Absence epilepsy in WAG/Rij rats was established in 1986 (van Luijtelaar and Coenen, 1986), and the first studies of sleep spindles in this model were published in 1990s (Drinkenburg et al., 1993; van Luijtelaar, 1997). The next chapter briefly introduces WAG/Rij rat model.
WAG/Rij RATS AS A MODEL OF ABSENCE EPILEPSY IN HUMANS
WAG/Rij rat strain is a well-known genetic model of typical absence epilepsy in humans (Coenen and van Luijtelaar, 2003). Absence epilepsy is a non-convulsive generalized type of epilepsy, which is characterized by a brief impairment of consciousness (so-called “absence”) with minimal myoclonic jerks of eyes and peri-oral automatisms (Panayiotopoulos, 2011). Typical absence seizures in patients are accompanied by generalized synchronous bilateral 3 Hz spike–wave complexes in EEG, never preceded by an aura and never associated with convulsions. EEG manifestation of absence epilepsy in WAG/Rij rats was described as “spike–wave complexes” (van Luijtelaar and Coenen, 1986): “a discharge of this type lasts at least 1 s and is characterized by a train of sharp spikes and slow waves. The spikes are directed upwards with an amplitude (mean 300 μV; range 100–450 μV) of at least twice the background EEG activity. The spike-wave complexes are asymmetric, the repetition of spikes within a burst varies from 7.5 to 9.5 Hz with a mean frequency of 8.7 Hz. The mean duration of the complexes is about 5 s (range 1–30 s) whereas the mean number of discharges per hour is 18 (range 4–33)” (p. 395). Nowadays, the term “spike–wave complexes” with reference to rat’s EEG was replaced by the term “spike–wave discharges” (SWD). SWD in WAG/Rij rats occurred spontaneously and their behavioral expression is similar to clinical manifestation of absence seizures in humans, such as immobility, minimal facial myoclonic jerks, and twitches (van Luijtelaar and Coenen, 1986; Coenen and van Luijtelaar, 2003; van Luijtelaar and Coenen, 2009). Figure 2 demonstrates an example of video-EEG recording of SWD in a WAG/Rij rat.
Screenshots of the video-EEG recording taken during spontaneous spike–wave discharges (absence seizures) in a female WAG/Rij rat (age = 8 months). EEG was recorded epidurally at the frontal (red track), parietal (blue track), and occipital (green track) areas (in mV). Note that the rat stays immobile during the entire seizure without any motor activity (such as convulsions). Time is indicated in format mm:ss
Citation: Sleep Spindles & Cortical Up States 2, 1; 10.1556/2053.01.2017.004
In WAG/Rij rats, the majority of SWD occurred during passive wakefulness, drowsiness, and light slow-wave sleep (Drinkenburg et al., 1991). During EEG seizures rats often stay immobile (Fig. 2, see Video 1.MP4 in Supplementary material), sometimes demonstrating vibrissal twitching, myoclonic jerks of the eyelids, and facial muscles (Video 2.MP4 in Supplementary material). The EEG profile of SWD in WAG/Rij rats is similar to that in human patients with absence epilepsy (Sitnikova and van Luijtelaar, 2007).
Absence epilepsy has a polygenic inheritance with many of the genes involved. In both patients and animal models, gene analysis revealed both mutations and susceptibility alleles in genes encoding GABA and Ca2+ receptor subunits (reviewed in Lagrange, 2006; Yalçın, 2012). A genome-wide scan in WAG/Rij rats identified a quantitative trait loc T1swd/wag (chromosome 5) for controlling duration of generalized SWD (Gauguier et al., 2004). Mutations in a family of genes encoding hyperpolarization-activated cyclic nucleotide-gated (HCN) channels are known to be involved in human epilepsy (references in DiFrancesco and DiFrancesco, 2015). The current through HCN channels (Ih current, “pacemaker current”) is known to play a key role in generation of cardiac and neuronal rhythmic activity. In vitro, in spontaneously spindling ferret lateral geniculate slices, activation and deactivation of Ih was shown to shape waxing and waning activity characterizing sleep spindles (Lüthi et al., 1998). In WAG/Rij rat model, no mutations in the coding sequence of HCN1-4 channels were identified so far. However, a decrease in currents corresponded to a 34% reduction in HCN1 protein was found in the somatosensory cortex (epileptic focus of SWD) in WAG/Rij rats (Strauss et al., 2004) and upregulation of HCN1 in the thalamus (Budde et al., 2005). The expression of HCN2-4 subunits did not change.
Absence epilepsy is a multifactorial (complex) disorder (Crunelli and Leresche, 2002), in which genetic and environmental factors are both important for the clinical manifestation of the disease. In particular, changes in early life environment (neonatal handling and maternal deprivation) in WAG/Rij rats resulted in 35% reduction of SWD duration in adulthood (Schridde et al., 2006). This effect was accompanied by an increase in HCN1 channels in somatosensory cortex and increase in lh current, assuming that neonatal manipulations influenced functional properties of HCN1 channels latter in life. In agreement with that we found that WAG/Rij rat pups fostered to Wistar dams during full weaning period showed significantly less absence seizures in EEG at adulthood (Sitnikova et al., 2015, 2016b).
Severity of absence epilepsy in rodent models seems to be influenced by environmental conditions (breeding, housing, etc.) and by genetic drift. The striking difference in seizure activity was found in Genetic Absence Epilepsy Rats from Strasbourg (GAERS) between four colonies established in Strasbourg, Grenoble, Melbourne, and Istanbul (Powell et al., 2014). In Melbourne’s colony, GAERS showed the least severe epilepsy phenotype, and GAERS in Grenoble’s colony exhibited four times more seizures than GAERS in Melbourne. Also in WAG/Rij rats, the breeding colony in Moscow (Institute of Higher Nervous Activity and Neurophysiology of Russian Academy of Sciences, Moscow, Russia) displayed differences in the development of SWD as compared to the original breeding colony in Nijmegen (Radboud University of Nijmegen, Nijmegen, the Netherlands). One of the first papers on absence epilepsy in WAG/Rij rats (Nijmegen’s colony) indicated that the first immature SWD appeared at the age of 75 days, and the number of seizures gradually increased until the age of 245 days (∼8.1 months; Coenen and van Luijtelaar, 1987). WAG/Rij rats were imported from Nijmegen to Moscow (Institute of Higher Nervous Activity) in 1995, and breeding colony has been maintained since then. In 2011, we found that Moscow’s population of WAG/Rij rats showed relatively late debut age of absence epilepsy (7–8 months) as compared to the original population (Sitnikova, 2011). Later on, in 2016, we found that some WAG/Rij rats that did not express any seizures in EEG until the age of 11–13 months and referred them as “asymptomatic” (or “non-epileptic”) rats (Sitnikova et al., 2016b). Some EEG features of sleep spindles in the asymptomatic WAG/Rij rats differed from that in symptomatic (Sitnikova et al., 2016b), and some features correlated with severity of absence epilepsy (high/low incidence; Sitnikova et al., 2012, 2014a, 2014b). This issue is summarized below.
SLEEP SPINDLES IN WAG/Rij RATS
Automatic detection of sleep spindles in WAG/Rij rats
Sleep spindles are short and non-stationary EEG events, therefore, the continuous wavelet transform appeared to be an optimal method for their time–frequency analysis (Hramov et al., 2015). Over the last decade, we use this method for the automatic detection and spectral analysis of sleep spindles in WAG/Rij rats (for more details, see Sitnikova et al., 2009; Pavlov et al., 2012; Hramov et al., 2015). We continuously improve spindle detection algorithm by using additional routines, such as “floating thresholding” and Hilbert–Huang transform (Grubov et al., 2017). We always use the complex Morlet wavelet as the “mother” function, because this function provides good resolution both in time and frequency domains.
Total amount and duration of sleep spindles in WAG/Rij rats
The higher incidence of epilepsy in drug-naïve WAG/Rij rats was associated with the lower duration of sleep spindles (Sitnikova et al., 2012). Taking into account the age-related increase of epileptic activity in WAG/Rij rats between 5 and 9 months of age, we examined age-dependent changes of time–frequency structure of sleep spindles in parallel to progressive development of absence seizures. Anterior sleep spindles and SWD were detected in the frontal EEG recorded at the age of 5, 7, and 9 months. The duration of epileptic activity increased with age (both mean duration of each attack and total seizure duration), but the mean duration of sleep spindles, in opposite, decreased (Sitnikova et al., 2012). In the same manner, untreated children with generalized SWD showed fewer spindles during stage 2 sleep in comparison to the healthy control. Noteworthy is that antiepileptic medication (monotherapy with valproic acid or ethosuximide) caused an increase in number of spindles during stage 2 sleep that reached the control values (Myatchin and Lagae, 2007).
Frequency parameters of sleep spindles in WAG/Rij rats
We found that the mean frequency of anterior sleep spindles in drug-naïve WAG/Rij rats with absence epilepsy was significantly lower than in non-epileptic Wistar rats: at the age of 7 months (11.2 vs. 13.1 Hz) as well as at the age of 9 months (11.3 vs. 13.2 Hz; Sitnikova et al., 2014a). More specifically, 38%–43% of sleep spindles in WAG/Rij rats appeared with the mean frequency 8–10 Hz, but percentage of these “slow” sleep spindles in Wistar rats was significantly lower (11%–17%), suggesting that absence epilepsy may be associated with the presence of slow (<10 Hz) sleep spindles in frontal EEG. Similarly, Drake et al. (1991) detected <10 Hz sleep spindles in patients with partial and generalized epilepsy during stage 2 sleep. These authors also showed a strong effect of medication: intrinsic frequency of sleep spindles may decrease due to polytherapy.
Earlier we found that the higher intensity of absence epilepsy in WAG/Rij rats was associated with some changes in time–frequency properties of sleep spindles (Sitnikova et al., 2012, 2014b, 2016a).
Intrinsic frequency of sleep spindles is an important characteristic of thalamocortical network activity with respect to generation of autonomous oscillations (reviewed in Urakami et al., 2012). In order to examine transient changes of frequency during sleep spindles, we used skeletons of wavelet surfaces (Sitnikova et al., 2012, 2014a, 2014b, 2015, 2016a). This technique was designed to extract the dominant and subdominant frequency components in EEG (Fig. 3) and study frequency dynamics during automatically detected sleep spindles.
An example of EEG track with automatically detected sleep spindle (marked in blue) and corresponding skeleton of wavelet surface. The bottom graph (skeleton of wavelet surface) shows the distribution of the highest values of wavelet energy over time in frequencies (2–16 Hz)
Citation: Sleep Spindles & Cortical Up States 2, 1; 10.1556/2053.01.2017.004
Analysis of the instantaneous (localized in time) frequency of sleep spindles was performed in WAG/Rij rats successively at the age of 5, 7, and 9 months and in non-epileptic Wistar rats at the age of 7 and 9 months (Sitnikova et al., 2014a, 2014b). Five-month-old WAG/Rij rats developed just a few and immature seizures, and their sleep spindles demonstrated an increase of instantaneous frequency from beginning to the end (Sitnikova et al., 2014a). Similar increase of intraspindle frequency was found in Wistar rats at the age of 7 and 9 months (Sitnikova et al., 2014b). At the age of 7 and 9 months, when WAG/Rij rats developed matured epilepsy, their sleep spindles did not display any changes of intrinsic frequency. In summary, age-dependent increase of epileptic activity in WAG/Rij rats disrupts an intrinsic frequency dynamics of sleep spindles.
In a group of asymptomatic and symptomatic WAG/Rij rats at the age of 9 months, the frequency of the dominant component in sleep spindles in the asymptomatic individuals was higher (12.7 Hz) than in the symptomatic ones (11.9 Hz; Sitnikova et al., 2016a).
Slow-wave components of sleep spindles in WAG/Rij rats
Sleep spindles are often superimposed with low-frequency EEG components (<9 Hz, Fig. 3), but surprisingly few reports have described this phenomenon. Andrillon et al. (2011) studied intracranial EEG in patients with pharmacologically intractable epilepsy and found large slow waves in depth EEG (either positive or negative peaks, corresponding to down and up states) before the occurrence of spindles. Mölle et al. (2002) investigated rhythmic activity coinciding sleep spindles using direct current (DC) EEG signals in healthy humans. They wrote: “the mean DC potential shows a distinct negative peak shortly before the rise in spindle activity. The negative peak is followed by a positive potential coinciding with the time of highest spindle power” (p. 10994). The highest spindle activity appeared ∼400 ms after the peak of slow negative half-waves. In our animals, we also noticed the same negatively oriented slow waves preceding sleep spindles (Figs 3 and 4), which can be associated with cortical down state.
Continuous wavelet transform of sleep spindles in frontal EEG in adult WAG/Rij rat. Axis of ordinates denotes EEG frequencies (Hz) in the logogriphic scale. Sleep spindles 9–14 Hz are superimposed with low-frequency components in the range of delta (1–4 Hz) and theta (5–9 Hz)
Citation: Sleep Spindles & Cortical Up States 2, 1; 10.1556/2053.01.2017.004
Sleep spindles in rodents are often superimposed with theta activity. In Wistar rats, “the anterior (frontal) spindles reach their maximum during deep slow sleep, when accompanied by theta activity” (Gandolfo et al., 1985, p. 151). In C57BL/6 mice, frontal sleep spindles exhibited an increase in spectral power in frequencies >4.5 Hz (Vyazovskiy et al., 2004). Wavelet analysis of sleep spindles in WAG/Rij rats revealed two types of low-frequency subdominant components in frequency range of delta (mean = 3.6 Hz) and theta (mean = 6.7 Hz; Sitnikova et al., 2016a). Wavelet surface in Fig. 4 demonstrated that sleep spindles coincide with low-frequency rhythmic EEG components.
Wavelet analysis of anterior sleep spindles in 9-month-old WAG/Rij rats showed that ∼71% of sleep spindles in asymptomatic individuals contained strong low-frequency components, and this percentage was significantly lower (∼58%) in the age-matched symptomatic rats (Sitnikova et al., 2016a). Basically speaking, low-frequency precursors of sleep spindles did not disappear with the onset of spindle activity, but they often overlap with consequent sleep spindles, therefore, we consider <9 Hz components as integrative parts of 10–16 Hz sleep spindles.
It is known that spindle and delta frequency oscillations are incompatible on the level of single neurons, because they occurred at different levels of membrane potentials of thalamocortical neurons (Nũnez et al., 1992), but on the level of EEG spindle and slow-wave activity may co-exist (Steriade, 2003). Moreover, a greater number of anterior sleep spindles coexisted with delta and theta activity in asymptomatic WAG/Rij rats in comparison to symptomatic suggesting that epileptic phenotype in WAG/Rij rats might related to suppression of slow-wave rhythmic activity during sleep spindles. We hypothesize that the higher amount of low-frequency components of sleep spindles associate with undeveloped seizures in subjects with genetic predisposition to absence epilepsy (Sitnikova et al., 2016a).
PRO-EPILEPTIC SPINDLE-LIKE OSCILLATIONS IN WAG/Rij RATS
Visual inspection of EEG in symptomatic WAG/Rij rats often reveals EEG waveforms comprising dual features of SWD and sleep spindles. It is difficult to classify these EEG transits because of great variety of their time–frequency structure. At the moment, we distinguished two kinds of pro-epileptic EEG patters: spike–wave spindles (spiky oscillations) and type 2 anterior sleep spindles.
Spike–wave spindles (spiky oscillations)
These EEG transients were characterized by sharp negative waves with fragments of spike–wave complexes (Fig. 5B), lasting from 1.5 to 3 s, showing a waxing–waning morphology (gradual elevation and descending of amplitude) and high amplitude in frontoparietal areas (Sitnikova, 2011). The waveform of spike–wave spindles appeared to be similar to the “spiky phenomena” described in 1993 by Drinkenburg et al. in WAG/Rij rats. The power (Fourier) spectrum of spike–wave spindles (spiky oscillations) displayed additional peaks corresponding to harmonics, which are integer multiples of the fundamental frequency. In Fig. 5(A and B), both SWD and spike–wave spindle demonstrated ∼8 Hz central frequency and the first harmonic at 16 Hz. In contrast, Fourier spectrum of sleep spindles (Fig. 5C) never revealed any harmonic frequencies (Sitnikova et al., 2009).
Fourier spectrum of spike–wave discharges (A), spike–wave spindle (B), and normal sleep spindle (C) in 8.6-month-old WAG/Rij rat as recorded in frontal EEG
Citation: Sleep Spindles & Cortical Up States 2, 1; 10.1556/2053.01.2017.004
Type 2 spindles
In order to identify aberrant forms of sleep spindles in EEG in WAG/Rij rats, we used adoptive wavelet analysis (Sitnikova et al., 2009). This technique aimed in constructing adoptive “spindle wavelet” functions using the native EEG signal and further applied “spindle wavelet” as basis functions in the continuous wavelet transform (Fig. 6A and B). The first type 1 “spindle wavelet” showed a strong congruity to the majority of sleep spindle in all subjects. More specifically, application of type 1 “spindle wavelet” as basis function for the automatic selection of sleep spindles resulted in 85%–90% true positive selections. Considering this result, type 1 “spindle wavelet” was regarded as a prototype of sleep spindles in WAG/Rij rats. The remaining sleep spindles (10%–15%) had an aberrant EEG structure and were selected using type 2 “spindle wavelet” that has to be chosen in each rat individually. In summary, adoptive wavelet analysis can be used for the automatic selection and classification of sleep spindles, but it was not appropriate for time–frequency EEG analysis.
Details of “spindle wavelet” study in WAG/Rij rats (Sitnikova et al., 2009). “Spindle wavelet” functions were constructed from the native EEG signal (A). The signal was normalized with Gaussian function (B) and used as basis function wavelet basis function for the automatic selection of sleep spindles. Type 1 “spindle wavelet” was universal, but type 2 “spindle wavelet” was selected individually. Fourier spectrum of type 2 “spindle wavelet” showed a strong 20–25 Hz frequency component (C) (Sitnikova et al., 2009).
Citation: Sleep Spindles & Cortical Up States 2, 1; 10.1556/2053.01.2017.004
Power spectrum analysis of “spindle wavelet” functions helped us to determine differences between two spindle prototypes (Sitnikova et al., 2009). The fundamental frequency of type 1 “spindle wavelet” in power spectrum (Fig. 6C) was centered at ∼12 Hz that corresponds well to the mean frequency of anterior sleep spindles in rats (Terrier and Gottesmann, 1978). In contrast, Fourier spectrum of type 2 “spindle wavelet” showed the main peak at 21 ± 3 Hz and additional peaks at 1, 3, and 16.7 Hz. The presence of “type 2” sleep spindles in WAG/Rij rats with a strong 16–25 Hz component might be elicited by epileptogenic processes.
CONCLUSION
In a genetic WAG/Rij rat model of absence epilepsy, EEG properties of anterior sleep spindles are distinguished from that in non-epileptic rats. A shorter duration of sleep spindles and lower intraspindle frequency may be considered as a hallmark of developing absence epilepsy in subjects with genetic predisposition. In addition, absence epilepsy is accompanied by pro-epileptic EEG waveforms comprising dual features of SWD and sleep spindles.
Majority of sleep spindles are superimposed with low-frequency rhythmic EEG components in epileptic and non-epileptic rats, and a powerful <9 Hz components of sleep spindles may associate with undeveloped absence seizures.
Conflict of interest
The author declares that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Acknowledgement
The part of this study was financially supported by the Russian Foundation for Basic Research (grant No. 16-04-00275).
REFERENCES
Andrillon T , Nir Y , Staba RJ , Ferrarelli F , Cirelli C , Tononi G , and Fried I (2011). Sleep spindles in humans: insights from intracranial EEG and unit recordings. J Neurosci 31(49):17821–17834. doi:10.1523/JNEUROSCI.2604-11.2011
Avanzini G , de Curtis M , Marescaux C , Panzica F , Spreafico R , and Vergnes M (1992). Role of the thalamic reticular nucleus in the generation of rhythmic thalamo-cortical activities subserving spike and waves. J Neural Transm Suppl 35:85–95.
Avoli M and Gloor P (1982). Interaction of cortex and thalamus in spike-and-wave discharges of feline generalized penicillin epilepsy. Exp Neurol 76(1):196–217. doi:10.1016/0014-4886(82)90112-1
Bodizs R , Gombos F , Ujma PP , and Kovacs I (2014). Sleep spindling and fluid intelligence across adolescent development: sex matters. Front Hum Neurosci 8:952. doi:10.3389/fnhum.2014.00952
Budde T , Caputi L , Kanyshkova T , Staak R , Abrahamczik C , Munsch T , and Pape HC (2005). Impaired regulation of thalamic pacemaker channels through an imbalance of subunit expression in absence epilepsy. J Neurosci 25(43):9871–9882. doi:10.1523/JNEUROSCI.2590-05.2005
Coenen AM and van Luijtelaar EL (1987). The WAG/Rij rat model for absence epilepsy: age and sex factors. Epilepsy Res 1(5):297–301. doi:10.1016/0920-1211(87)90005-2
Coenen AML and van Luijtelaar ELJM (2003). Genetic animal models for absence epilepsy: a review of the WAG/Rij strain of rats. Behav Genet 33(6):635–655. doi:10.1023/A:1026179013847
Crunelli V and Leresche N (2002). Childhood absence epilepsy: genes, channels, neurons and networks. Nat Rev Neurosci 3(5):371–382. doi:10.1038/nrn811
De Gennaro L and Ferrara M (2003). Sleep spindles: an overview. Sleep Med Rev 7(5):423–440. doi:10.1053/smrv.2002.0252
De Gennaro L , Ferrara M , Vecchio F , Curcio G , and Bertini M (2005). An electroencephalographic fingerprint of human sleep. Neuroimage 26(1):114–122. doi:10.1016/j.neuroimage.2005.01.020
Destexhe A and Sejnowski, TJ (2001). Thalamocortical assemblies. Oxford, UK: Oxford University Press.
DiFrancesco JC and DiFrancesco D (2015). Dysfunctional HCN ion channels in neurological diseases. Front Cell Neurosci 9:71. doi:10.3389/fncel.2015.00071
Drake ME Jr , Pakalnis A , Padamadan H , Weate SM , and Cannon PA (1991). Sleep spindles in epilepsy. Clin Electroencephalogr 22(3):144–149. doi:10.1177/155005949102200305
Drinkenburg WH , Coenen AM , Vossen JM , and Van Luijtelaar EL (1991). Spike-wave discharges and sleep-wake states in rats with absence epilepsy. Epilepsy Res 9(3):218–224. doi:10.1016/0920-1211(91)90055-K
Drinkenburg WH , van Luijtelaar EL , van Schaijk WJ , and Coenen AM (1993). Aberrant transients in the EEG of epileptic rats: a spectral analytical approach. Physiol Behav 54(4):779–783. doi:10.1016/0031-9384(93)90092-T
Fogel SM and Smith CT (2011). The function of the sleep spindle: a physiological index of intelligence and a mechanism for sleep-dependent memory consolidation. Neurosci Biobehav Rev 35(5):1154–1165. doi:10.1016/j.neubiorev.2010.12.003
Gandolfo G , Glin L , and Gottesmann C (1985). Study of sleep spindles in the rat: a new improvement. Acta Neurobiol Exp (Wars) 45(5–6):151–162.
Gauguier D , van Luijtelaar G , Bihoreau MT , Wilder SP , Godfrey RF , Vossen J , Coenen A , and Cox RD (2004). Chromosomal mapping of genetic loci controlling absence epilepsy phenotypes in the WAG/Rij rat. Epilepsia 45(8):908–915. doi:10.1111/j.0013-9580.2004.13104.x
Gloor P (1968). Generalized cortico-reticular epilepsies: some considerations on the pathophysiology of generalized bilaterally synchronous spike and wave discharge. Epilepsia 9:249–263.
Grubov VV , Sitnikova E , Pavlov AN , Koronovskii AA , and Hramov AE (2017). Recognizing of stereotypic patterns in epileptic EEG using empirical modes and wavelets. Physica A Statist Mech Appl 486:206–217. doi:10.1016/j.physa.2017.05.091
Hramov AE , Koronovskii AA , Makarov VA , Pavlov AN , and Sitnikova E (2015). Wavelets in neuroscience. Berlin/Heidelberg, Germany: Springer.
Jasper HH and Drooglever-Fortuyn J (1946). Experimental studies on the functional anatomy of petit mal epilepsy. Proc Assoc Res Nerv Ment Dis 26:272–298.
Kandel A and Buzsaki G (1997). Cellular-synaptic generation of sleep spindles, spike-and-wave discharges, and evoked thalamocortical responses in the neocortex of the rat. J Neurosci 17:6783–6797.
Kellaway P , Frost JD , and Crawley JW (1990). The relationship between sleep spindles and spike-and-wave bursts in human epilepsy. In M Avoli, P Gloor, G Kostopoulos, R Naquet (Eds.), Generalized epilepsy (pp. 36–48). Boston, MA: Birkhäuser. doi:10.1007/978-1-4684-6767-3_4
Kokkinos V , Koupparis AM , Stavrinou ML , Kostopoulos GK , and Garganis K (2011). Sleep spindle alterations following-up a treated childhood absence epilepsy case. Epileptologia. Int J Clin Exp Res 19(2):73–83.
Kostopoulos GK (2000). Spike-and-wave discharges of absence seizures as a transformation of sleep spindles: the continuing development of a hypothesis. Clin Neurophysiol 111(Suppl. 2):S27–S38. doi:10.1016/S1388-2457(00)00399-0
Lagrange A (2006). Genetic variants in absence epilepsy: a contextual consideration of calcium current kinetics. Epilepsy Curr 6(3):99–101. doi:10.1111/j.1535-7511.2006.00111.x
Leresche N , Lambert RC , Errington AC , and Crunelli V (2012). From sleep spindles of natural sleep to spike and wave discharges of typical absence seizures: is the hypothesis still valid?. Pflugers Arch 463(1):201–212. doi:10.1007/s00424-011-1009-3
Loomis AL , Harvey EN , and Hobart G (1935). Potential rhythms of the cerebral cortex during sleep. Science 81(2111):597–598. doi:10.1126/science.81.2111.597
Lüttjohann A , Schoffelen JM , and van Luijtelaar G (2013). Peri-ictal network dynamics of spike-wave discharges: phase and spectral characteristics. Exp Neurol 239:235–247. doi:10.1016/j.expneurol.2012.10.021
Lüttjohann A and van Luijtelaar G (2015). Dynamics of networks during absence seizure’s on- and offset in rodents and man. Front Physiol 6:16. doi:10.3389/fphys.2015.00016
Lüthi A , Bal T , and McCormick DA (1998). Periodicity of thalamic spindle waves is abolished by ZD7288, a blocker of Ih. J Neurophysiol 79:3284–3289.
Meeren HK , Pijn JP , Van Luijtelaar EL , Coenen AM , and Lopes da Silva FH (2002). Cortical focus drives widespread corticothalamic networks during spontaneous absence seizures in rats. J Neurosci 22:1480–1495.
Meeren HK , Veening JG , Möderscheim TA , Coenen AM , and van Luijtelaar G (2009). Thalamic lesions in a genetic rat model of absence epilepsy: dissociation between spike-wave discharges and sleep spindles. Exp Neurol 217(1):25–37. doi:10.1016/j.expneurol.2009.01.009
Mölle M , Marshall L , Gais S , and Born J (2002). Grouping of spindle activity during slow oscillations in human non-rapid eye movement sleep. J Neurosci 22(24):10941–10947.
Morin D and Steriade M (1981). Development from primary to augmenting responses in the somatosensory system. Brain Res 205(1):49–66. doi:10.1016/0006-8993(81)90719-8
Morison RS and Dempsey EW (1942). A study of thalamo-cortical relations. Am J Physiol 135:281–292.
Myatchin I and Lagae L (2007). Sleep spindle abnormalities in children with generalized spike-wave discharges. Pediatr Neurol 36:106–111. doi:10.1016/j.pediatrneurol.2006.09.014
Nielsen T , Carr M , Blanchette-Carrière C , Marquis L-P , Dumel G , Solomonova E , Julien S-H , Picard-Deland C , and Paquette T (2017). NREM sleep spindles are associated with dream recall. Sleep Spindles & Cortical Up States 1(1):27–41. doi:10.1556/2053.1.2016.003
Nũnez A , Curró Dossi R , Contreras D , and Steriade M (1992). Intracellular evidencefor incompatibility between spindle and delta oscillations in thalamocorticalneurons of cat. Neuroscience 48(1):75–85. doi:10.1016/0306-4522(92)90339-4
Panayiotopoulos CP (2011). The new ILAE report on terminology and concepts for organization of epileptic seizures: a clinician’s critical view and contribution. Epilepsia 52(12):2155–2160. doi:10.1111/j.1528-1167.2011.03288.x
Pavlov AN , Hramov AE , Koronovskii AA , Sitnikova EYu , Makarov VA , and Ovchinnikov AA (2012). Wavelet analysis in neurodynamics. Phys Usp 55(9):845–875. doi:10.3367/UFNe.0182.201209a.0905
Polack PO , Guillemain I , Hu E , Deransart C , Depaulis A , and Charpier S (2007). Deep layer somatosensory cortical neurons initiate spike-and-wave discharges in a genetic model of absence seizures. J Neurosci 27:6590–6599. doi:10.1523/JNEUROSCI.0753-07.2007
Powell KL , Tang H , Ng C , Guillemain I , Dieuset G , Dezsi G , Çarçak N , Onat F , Martin B , O’Brien TJ , Depaulis A , and Jones NC (2014). Seizure expression, behavior, and brain morphology differences in colonies of Genetic Absence Epilepsy Rats from Strasbourg. Epilepsia 55(12):1959–1968. doi:10.1111/epi.12840
Rechtschaffen A and Kales AA (1968). Manual of standardized terminology, technique and scoring system for sleep stages of human subjects. Washington, DC: Public Health Service, U.S. Government Printing Office.
Schridde U , Strauss U , Bräuer AU , and van Luijtelaar G (2006). Environmental manipulations early in development alter seizure activity, Ih and HCN1 protein expression later in life. Eur J Neurosci 23(12):3346–3358. doi:10.1111/j.1460-9568.2006.04865.x
Sitnikova E (2010). Thalamo-cortical mechanisms of sleep spindles and spike-wave discharges in rat model of absence epilepsy (a review). Epilepsy Res 89(1):17–26. doi:10.1016/j.eplepsyres.2009.09.005
Sitnikova E (2011). Neonatal sensory deprivation promotes development of absence seizures in adult rats with genetic predisposition to epilepsy. Brain Res 1377:109–118. doi:10.1016/j.brainres.2010.12.067
Sitnikova E , Grubov VV , Khramov AE , and Koronovskiĭ AA (2012). [Age-related changes in time-frequency structure of sleep spindles in EEG in rats with genetic predisposition to absence epilepsy (Wag/Rij)]. Zh Vyssh Nerv Deiat Im I P Pavlova 62(6):733–744 (in Russian).
Sitnikova E , Hramov AE , Grubov V , and Koronovsky AA (2014a). Age-dependent increase of absence seizures and intrinsic frequency dynamics of sleep spindles in rats. Neurosci J 2014:1–6. doi:10.1155/2014/370764
Sitnikova E , Hramov AE , Grubov V , and Koronovsky AA (2014b). Time-frequency characteristics and dynamics of sleep spindles in WAG/Rij rats with absence epilepsy. Brain Res 1543:290–299. doi:10.1016/j.brainres.2013.11.001
Sitnikova E , Hramov AE , Grubov V , and Koronovsky AA (2016a). Rhythmic activity in EEG and sleep in rats with absence epilepsy. Brain Res Bull 120:106–116. doi:10.1016/j.brainresbull.2015.11.012
Sitnikova E , Hramov AE , Koronovskii AA , and van Luijtelaar G (2009). Sleep spindles and spike-wave discharges in EEG: their generic features, similarities and distinctions disclosed with Fourier transform and continuous wavelet analysis. J Neurosci Methods 180(2):304–316. doi:10.1016/j.jneumeth.2009.04.006
Sitnikova E , Rutskova EM , and Raevsky VV (2015). Reduction of epileptic spike-wave activity in WAG/Rij rats fostered by Wistar dams. Brain Res 1594:305–309. doi:10.1016/j.brainres.2014.10.067
Sitnikova E , Rutskova EM , and Raevsky VV (2016b). Maternal care affects EEG properties of spike-wave seizures (including pre- and post ictal periods) in adult WAG/Rij rats with genetic predisposition to absence epilepsy. Brain Res Bull 127:84–91. doi:10.1016/j.brainresbull.2016.08.019
Sitnikova E and van Luijtelaar G (2007). Electroencephalographic characterization of spike-wave discharges in cortex and thalamus in WAG/Rij rats. Epilepsia 48(12):2296–2311.
Spencer WA and Brookhart JM (1961). A study of spontaneous spindle waves in sensorimotor cortex of cat. J Neurophysiol 24:50–65.
Steriade M (2003). The corticothalamic system in sleep. Front Biosci 8(4):d878–d899. doi:10.2741/1043
Strauss U , Kole MH , Bräuer AU , Pahnke J , Bajorat R , Rolfs A , Nitsch R , and Deisz RA (2004). An impaired neocortical Ih is associated with enhanced excitability and absence epilepsy. Eur J Neurosci 19(11):3048–3058. doi:10.1111/j.0953-816X.2004.03392.x
Terrier G and Gottesmann C (1978). Study of cortical spindles during sleep in the rat. Brain Res Bull 3(6):701–706. doi:10.1016/0361-9230(78)90021-7
Ujma PP , Bódizs R , Gombos F , Stintzing J , Konrad BN , Genzel L , Steiger A , and Dresler M (2015). Nap sleep spindle correlates of intelligence. Sci Rep 5(1):17159. doi:10.1038/srep17159
Urakami Y , Ioannides AA , and Kostopoulos GK (2012). Sleep spindles – as a biomarker of brain function and plasticity. In IM Ajeena (Ed.), Advances in clinical neurophysiology (Chapter 4, pp. 73–108). Oxford, UK: InTech.
van Luijtelaar EL (1997). Spike-wave discharges and sleep spindles in rats. Acta Neurobiol Exp (Wars) 57(2):113–121.
van Luijtelaar G and Bikbaev A (2007). Midfrequency cortico-thalamic oscillations and the sleep cycle: genetic, time of day and age effects. Epilepsy Res 73(3):259–265. doi:10.1016/j.eplepsyres.2006.11.002
van Luijtelaar EL and Coenen AM (1986). Two types of electrocortical paroxysms in an inbred strain of rats. Neurosci Lett 70(3):393–397. doi:10.1016/0304-3940(86)90586-0
van Luijtelaar EL and Coenen AM (1988). Circadian rhythmicity in absence epilepsy in rats. Epilepsy Res 2(5):331–336.
van Luijtelaar G and Coenen A (2009). Genetic models of absence epilepsy: new concepts and insights. In PA Schwartzkroin (Ed.), Encyclopedia of basic epilepsy research (Vol. 1, pp. 1–8). Oxford, UK: Elsevier.
van Luijtelaar G and Sitnikova E (2006). Global and focal aspects of absence epilepsy: the contribution of genetic models. Neurosci Biobehav Rev 30(7):983–1003. doi:10.1016/j.neubiorev.2006.03.002
Vyazovskiy VV , Achermann P , Borbély AA , and Tobler I (2004). The dynamics of spindles and EEG slow-wave activity in NREM sleep in mice. Arch Ital Biol 142(4):511–523.
Westmijse I , Ossenblok P , Gunning B , and van Luijtelaar G (2009). Onset and propagation of spike and slow wave discharges in human absence epilepsy: a MEG study. Epilepsia 50(12):2538–2548. doi:10.1111/j.1528-1167.2009.02162.x
Yalçın O (2012). Genes and molecular mechanisms involved in the epileptogenesis of idiopathic absence epilepsies. Seizure 21(2):79–86. doi:10.1016/j.seizure.2011.12.002
Electronic Supplementary Material (ESM)
Electronic Supplementary Material (ESM) associated with this article can be found at the website www.akademiai.com/doi/suppl/10.1556/2053.01.2017.004
