Download Csercsa Richárd

Survey
yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Document related concepts

Cognitive neuroscience wikipedia, lookup

Environmental enrichment wikipedia, lookup

Human brain wikipedia, lookup

Eyeblink conditioning wikipedia, lookup

Theta wave wikipedia, lookup

Premovement neuronal activity wikipedia, lookup

Electrophysiology wikipedia, lookup

Time perception wikipedia, lookup

Neuroscience in space wikipedia, lookup

Aging brain wikipedia, lookup

Emotion and memory wikipedia, lookup

Connectome wikipedia, lookup

Limbic system wikipedia, lookup

Single-unit recording wikipedia, lookup

Biology of depression wikipedia, lookup

Development of the nervous system wikipedia, lookup

Neuroeconomics wikipedia, lookup

Brain Rules wikipedia, lookup

Nonsynaptic plasticity wikipedia, lookup

Neuroanatomy wikipedia, lookup

Nervous system network models wikipedia, lookup

Optogenetics wikipedia, lookup

Sleep wikipedia, lookup

De novo protein synthesis theory of memory formation wikipedia, lookup

Neuroanatomy of memory wikipedia, lookup

Neuroplasticity wikipedia, lookup

Neural oscillation wikipedia, lookup

Feature detection (nervous system) wikipedia, lookup

Holonomic brain theory wikipedia, lookup

Activity-dependent plasticity wikipedia, lookup

Delayed sleep phase disorder wikipedia, lookup

Insomnia wikipedia, lookup

Synaptic gating wikipedia, lookup

Spike-and-wave wikipedia, lookup

Rapid eye movement sleep wikipedia, lookup

Memory consolidation wikipedia, lookup

Sleep apnea wikipedia, lookup

Sleep paralysis wikipedia, lookup

Neuroscience of sleep wikipedia, lookup

Sleep deprivation wikipedia, lookup

Metastability in the brain wikipedia, lookup

Sleep medicine wikipedia, lookup

Sleep and memory wikipedia, lookup

Neural correlates of consciousness wikipedia, lookup

Effects of sleep deprivation on cognitive performance wikipedia, lookup

Neuropsychopharmacology wikipedia, lookup

Start School Later movement wikipedia, lookup

Clinical neurochemistry wikipedia, lookup

Transcript
Laminar analysis of the slow wave activity
in humans
Doctoral thesis
Csercsa Richárd
Semmelweis University
János Szentágothai School of Neurosciences
Functional Neurosciences
Supervisor: Dr. Ulbert István Ph.D.
Official opponents:
Dr. Kamondi Anita C.Sc.
Dr. Barthó Péter Ph.D.
President of committee:
Members of committee:
Dr. Szirmai Imre Ph.D., D.Sc.
Dr. Bódizs Róbert Ph.D.
Dr. Gál Viktor Ph.D.
Budapest
2010
Csercsa Richárd
TABLE OF CONTENTS
TABLE OF CONTENTS ............................................................................................... 2
LIST OF ABBREVIATIONS ........................................................................................ 4
INTRODUCTION .......................................................................................................... 6
Sleep – general introduction ......................................................................................... 6
Sleep patterns ................................................................................................................ 7
Sleep stages .............................................................................................................. 7
Development ............................................................................................................. 9
Neural basis of arousal ............................................................................................... 10
Passive or active ..................................................................................................... 11
Deafferentation ....................................................................................................... 11
Ascending reticular activating system .................................................................... 12
Ventrolateral Preoptic Nucleus .............................................................................. 14
Sleep regulatory substances (SRS) ......................................................................... 16
Local sleep .............................................................................................................. 17
Sleep deprivation ........................................................................................................ 21
Functions of sleep ....................................................................................................... 22
Recovery and restoration........................................................................................ 22
Energy conservation ............................................................................................... 22
Plasticity, memory, learning................................................................................... 23
Memory consolidation during sleep ........................................................................... 24
Memory categories ................................................................................................. 24
Memory formation .................................................................................................. 26
Synaptic consolidation............................................................................................ 26
System consolidation .............................................................................................. 28
Slow wave sleep ......................................................................................................... 29
Slow oscillation ...................................................................................................... 29
Slow oscillation generation .................................................................................... 31
Propagating wave and source localization ............................................................ 34
Stimulation evoked slow oscillation ....................................................................... 36
2
Laminar analysis of the slow wave actvity in humans
Modeling of slow oscillation .................................................................................. 37
Memory consolidation during SWS ........................................................................ 40
Slow waves and K-complexes ................................................................................. 47
OBJECTIVES ............................................................................................................... 49
METHODS .................................................................................................................... 50
Patients and electrodes................................................................................................ 50
Histology .................................................................................................................... 50
Recordings .................................................................................................................. 52
Slow wave activity detection ...................................................................................... 54
Current source density analysis .................................................................................. 56
Statistical analysis of electrophysiology and histology .............................................. 57
Single and multiple unit activity analysis ................................................................... 58
RESULTS ...................................................................................................................... 60
Slow oscillation in humans ......................................................................................... 60
Supragranular activity................................................................................................. 65
Up-state onset ............................................................................................................. 69
Frequency of slow oscillation ..................................................................................... 71
Sparse firing................................................................................................................ 72
DISCUSSION ................................................................................................................ 75
Epilepsy and slow waves ............................................................................................ 75
Supragranular activity in SWS ................................................................................... 77
Frequency of the slow waves ..................................................................................... 81
Role of calretinin positive neurons ............................................................................. 82
Firing rate in SWS ...................................................................................................... 83
SUMMARY ................................................................................................................... 87
BIBLIOGRAPHY......................................................................................................... 88
PUBLICATIONS ........................................................................................................ 107
ACKNOWLEDGEMENTS ....................................................................................... 108
3
Csercsa Richárd
LIST OF ABBREVIATIONS
5-HT
Serotonin (5-Hydroxytryptamine)
Ach
Acetylcholine
AMPA
α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid
ANOVA
Analysis Of Variance
ARAS
Ascending Reticular Activating System
BOLD signal
Blood Oxygenation Level Dependent signal
CEM
Competitive Expectation-Maximization
CR
Calretinin
CSD
Current Source Density
DAB
3,3'-Diaminobenzidine
ECoG
Electrocorticography
EEG
Electroencephalography
EPSP
Excitatory Postsynaptic Potential
FFT
Fast Fourier Transform
fMRI
Functional Magnetic Resonance Imaging
GABA
Gamma-Aminobutyric Acid
GFAP
Glial Fibrillar Acidic Protein
IPSP
Inhibitory Postsynaptic Potential
LFP
Local Field Potential
LFPg
Local Field Potential gradient
4
Laminar analysis of the slow wave actvity in humans
LTD
Long-Term Depression
LTP
Long-Term Potentiation
ME
Multielectrode
MEG
Magnetoencephalography
MRI
Magnetic Resonance Imaging
MUA
Multiunit Activity
NeuN
Neuronal Nuclei staining
NMDA
N-methyl-D-aspartic acid
NREM
Non Rapid Eye Movement
PB
Phosphate Buffer
PET
Positron Emission Tomography
PGO wave
Ponto-Geniculo-Occipital wave
REM
Rapid Eye Movement
SD
Standard Deviation
SO
Slow Oscillation
STDP
Spike-Timing-Dependent Plasticity
SUA
Single Unit Activity
SWA
Slow Wave Activity
SWS
Slow Wave Sleep
TMS
Transcranial Magnetic Stimulation
VLPO
Ventrolateral Preoptic Nucleus
5
Csercsa Richárd
INTRODUCTION
SLEEP – GENERAL INTRODUCTION
Sleep seems to be essential for every living creature with a nervous system,
though its function is not clear yet. It can be observed in most mammals and birds, and
also in many insects (e.g. cockroach, bee, scorpion, or Drosophila), reptiles (e.g. turtle),
amphibians (e.g. tree frog), and fish (e.g. zebrafish or perch) (Cirelli and Tononi, PLoS
Biol, 2008). There are species that do not seem to meet the criteria of sleeping, e.g.
bullfrog, or coral reef fish (Siegel, Trends Neurosci, 2008).
Sleep is a condition during which animals are resting and their responses to
external sensory stimuli (sounds, touch, smell) and processing of the information these
stimuli carry are altered compared to the waking state. Most of these stimuli don’t reach
the conscious level, they fade in the neuronal silence of the brain. Opposing this
function that maintains sleep, there is need for waking, when external stimuli represent
for example imminent danger. The brain needs to uphold a state that protects the
physiologically important sleep but aborts it when necessary.
Circadian rhythmicity and homeostatic regulation are both important criteria for
determining sleep (Borbely, Hum Neurobiol, 1982). Circadian rhythmicity means the
sleep need is elevated in certain phases of the day. In diurnal animals it is increased
during the night, while nocturnal animals tend to sleep during the day. In humans,
besides the elevation of sleep need during the night, there is an increase of sleep
willingness in the early afternoon also. A behavioral state cannot be called sleep if there
is no homeostatic regulation, i.e. the amount of sleep should be proportional to the
amount of waking and this is maintained by the nervous system itself. It is signaled by
the compensatory rebound after sleep deprivation, when subjects tend to sleep more.
(Bódizs, Kognitív idegtudomány, 2003)
6
Laminar analysis of the slow wave actvity in humans
SLEEP PATTERNS
Sleep can be divided into different stages, which follow a specific pattern in
animals and humans alike, though the duration and properties of these stages may vary
from species to species.
SLEEP STAGES
In humans the two main types of sleep are Rapid Eye Movement (REM) and NonRapid Eye Movement (NREM) sleep. NREM can be further divided into three or four
stages which go from light to deep sleep. During the night, these stages follow each
other in a more or less determined manner, which can be described on a hypnogram
(Figure 1). The hypnogram shows the cyclic alternations of sleep stages during the
whole night. Stages are characterized by specific brain electrical activity, muscle tone,
etc., that became the main criteria for determining them. These criteria were
standardized in 1968 by Rechtschaffen and Kales. Their classification contained four
stages of NREM sleep. They were used for sleep staging until 2007, when the American
Academy of Sleep Medicine (AASM) issued a manual for the scoring of sleep, in which
NREM stages 3 and 4 were merged into stage 3 (Iber et al., 2007).
Figure 1. Sleep hypnogram. Change in sleep stages (vertical axis) during the night (horizontal
axis). It shows that the first half of the night consists mostly of NREM sleep, while the second
half is rich in REM sleep. Modified from (Siegel, 2002)
7
Csercsa Richárd
NREM SLEEP
The dominant sleep stage during the first half of a night’s sleep is the NREM sleep
that can be further divided to three stages according to the new AASM guidelines that
correspond to the levels of awakability, with stage 1 being the lightest and stage 3 the
deepest stage (Figure 2).
•
Stage 1 is the onset of sleep, a transition from waking that takes 510% of total sleep time. Alpha waves (8-12 Hz) that are
characteristic for the drowsy waking state are replaced with theta
waves (4-8 Hz). Sudden twitches and hypnic jerks may also occur.
•
Stage 2 occupies 45-55% of total sleep and it is characterized by
sleep spindles (around 12 Hz) generated by thalamocortical loops
and K-complexes. Muscle tone decreases and conscious awareness
of the external environment disappears.
•
Stage 3 (15-25% of sleep time) is commonly referred to as deep
sleep or slow wave sleep (SWS), because during this stage delta
frequency (1-4 Hz) and slow (below 1 Hz) oscillations appear.
Deep sleep is considered to be a state when the mind is detached
from the external world, and sensory stimuli are rarely able to reach
the cortex for further processing. Recent findings show a more
complex image of the processes taking place during deep sleep.
This is also the stage when night terrors and sleepwalking occur.
8
Laminar analysis of the slow wave actvity in humans
Figure 2. The EEG graphoelements of sleep stages. The awake EEG consists of highfrequency, low-amplitude waves, while during deep sleep low-frequency, high-amplitude
oscillations can be observed. REM sleep EEG pattern is similar to waking. Modified from
(Pace-Schott and Hobson, Nat Rev Neurosci, 2002).
REM SLEEP
Rapid Eye Movement (REM) sleep was first described in 1953 by Aserinsky and
Kleitman (Aserinsky and Kleitman, Science, 1953). It accounts for 20-25% of total
sleep time and it is mainly characterized by fast movements of the eyes, desynchronized
high-frequency EEG activity, muscle atonia, PGO waves (recorded from the pons, the
geniculatum, and the occipital cortex), and theta frequency (4-10 Hz) oscillation in the
hippocampal formation. Brain activity largely resembles that of the waking activity,
thus this stage is also called paradoxical sleep. REM occurs more often during the
second half of sleep. It is thought to have important function in spatial and procedural
(implicit) memory consolidation, and dreaming occurs during REM sleep.
DEVELOPMENT
Sleep patterns also change with the life span. In humans, sleep time decreases with
age. Infants sleep up to 13-14 hours a day, which decreases to 7-8 hours in adolescence,
and further to 5-6 hours at the age of 70. The proportion of REM sleep to NREM also
decreases from 50% in infants to 20% in adolescents and 15% in old age (Détári, 2008).
9
Csercsa Richárd
Slow wave activity (SWA) during NREM sleep also undergoes prominent
changes, it increases during the first years of life, reaches its peak around the age of 8
and after that it declines, forming an inverted U-shape curve. It has been suggested that
SWA decrease is correlated with thinning of gray matter, therefore decline in synaptic
density (Campbell and Feinberg, Proc Natl Acad Sci U S A, 2009). At the age of 2-5,
SWA is dominant in the occipital cortex and shifts to frontal cortex by the age of 20
(Kurth et al., J Neurosci, 2010). This shift is also consistent with studies of cortex
maturation. These findings support the hypothesis of a strong connection between SWA
and synaptic strength (Tononi and Cirelli, Sleep Med Rev, 2006).
Interestingly, genetics seem to largely influence the individual sleep profile. It has
been
shown
in
a
study
of
monozygotic
and
dizygotic
twins
that
the
electroencephalographic “fingerprint” (power spectra between 8 and 16 Hz during
NREM sleep, which allows a correct discrimination between individuals) is genetically
determined. It shows a heritability estimate of 96%, making the sleep EEG profile one
of the most heritable traits of humans (De Gennaro et al., Ann Neurol, 2008).
NEURAL BASIS OF AROUSAL
Despite sleep is an everyday phenomenon, we still don’t entirely understand its
functional significance. Neural elements and processes related to sleep may give us a
clue for its functions. Unfortunately, the neural mechanisms underlying the generation,
maintenance, and termination of sleep are still not clear. Theories and the current
conceptions of these mechanisms are described in this section.
10
Laminar analysis of the slow wave actvity in humans
PASSIVE OR ACTIVE
There are several theories of what could cause the change in arousal as well as in
cortical excitability during sleep. There are two classical conceptions: the passive and
the active hypothesis. According to the passive hypothesis, the default arousal state is
sleep, and it is a result of a decrease in sensory stimulation. The active hypothesis states
that sleep is the result of an active process mediated by some sleep-promoting factors in
the nervous system. According to our present knowledge, sleep is far too complex to be
regarded as a passive state. There are indeed chemical substances that induce sleep such
as sedatives or cytokines (e.g. TNF and IL-1) (Krueger et al., Nat Rev Neurosci, 2008),
supporting the active hypothesis. On the other hand, both sleep-promoting (von
Economo, Ergebn. Physiol. (München), 1929) and wake-promoting centers (Moruzzi
and Magoun, Electroencephalogr Clin Neurophysiol, 1949) have been described.
DEAFFERENTATION
An important milestone in the history of sleep research is the experiment by
Frederick Bremer in 1935 (Bremer, C R Soc Biol, 1936). He surgically isolated the
brain from the body by cutting the neural axis between the brain stem and the spinal
cord at the level of the first or second cervical vertebra. This preparation is called the
encéphale isolé (Figure 3, left). After the lesion he experienced normal sleep-wake
cycles. Furthermore, his second surgical preparation, the cerveau isolé (Figure 3, right),
when he isolated the forebrain from the brain stem, resulted in constant sleep. From
these experiments, Bremer concluded that waking requires sensory stimulation of the
brain and sleep is the result of sensory deafferentation. This indicates that sleep is the
default state of arousal and the nervous system reverts to the sleep mode as a passive
process after sensory stimuli cease to reach the forebrain. Despite recent theories think
of sleep rather as an active process generated by complex mechanisms in distributed
systems, the importance of deafferentation cannot be neglected. When sleeping time
11
Csercsa Richárd
comes, all animals and humans look for a dark, silent place with as little external stimuli
as possible. Another interpretation of these findings could be that brain areas necessary
for waking are below the lesion of cerveau isolé, but this was not suggested at that time.
Figure 3. The surgically isolated brain experiment setups by Bremer. Left: encephalé isolé
(cut between the brain stem and the spinal cord, normal sleep-wake cycles), right: cerveau
isolé (lesion above the brain stem, results in constant sleep). From (Bremer, C R Soc Biol,
1936)
ASCENDING RETICULAR ACTIVATING SYSTEM
The next big step in understanding sleep processes was the discovery of the
Ascending Reticular Activating System (ARAS) by Moruzzi and Magoun in 1949
(Moruzzi and Magoun, Electroencephalogr Clin Neurophysiol, 1949). They showed that
sleep activity induced by anesthesia or the encéphale isolé preparation can be turned
into waking activity by stimulating the reticular formation. Interestingly, this waking
activity could be observed throughout the cortex, and lasted as long as the stimulation
period (Figure 4).
Figure 4. Waking activity due to brainstem stimulation. Electrical stimulation of the reticular
formation (horizontal bar) turns sleeping EEG activity into waking. From (Moruzzi and
Magoun, Electroencephalogr Clin Neurophysiol, 1949)
12
Laminar analysis of the slow wave actvity in humans
The ARAS consists of pathways ascending from the brain stem to the thalamus
and the neocortex, regulating the release of different neurotransmitters. The cholinergic
pathway originates in the nucleus pedunculopontine tegmentum and the nucleus
laterodorsalis tegmentum, the noradrenergic pathway in the locus coeruleus in the pons,
the serotonergic system in the raphe nuclei of the medulla, pons, and midbrain, and the
histaminergic system in the tuberomammillary nucleus in the hypothalamus (Figure 5).
The interaction of these distributed systems is thought to promote arousal. During
waking, both the cholinergic and the monoaminergic (noradrenergic, serotonergic and
histaminergic) systems are active. During NREM sleep, the influence of both decreases,
while during REM sleep, the monoaminergic system is totally inactive. These nuclei
project to the thalamus and excite the cortex broadly.
Figure 5. Ascending reticular activating system. The origins of neuromodulators affecting
arousal and their projections. From (Saper et al., Trends Neurosci, 2001)
In addition, a very small population of cells in the lateral hypothalamus that
produce orexin (or hypocretin as it is also called) is also thought to promote
wakefulness. These neurons provide excitatory input to the arousal system as well as to
the cerebral cortex. Orexin plays a critical role in preventing abnormal behavioral state
transitions, particularly into REM sleep, thus being a major factor in narcolepsy. In
13
Csercsa Richárd
experiments in which the gene for orexin was knocked out in mice, the animals became
narcoleptic (Chemelli et al., Cell, 1999). In humans with narcolepsy, the orexin level in
the cerebrospinal fluid is abnormally low (Nishino et al., Lancet, 2000). It has been
found that histamine-synthesizing cells are excited by orexin and orexin cells are
synapsed upon by histaminergic cells (Eriksson et al., J Neurosci, 2001), so it seems
that there are two cooperating neuronal populations in the hypothalamus regulating
arousal level. (Siegel, 2002)
VENTROLATERAL PREOPTIC NUCLEUS
Besides the distributed activating system, a sleep-promoting center situated in the
posterior hypothalamus has been described. The preoptic area of hypothalamus was
proposed to play a critical role in sleep by von Economo in 1929 (von Economo,
Ergebn. Physiol. (München), 1929). He examined patients with encephalitis lethargica
and found that lesions in the preoptic area caused insomnia (Triarhou, Brain Res Bull,
2006). This finding was verified in rats by Nauta in 1946 (Nauta, J Neurophysiol, 1946)
and in cats by Sterman and Clemente in 1962 (Wyrwicka et al., Science, 1962).
Sleep promoting neurons were not localized precisely until 1996, when Sherin et
al. found in rats that after long sleep, c-fos, an indirect marker of neural activity was
densely packed in the Ventrolateral Preoptic nucleus (VLPO), a small group of neurons
in the preoptic area. It has also been demonstrated that the density of neurons stained for
c-fos in VLPO positively correlates with the amount of sleep, and that the activity of
VLPO and suprachiasmatic nucleus (critical in circadian regulation) is synchronized
(Sherin et al., Science, 1996).
Electrophysiological recordings revealed that VLPO neurons start firing seconds
before the onset of SWS and remain active, with their firing rate increasing when sleep
gets deeper. Furthermore, the lesion of VLPO neurons in rats caused long-lasting
insomnia. These results point to the VLPO as a sleep-promoting center in the brain.
14
Laminar analysis of the slow wave actvity in humans
VLPO is reciprocally connected to histaminergic, orexinergic, serotoninergic,
noradrenergic, and cholinergic neurons of the reticular activating system (Figure 6).
These reciprocal connections are inhibitory in both directions, i.e. the sleep-promoting
VLPO inhibits the wake-promoting systems, and vice versa. During sleep, VLPO
neurons increase their activity, inhibiting wake-promoting neurons, thus disinhibiting
themselves. During waking, VLPO neurons are inhibited by the activating system,
reinforcing the firing of wake-promoting neurons.
Figure 6. VLPO projections. The sleep-promoting VLPO reciprocally innervates the wakepromoting ARAS with inhibitory connections, thus forming a ’flip-flop’. From (Saper et al.,
Trends Neurosci, 2001)
This structure of mutual negative feedback resembles a piece of electrical
circuitry, called the ‘flip-flop’ (Figure 7). This bistable switch ensures that any time
during the day, only one of the two arousal states is present, and there are no overlaps
between the two (Saper et al., Trends Neurosci, 2001). Studies with narcolepsy, when
sleep suddenly breaks into waking, point to orexin as a regulating factor of the switch.
A recent study indicates that the ‘flip-flop’ can be switched by the high-frequency
bursting of a single cortical neuron (Li et al., Science, 2009). This activity of a single
neuron may affect arousal states through cortico-cortical connections, the thalamo-
15
Csercsa Richárd
cortico-reticular loop, or by descending projections from the cortex to the reticular
activating system. (Fort et al., Eur J Neurosci, 2009)
Figure 7. The ’flip-flop’ of sleep and waking. Several neuromodulators play roles in its
operation, with orexin as the main switch. From (Saper et al., Trends Neurosci, 2001)
SLEEP REGULATORY SUBSTANCES (SRS)
The theory that sleep is induced by humoral factors rather than neural networks
was first proposed in 1909 by Kuniomi Ishimori in Japan, and in the same time, but
independently, by Henri Piéron in France. They both took samples from the
cerebrospinal fluid of sleep-deprived dogs and injected it in the brain of dogs with
normal sleep-wave cycle, and the dogs fell asleep for 2-6 hours after the injection. In
1967, J. R. Pappenheimer repeated the experiment, injecting the cerebrospinal fluid of
sleep-deprived goats in cats and rats, inducing sleep-like behavior that could be broken
by noise or handling, but reverted back to sleep if the animal was left undisturbed
(Pappenheimer et al., Proc Natl Acad Sci U S A, 1967).
16
Laminar analysis of the slow wave actvity in humans
Many substances have been proposed as sleep regulatory. The main criteria that a
viable candidate should fulfill include: the substance should enhance sleep, it should
reduce spontaneous sleep if inhibited, its brain levels should vary with sleep-wake
history, and it should act on sleep-regulatory circuits (Krueger et al., Nat Rev Neurosci,
2008). For example adenosine, nitric oxide, prostaglandin D2, tumor necrosis factor
(TNF), interleukin-1 (IL1), and growth hormone releasing hormone (GHRH) meet these
criteria and have been proposed as SRSs.
The idea is that ATP, released during neurotransmission, acting on glia releases
IL1 and TNF that in turn act on neurons to change their electrical properties. These
events happen locally and thereby alter the state of the neural network (Krueger, Curr
Pharm Des, 2008). Cytokines like IL1 and TNF have long evolutionary history, dating
back to invertebrates. This may explain the universality of sleep among mammals.
LOCAL SLEEP
Sleep is traditionally seen as a global process, based on that changes in behavior
and physiological properties are present in the whole organism. The systemic nature and
top-down control of sleep seems to be obvious if we consider that the animal’s posture,
hormone secretion, metabolic rate, and brain activity all change during the transition
from waking to sleep and from sleep to waking.
But if we examine it on a finer scale of time and space, at the level of milliseconds
and cortical columns, we find a bottom-up organization of ‘sleeping’ local networks.
SLEEP IN GROUPS OF NEURONS
There are several observations that conflict with the global nature of sleep. First, it
was shown by Mukhametov in 1977 (Mukhametov et al., Brain Res, 1977), that marine
mammals like dolphins, whales, and manatees sleep unihemispherically, i.e. they almost
17
Csercsa Richárd
never exhibit slow wave activity in both hemispheres at the same time (they are the only
mammals without REM sleep, too). The eye contralateral to the hemisphere with the
slow waves is closed, while the other is open. The immobility, associated with the sleep
of terrestrial mammals, is not a definitive criterion for sleep in marine mammals, since
many of them keep moving during their whole life. (Siegel, Nature, 2005).
It has been proposed that human subjects during sleep walking are awake and
asleep simultaneously (Mahowald and Schenck, Nature, 2005). Furthermore, no case of
survivable brain lesion has been reported that caused complete insomnia. This suggests
that sleep is a fundamental self-organizing property of any group of neurons (Krueger et
al., Nat Rev Neurosci, 2008).
SLEEP IS USE-DEPENDENT/EXPERIENCE-DEPEDENT
Another observation that points to the emergence of sleep in local networks is that
the activation of specific brain regions during wakefulness leads to an increase in slow
wave activity during sleep in the same brain areas.
In 1993, Krueger and Obál suggested that sleep begins as a local neuronal group
event, serving to preserve a constancy of a synaptic superstructure by the use of
synapses unstimulated during waking (Krueger and Obal, J Sleep Res, 1993). According
to their theory, neural activity leads to the enhanced production of sleep factors acting
locally, modifying the neural activity that results in the amplification of synaptic
strength (increase in postsynaptic potentials) of active synapses. They also hypothesized
the existence of higher order neuronal groups that globally coordinate sleep either
through their extensive projections or via humoral agents.
One of the consequences of this theory was justified when Rector et al. showed
that a sleep-like state can exist in local networks in the brain, specifically in cortical
columns, that are thought to be the basic processing neuronal assemblies (Rector et al.,
Brain Res, 2005). The response evoked by afferent sensory stimulation depends on the
state of the column that can be either waking or sleeping. During sleep escalated to the
18
Laminar analysis of the slow wave actvity in humans
whole organism, responses from most columns indicate that it is sleeping, while during
waking most columns are in waking mode. However, some columns are in waking
mode during global sleep and some are sleeping while the animal is awake. This
suggests that the basic element of sleep is a neuronal assembly and they can go into
sleeping or waking state independently from each other.
Moreover, this local sleep seems to be homeostatically regulated, which is another
direct consequence of Krueger and Obál’s theory. The likelihood a column goes to sleep
depends on the amount of time it spent awake (Rector et al., Brain Res, 2005).
Another crucial consequence of the theory is that the amount of sleep in a column
or a brain area measured by the slow wave activity seems to be dependent on the level
of activity it was exposed to during waking. This property has been tested in several
configurations.
Kattler et al. applied vibratory stimuli to a subject’s left or right hand for 6 hours.
They found that during subsequent sleep, EEG power in the delta frequency range was
shifted to the hemisphere contralateral to the stimulated hand (Kattler et al., J Sleep Res,
1994).
In a rat experiment, sensory inputs were reduced to one side by cutting the
contralateral whiskers. After 6 hours spent awake in an enriched environment, spectral
power in the NREM sleep EEG in the 0.75-6 Hz range exhibited an interhemispheric
shift towards the side opposite to the intact whiskers (Vyazovskiy et al., J Sleep Res,
2000).
In a human study using 256-channel EEG recordings, the subject had to perform a
visuomotor task either with rotation of the target (learning task) or without rotation (no
learning). A local increase in slow wave activity (1-4 Hz) appeared during the first half
hour of sleep immediately after the task. This increase was confined to brain areas
associated with spatial tasks and spatial coordinate transformation and could be
observed only after the learning task. The same group also found a correlation between
local increase in slow wave activity and post-sleep task performance improvement
(Huber et al., Nature, 2004).
19
Csercsa Richárd
If an increase in synaptic strength due to learning leads to an increase in slow
wave activity, the decrease in synaptic strength should do the opposite. A subject’s arm
was immobilized during the day, reducing synaptic strength locally as indicated by an
amplitude reduction in somatosensory evoked potentials (electrical median nerve
stimulation) and in motor evoked potentials (transcranial magnetic stimulation). This
resulted in a drop in motor performance. During subsequent sleep, slow wave activity
was reduced over the same cortical area (contralateral to the immobilized arm). Thus,
local sleep regulation can be linked to cortical plasticity even without learning (Huber et
al., Nat Neurosci, 2006).
These results show that the amount of slow wave activity is correlated with
synaptic weights. The bigger the synaptic strength, the more local slow waves exist, and
vice versa. Tononi and Cirelli suggested that the purpose of slow waves is to depress the
synapses, thus maintaining a homeostatic regulation of synaptic strength (Tononi and
Cirelli, Sleep Med Rev, 2006). Synaptic downscaling is thought to have a beneficial
effect on memory that is described in a further chapter. Other hypothesized advantages
of downscaling include energy and space saving. The gray matter is the most energydemanding part of the brain, and the majority of the energy requirement is due to the
repolarization following postsynaptic potentials and spike generation. The number of
spikes increases with stronger synaptic weights, so downscaling would significantly
reduce energy demand. Synaptic strengthening probably increases the number and the
size of terminal boutons. Therefore, downselection is crucial in maintaining the space
requirements of synapses (Tononi and Cirelli, Sleep Med Rev, 2006).
Cytokines have been proposed to play an important role in promoting sleep in
local circuits (Krueger, Curr Pharm Des, 2008). Furthermore, the role of cytokines in
synaptic plasticity is crucial. It has been shown that TNF-α mediates synaptic scaling
through changes in the number of AMPA receptors and that glia cells are the source of
this form of plasticity (Stellwagen and Malenka, Nature, 2006). This might link the
theory of synaptic downscaling as the cause of local sleep to the theory of cytokines
provoking sleep in neuronal assemblies.
The state-of-the-art idea is thus that on macroscopic scale sleep is global, with
physiological and behavioral changes having impact on the whole organism, while on
20
Laminar analysis of the slow wave actvity in humans
the level of neuronal assemblies, cortical columns can exhibit slow wave activity
individually, depending on past activity.
SLEEP DEPRIVATION
Sleep is essential and executes numerous vital functions. We can come to that
conclusion if we notice that sleep deprivation can be harmful. Rats kept awake for a
prolonged time developed an increase in metabolic rate and decrease in body weight,
which culminated in death. It can lead to death in flies and cockroaches, too. Pigeons,
however, appear capable of surviving prolonged sleep deprivation (Cirelli and Tononi,
PLoS Biol, 2008). In humans, staying awake for longer than usual leads to a feeling of
sleepiness, microsleep episodes (intrusions of sleep into waking), cognitive impairment,
tremors, hallucinations, and even death. A well-documented case is an experiment from
1965, where the condition of Randy Gardner, a 17-year old high school student from
San Diego, was monitored by sleep researcher William Dement during sleep deprivation
for 11 days. Except for a few illusions he demonstrated no unusual behavior and his
physiological conditions remained within healthy limits (Siegel, 2002) showing large
individual differences in sleep demand.
Sleeping too much may also be harmful. People who sleep too much have a more
unstable mood, they are more anxious and distressed than short sleepers. In contrast,
staying awake during the second half of the night can induce a positive change in the
mood of depressed patients. There is also a correlation between average daily sleep time
and death rate. People who sleep very short or very long have higher death rates than
people who sleep the average, daily 7-8 hours (Bódizs, Kognitív idegtudomány, 2003).
21
Csercsa Richárd
FUNCTIONS OF SLEEP
There has been extensive search for the roles of sleep, but they still remain
elusive. Sleep is supposed to serve fundamental functions, since it occurs very
frequently, despite putting the sleeper in danger, and deprivation of sleep may have
lethal consequences. Functions sleep may execute include recovery, conserving energy,
and recent findings indicate that it plays an important role in memory consolidation.
RECOVERY AND RESTORATION
The theory that recovery is a function of sleep comes from the observation that we
cannot execute either physical or mental tasks after a considerable amount of waking as
well as immediately after sleep. This means that sleep recovers the body and restores
the same state of its functionality as the day before.
One of the conditions thought to be restored during sleep is the synaptic weights
between neurons. In their synaptic downscaling theory, Tononi and Cirelli proposed that
synaptic weights are strengthened during waking due to learning and downscaled during
deep sleep in order to keep newly formed memories, and make place for even newer
ones (Tononi and Cirelli, Sleep Med Rev, 2006).
ENERGY CONSERVATION
During deep sleep, metabolism is decreased with about 10% compared to waking,
but the energy conservation at lower temperatures may be even higher. The metabolic
reduction in an unclothed human subject at the room temperature of 21 °C may reach
40% after sleep onset. (Velluti, 2008)
22
Laminar analysis of the slow wave actvity in humans
Energy conservation may be particularly important in newborns. They have a high
ratio of surface area to body mass, thus they need more energy to maintain their body
temperature. Because their sensory-motor system is immature, they also benefit from
the less exposure to danger during sleep, since they are not moving around exposing
themselves, but lie quietly in a protective environment (Siegel, Nature, 2005).
Furthermore, the high amount of sleep in infants can be explained by sleep’s role in
memory consolidation and learning.
However, energy conservation cannot be the only function of sleep since quite
wakefulness is also a state with reduced energy consumption without the danger that is
due to the loss of consciousness. Moreover, arctic ground squirrels periodically interrupt
their hibernation and consistently sleep during these arousals with their EEG showing a
progressive decrease in slow wave activity, indicating a declining requirement for sleep
(Daan et al., Neurosci Lett, 1991). Since hibernation is already an energy conserving
state, this points to that sleep serves more functions than simply conserving energy,
functions that are important enough to interrupt hibernation from time to time.
PLASTICITY, MEMORY, LEARNING
It is a widely accepted theory that sleep benefits learning (Maquet, Science, 2001),
(Frank and Benington, Neuroscientist, 2006). Several studies show that performance of
diverse memory tasks is significantly improved after sleep. But it is still debated which
sleep stage contributes to the consolidation of which type of memories and what the
underlying mechanisms are.
Some researchers question this theory (Siegel, Science, 2001), (Vertes and Siegel,
Sleep, 2005). They argue that there is no clear evidence for a major role of REM sleep
in memory consolidation, instead they bring the example of patients with REM sleep
deprivation induced by drugs (serotonin re-uptake inhibitors) or lesion who did not
develop learning deficit (Rasch et al., Nat Neurosci, 2009).
23
Csercsa Richárd
MEMORY CONSOLIDATION DURING SLEEP
Memory consolidation could well be the reason for the loss of consciousness
during sleep, making it the main candidate in the search for the function sleep is
essential for. The same neuronal networks in the brain are used for storing and
processing information, two processes that may not be executed in the same time. Thus
two modes of operation seem to be necessary: an 'online' mode for information
processing (waking) and an 'offline' mode for memory formation (sleep).
The idea that sleep benefits memory dates back to 1801, when David Hartley, a
British psychologist proposed that dreaming might change the association links of
memory. However, the first study examining the connection between sleep and learning
was performed only in 1924 by Jenkins and Dallenbach (Jenkins and Dallenbach,
American Journal of Psychology, 1924). They showed that memory recall improved
after a night of sleep. They theorized that this improvement resulted from a lack of
sensory interference during sleep. Instead of this passive process theory, the modern
conception is that sleep actively modifies synaptic structure, thus consolidates memory
traces (Walker and Stickgold, Annu Rev Psychol, 2006).
MEMORY CATEGORIES
Memory is the collective term of processes, not a unitary entity. In humans it can
be classified in two main categories: declarative and non-declarative (Figure 8) (Cohen
and Squire, Science, 1980).
Non-declarative memory is regarded as non-conscious, implicitly learned,
including procedural memory that is the learning of perceptual and motor skills
(‘knowing how’). They are acquired gradually by repeated practice, but remain fairly
stable. Their consolidation is dependent on cortico-striatal circuitry.
24
Laminar analysis of the slow wave actvity in humans
Declarative memories can be considered as explicitly learned, consciously
accessible memories of facts and events (‘knowing what’). It can be further divided to
episodic memory (autobiographical memory for events in the past) and semantic
memory (memory for general knowledge not related to specific events). They are
rapidly encoded, but sensitive for attenuation and interference. Their encoding and
short-term retrieval essentially rely on structures in the medial temporal lobe, especially
the hippocampus. In 1957, Scoville and Milner reported the case of HM, a patient with
epilepsy, whose hippocampus on both sides was removed surgically and as a result he
developed anterograde amnesia. He could not store long-term declarative memories, but
his working memory and procedural skill learning remained intact (Scoville and Milner,
J Neurol Neurosurg Psychiatry, 1957). According to the standard model, the retrieval of
declarative memories becomes independent of hippocampus over time, possibly due to a
hypothesized information transfer to the neocortical stores. However, the multiple
memory trace theory suggests that while semantic information can be established in the
neocortex, hippocampus is always necessary for the retrieval of episodic and spatial
information (Nadel et al., Hippocampus, 2000).
(Walker and Stickgold, Annu Rev Psychol, 2006), (Stickgold, Nature, 2005),
(Marshall and Born, Trends Cogn Sci, 2007), (Born et al., Neuroscientist, 2006)
Figure 8. Memory categories. Memory consists of two main categories, declarative and nondeclarative, which can be further distinguished. From (Stickgold, Nature, 2005)
25
Csercsa Richárd
MEMORY FORMATION
The formation of memory consists of three fundamental stages: acquisition,
consolidation, and retrieval.
Acquisition is the learning process when new information is encoded into a neural
trace, its representation as a neural ensemble.
Consolidation is the stabilization and enhancement of memories. Stabilization
means that memory becomes resistant to interference, while enhancement is the active
maintenance of a memory. During consolidation, the brain has to integrate newly
encoded memories with pre-existing long-term memories. In this sense, we can
differentiate two levels of consolidation. ‘Synaptic consolidation’ is accomplished
within hours after learning and involves mainly the stabilization of synaptic changes.
‘System consolidation’ means that the memory trace undergoes a reorganization so it
becomes represented by different neural networks, coming into context with related
memories. This process can take days or months to complete.
Retrieval refers to the recall of stored memories. The act of memory recall is
believed to destabilize the memory representation therefore reconsolidation is needed to
prevent degradation. (Walker and Stickgold, Annu Rev Psychol, 2006), (Born et al.,
Neuroscientist, 2006)
SYNAPTIC CONSOLIDATION
Since the recognition that the number of neurons in the adult brain is more or less
constant, it became obvious that learning and forming new memories cannot be
attributed to new neurons arising. At the end of the 19th century, Santiago Ramón y
Cajal suggested that learning is the result of a change in neuronal connections,
modifying their communication. In 1949, Donald Hebb postulated that if a cell fires
persistently soon enough after another cell’s firing, then there will be a change in their
26
Laminar analysis of the slow wave actvity in humans
connection due to some growth process or metabolic change so that the efficiency of
their communication will be increased (‘cells that fire together, wire together’) (Hebb,
1949).
The microbiological basis of Hebbian learning was discovered by Lømo in 1966.
He experienced that excitatory postsynaptic potentials elicited by single electrical pulses
applied to presynaptic cells were increased if preceded by the application of a highfrequency spike train to the presynaptic cell, and that the increase was long-lasting
(Lomo, Acta Physiologica Scandinavia, 1966). This effect is called long-term
potentiation (LTP).
LTP can be induced not only by high-frequency spiking but by the precise timing
of pre- and postsynaptic action potentials (spike-time dependent plasticity, STDP). If
the firing of the presynaptic cell precedes the firing of the postsynaptic cell, the
connection between them is strengthened. However, if the postsynaptic cell fires first,
the connection becomes weaker (long-term depression, LTD) (Figure 9).
Figure 9. Time dependency of firing during STDP. If presynaptic firing precedes postsynaptic
firing, it causes LTP, while the opposite order results in LTD. From (Nevian and Sakmann, J
Neurosci, 2006)
LTP has an early phase, when the synapses become stronger, and a late phase,
when this conformation is maintained. Early LTP depends on the calcium concentration
of the postsynaptic cell. The concentration has to exceed a certain threshold, which
makes the postsynaptic N-methyl-d-aspartate (NMDA) receptors essential. Late LTP
requires gene transcription and protein synthesis.
27
Csercsa Richárd
It was shown that the time and threshold necessary for LTP induction vary on
different synapses, but once it is established, the response is rapid and digital. It means
that there is a step increase in postsynaptic currents after the LTP and stays at this
elevated level even for further potentiation. The advantage of this digital type of
memory storage may be its robustness in a noisy environment. If there is a slight change
in the synaptic weight due to noise, it might affect the graded value greatly, but it would
take a considerable amount of noise to affect the digital value. Even if there is
variability (as there often is in biological systems), it will not lead to errors (Petersen et
al., Proc Natl Acad Sci U S A, 1998) (Lynch, Physiol Rev, 2004).
SYSTEM CONSOLIDATION
Two types of memory stores is thought to be present in the brain: a short term
storage for keeping information from seconds to hours, and a long term storage from
hours to months or even the whole lifetime. The theory that new learning takes some
time dates back to the 19th century. But it was Atkinson and Shiffrin who proposed a
model that included short term and long term memory stores (Atkinson and Shiffrin,
The psychology of learning and motivation: II, 1968). The basic idea is that newly
acquired, labile memory traces are converted into long-lasting forms during different
modes of operation. Marr suggested that the archicortex (hippocampus) and neocortex
are ideal candidates for these two stores (Marr, Proc R Soc Lond B Biol Sci, 1970),
(Marr, Philos Trans R Soc Lond B Biol Sci, 1971), (Buzsaki, Neuroscience, 1989),
(Squire, Psychol Rev, 1992). The state-of-the-art concept of the model is that the
synaptic weight structure is reorganized due to learning during waking (online mode),
by the sequential firing of neurons in an assembly, forming memory traces (encoding).
These traces are re-activated during sleep (offline mode) in order to strengthen the new
structure of connections (synaptic consolidation) and transfer the new information from
temporary storage to long term storage, from the hippocampus to the neocortex (system
consolidation) (McClelland et al., Psychol Rev, 1995).
28
Laminar analysis of the slow wave actvity in humans
The neurobiological mechanisms underlying memory consolidation during sleep is
still not known, though several theories exist. A widely accepted theory states that
declarative, hippocampus-dependent memories are mainly consolidated during SWS,
while consolidation of procedural memories independent of hippocampus is supported
mainly by REM sleep (Born et al., Neuroscientist, 2006), (Diekelmann et al., Sleep
Med Rev, 2009), (Marshall and Born, Trends Cogn Sci, 2007), though the latter seems
to be inconsistent with the finding that antidepressant drugs (selective serotonin reuptake inhibitors) suppressing REM sleep slightly improved skill memory consolidation
(Rasch et al., Nat Neurosci, 2009), (Vertes and Siegel, Sleep, 2005), and studies
showing that sleep does not benefit implicit sequence-specific motor learning (Song et
al., J Neurosci, 2007), (Nemeth et al., Exp Brain Res, 2010). Other studies suggest that
proper memory consolidation takes place only when the whole night’s sleep is
undisturbed, meaning that REM sleep follows SWS (Gais et al., Nat Neurosci, 2000).
SLOW WAVE SLEEP
Slow wave sleep is considered the deepest phase of sleep, dominant during the
first half of the night. It is accompanied by maintained muscle tone, lower awakability,
and high-amplitude, low-frequency (< 2Hz) waves on the EEG.
SLOW OSCILLATION
Delta waves (0-4 Hz) during sleep were reported in 1937 (Blake and Gerard, Am J
Physiol, 1937), but slow oscillations (SO, < 1 Hz), the main characteristic
electrophysiological features of deep sleep were first described in anesthetized cats by
Mircea Steriade in 1993 (Steriade et al., J Neurosci, 1993a). They were later detected in
other mammals and humans, too (Achermann and Borbely, Neuroscience, 1997), during
various arousal states like anesthesia, natural sleep (Destexhe et al., J Neurosci, 1999),
29
Csercsa Richárd
and quiet wakefulness (Petersen et al., Proc Natl Acad Sci U S A, 2003). Furthermore,
SO can be induced in cortical slices in vitro (Sanchez-Vives and McCormick, Nat
Neurosci, 2000).
SO consist of the alternation of two states in the membrane potential of the
neurons: a depolarized state, during which the membrane potential is closer to the firing
threshold and neurons generate action potentials (also called active or up-state), and a
hyperpolarized state, during which the membrane potential is more negative with
virtually no firing (passive or down-state) (Steriade and Timofeev, Neuron, 2003). The
cortical local field potential recorded in cats with a laminar extracellular electrode is
positive superficially during up-states that turns into negativity in deep layers. During
down-states a superficial negativity and deep positivity can be detected. During upstates, fast oscillations emerge, while during down-states they disappear.
In human SWS, the alpha and beta power is increased during the surface positive
half-wave (up-state) compared to the surface negative half-wave (down-state),
suggesting that their basic neurophysiology may be similar to animal findings (Molle et
al., J Neurosci, 2002), (Massimini et al., J Neurosci, 2004). While the SO in animals is
limited to below 1 Hz (Steriade et al., J Neurosci, 1993a), the recent AASM guidelines
suggest the 0.5-2 Hz range for slow wave activity (SWA) in humans (Iber et al., 2007).
The fine scale laminar structure of neuronal activity was analyzed in ferret cortical
slice preparations, revealing that firing during the up-state is the earliest in the
infragranular layers and spread towards the superficial layers with a long ~100ms interlaminar delay (Sanchez-Vives and McCormick, Nat Neurosci, 2000). In intact animals
the up-state onset related initial firing, intracellular membrane potential and LFP
changes could be detected in any layer in a probabilistic manner, with a short interlaminar delay (~10ms), however, on average, the earliest activity was found in the
infragranular layers (Sakata and Harris, Neuron, 2009), (Chauvette et al., Cereb Cortex,
2010).
Current source density (CSD) analysis of the low frequency (<1 Hz) components
of the anesthesia induced SO in cats (Steriade and Amzica, Proc Natl Acad Sci U S A,
1996), revealed sink in the middle layers during the up-state and source during the
30
Laminar analysis of the slow wave actvity in humans
down-state. This indicates current flow during up-states into the cell, probably in the
somatic region (positively charged ions traveling through the membrane into the cell or
negative ions to the outside), and current flow out of the cell during down-states
(positive ions flowing out to the extracellular space or negative ions flowing into the
cell). The most prominent up-state related sinks were localized in the middle and
deepest cortical layers (most probably layer III-VI). In contrast, the fast (30-40 Hz)
components were more distributed, composed of “alternating microsinks and
microsources” along the whole cortical depth (Steriade and Amzica, Proc Natl Acad Sci
U S A, 1996). During spontaneous and evoked K-complexes, a massive up-state related
sink was reported in layers II-III besides weaker ones in the deeper layers (Amzica and
Steriade, Neuroscience, 1998). In still another cat study, the maximal sink during the
up-states in natural sleep was located in the middle and deep layers (Chauvette et al.,
Cereb Cortex, 2010). In the rat primary auditory cortex, the laminar distribution of the
major up-state related sink was variable (Sakata and Harris, Neuron, 2009). On average
across animals, it was located in middle and deep layers (most probably layer III-V) in
natural sleep, whereas it was located in superficial layers (most probably layer II-III)
under urethane anesthesia (Sakata and Harris, Neuron, 2009). Although the cellular and
synaptic/trans-membrane mechanisms of slow waves during natural sleep are thus under
intense investigation in animals, these mechanisms have not previously been studied in
humans.
Up-states are proposed to be micro-wake ‘fragments’, similar to the activated state
of waking (Destexhe et al., Trends Neurosci, 2007), while down-states are considered
silent states filtering stimuli to the cortex.
SLOW OSCILLATION GENERATION
Slow oscillation can be detected in several brain structures:
•
all neocortical areas (in all types of neurons and glia cells),
•
thalamus (in thalamocortical and reticular neurons as well),
31
Csercsa Richárd
•
hippocampus (Moroni et al., PLoS One, 2007),
•
basal ganglia (Tseng et al., J Neurosci, 2001),
•
cerebellum (Ros et al., J Neurosci, 2009).
Despite of the above findings, SO is traditionally considered an oscillation
generated by neocortical networks. The main arguments for its neocortical origin are
that it is present in the neocortex after thalamectomy (Steriade et al., J Neurosci,
1993b), but absent in the thalamus after decortication (Timofeev and Steriade, J
Neurophysiol, 1996), and the disconnection of intracortical synaptic linkages resulting
in the disruption of its long-range synchronization (Amzica and Steriade, J Neurosci,
1995).
Recently, the pure neocortical origin of SO has been disputed. Crunelli and
Hughes pointed out the crucial role of thalamus in the generation of SO (Crunelli and
Hughes, Nat Neurosci, 2010). They argue that SO in the isolated neocortex was not
identical to those present when all cortical connections were intact. In fact, SO-like
activity can be detected in thalamic slices if the metabotropic glutamate receptors
(mGluR) of thalamocortical or reticular neurons were activated. They suggest that SO
can be generated in three structures:
•
cortical circuits (due to synaptic network properties),
•
thalamocortical neurons (intrinsically, depending on glutamatergic
input from the cortex),
•
reticular neurons (intrinsically, depending on glutamatergic input from
the cortex),
and the interplay between these oscillators is necessary for the full manifestation of the
slow oscillation. Furthermore, they insist that high-frequency (150-300 Hz) bursts of
thalamocortical neurons mediated by a low-threshold Ca2+ potential consequently
precede the firing of cortical cells and the depth-negative peak of the LFP in the cortex,
pointing out that cortical up-state generation may be triggered by thalamic input.
32
Laminar analysis of the slow wave actvity in humans
The activation of persistent sodium current that is essential for up-state initiation
(Mao et al., Neuron, 2001) may be achieved by thalamic input, spontaneously occurring
coincidence of mini EPSPs, or the action of some neurons that have a slightly lower
spiking threshold. But subthreshold membrane potential fluctuations clearly precede
neuronal firing at up-state onset, thus, firing may be the consequence rather than the
cause of up-state initiation (Chauvette et al., Cereb Cortex, 2010).
Mechanisms generating and maintaining down-states are also debated. Input
resistance was found to be the highest during the down-states and lowest during the
early part of up-states, increasing until its end (Destexhe et al., Nat Rev Neurosci,
2003). Furthermore, during down-states, inhibitory neurons are silent, too. Thus, instead
of an active inhibition, the hyperpolarization is thought to be the result of disfacilitation
i.e. the lack of firing of input neurons (Contreras et al., J Physiol, 1996), or the buildup
of activity-dependent potassium conductances (Sanchez-Vives and McCormick, Nat
Neurosci, 2000) or both (Hill and Tononi, J Neurophysiol, 2005). K+ currents are
believed to be the main factor in generating down-states since the application of Cs+,
blocker of K+ currents, reduces hyperpolarization (Timofeev et al., Proc Natl Acad Sci
U S A, 2001). However, it has been shown that GABAB has an important role in
terminating up-states (Mann et al., J Neurosci, 2009). Indeed, the observation that the
onset of down-states is synchronized more precisely than the onset of up-states suggests
a long-range synaptic mechanism (Volgushev et al., J Neurosci, 2006).
33
Csercsa Richárd
PROPAGATING WAVE AND SOURCE LOCALIZATION
Unlike oscillations during waking or REM sleep, slow oscillations during slow
wave sleep are correlated over larger cortical areas (7 mm distance in cats) (Destexhe et
al., J Neurosci, 1999). Slow oscillation was described as a traveling wave in rats
(Luczak et al., Proc Natl Acad Sci U S A, 2007) and humans (Massimini et al., J
Neurosci, 2004). It was shown that in humans it can start on any recording site of the
scalp EEG and travel in arbitrary direction, though the most frequent site of origin is the
frontal region and the most probable direction of propagation is anteroposterior (Figure
10). Using high-density EEG recordings (256 electrodes) and source localization
methods, slow waves in general were found to travel through definite cortical structures,
despite the unique localization of the individual waves (Murphy et al., Proc Natl Acad
Sci U S A, 2009). Waves with multiple peaks may have multiple origins (Riedner et al.,
Sleep, 2007). The hot spots of wave initiation are the lateral sulcus (including insular,
temporal, frontal, and parietal cortices) and the medial cingulate gyrus. The main areas
involved in the propagation include the anterior cingulate, the cingulate, the posterior
cingulate, the precuneus, and the left inferior frontal gyrus and insula. Although most of
the waves travel in the anterioposterior direction, propagation in the opposite direction
also occurs (Murphy et al., Proc Natl Acad Sci U S A, 2009). The same structures were
found to be activated by simultaneous EEG and fMRI recordings, with the addition of
the pontine tegmentum, the cerebellum, and the parahippocampal gyrus (Dang-Vu et
al., Proc Natl Acad Sci U S A, 2008).
Cerebral structures active during SWS have been associated with the so called
default-mode network (medial prefrontal cortex, anterior cingulate, posterior cingulate,
precuneus, left inferior parietal cortex) (Raichle et al., Proc Natl Acad Sci U S A, 2001),
(Greicius et al., Proc Natl Acad Sci U S A, 2003), (Raichle and Snyder, Neuroimage,
2007). The default system is a network of functionally connected brain structures that
show decreased activity in PET and fMRI images during goal-oriented cognitive tasks
compared to resting. The energy requirement of the brain’s responses to external stimuli
is only a small fraction of the amount required for ongoing activity. It is not probable
that the human body can afford expending this vast amount of energy (20% of the
34
Laminar analysis of the slow wave actvity in humans
body’s total energy consumption for an organ weighing 2% of total body weight) on
idle behavior. The purpose of this expenditure may be maintaining balance of
excitability, generating predictions about the future or spontaneous cognition (Raichle,
Science, 2006). Tasks activating regions in the default network include internal
mentation, theory of mind, or self-referential decision making. Recent experiments
suggest that the default network has two subsystems with distinct functions: a medial
temporal lobe subsystem that shows increased activity when participants made decisions
based on episodic memories, and a dorsal medial prefrontal cortex subsystem
preferentially active when they considered the mental states of themselves or others
(Andrews-Hanna et al., Neuron, 2010).
Figure 10. Slow oscillation as a traveling wave. Upper traces show superimposed recordings
from EEG channels during the same time period. Below, propagation of individual slow
waves. Red asterisk signals the initiation of the wave and blue streamlines indicate the path of
propagation. From (Massimini et al., J Neurosci, 2004)
The significance of the overlap between the structures involved in slow oscillation
generation and propagation and the structures assigned to the default network is not yet
known. The similarity between introspection during wakefulness and the brain’s
dissociation from external stimuli during sleep may explain this overlap. But activation
of similar regions does not imply that similar neural processes take place. Furthermore,
spontaneous fluctuations in the BOLD signal have a frequency of < 0.1 Hz, while slow
oscillations are < 1 Hz, which is a substantial difference. Though, BOLD fluctuations
35
Csercsa Richárd
may reflect infraslow EEG oscillations that modulate excitability and group slow
oscillations (Vanhatalo et al., Proc Natl Acad Sci U S A, 2004).
STIMULATION EVOKED SLOW OSCILLATION
Slow oscillations occur spontaneously during anesthesia and deep natural sleep
and they can be evoked with external stimulation, too. Solitary SO-like events, called Kcomplexes can be evoked during sleep by sensory stimuli, and though the mechanisms
underlying their generation are likely similar to those of the hyperpolarized phase of SO
(Cash et al., Science, 2009), they are isolated phenomena not constituting an ongoing
oscillation that may distinguish them from the fully developed slow oscillation (see in
section Slow waves and K-complexes). It has been shown that sound stimuli can entrain
the slow oscillation in the thalamus (He, J Neurosci, 2003), (Gao et al., J Neurosci,
2009), furthermore, SO can be induced by transcranial magnetic stimulation (TMS) in
humans (Massimini et al., Proc Natl Acad Sci U S A, 2007), and by electrical
stimulation in rats (Vyazovskiy et al., J Neurophysiol, 2009), (Ozen et al., J Neurosci,
2010) and humans (Entz et al., Society for Neuroscience, 2009).
Massimini et al. showed that slow waves triggered by transcranial magnetic
stimulation resemble spontaneous waves in their morphology and propagation
properties (Massimini et al., Proc Natl Acad Sci U S A, 2007). Every TMS pulse
evoked a slow wave that started under the stimulator and spread over the scalp.
Interestingly, while slow waves could be evoked in any phase of NREM sleep,
responses during waking had a low-amplitude, complex wave shape. Sleep TMS pulses
triggered a stereotyped down-state, either local (indicating the breakdown of
connectivity during sleep) or global (an aspecific response spreading away from the
stimulation site). On the other hand, awake responses were long-range, spatially, and
temporally differentiated. These findings suggest that the sleeping brain, despite being
active and reactive, loses its ability of entering states that are both integrated and
differentiated. Both features may be important to account for the fading of
36
Laminar analysis of the slow wave actvity in humans
consciousness during the early phase of sleep (Massimini et al., Proc Natl Acad Sci U S
A, 2007).
Entz et al. showed that slow waves can be evoked in humans by cortical electrical
stimulation and characterized the laminar profile of evoked waves (Entz et al., Society
for Neuroscience, 2009). They showed that the local field potential, multiunit activity,
spectrogram, and current-source density attributes of intracortical responses are similar
to those measured during natural SO. Responses triggered in different vigilance states
(awake, sleeping, and anesthetized) differed only in the amplitude and latency of evoked
potential components derived from subdural electrodes (Entz et al., FENS Forum,
2008).
MODELING OF SLOW OSCILLATION
Computational models and computer simulations are useful complementary
approaches to traditional experimental techniques in understanding information
processing in neural systems (Dayan and Abbott, 2001). Several models have been
developed in order to understand the cellular and network properties of the sleep slow
oscillation.
The model by Bazhenov et al. (Bazhenov et al., J Neurosci, 2002) includes
Hodgkin-Huxley-type principal cells and interneurons from the thalamus and the cortex
(100 pyramidal cells, 25 interneurons, 50 thalamocortical, 50 reticular cells) (Figure
11). They found that down-up transitions are mainly caused by mini EPSPs and downup transitions by Ca-dependent K current, synaptic depression, and inhibition from
interneurons.
37
Csercsa Richárd
Figure 11. Thalamocortical model by Bazhenov et al. PY: pyramidal cells, IN: interneurons,
RE: reticular cells, TC: thalamocortical neurons. Lines indicate connections, filled circle:
excitatory, empty circle: inhibitory. From (Bazhenov et al., J Neurosci, 2002).
The network of Compte et al. (Compte et al., J Neurophysiol, 2003) consists of
1024 pyramidal neurons and 256 interneurons placed equidistantly on a line of ~5 mm
(Figure 12). Pyramidal cells are two-compartment, interneurons single compartment
Hogkin-Huxley models. They analyzed the propagation of evoked slow waves and
found that the firing of inhibitory neurons precedes the firing of pyramidal cells. The
membrane resistance is the lowest at the onset of the up-state and increases until the end
of the down-state. The inhibitory and excitatory synaptic conductances change over
time, but their ratio remains constant, around 4:1. They theorized that spontaneous
firing causes up-states, and the accumulating Na-dependent K current causes downstates. Wave propagation in the model was similar to propagation observed in
experiments only if weak, long-range, ‘patchy’ connections were established. There is
example for this type of connectivity in the neocortex, especially in the supragranular
layers (Chalupa and Werner, 2001).
38
Laminar analysis of the slow wave actvity in humans
Figure 12. Model of Compte et al. Empty circles: pyramidal cells, filled circles: interneurons.
Triangle: excitatory, circle: inhibitory connections. Distribution of inhibitory connections is
narrower than distribution of excitatory connections. From (Compte et al., J Neurophysiol,
2003).
Hill et al. constructed a large-scale model of the thalamocortical system (Hill and
Tononi, J Neurophysiol, 2005) with three-layered cortex (supragranular, granular,
infragranular) and two-segment thalamus (reticular nucleus and dorsal thalamus). The
model in total consists of about 53,000 cortical and 10,000 thalamic single compartment
Hodgkin-Huxley neurons (Figure 13). They found that up-state generation depends on
the activation of persistent Na current that can be produced via either mini EPSPs,
synaptic input, or a hyperpolarization-activated cation current. Up-states can be
terminated
by depolarization-dependent
K
current
and
synaptic
depression.
Furthermore, network connections are crucial in synchronizing the oscillation. The
important role of horizontal connections in synchronization is in accordance with the
finding that the frequency of the slow oscillation depends on the size of the
corticocortical network (Timofeev et al., Cereb Cortex, 2000).
39
Csercsa Richárd
Figure 13. Large-scale thalamocortical model of Hill et al. Vp: three-layered primary visual
cortex, Vs: three-layered secondary visual cortex, R: reticular nucleus, T: dorsal thalamus.
White circles indicate exciatatory, black circles inhibitory cells. From (Hill and Tononi, J
Neurophysiol, 2005).
MEMORY CONSOLIDATION DURING SWS
Oscillations during SWS like SO, spindles, sharp waves, and ripples may have a
critical role in memory consolidation. Several studies show correlation between the
amount of slow wave sleep and learning performance. Cortical areas activated during
learning exhibit larger slow wave activity during SWS and the increase in task
performance was correlated with the amount of slow wave activity (Huber et al.,
Nature, 2004), (Tucker and Fishbein, J Sleep Res, 2009).
40
Laminar analysis of the slow wave actvity in humans
Presenting odors while learning hippocampus-dependent declarative memories
and re-exposure to them during subsequent slow wave sleep improved memory
retention (Rasch et al., Science, 2007). In contrast, re-exposure during REM sleep or
presenting odors during SWS without prior association did not result in memory
improvement. In case of hippocampus-independent procedural memories, cuing during
either SWS or REM sleep did not benefit memory consolidation.
Not only spontaneous but induced slow oscillations may benefit memory. The
transcranial application of potentials oscillating at 0.75 Hz enhanced the retention of
declarative memories in humans (Marshall et al., Nature, 2006).
Another recent concept that gained wide acceptance is the synaptic downscaling
model by Tononi and Cirelli (Tononi and Cirelli, Sleep Med Rev, 2006). According to
this theory, the global downscaling of synaptic strengths during slow oscillations
improves signal to noise ratio and thus enhances important memory traces while
eliminating the weaker ones. It also provides a baseline state of synaptic strengths after
sleep, preparing the brain for the acquisition of new memories.
Re-activation theory assumes memory consolidation is an active process, while
synaptic downscaling results in consolidation passively. Though these two theories can
be combined (Diekelmann and Born, Nat Rev Neurosci, 2010). Re-activation might
select memories worth keeping and transfer them to long-term storage, while
downscaling creates an environment necessary for the acquisition of new memories.
MEMORY TRANSFER (SYSTEM CONSOLIDATION)
Following the online mode of acquisition, an offline mode of operation is
necessary for memory consolidation to avoid interference between new and existing
memories. In the offline mode, memories are placed in context and transferred to the
long-term store.
This concept is corroborated by the observations that after lesion of the
hippocampus, memories acquired just before the lesion were forgotten, while memories
41
Csercsa Richárd
learned earlier could be recalled (Zola-Morgan and Squire, Science, 1990). It suggests
that the retrieval of hippocampus-dependent memories gradually became hippocampusindependent over time.
Neuroimaging studies showed that recall of new declarative memories depended
on hippocampus, but over time BOLD activity during recall exponentially decreased in
hippocampus and increased in the medial prefrontal cortex (Takashima et al., Proc Natl
Acad Sci U S A, 2006). The expression of immediate early genes required for synaptic
plasticity increased significantly 30 days after memory acquisition in prefrontal and
anterior cingulate cortices in mice, compared to control subjects, while their expression
in the hippocampus showed a decrease (Maviel et al., Science, 2004).
These results point to the hippocampus as short-term storage, that is active in the
early stages of encoding and recall, like a librarian that searches for the right
information, and neocortex as long-term storage like the library itself that stores the
books (Buzsáki, 2006).
Theoretical models also support the idea of memory transfer from hippocampus to
neocortex (Kali and Dayan, Nat Neurosci, 2004). This model also predicts that retrieval
of episodic memories should become impaired without hippocampus-initiated replay,
due to plasticity in the neocortex. Semantic knowledge, on the other hand, would not be
disrupted by neocortical plasticity, since it remains consistent with the observed
statistical structure of the world. This is in contrast with episodes, whose instability
arises because they correspond to extremely peaked probability distributions that are not
statistically representative.
RE-ACTIVATION
Re-activation of memory traces is necessary for synaptic consolidation, since the
co-activation of neurons strengthens the synapses between them (‘fire together – wire
together’, Hebbian-rule). It is also required for system consolidation, since for the
transfer of memories they need to be re-activated again.
42
Laminar analysis of the slow wave actvity in humans
An experiment in rats supporting re-activation showed that hippocampal cell pairs
whose activity was correlated when the animal was awake will be correlated during
subsequent sleep (Wilson and McNaughton, Science, 1994) (Figure 14). The same was
found in macaque neocortex, within and across all examined brain regions (sensory,
motor, association areas), but not the dorsal prefrontal cortex (Hoffman and
McNaughton, Science, 2002).
Figure 14. Replay of cortical activity associated with learning. Each dot represents a cell
and lines indicate correlation between cell pairs. On the left (PRE) shows correlations
during sleep before learning task, in the middle (RUN) during task, and on the right
(POST) during sleep after learning. Cell pairs active during learning task are active
during post-learning sleep. From (Wilson and McNaughton, Science, 1994)
Not only the firing of cell pairs, but spike sequences of neuron assemblies are
repeated during sleep, too (Nadasdy et al., J Neurosci, 1999). Cell assembly sequences
are thought to represent episodic memories with a temporal order and support planning
actions and goals (Pastalkova et al., Science, 2008).
Beyond correlation, the replay of the temporal order of cell ensemble firing
sequences has been shown in rat hippocampus (Lee and Wilson, Neuron, 2002) and
neocortex (Ji and Wilson, Nat Neurosci, 2007). Neuron ensembles repeated the firing
sequence of awake running experience during subsequent slow wave sleep frames (upstates) (Ji and Wilson, Nat Neurosci, 2007) (Figure 15). About 8 percent of all
examined sleep frames contained firing sequences similar to prior awake frames, which
is significantly above chance level. Furthermore, cortical and hippocampal replaying
43
Csercsa Richárd
frames occurring together could be observed. Their number was also above chance
level. Such replays possibly occur more frequently, but the number of electrodes, hence
the number of recordable cells is limited. The repeated pattern of activation during sleep
is compressed in time by a factor of 20 in the hippocampus (Lee and Wilson, Neuron,
2002) and of 6 to 7 in the neocortex (Euston et al., Science, 2007). Fast replay may
reflect the speed of the brain’s intrinsic dynamics (e.g., conduction speeds, synaptic
delays, etc.) when not constrained by behavioral events.
Figure 15. Re-activation of cell assemblies during sleep. The temporal order of cell firing was
preserved from learning task (RUN) to post-learning sleep (POST), in the cortex (left, CTX)
and in the hippocampus (right, HP). From (Ji and Wilson, Nat Neurosci, 2007)
HIPPOCAMPUS-NEOCORTEX INTERACTION
The transfer of memory traces from hippocampus to neocortex for long-term
storage requires functional connection between the two structures during SWS. Siapas
and Wilson found a temporal correlation between hippocampal ripples and neocortical
spindles (Siapas and Wilson, Neuron, 1998), but the direction of information flow was
not clarified. Buzsáki suggested that information flows from cortex to hippocampus
during wakefulness and from hippocampus to cortex during sleep (Buzsaki, J Sleep Res,
1998). Several studies indicate that neocortex initiates the communication during SWS.
Ji and Wilson showed that neocortical MUA precede hippocampal neuronal population
activity by about 40 ms (Ji and Wilson, Nat Neurosci, 2007). Mölle et al. came to a
similar conclusion, finding that prefrontal MUA precedes hippocampal sharp waves by
about 30 ms (Molle et al., J Neurophysiol, 2006). In vivo whole-cell recordings from
44
Laminar analysis of the slow wave actvity in humans
hippocampal CA1 interneurons and parietal cortex neurons in urethane-anesthetized
mice showed a consistent lag of hippocampal activity with a latency of 160 ± 50 ms
(Hahn et al., Nat Neurosci, 2006). Wierzynski et al., on the other hand, found that
prefrontal cells fire in a 100 ms window following hippocampal cell firing (Wierzynski
et al., Neuron, 2009), which might be the response of the hippocampus to the call of the
neocortex. All these studies were conducted in rats. Human intracranial EEG
examinations revealed a bidirectional coupling, with the influence of neocortex on
hippocampus significantly increased during sleep compared to waking (Wagner et al.,
Cortex, 2010). These results altogether indicate that the two structures are synchronized
by neocortical slow oscillations, during which memory traces are transferred.
The exact mechanisms underlying the memory transfer are unclear, but special
oscillations seem to be involved. High-frequency sharp wave-ripple complexes in the
hippocampus (Chrobak and Buzsaki, J Neurosci, 1996), present during waking and
SWS and thalamocortical spindles during NREM sleep are thought to play a critical
role. A temporal correlation between hippocampal ripples and cortical spindles was
observed (Siapas and Wilson, Neuron, 1998), indicating that a link is established
between the two structures during these episodes. It was also shown that neocortical
slow oscillation groups spindles (Contreras and Steriade, J Neurosci, 1995), (Molle et
al., J Neurosci, 2002) and sharp waves (Molle et al., J Neurophysiol, 2006), (Clemens et
al., Brain, 2007) in animals and humans alike. Furthermore, neocortical replay of neural
patterns that appeared during learning, preferentially occur during hippocampal sharpwaves (Peyrache et al., Nat Neurosci, 2009).
The importance of these oscillations in memory consolidation is indicated by
several findings: the density of spindles is increased after learning (Gais et al., J
Neurosci, 2002), memory retention performance is correlated with the number of
spindles (Clemens et al., Neuroscience, 2005), and on the molecular level, spindle
stimulation patterns consisting of spike trains extracted from a slow oscillation up-state
induced LTP (Rosanova and Ulrich, J Neurosci, 2005).
Hippocampal ripples are correlated with memory performance (Axmacher et al.,
Brain, 2008) and their elimination results in memory impairment (Girardeau et al., Nat
Neurosci, 2009). They occur most frequently during waking (Axmacher et al., Brain,
45
Csercsa Richárd
2008), which corroborates the idea that events experienced during waking replay during
sleep.
DOWNSCALING
According to the model of Tononi and Cirelli (Tononi and Cirelli, Sleep Med Rev,
2006), the processes of synaptic downscaling are: (1) synapses are potentiated due to
learning during waking, (2) slow wave activity increases due to potentiated synapses,
(3) synapses are depressed due to slow wave activity, (4) slow wave activity decreases
as synaptic strengths return to a ‘baseline’ due to depression, leading to the beneficial
effects of sleep. They hypothesize that downscaling depresses all synapses in the region
proportionally, thus eliminating weaker ones therefore improving the signal-to-noise
ratio of others (Figure 16).
Figure 16. Synaptic downscaling. Synapses are strengthened due to experiences during
waking, and downscaled during sleep. Downscaling causes weak synapses to disappear and
improves signal-to-noise ratio. From (Diekelmann and Born, Nat Rev Neurosci, 2010)
There are experiments supporting processes (1), (2) (Kattler et al., J Sleep Res,
1994), (Vyazovskiy et al., J Sleep Res, 2000), (Huber et al., Nature, 2004), and (4)
(Huber et al., Nat Neurosci, 2006), though it is not easy to prove point (3) directly. It
has been shown that protein levels of key components of synapses are high after waking
and low after sleep in the rat (Vyazovskiy et al., Nat Neurosci, 2008), (Faraguna et al., J
Neurosci, 2008) and in Drosophila (Gilestro et al., Science, 2009), independent of the
46
Laminar analysis of the slow wave actvity in humans
time of the day. Furthermore, the number of synaptic terminals is reduced after sleep
(Donlea et al., Science, 2009). In addition, mini EPSCs are increased after waking and
decreased after sleep (Liu et al., J Neurosci, 2010), also indicating synaptic depression
during SWS, supported by a computational model too (Olcese et al., J Neurophysiol,
2010).
SLOW WAVES AND K-COMPLEXES
K-complexes are large graphoelements of the sleep EEG, occurring mostly in
stage 2 of NREM sleep. They are isolated events consisting of two or three phases: first,
a short surface-positive transient, then a slow and large surface-negative complex with
peaks at 350 and 550 ms (also called N550), which may be followed by a positivity at
900 ms. K-complexes are sometimes followed by 10-14 Hz spindles. They can be
evoked by stimuli of multiple modalities or occur spontaneously (maybe due to internal
stimuli, like changes in respiration, limb movement, gastric noises, or because of
spontaneous neuronal activity). (Loomis et al., J Neurophysiol, 1938), (Halasz, Sleep
Med Rev, 2005), (Colrain, Sleep, 2005)
K-complexes have been suggested to represent either arousal response or sleep
protecting mechanisms. In the first case, K-complexes are hypothesized to be the
cortical responses to potentially arousing stimuli, accompanied by peripheral
sympathetic activation, and elevation in heart rate and blood pressure. On the other
hand, Amzica and Steriade suggested that they are the expressions of the sleep slow
oscillation (Amzica and Steriade, Neuroscience, 1998). Recent experiments revealing
the laminar profile of K-complexes in humans (Cash et al., Science, 2009) support this
idea. During the main component (i.e. the slow and prominent surface-negative wave)
there was an active source in layer III and a passive sink close to the surface. The
multiunit activity in all cortical layers decreased, which, together with the CSD results,
indicate a hyperpolarizing current in layer III. Spectral power in the gamma frequency
range (20-100 Hz) was also decreased during the surface negativity. These findings
altogether largely resemble the laminar profile of the hyperpolarized phase (down-state)
47
Csercsa Richárd
of the slow oscillation. These findings support the sleep protecting function of the Kcomplex, because it suppresses neuronal activity and arousal to sensory stimuli (Cash et
al., Science, 2009).
48
Laminar analysis of the slow wave actvity in humans
OBJECTIVES
Sleep is a general behavioral state in the animal kingdom, and also a unique
mode of brain operation. Brain waves during sleep are markedly different from awake
oscillations, as visible on scalp EEG recordings. Unfortunately, until recently, EEG was
the only way of studying brain activity in humans. While in animal models,
microelectrodes with multiple contacts could be inserted into the cortex, in vivo
electrophysiological experiments were restricted to scalp recordings in humans.
Slow wave sleep is the deepest stage of sleep, named after the high-amplitude,
low-frequency waves that characterize it on scalp derivations in humans and
intracortical recordings in animals, called slow oscillations. This stage is thought to
underlie essential restorative processes and facilitate memory consolidation. Thus,
understanding the mechanisms generating and maintaining slow oscillations is of great
importance in sleep research as well as in other areas of neuroscience.
Our objectives were to derive local field potentials and action potentials from
within the cortex of humans during slow wave sleep, and give a description of slow
wave activity in humans on the level of cortical cells and layers.
Animal models show increased firing, increased wide-band spectral power and
inward transmembrane currents during surface positive LFP (up-state), while decreased
firing, decreased spectral power and outward currents during surface negative waves
(down-state). Current-source density analysis localized the most prominent up-state
related sinks to middle or deep cortical layers in animals and showed deep layers to
initiate firing at the onset of an up-state.
Since the human prefrontal cortex is more than a hundred times larger than the
prefrontal cortex of cats and rats, with a striking increase in pyramidal cell dendritic
complexity, some differences were expected in the cellular and laminar properties of
human deep sleep compared to animals. We implanted laminar multielectrodes
chronically in the frontal and parietal cortical areas of epileptic patients in order to
examine the electrophysiological properties of deep sleep in humans.
49
Csercsa Richárd
METHODS
PATIENTS AND ELECTRODES
Five patients with intractable epilepsy underwent chronic clinical subdural grid
and strip electrode implantation as a standard procedure for localization of the seizure
focus and eloquent areas. Fully informed consent was obtained from each subject under
the auspices of the Hungarian Medical Scientific Council and local ethical committee;
National Institute of Neuroscience, Budapest, Hungary according to the World Medical
Association
Declaration
of
Helsinki.
Conventional
clinical
subdural
electrocorticography (ECoG) electrode strips and grids were implanted to cover the
frontal, temporal and parietal gyri. In addition to the surface electrodes, a 350 µm
diameter, 24 contact experimental laminar multichannel microelectrode array (ME) was
implanted perpendicular to the cortical surface, underneath the clinical grids (Ulbert et
al., J Neurosci Methods, 2001), (Ulbert et al., Hum Brain Mapp, 2001), (Ulbert et al.,
Epilepsia, 2004), (Cash et al., Science, 2009), (Keller et al., J Neurosci Methods, 2009).
The 40 µm diameter Pt/Ir contacts were spaced evenly at 150 µm providing LFP
recordings from a vertical, 3.5mm long cortical track, spanning from layer I to layer VI.
A silicone sheet attached to the top of the ME shank prevented the first contact from
sliding more than 100 µm below the pial surface (Ulbert et al., J Neurosci Methods,
2001). In each case, the explanted ME was visually inspected under a microscope for
structural damage, and we did not find any alteration, indicating intact structure
throughout the recordings. The location and duration of the clinical electrode
implantation were determined entirely by clinical considerations, the ME was placed in
cortex that was likely to be removed at the definitive surgery.
HISTOLOGY
The positions of the electrodes were confirmed by intraoperative navigation, colocalization of intraoperative photographs, pre- and postoperative MRI scans and 3D
MRI reconstructions. Photographs were also taken during the resective surgery to
50
Laminar analysis of the slow wave actvity in humans
confirm that the surface electrodes did not shift during monitoring. The brain tissue
containing the electrode track in Patients 4 and 5 was removed en-bloc for further
anatomical analysis (Ulbert et al., Exp Neurol, 2004), (Fabo et al., Brain, 2008). It was
cut into 2-5 mm blocks and immersed into a fixative containing 4% paraformaldehyde,
0.1% glutaraldehyde and 0.2% picric acid in 0.1 M phosphate buffer (PB, pH 7.4). The
fixative was changed every hour to a fresh solution during constant agitation for 6
hours, and then the blocks were post-fixed in the same fixative overnight. Vibratome
sections (60 µm thick) were cut from the blocks, and photographs were taken from the
electrode tracks. Following washing in PB, sections were immersed in 30% sucrose for
1–2 days, and then frozen three times over liquid nitrogen. Endogenous peroxidase was
blocked by 1% H2O2 in PB for 10 min. Sections containing the electrode track were
processed for immunostaining against the neuron marker NeuN (Figure 17A), calretinin
(CR) (Figure 17B, reconstructed from camera lucida), SMI-32 (Figure 17C), glial
fibrillar acidic protein (GFAP) (Figure 17D) to stain every neuron, a subset of
interneurons, pyramidal cells and glia respectively. PB was used for all the washes
(3x3-10 min between each step) and dilution of the antisera. Non-specific
immunostaining was blocked by 5% milk powder and 2% bovine serum albumin.
Monoclonal mouse antibodies against NeuN (1:3000, Chemicon, Temecula, CA, USA),
SMI-32 (1:3000, Covance, Princeton, NJ, USA), GFAP (1:2000, Novocastra,
Newcastle, upon Tyne, UK) and CR (1:5000, SWANT, Bellinzona, Switzerland) were
used for 2 days at 4 °C. Specificity of the antibodies has been thoroughly tested by the
manufacturers. For visualization of immunopositive elements, biotinylated anti-mouse
immunoglobulin G (1:300, Vector) was applied as secondary antiserum followed by
avidin-biotinylated horseradish peroxidase complex (ABC; 1:300, Vector). The
immunoperoxidase
reaction
was
developed
by
3,3’-diaminobenzidine
tetrahydrochloride (DAB; Sigma), as a chromogen. Sections were then osmicated
(0.25% OsO4 in PB, 30 min) and dehydrated in ethanol (1% uranyl acetate was added at
the 70% ethanol stage for 30 min) and mounted in Durcupan (ACM, Fluka). Layers of
the neocortex were outlined using all of the above stains and a shrinkage correction
factor published earlier (Turner et al., J Comp Neurol, 1995), (Wittner et al., Eur J
Neurosci, 2006).
51
Csercsa Richárd
Figure 17. Electrode track histology, representative examples from Patient 4. (A) The ME
electrode track (black contour line) and inferred contact locations (black dots) are shown
relative to cortical layers (Roman numbers) revealed by the NeuN stain. Laminarization
appears to be intact. (B) Camera lucida reconstruction of CR+ cell bodies and processes next
to the electrode track. (C) Well preserved pyramidal (SMI-32 stain) and (C) glial cells (GFAP
stain) next to the electrode track.
Cell counting was performed in Patients 4 and 5 using camera lucida drawing
(Figure 17B) of CR immunopositive (CR+) cells (two sections per patient). The
normalized (between 0 and 1) CR+ cell density laminar depth profile (number of cells
over unit area of cortex) was calculated in each consecutive 150 µm wide and variable
length (1-3 mm) horizontal cortical stripes to match the depth structure of the
electrophysiology measurements.
RECORDINGS
After electrode placement, the patients were transferred to the intensive
monitoring unit for 5-7 days, where continuous 24 hour video-EEG observation took
place in order to localize the seizure focus. ECoG from clinical strip and grid electrodes
(32-92 channels, mastoid reference) was recorded concurrently with patient video using
the standard hospital system (band-pass: 0.1-200 Hz, acquisition rate: 400-5000 Hz / 16
52
Laminar analysis of the slow wave actvity in humans
bit). Video-EEG data for the duration of monitoring were stored on hard disks for later
analysis. Spatial LFP gradient (LFPg), the voltage difference between consecutive
laminar electrode contacts was provided by a special preamplifier placed inside the head
bandage of the patient (Ulbert et al., J Neurosci Methods, 2001). For simplicity,
throughout the text the spatial potential gradient is expressed in µV unit rather than the
formally correct µV per inter-contact distance (150 µm). This reference independent
measurement method was proven to be effective in minimizing the motion related and
electro-magnetic artefacts (Ulbert et al., J Neurosci Methods, 2001). The LFPg was split
to EEG range (0.1-300 Hz) and single (SUA), multiple unit activity (MUA) frequency
range (300-5000 Hz) by analogue band-pass filtering at the level of a custom made main
amplifier (Ulbert et al., J Neurosci Methods, 2001). EEG range signal was sampled at 2
kHz / 16 bit; MUA range was sampled at 20 kHz / 12 bit and stored on a hard drive.
Figure 18. Histology and representative continuous traces of all 23 channels of LFPg in SWS
from Patient 4. One minute long raw data without additional filtering on the right. Histology
slide from the vicinity of the electrode track on the left. Numbers from 1 to 23 on the slide
represent the LFPg channels from the surface to the depth of the cortex. Roman numerals
represent the cortical layers.
53
Csercsa Richárd
SLOW WAVE ACTIVITY DETECTION
We have analyzed the LFPg, MUA, SUA and ECoG data acquired from each
patient during one to three nocturnal recording sessions. Since the sleep of the patients
was fragmented due to medical care, distress from the hospitalization and head wound,
we cannot provide standard hypnograms that are usually obtained from healthy subjects
without a recent craniotomy. Craniotomies may also distort the scalp distribution of the
EEG due to lack of bone and excessive fluid accumulation below the scalp, furthermore
scalp electrodes if placed close to the frontal craniotomy wounds may induce infection,
this why we avoided placing more than two frontal scalp EEG electrodes. Partial sleep
staging was performed based on readings of the available scalp EEG and ECoG
electrodes by expert neurologists. In this study, we have analyzed electrophysiological
data obtained only from the deepest stage of NREM sleep (N3, or SWS) (Iber et al.,
2007). Behavioral sleep was confirmed by the video recording, while SWS was
electrographically identified in accordance with the recent AASM guidelines (Iber et
al., 2007). SWS periods were identified when 20% or more of an epoch consisted of
SWA (waves in the 0.5–2 Hz frequency range with peak-to-peak amplitude larger than
75 µV, measured over the frontal regions) (Iber et al., 2007). Data containing interictal
spikes (within 1 min) and seizures (within 60 min) were excluded from the study to
avoid epileptic contamination.
In addition to spectral (Figure 23A) and autocorrelation analyses (Figure 23B),
SWA cycle detection was based on phase and amplitude information, extracted from the
narrow-band filtered (0.3-3 Hz, 24 dB/octave, zero phase shift) layer II LFPg (Figure
19A) and ECoG (Figure 19B) data. Instantaneous phase of the filtered signal was
calculated by the Hilbert transformation. The discrete-time analytic signal was
computed with the help of Hilbert transform, then the instantaneous phase was
calculated as the arctangent of the quotient of the imaginary and real part of the analytic
signal. In our implementation, a single SWA cycle was defined between -180 ° and
+180 ° phase. The -180 ° phase value corresponded to the trough of the negative halfwave (down-state) preceding the 0 ° phase, which corresponded to the peak of the
positive half-wave (up-state) and finally the +180 ° phase corresponded to the following
54
Laminar analysis of the slow wave actvity in humans
negative half-wave trough (down-state). At each +180 ° crossing the phase was wrapped
-360 ° for better visualization. To avoid the detection of higher frequency (e.g.: theta)
oscillations, waves with shorter than 500 ms cycle lengths (corresponding to higher than
2 Hz frequency) were excluded from the analysis. Waves with non-monotonic phase
runs were also excluded, since phase inversions may also indicate higher frequency
contamination. In addition to phase constraints, valid SWA cycles had to fulfil the
following amplitude criteria: the up-state peak amplitude had to be more positive than
+50 µV and the preceding or following down-state trough amplitude had to be more
negative than -50 µV. The SWA detection algorithm parameters were tuned and
carefully validated by expert electroencephalographers. Similar algorithmic parameters
were used for all of the patients. To facilitate comparison of our results with previous
animal studies, the threshold level was set to +50 µV on the filtered (0.3-3 Hz, 24
dB/octave, zero phase shift) upper layer III LFPg, and the wave triggered (up-state
locked) averages were calculated on the un-filtered LFPg and MUA (Figure 24).
Figure 19. Similarity between LFPg recorded with ME within the cortex and ECoG recorded
from macrocontacts subdurally. (A) Upper: ‘raw’ traces of single sweeps containing SWA;
broadband (0.1-300 Hz) LFPg from layer II. Middle: ‘filtered’ traces after band pass (0.3-3
Hz) filtering. Lower: ‘phase’ traces showing the instantaneous phase of the ‘filtered’ trace
above derived by the Hilbert transform. Grey rectangles indicate automatically detected up-
55
Csercsa Richárd
states (surface positive half-waves). Patient 4. (B) Same as (A), but recorded from
neighbouring ECoG contacts. (C) Colour map of LFPg and (D) ECoG spectral power during
the SWA shown in (A) and (B). X-axis: time, Y-axis: frequency, Z-axis: colour coded relative
spectral power in dB, compared to the mean of the entire interval in each frequency band
(relative spectrogram).
To quantify and compare SWA parameters with other studies, the frequency of
SWA occurrence (detected valid cycles per minute), the interdetection interval
histogram (Figure 23C) and the cycle length histogram (Figure 23D) were calculated
(Massimini et al., J Neurosci, 2004). The single sweep time-frequency content of the
SWA signal (Figure 19C-D) was also computed using wavelet transforms (Delorme and
Makeig, J Neurosci Methods, 2004). In addition, we attempted to describe the laminar
distribution of the SWA in more detail using the LFPg FFT power spectrum depth
profile (Figure 27), the pairwise linear coherence between each LFPg trace in the SWA
(0.3-3 Hz) frequency range (Figure 28), and the depth profile of the LFPg
autocorrelation (Figure 29).
CURRENT SOURCE DENSITY ANALYSIS
CSD analysis identifies synaptic/trans-membrane generators of LFP in laminated
neural structures (Nicholson and Freeman, J Neurophysiol, 1975), (Freeman and
Nicholson, J Neurophysiol, 1975). According to Poisson’s equation,
߲ଶΦ
߲ଶΦ
߲ଶΦ
‫ܫ‬௠ = − ቆߪ௫ ∙
+ ߪ௬ ∙
+ ߪ௭ ∙
ቇ
߲‫ ݔ‬ଶ
߲‫ ݕ‬ଶ
߲‫ ݖ‬ଶ
the current source density equals the negative of the second spatial derivative of the
potential. Under the assumptions that cortical activity is constant in planes parallel to
the surface, and the extracellular conductivity is homogeneous, this equation reduces to
the second spatial derivative of the local field potentials in the z direction (perpendicular
to the surface), which we can record with the laminar multielectrode. The result will
closely approximate the macroscopic current density over unity cell membrane area.
56
Laminar analysis of the slow wave actvity in humans
Since LFPg (Figure 30A) is the first spatial derivative of LFP, one additional spatial
derivation resulted in the CSD (Figure 30C) for the EEG range (0.1-300 Hz) data.
Inhomogeneous conductivity and electrode spacing were not taken into account (both
were substituted by the dimensionless number 1 in the calculations); high spatial
frequency noise and boundary effects were reduced by Hamming-window smoothing
and interpolation (Ulbert et al., J Neurosci Methods, 2001) thus CSD was expressed in
µV units. It was shown previously that our recording and analysis techniques can
reliably detect CSD activity in each layer of the human cortex and hippocampus (Ulbert
et al., J Neurosci Methods, 2001), (Ulbert et al., Hum Brain Mapp, 2001), (Ulbert et al.,
Epilepsia, 2004), (Ulbert et al., Exp Neurol, 2004), (Wang et al., J Neurosci, 2005),
(Halgren et al., Neuroimage, 2006), (Knake et al., Neuroimage, 2007), (Fabo et al.,
Brain, 2008), (Cash et al., Science, 2009), (Steinvorth et al., Hippocampus, 2010),
(Wittner et al., Brain, 2009).
STATISTICAL ANALYSIS OF ELECTROPHYSIOLOGY AND HISTOLOGY
ANOVA with Tukey’s HSD (honestly significant difference) test were applied
to the normalized values (LFPg and CSD: between -1 and +1; MUA, gamma band LFPg
and CSD: between 0 and 1). Normalized values were grouped by layers (I-VI), and the
grand average (across all patients) of LFPg, MUA, CSD, gamma band (30-150 Hz)
LFPg and gamma band CSD power depth profile at the up-state peak were tested to
determine statistically significant differences (p<0.01) between electrophysiological
activations in different layers of the cortex (Figure 30E-H). Results are depicted on boxwhisker plots; small box: mean, big box: standard error (SE), whisker: standard
deviation (SD).
In Patients 4 and 5 (with available histology), the averaged, normalized CR+ cell
density (Figure 38A) and averaged, normalized electrophysiology depth profiles
(consecutive values at each cortical depth) were constructed. Average depth profiles of
LFPg (Figure 38B) and CSD (Figure 38C) at the peak of the up-state were normalized
between -1 and +1, while average depth profiles of MUA (Figure 38D) and spectral
measures (Figure 38E-F) (gamma band LFPg and CSD) at the peak of the up-state were
57
Csercsa Richárd
normalized between 0 and 1. CR+ cell density depth profile was normalized between 1
and 0. SE is marked by whisker in this case. The normalized cell density and
electrophysiology measures were compared using the Pearson r correlation method with
p<0.01 significance level criterion.
SINGLE AND MULTIPLE UNIT ACTIVITY ANALYSIS
A continuous estimate of population neuronal firing rate was calculated from the
MUA range (300-5000 Hz) data. The signal was further filtered (500-5000 Hz, zero
phase shift, 48 dB/octave), rectified and decimated at 2 kHz, applying a 0.5 ms sliding
average rectangular window, followed by a final, smoothing low-pass filter (20 Hz, 12
dB/octave) (Figure 30B). Putative single units (Figure 20, Figure 34B) were analyzed
by conventional threshold detection and clustering methods using Dataview and
Klustawin (Heitler, 2006) and custom made Matlab software. Putative single units from
Patients 1, 4 and 5 were included in the analyses, which were recorded stably for at least
600-1000 sec in SWS. In Patient 2, MUA was not recorded for technical reasons, while
Patient 3 showed no discriminable single units. After threshold recognition (mean ±3-5
SD) (Csicsvari et al., Neuron, 1998) at a given channel, three representative amplitude
values were assigned to each un-clustered spike waveform. These triplets constituted the
features by which spikes were categorized into different clusters. They were projected
into 3D space, and a CEM (competitive expectation-maximization) based algorithm
(Harris et al., J Neurophysiol, 2000) was used for cluster cutting (Heitler, 2006). If the
autocorrelogram of the resulting clusters contained spikes within the 2 ms refractory
interval, it was re-clustered. If re-clustering did not yield a clean refractory period, the
cell was regarded as multiple units and omitted from the single cell analysis. Given the
gradient recording, spikes at neighbouring traces appeared as mirror images, thus from
adjacent channels (150 µm apart) only one channel (the one that yielded the better
signal to noise ratio) was included in the analysis. To reveal double detection, the crosscorrelogram was constructed (Staba et al., Hippocampus, 2002) for next to adjacent
(300 µm apart) pairs of putative single cells. No coincident interactions (99%
confidence limit at 0 ms (Staba et al., Hippocampus, 2002)) were found. A spike train
58
Laminar analysis of the slow wave actvity in humans
was determined as a burst, if at least three consecutive spikes occurred within a
maximum 20 ms long interval, which was preceded and followed by at least 20 ms long
intervals with neuronal silence (Staba et al., J Neurosci, 2002).
Phase dependence of single cell firing rate (Figure 36) was computed for 30°
phase bins; the total number of firing in a given bin was divided by the total time that
the cortex spent in that phase bin (thus producing a phase histogram). The Rayleigh test
(p<0.01) was used to judge if the resulting circular distribution was significantly
different from the uniform distribution.
We have shown previously that our SUA, MUA recording and analysis
techniques can reliably detect task or epilepsy related modulation of neuronal firing
from each layer of the human cortex and hippocampus (Ulbert et al., J Neurosci
Methods, 2001), (Ulbert et al., Hum Brain Mapp, 2001), (Ulbert et al., Epilepsia, 2004),
(Ulbert et al., Exp Neurol, 2004), (Wang et al., J Neurosci, 2005), (Halgren et al.,
Neuroimage, 2006), (Fabo et al., Brain, 2008), (Cash et al., Science, 2009), (Wittner et
al., Brain, 2009).
Figure 20. Spike sorting. (A) Upper trace shows band pass filtered (500-5000 Hz, 24 dB/oct)
MUA from Pt. 1 with high signal to noise ratio. Besides the prominent single unit activity,
some less clearly differentiated action potentials are also evident. Lower left inset (B) shows
three well separated unit clusters, with single action potential waveforms (C) from Pt. 4.
59
Csercsa Richárd
RESULTS
Previous studies of SWA in humans (Achermann and Borbely, Neuroscience,
1997), (Massimini et al., J Neurosci, 2004), (Molle et al., Proc Natl Acad Sci U S A,
2004), (Massimini et al., Science, 2005), (Marshall et al., Nature, 2006), (Massimini et
al., Proc Natl Acad Sci U S A, 2007) have been limited to macroelectrode recordings
that superimpose activity from several cm2 of cortex. These recordings are ambiguous
as to the circuits involved, are not sensitive to neuronal firing, and do not distinguish
between excitatory and inhibitory mechanisms. We used laminar multichannel
microelectrode array recordings (Ulbert et al., J Neurosci Methods, 2001), (Ulbert et al.,
Hum Brain Mapp, 2001), (Ulbert et al., Epilepsia, 2004), (Ulbert et al., Exp Neurol,
2004), (Wang et al., J Neurosci, 2005), (Halgren et al., Neuroimage, 2006), (Knake et
al., Neuroimage, 2007), (Fabo et al., Brain, 2008), (Cash et al., Science, 2009),
(Steinvorth et al., Hippocampus, 2010), (Wittner et al., Brain, 2009) to estimate
neuronal firing and synaptic/trans-membrane currents in different cortical layers. Since
cortical neuronal populations and synaptic inputs are organized into distinct layers,
these recordings allowed us to resolve the cortical generators underlying SWA in
humans.
SLOW OSCILLATION IN HUMANS
I.
Human slow wave activity reflects a rhythmic oscillation in the
extracellular local field potential: the surface positive half-wave
corresponds to the depolarized up-state with increased cell
firing and the surface negative half-wave corresponds to the
hyperpolarized down-state with neuronal silence.
Clinical subdural strip and grid electrodes and MEs were implanted into the
frontal and parietal cortices of patients (n=5) with intractable epilepsy (Figure 21) in
60
Laminar analysis of the slow wave actvity in humans
order to identify the seizure focus and eloquent cortex prior to surgical therapy. The
focus was eventually localized to the frontal lobe in four patients, aand
nd to the temporal
lobe in one.
Figure 21. Grid, strip and multichanne
multichannell microelectrode array (ME) locations in all patients.
Locations are superimposed on 3D reconstructions of MRIs taken with the electrodes in place,
aided by intraoperative navigation and photographs. Grid and strip electrode contacts are
depicted in red and
d blue colors; the first grid contact is marked with G1. ME locations of
Patient 1 (blue): left Brodmann area (BA) 9, Patient 2 (red): right BA 2, Patient 3 (green): left
BA 46, Patient 4 (black): right BA 9, Patient 5 (orange): right BA 8 are marked with circles.
Grid and superficial ME recordings showed great similarity. LFPg recorded in layer II
closely resembled to thee locally recorded ECoG
ECoG, with Pearson r>0.9 (p<0.01)
<0.01) in all
patients (Figure 22). Time-locked
locked averages to the peak of the surface positive half-wave
half
(up-state)
state) showed similar LFPg and ECoG waveforms regardless if time locking was
based on the LFPg or ECoG ((Figure 27A
A red and green traces). Both LFPg and ECoG
showed broadband (10-200
200 Hz) spectral increases during up
up-states
states and decreases during
down-states (Figure 19C-D).
D).
61
Csercsa Richárd
Figure 22. Similarity between SWA recorded simultaneously within the cortex by
microcontacts (LFPg) and subdurally with macrocontacts (ECoG). ECoG and LFPg traces
during SWS in three patients (Patients 3, 4 and 5). Upper, colour SWA traces mark ECoG
from four grid channels (e.g.G14 etc…) surrounding the ME. Black SWA traces below show
layer II local field potential gradient (LFPg) from each ME. Overlay in the bottom panel
indicates good correspondence between ECoG and LFPg. Positive is depicted upwards on all
figures. Positive in potential gradient recordings indicates relative positivity in the more
superficial lead.
Histology of the ME penetration track was recovered in two patients. It showed
intact laminarization (Figure 17A), well preserved interneurons, pyramidal cells and glia
(Figure 17B-D), indicating no discernable epilepsy or implantation related damage of
the examined cortex. None of the patients had pre-operative pathological MRI findings
in the 1-2 cm vicinity of the ME implantation site. From the histology we could
determine the layers each channel was recording from (Figure 18).
Automatic SWA cycle detection was based on amplitude and phase information
using LFPg (Figure 19A) and ECoG (Figure 19B) recorded during SWS. On average,
20 SWA cycles (mean=20 1/min, range=12-26 1/min, SD=7 1/min) were detected per
minute (Figure 25, Figure 26). Cycle length peaked on average at 0.8 sec (mean=0.8
sec, range=0.6-1.4 sec, SD=0.3 sec) (Figure 23D). Interdetection interval (Figure 23B)
peaked on average at 1.1 sec (mean=1.1 sec, range=0.8-1.2 sec, SD=0.4 sec), all
comparable to healthy subjects (Massimini et al., J Neurosci, 2004). LFPg (Figure 23A,
C) and ECoG (not shown) FFT power spectrum and autocorrelation (Figure 23A-B,
Figure 27G-H) also corresponded well to previous human (Achermann and Borbely,
Neuroscience, 1997) and animal (Isomura et al., Neuron, 2006) findings, indicating
correct SWA cycle identification and relatively normal SWA production.
62
Laminar analysis of the slow wave actvity in humans
Figure 23. Spectral and temporal properties of SWA cycles. (A) Representative examples of
FFT power spectrum and (B) autocorrelation of supragranular LFPg in Patients 3 and 5. (C)
entative example
Wave triggered SWA averages of LFPg and MUA show increased neuronal
activity during up-states and decreased firing during down-states, similar to animal
models (Figure 24).
Figure 24. Wave triggered SWA averages of LFPg and MUA in upper layer III in Patient 4.
To facilitate comparison of our results with animal studies, the threshold level was set to +50
µV on the filtered (0.3-3 Hz, 24 dB/octave, zero phase shift) upper layer III LFPg, and the
wave triggered (up-state locked) averages were calculated on the un-filtered LFPg (upper
trace) and MUA (lower trace). Note, that in our case the threshold crossing was calculated as
a triggering point instead of peak detection, this explains why the up-states in our analysis
have a less sharply contoured peak as compared to the up-states in cats, where peak-locked
averages were calculated.
63
Csercsa Richárd
Layer II LFPg (lower row of Figure 25) and the closest neighbour grid ECoG
trace (upper row of Figure 25) were used with similar parameters to test the SWA
detection algorithm. The average number of SWA cycles was calculated, which clearly
showed deepening sleep as the detection frequency for both ECoG and LFPg increases
towards the end (Figure 25, first column). For the average SWA cycle length, similar
values were found for ECoG and LFPg, with consistent duration across the 600 s epoch
(Figure 25, second column). Both the interdetection interval and the cycle length
distribution were very similar between LFPg and ECoG, further indicating strong
coupling between surface and intracortical potentials (Figure 25, third and fourth
column). In addition, the distribution of cycle lengths and interdetection intervals shown
here are similar to those found in human scalp recordings during SWS. We also found
good temporal coincidence between ECoG and LFPg based detection as reflected in the
cross-correlogram between ECoG and LFPg detected up-state time stamps (Figure 25,
on the right).
Figure 25. Uniformity of SWA cycle detection in ECoG and LFPg data. Representative sample
from Patient 3. Layer II LFPg (lower row of figures) and the closest neighbour grid ECoG
trace (upper row). Detection frequency (Y-axis) was calculated in cycles/minute as indicated
in the first column. Average number of SWA cycles was calculated over a 30 sec interval bin,
with 15 sec overlap for a total duration of 600 sec (X-axis). The second column shows the
average SWA cycle length (Y-axis) in ms versus elapsed time in seconds (X-axis). Third
column illustrates the interdetection interval (Y-axis: counts, X-axis: interval, in 250 ms bins)
distribution. Fourth column shows the distribution of cycle lengths (Y-axis: counts, X-axis:
duration, in 62.5 ms bins). The fifth column depicts the cross-correlogram between ECoG and
LFPg detected up-state time stamps (Y-axis: counts, X-axis: lag between ECoG and LFPg, 50
ms bins).
64
Laminar analysis of the slow wave actvity in humans
The uniformity of SWA detection is illustrated on Figure 26. The detection frequency,
the interdetection interval, and the cycle lengths were similar in all patients.
Figure 26. Representative samples of detection parameters of SWA cycles in Patients 1, 2 and
3. Upper row: SWA detection frequency in a 240 sec interval (divided into 30 bins) during
SWS (X-axis: time, Y-axis detected cycles per minute, 1/min). Middle row: histogram of the
interdetection intervals (X-axis: time between valid SWA cycle detections, in 166 ms bins, Yaxis: counts). Lower row: cycle length histogram (X-axis: valid cycle lengths, in 33 s bins, Yaxis: counts).
SUPRAGRANULAR ACTIVITY
II.
The increase in wideband (0.3-200 Hz) spectral power,
multiunit and single unit activity, and inward transmembrane
currents, associated with the up-state; and the decrease in
spectral power, unit activity, and outward transmembrane
currents, associated with the down-state, are mainly localized to
the supragranular layers.
65
Csercsa Richárd
To estimate the laminar contribution of various activities, ME channels were
assigned into six putative layers (I-VI) based on the histological findings (Figure 17A)
when available, and cortical depth when not. This analysis revealed substantial
concentration of the 0.3-3 Hz band LFPg FFT power within layers I-III (Figure 27) in
each patient, indicating strong supragranular synaptic/trans-membrane activity. The
SWA shape similarities between electrode contacts were significantly greater in
supragranular versus infragranular layers in each patient, (0.634 versus 0.423, grand
average pairwise coherence, Kruskal-Wallis ANOVA, p<0.01) (Figure 28), while
autocorrelation profiles revealed a more precisely paced rhythm supragranularly (Figure
29) in each patient.
Figure 27. Depth distribution profile of LFPg FFT power spectrum in Patient 3 and 5 (EEG
range: 0.1-300Hz data, no additional digital filtering was used, X-axis: frequency, Y-axis:
cortical depth, with corresponding layers, Z-axis: colour coded FFT power).
Figure 28. Depth distribution profile of pairwise coherence of LFPg channels in different
cortical layers in Patient 3 and 5. X-axis: cortical depth, with corresponding layers, Y-axis:
cortical depth, with corresponding layers, Z-axis: colour coded pairwise coherence of band
pass (0.3-3 Hz) LFPg.
66
Laminar analysis of the slow wave actvity in humans
Figure 29. Depth distribution profile of LFPg autocorrelation in Patient 4. X-axis: time, Yaxis: cortical depth, with corresponding layers, Z-axis: colour coded autocorrelation of the
LFPg.
Several measurements, both in individual patients (Figure 30A-D) and in grand
averages (Figure 30E-H), reflecting different aspects of population synaptic/transmembrane and firing activity, were maximal in supragranular at the up-state peak.
Normalized, grand average depth profiles of LFPg (Figure 30E, Figure 38B) were
marked by maximally positive deflections in layers I-III, inverting in layers V-VI into a
small negativity. MUA was also maximal in layer III (Figure 30F, Figure 38D). The
CSD depth profile at the peak of the SWA up-state showed a maximal source (outward
current) in layer I and maximal sink (inward current) in layers II-III, only very small
CSD deflections were observed infragranularly (Figure 30G, Figure 38C). In contrast,
the CSD depth profile of a population of interictal spikes detected manually and locked
to the surface positive LFPg (similarly as in the case of the up-state), were markedly
different, exhibiting a large sink-source pair in the infragranular layers (Figure 37).
Significant increases (bootstrap analysis (Delorme and Makeig, J Neurosci Methods,
2004), p<0.01) in LFPg spectral power were detected in all layers at 10-100 Hz
frequencies during up-states (Figure 30D). Gamma power of LFPg and CSD was
maximal in layer III (Figure 30H, Figure 38E-F). Detailed statistical analysis of
normalized grand averages is shown on Table 1.
67
Csercsa Richárd
Figure 30. Role of supragranular layers in SWA generation. Representative depth profile map
examples from Patient 4 (A-D) and grand averages of all patients (E-H). (A) LFPg, (B) MUA,
and (C) CSD depth profile maps. X-axis: time, Y-axis: cortical depth, with corresponding
laminarization, Z-axis: colour coded amplitude of LFPg, MUA and CSD units. Positive values
are red, negative are blue, except for CSD, where sink is depicted in red and source in blue.
(D) LFPg spectrograms from nine representative channels in layers I-VI. X-axis: time, Y-axis:
frequency, Z-axis: colour coded averaged relative spectral power in dB. Box-whisker plots of
(E) LFPg, (F) MUA, (G) CSD, (H) LFPg (red) and CSD (blue) gamma power (30-150Hz);
normalized grand average of all patients at the peak of the up-state in each layer. Mean: small
box, standard error (SE): large box, standard deviation (SD): whisker.
68
Laminar analysis of the slow wave actvity in humans
UP-STATE ONSET
III.
Action potentials at up-state onset are synchronized within ±10
ms across all cortical layers, suggesting that any layer could
initiate firing.
The time courses of MUA were examined to determine if one layer may lead
others. It was shown in ferret slices, (Sanchez-Vives and McCormick, Nat Neurosci,
2000) that layer V MUA consistently led layers II-III by an average of over 100 ms. In
our study, the up-state associated MUA peak-locked averages indicated no evident
timing difference in any of the patients, between layers III and V, regardless of whether
peak alignment was based on layer III or layer V activity (Figure 31).
Figure 31. Simultaneity of MUA response in supra- and infragranular layers of Patient 4 and
Patient 5. MUA from layers III (red trace) and V (blue) are shown when aligned and averaged
on the up-state associated MUA peak detected in layer III and in layer V. There is no visible
MUA delay between layers III and V regardless of which layer is used for time locking.
69
Csercsa Richárd
To further characterize the MUA timing between different layers, it was crosscorrelated (3 SD threshold, 10 ms bin size) between each pair of channels, within 200ms
of every up-state onset. In agreement with animal studies (Sakata and Harris, Neuron,
2009), (Chauvette et al., Cereb Cortex, 2010), delay maps and histograms (Figure 32)
indicated a short inter-laminar MUA timing difference at up-state onset; most of the
delays were within the ±10 ms bin. We also calculated how often (in percentage of all
sweeps) any given MUA channel shows the earliest firing at up-state onset. In all
patients (where MUA was available), the initial firing was quite uniformly distributed
across cortical depths. Unlike in a ferret in vitro study (Sanchez-Vives and McCormick,
Nat Neurosci, 2000), we found no evidence for long (~100 ms) lead or lag times
between different layers (Figure 32).
Figure 32. (A) MUA cross-correlation peak latencies (X-axis versus Y-axis) between each
pair of channels in Patient 1. Positive latencies (red) indicate that X channel leads over Y
channel, while negative latencies (blue) represent lagging. (B) Histogram of leading and
lagging values from (A). (C-D) Percentage of a given MUA channel showing the earliest
firing at up-state onset, representative data from Patients 1 and 3.
70
Laminar analysis of the slow wave actvity in humans
FREQUENCY OF SLOW OSCILLATION
IV.
The laminar mechanisms generating slow oscillation are
similar within frequency bands 0.6-2 Hz.
Separate averages of different SWA cycle lengths corresponding to appropriate (0.6-0.8
Hz, 0.8-1 Hz, 1-1.3 Hz and 1.3-2 Hz) oscillation frequencies also yielded qualitatively
similar LFPg, spectral LFPg, MUA and CSD distribution (Figure 33). We have found
no statistically significant differences in any layers (ANOVA, Tukey’s HSD post-hoc
test, p>0.3) in the CSD or MUA at the peak of the up-state between any of the four
frequency bands indicating similar cortical generator mechanisms above (up to 2 Hz)
and below 1 Hz (down to 0.6 Hz).
Figure 33. Depth profiles at different SWA frequencies. Up-state locked averages of LFPg,
LFPg spectrogram, MUA and CSD in Patient 3 and Patient 4 at four different SWA
frequencies. Frequencies 1.3-2Hz, correspond to a cycle length of: 500-750 ms; 1-1.3 Hz to
750-1000 ms; 0.8-1 Hz to 1000-1250 ms; and 0.6-0.8 Hz to 1250-1500 ms. Roman numerals
mark putative cortical layers. Colour calibrations are on the bottom. CSD sink is depicted in
red, source in blue. Each spectrogram (SPC) window shows the spectral content (Z-axis,
colour coded) versus time (X-axis) of a representative LFPg channel from a given layer from
1-100 Hz (Y-axis), measures are expressed in dB relative to a distant baseline (-2500 to 1500 ms).
71
Csercsa Richárd
SPARSE FIRING
V.
Neuronal firing in the up
up-state
state is sparse compared to
extracellular
llular recordings in animals.
Recordings from three patients yielded good quality single unit activity (Figure
(
20, Figure 34, Figure 35).
). Epochs (~1000 sec) showin
showing
g the largest SWA detection
frequencies were selected for analysis from the first sleep cycle. Overall 33 single units
were clustered (9, 12 and 12 from Patients 1, 4 and 5) with mean firing rate of 0.66 Hz
(range=0.12-2.0
2.0 Hz, SD=0.48) and mean burst frequ
frequency
ency of 3.1 1/minute (range=0-14
(range=0
1/minute, SD=3.6). Both the average firing rate and the spontaneous burst rate were
well below the reported epileptic threshold found in cortical and hippocampal structures
(Staba
Staba et al., J Neurosci, 2002
2002).
Figure 34.. Single unit firing in SWA. (A) Superimposed (40 consecutive sweeps) and (B)
individual single sweeps of simultaneously recorded supragranular LFPg and MUA/SUA
(Patient 5). Solid and dashed red lines represent LFPg mean and standard deviation.
Nearly all cells (31 of 33) showed si
significantly non-uniform
uniform spiking (Figure
(
34,
Figure 35, Figure 36)) over the SWA cycle (Rayleigh test, p<0.01), with peak up-state
up
firing rate mean of 1.63 Hz (range=0.45
(range=0.45-4.6
4.6 Hz, SD=0.96). We found no significant
differences between patients in mean firing rates (Kruskal
(Kruskal-Wallis
Wallis ANOVA, p>0.2),
72
Laminar analysis of the slow wave actvity in humans
indicating homogeneous distribution. Although mean firing rates grouped by supraversus infragranular layers showed no significant differences (p>0.1), supragranular
peak up-state firing rates were significantly higher (2.2 Hz versus 1.2 Hz, KruskalWallis ANOVA, p<0.01) than the same measure for infragranular layers. We found the
proportion of firing cells and the rate at which they fire in any given up-state (Figure 35,
Figure 39) remarkably low. On average, only 27% of the clustered cells were active
(firing at least one spike) during any given up-state (20%, 25% and 36% in each
patient). Thus, an average neuron fired in every third to fifth up-state. Moreover, on
average, each neuron fired only 0.32 spikes per up-state (0.44, 0.2 and 0.32 in each
patient). As an example, out of the 12 clustered neurons in Patient 4, the most probable
number of active cells in a given up-state was 2 (Figure 39C), and the most probable
number of overall spikes the 12 cells fired within a given up-state was also 2 (Figure
39B).
Figure 35. Representative single unit firing in SWA. (A) Phase raster plots of two well
separated units (cell1, cell2) in SWS. Columns represent individual SWA cycles (1-385), rows
represent phase from -180° to +180° (top to bottom) in 30° bin increments. Blue: no firing in
the given phase bin, green: one, yellow: two, red: three or more discharges. (B) Phase
histograms of action potentials for cell 1 (upper) and 2 (lower), firing rate in Hz versus phase,
in 30° bins.
73
Csercsa Richárd
Figure 36. Representative normalized (from 0 to 1) firing rate versus phase histograms (from 180° to +180°, in 30° bins) of clustered cells from different layers and patients. Red line:
positive half-wave (up-state), green line: negative half-wave (down-state).
74
Laminar analysis of the slow wave actvity in humans
DISCUSSION
Our results establish a close similarity between the human SWA and the animal
SO at the level of field potential, cellular firing activity and spectral measurements
(Steriade, Neuroscience, 2006), but also reveal a number of novel, unexpected findings.
Consistent with prior studies in animals, we have shown in humans that the up-state was
associated with increased firing and elevated spindle, alpha, beta, gamma and ripple
power during the surface positive LFP half-wave, while the down-state was
characterized by the widespread surface negative LFP half-wave with decreased firing,
and oscillatory activity (Cash et al., Science, 2009). Differences from prior studies were
found in the laminar distribution of the up-state, average firing rates during the up-state,
and the consistency of generators for oscillations above versus below one Hertz. These
contrasts could reflect cortical cytoarchitectonic differences or they could be due to the
circumstances of the recordings, including natural sleep versus different types of
anesthesia, or in vivo versus in vitro preparations. They could also be due to epileptic
pathology or to phylogenetic differences.
EPILEPSY AND SLOW WAVES
Epilepsy is a multi causal disease with diverse etiology. Focal epilepsies have
circumscribed seizure initiating regions without wide spread pathological alterations in
other areas. Surgical candidates are selected exclusively from this patient group during
the careful pre-operative evaluation, based on several diagnostic findings (CT, MRI,
fMRI, PET, SPECT, video-EEG, MEG, functional neurophysiological tests, Wada-test
and seizure semiology). In the present study we included only patients with the focal
disease and carefully avoided to analyze data from the spatial and temporal proximity of
the seizure focus and epileptiform events.
The laminar LFPg, CSD, MUA and spectral profile of interictal activity in vivo
and in vitro has already been established by our group (Ulbert et al., Epilepsia, 2004),
75
Csercsa Richárd
(Ulbert et al., Exp Neurol, 2004), (Fabo et al., Brain, 2008), (Wittner et al., Brain,
2009). We have shown that the initial phase of the interictal discharges are large
amplitude brief events characterized by substantial action potential, LFPg, CSD, MUA
and spectral surges, often emerging from the granular and infragranular layers of the
cortex. These events are clearly distinctive from the background activity and exquisitely
visible in single sweeps (Figure 37). Based on our prior knowledge we carefully
excluded any suspicious pathological events from the analysis presented in here.
Figure 37. CSD depth profile of the up-state and a population of interictal spikes in Patient 4.
Boxes indicate the mean normalized (between +1 and -1) CSD in layers I-VI, whiskers
indicate the standard error. Up-state depth profile is depicted in blue, interictal spike depth
profile in red. Interictal spikes were detected manually, such a way that the peak of the
surface positivity was designated as time zero, such as with the up-states. While up-states
exhibit a major sink-source pair in the superficial layers, the calculated population of
interictal spikes exhibit a large sink-source pair in the deep layers.
Several other considerations suggest that the current findings on the neuronal
mechanisms underlying SWA, although recorded in epileptic patients, might also apply
to healthy subjects. Our SWA morphology corresponded well to those collected from
standard scalp EEG recordings from healthy subjects. Similarities included not only the
SWA frequency and rhythmicity (Achermann and Borbely, Neuroscience, 1997), but
76
Laminar analysis of the slow wave actvity in humans
the asymmetric shape, the briefer and sharper deflection in the down-state (Massimini et
al., J Neurosci, 2004) and higher beta power content in the up-state (Molle et al., J
Neurosci, 2002). Our results of detection frequency, cycle length and inter-detection
interval histograms are all comparable with previous findings from healthy subjects
(Massimini et al., J Neurosci, 2004), despite that both the recording and analysis
methodology were different. Minor deviations in the exact numbers are therefore natural
and may reflect methodological differences rather than disease related alterations. In
addition, neither the firing rate, nor the burst rate exceeded the pathological criteria
found for single neurons in SWS (Staba et al., J Neurosci, 2002), (Staba et al.,
Hippocampus, 2002). Finally, the lack of any MRI abnormalities and intact
laminarization of the excised tissue strongly suggests that we recorded from structurally
intact regions, free of gross functional alterations.
Nevertheless, there are some observations in our study that may be related to the
pathology. Out of the 33 clustered units in 3 patients, 2 single cells in one patient
(Patient 1) showed uniform firing during the SWA cycle (layer III unit #3, average
firing rate=1.17 Hz, layer V unit #6, average firing rate=0.54 Hz) . While it was
expected due to biological (short up-states in a long down-state) and/or methodological
reasons (inaccuracy of the state detection algorithm) that some of the detected downstates may contain firing activity, it was unexpected that cells did not show significant
modulation during the two-state oscillation at all. We are not aware of similar effects in
humans or in animal models. This finding may reflect an excessive resistance of a small
subgroup of cortical neurons to the network-wide hyperpolarisation occurring in the
down-state and needs further investigation.
SUPRAGRANULAR ACTIVITY IN SWS
A consistent CSD pattern of our study was the prominent sink-source pair in the
supragranular layers compared to the weak infragranular activation. This localization
was true for both the low (0.5-2 Hz) and the high frequency (10-200 Hz) oscillations in
all patients, in frontal and parietal areas.
77
Csercsa Richárd
In our interpretation, the prominent layer II-III sink during the up-state reflects
the large active inward currents flowing across the distal dendritic membrane
compartments of layer V-VI pyramidal cells and distal, proximal and basal dendritic
membrane compartments and perhaps on the somatic membrane of layer II-III
pyramidal cells. The corresponding passive, return, source currents are flowing in layer
I, across the most superficial apical dendritic membrane compartments of the pyramidal
cells. The spatial CSD pattern in the down-state is inverted, exhibiting a large active
current source in layer II-III due to hyperpolarizing currents (likely outward potassium
flows from the pyramidal cells) and a passive return sink in layer I (Cash et al., Science,
2009).
A
I
II
III
IV
V
VI
0.766085
0.000124
0.000124
0.000124
0.000124
0.001666
0.000124
0.000124
0.000124
0.688070
0.000124
0.000124
0.000124
0.949348
0.957987
I
II
III
IV
V
VI
0.766085
0.000124
0.000124
0.000124
0.000124
0.000124
0.000124
0.000124
0.000124
0.001666
0.000124
0.000124
0.688070
0.949348
0.957987
B
I
II
III
IV
V
I
II
III
IV
V
VI
C
I
II
III
IV
V
VI
0.000125
0.000125
0.000125
0.000125
0.000125
0.000125
I
0.999849
0.000126
0.000125
0.000125
II
0.085258
0.085258
0.000143
0.095084
0.002729
0.341979
0.435127
1.000000
0.971067
0.786468
0.000125
0.999849
0.000125
0.000125
0.000125
III
0.000143
0.435127
0.400887
0.778203
0.000478
78
0.000125
0.000126
0.000125
0.607915
0.180558
IV
0.095084
1.000000
0.400887
0.961539
0.816120
0.000125
0.000125
0.000125
0.607915
VI
0.000125
0.000125
0.000125
0.180558
0.927437
0.927437
V
0.002729
0.971067
0.778203
0.961539
0.090742
VI
0.341979
0.786468
0.000478
0.816120
0.090742
Laminar analysis of the slow wave actvity in humans
D
I
II
III
IV
V
VI
E
I
II
III
IV
V
VI
I
II
0.023491
0.023491
0.000135
0.079675
0.988682
0.879812
I
0.701007
0.972408
0.023824
0.000159
II
0.012708
0.012708
0.000226
0.432125
0.999551
0.965169
0.980461
0.424561
0.004422
0.000248
III
0.000135
0.701007
0.105680
0.000124
0.000123
IV
0.079675
0.972408
0.105680
0.087117
0.000210
III
0.000226
0.980461
0.026647
0.000126
0.000125
IV
0.432125
0.424561
0.026647
0.376114
0.027556
V
0.988682
0.023824
0.000124
0.087117
VI
0.879812
0.000159
0.000123
0.000210
0.185144
0.185144
V
0.999551
0.004422
0.000126
0.376114
VI
0.965169
0.000248
0.000125
0.027556
0.697058
0.697058
Table 1. Detailed statistical tables of (A) LFPg, (B) CSD, (C) MUA, (E) LFPg gamma power
(30-150Hz) and (F) CSD gamma power. Normalized grand average of all patients at the peak
of the up-state, grouped by layers I-VI. Significant differences (ANOVA with Tukey HSD
test, p<0.01) are depicted in red.
Our CSD findings from the frontal and parietal areas in natural sleep are in
contrast with a study in the cat suprasylvian area, albeit under ketamine/xylazine
anesthesia (Steriade and Amzica, Proc Natl Acad Sci U S A, 1996). At low frequencies,
(~1 Hz) the maximal up-state related sink in the cat was located in the middle rather
than in the superficial layers, surrounded by not only a superficial but also a large deep
source. In addition, a substantial up-state related sink in the deepest layer was present in
the cat, which was practically invisible in our human recordings. The same authors also
showed a massive supragranular (layer II-III) up-state related sink besides one or two
deeper and weaker sinks, during the spontaneous and evoked K-complex (Amzica and
Steriade, Neuroscience, 1998). Moreover, at higher frequencies, (~35 Hz) and during
the K-complex, a series of “alternating microsinks and microsources” was found
throughout the depth of the cat suprasylvian area (Steriade and Amzica, Proc Natl Acad
Sci U S A, 1996), (Amzica and Steriade, Neuroscience, 1998). Such alternating patterns
are best explained by insufficient spatial resolution in the LFP sampling (8 contact, 250
µm spacing electrode array) and corresponding spatial aliasing error, and not by
neuronal sources (Tenke et al., Exp Brain Res, 1993).
79
Csercsa Richárd
A recent study in the cat suprasylvian area in natural sleep with adequate spatial
sampling (100 µm) revealed alternation free middle and deep layer sinks and a
superficial source during the up-state (Chauvette et al., Cereb Cortex, 2010). Thus,
besides cytoarchitectonic differences between different types of cortices (e.g.: frontal
and parietal areas in humans versus suprasylvian area in the cat), and methodological
errors, it is also plausible to assume that neuronal mechanisms of natural sleep and
ketamine/xylazine anesthesia may be different, further accounting for the divergent
findings. Another recent study on the rat primary auditory cortex using high spatial
resolution (50 µm) CSD mapping showed the maximal up-state related sink to be in the
presumed supragranular layers under urethane anesthesia, while in natural sleep it is
rather the presumed granular and probably infragranular layers that exhibited the largest
up-state related sink (see average data, Supplementary Figures S10 and S16 in (Sakata
and Harris, Neuron, 2009)). Our CSD results in natural sleep are quite close to the
results of Sakata and Harris obtained under urethane anesthesia, except for the large
source deep in the infragranular layers.
Some differences between humans and cats or rats might also be expected given
that our prefrontal cortex is more than hundred times larger, with a striking increase in
pyramidal cell dendritic complexity (Elston, Cereb Cortex, 2003), (Elston et al., Anat
Rec A Discov Mol Cell Evol Biol, 2006). Our finding that the SWA and corresponding
high frequency rhythms including spindle, alpha, beta, gamma and ripple oscillations
may involve supragranular layers is consistent with this massive cortical expansion of
neuronal number (Herculano-Houzel et al., Proc Natl Acad Sci U S A, 2007), inasmuch
as increased dendritic complexity and cortico-cortical association fibres predominantly
target these layers (Gonzalez-Burgos et al., Cereb Cortex, 2000).
The strong supragranular oscillatory activity in sleep may be beneficial for the
local, higher order processing of sensory experience and perhaps memory consolidation,
since these layers are interconnected by dense cortico-cortical projections forming finescale functional networks to perform integrative functions (Yoshimura et al., Nature,
2005). The weaker infragranular activity may reflect the relatively suppressed cortical
executive, output functions, which may prevent effective connectivity between distant
cortical areas from developing in SWS (Massimini et al., Science, 2005).
80
Laminar analysis of the slow wave actvity in humans
FREQUENCY OF THE SLOW WAVES
According to our observations, the SWA depth profile, represented by the upstate peaks, was similar between the four investigated frequency ranges including the
slow (<1 Hz) and delta band (up to 2 Hz). In our opinion, these frequency bands are
thus substantially overlapping, hence a less strict distinction should be applied between
activities above versus below 1 Hz. We agree that the slow activity and thalamic delta
have obviously different neuronal mechanisms, but it seems that these waves cannot be
distinguished using exclusively a frequency band criterion.
We have found similar signs of elementary hierarchical organization of low and
high frequency oscillations in humans as it was shown in animal models (Lakatos et al.,
J Neurophysiol, 2005). The organizing substrate was the up-state of the SWA, which
gave rise to a wide variety of higher frequency activity including spindle, alpha, beta,
gamma and ripple oscillations. Each of these high frequency oscillatory bursts was quite
different from sweep to sweep, showing for example occasional spindle sequences or
marked gamma or ripple band enhancements at various peak frequencies. These
observations suggest that each SWA cycle with unique oscillatory signature reflect
individual information content coded differently in the oscillatory process. Given the
variability of the high frequency oscillatory activity during up-states, it is plausible to
assume that different underlying neuronal populations might be responsible for the
generation of each specific oscillatory pattern. This strategy might be beneficial in the
configuration of functional connectivity between neurons to form stable ensembles that
may promote the consolidation of memory in sleep.
Paroxysmal activity is known to emerge more often from NREM sleep
compared to REM (Steriade and Timofeev, Neuron, 2003). Animal studies revealed that
cortical hyperexcitability associated with ripple oscillations are often result in
pathological synchronization leading to epileptic seizures (Grenier et al., J
Neurophysiol, 2003). We have shown that up-states are characterized by a large
increase in cortical excitability reflected in the increased power of gamma and ripple
oscillations. Thus, we hypothesize that the active state of the slow oscillation may play
81
Csercsa Richárd
an important role in the generation of seizures and other paroxysmal signs in the cortical
epileptic network.
ROLE OF CALRETININ POSITIVE NEURONS
Calretinin positive (CR+) cell density depth profiles (Figure 38A) were
calculated in two patients and correlated with the depth profile at up-state peak of LFPg,
CSD, MUA, LFPg and CSD gamma power (Figure 38B-F). CR+ cell density between
Patient 4 and 5 showed high similarity (r=0.95, p<0.01). The highest positive
correlation was found between CR+ cell density and CSD gamma power (r=0.85,
p<0.01).
Figure 38. Depth profiles of CR+ cell density and SWA. (A) Averaged normalized CR+ cell
density profile of Patients 4 and 5, whiskers represent standard errors. (B) LFPg, (C) CSD,
(D) MUA, (E) LFPg gamma power and (F) CSD gamma power of averaged normalized depth
profile of up-state in Patients 4 and 5 with standard error (whisker). Number in the upper right
corner indicates the Pearson r correlation between CR+ density and (B-F).
The relatively high correlation in laminar location between LFPg, CSD, MUA
and gamma power during up-states and CR+ cell density may provide additional
insights into the mechanism for the predominance of oscillatory activity in
supragranular layers. CR+ cells are relatively numerous for inhibitory cells, comprising
~8% of the total number of prefrontal neurons, and ~14.2-17.6% of layer II-III neurons,
82
Laminar analysis of the slow wave actvity in humans
in human prefrontal cortex (Gabbott et al., J Comp Neurol, 1997). CR+ cell density in
the human, monkey, cat, and rat cortex is highest in the supragranular layers (Fonseca
and Soriano, Brain Res, 1995), (Gabbott et al., J Comp Neurol, 1997), (Schwark and Li,
Brain Res Bull, 2000). The layer II-III population of CR+ cells (78% of all CR+ cells) is
significantly more numerous (+31%) in humans than in the rat (Gabbott et al., J Comp
Neurol, 1997), and the supragranular CR+ predominance in humans is not affected by
epilepsy (Barinka et al., Epilepsy Res, 2010). The relatively high density and
remarkable vertically oriented dendritic alignment (Gabbott et al., J Comp Neurol,
1997) of layer II-III CR+ cells (Figure 17B) suggest that this population on its own may
contribute significantly to the CSD. Unlike basket cells that target principal cells
(Somogyi et al., Neuroscience, 1983) establishing local negative-feedback circuits,
layer II-III CR+ interneurons preferentially target other inhibitory cells locally in layer
II-III and target pyramidal cells in layer V (Meskenaite, J Comp Neurol, 1997) forming
local positive-feedback circuits (Dantzker and Callaway, Nat Neurosci, 2000) in the
supragranular layers and negative-feedback circuits between supra- and infragranular
layers. The negative-feedback imposed by the population of CR+ cells may relatively
attenuate the infragranular activation, while the positive-feedback disinhibition (Tamas
et al., J Neurosci, 1998) may amplify the supragranular synaptic/trans-membrane
oscillations in the gamma band and cellular activity (Whittington et al., Nature, 1995).
In addition, GABAB receptors are more concentrated in the upper layers of the
cortex, at least in rodents (Lopez-Bendito et al., Eur J Neurosci, 2002), (Tamas et al.,
Science, 2003), which might also contribute to the potassium current that may play an
important role in the down-state generation (Timofeev et al., Proc Natl Acad Sci U S A,
2001).
FIRING RATE IN SWS
In vitro slice studies in animals found that firing in infragranular layers
consistently lead supragranular layers by over 100 ms at the onset of the up-state
(Sanchez-Vives and McCormick, Nat Neurosci, 2000). In contrast, we found that the
onset of activity during up-states differs less than ±10 ms between layers. Although the
83
Csercsa Richárd
slice preparation is a powerful tool, it severs connections that are present in the intact
animal, removing background synaptic input and placing the cell in an artificial
medium.
In healthy humans, it is believed that each individual slow wave cycle has
distinct origin and propagates uniquely across a number of brain areas (Murphy et al.,
Proc Natl Acad Sci U S A, 2009). Similar patterns were also found in animal models
(Ferezou et al., Neuron, 2007), (Mohajerani et al., J Neurosci, 2010). Thus, variable
projections may be involved in its propagation, terminating in variable cortical layers
making the laminar distribution of the initial unit firing also variable, as it was shown in
our study.
Figure 39. Single unit data from Patients 1, 4 and 5 indicating sparse firing in up-states. (A)
Colour raster plot illustrates the firing of clustered neurons in up-states. Columns represent
individual SWA cycles (221 in Patient 1, 252 in Patient 4, and 222 in Patient 5), rows
represent clustered neurons (cell 1-7 in Patient 1, cell 1-12 in Patient 4 and 5), colour
represents the firing of a given cell. Blue: no firing in the given up-state for the given cell,
green: one, yellow: two, red: three or more action potentials. (B) Histogram of the total
number of fired action potentials (spike number count) from all clustered cells during a given
up-state (X-axis: total number of action potentials fired by all cells, Y-axis: number of upstates with that number of action potentials, count). (C) Histogram of the number of active
clustered cells (active cell count) during the up-states (X-axis: number of cells that fired at
least one action potential, Y-axis: number of up-states, count). These data illustrate sparse
firing in up-states, only a small fraction of the clustered cells fire and these cells together
generate only a few action potentials.
84
Laminar analysis of the slow wave actvity in humans
The low average firing rate (0.6 Hz) found in here is consistent with a prior
human report (Ravagnati et al., Sleep, 1979). Animal studies using extracellular silicon
probes and intracellular sharp electrodes report average firing rates in the 2-20 Hz range
from the entire depth of the cortex (Steriade et al., J Neurophysiol, 2001), (Isomura et
al., Neuron, 2006), (Luczak et al., Proc Natl Acad Sci U S A, 2007), cell-attached and
whole-cell patch-clamp studies from layer II-III neurons report average firing rates in
the 0.01-0.3 Hz range (Margrie et al., Pflugers Arch, 2002), (Waters and Helmchen, J
Neurosci, 2006), (Hromadka et al., PLoS Biol, 2008). Some of these differences may be
related to the laminar location of the neurons and to ‘collateral damage’ inherent to the
different techniques. Extracellular recordings might be biased towards higher average
firing rates, because of the use of a minimum spontaneous firing rate (1-2 Hz) constraint
(Luczak et al., Proc Natl Acad Sci U S A, 2007). We did not use such correction in our
unit analysis, thus slower firing cells were also included. Sharp electrode intracellular
recordings (Steriade et al., J Neurophysiol, 2001), (Isomura et al., Neuron, 2006)
disrupt the cell membrane and introduce leakage current, which may also alter the firing
rate. In cell-attached recordings (Margrie et al., Pflugers Arch, 2002), (Hromadka et al.,
PLoS Biol, 2008) the membrane is partially covered by the recording pipette causing
substantial mechanical stress, receptor, ion channel masking and membrane capacitance
changes, while the whole-cell configuration (Waters and Helmchen, J Neurosci, 2006)
disrupts the membrane and causes cell dialysis. Indeed, when establishing the wholecell configuration the spontaneous firing rate may double compared to the cell-attached
state (Margrie et al., Pflugers Arch, 2002). Given these apparently strong effects of
recording methodology on cell firing, it is hard to definitively relate the present findings
to animal experiments. However, it seems reasonable to expect that any technique
which physically contacts the cell would alter the firing rate to a greater extent than
techniques which do not. To further elucidate these differences, unbiased extracellular
action potential techniques need to be implemented in different animal models.
The apparently lower firing rates of human cortical neurons during up-states
may also reflect an adaptation to prevent runaway excitation of our larger and more
densely interconnected neocortex, or it may better support the sparse representation of
our extensive long term memories (Quiroga et al., Nature, 2005).
85
Csercsa Richárd
Finally we conclude, that it is extremely important to further characterize the
above discussed influences and establish an integrated view that takes into account that
the neuronal mechanisms underlying slow wave generation may be depending on
several parameters including for example the neurological condition, the observed
cortical areas, the experimental preparation, the type of sleep induction and perhaps
species differences.
86
Laminar analysis of the slow wave actvity in humans
SUMMARY
The cortical slow wave activity emerging during the deepest stages of non-rapid
eye movement sleep is thought to underlie essential restorative processes and facilitate
the consolidation of declarative memories. In animals the slow oscillation consists of
two rhythmically recurring phases: one of them is characterized by widespread,
increased cellular and synaptic activity, referred to as active- or up-state, followed by
cellular and synaptic inactivation, referred as silent- or down-state. However, its neural
mechanisms in humans are poorly understood since the traditional intracellular
techniques used in animals are inappropriate for investigating the cellular and
synaptic/trans-membrane events in humans.
For the examination of the neuronal properties of the sleep slow oscillation, we
recorded intracortical laminar local field potential gradient, multiple and single unit
activities with laminar multichannel microelectrodes and simultaneous surface
potentials with subdural grid electrodes chronically implanted into the cortex of patients
with drug resistant focal epilepsy undergoing cortical mapping for seizure focus
localization. We also analyzed the current source density and spectral features of the
recorded signals.
We found that slow wave activity in humans reflects a rhythmic oscillation (0.6
– 2 Hz) between neuronal activation and silence. Similar to animal studies, cortical
activation was demonstrated as increased wideband (0.3-200 Hz) spectral power,
increased multiple and single unit activity, and powerful inward transmembrane
currents, all mainly localized to the supragranular layers. Neuronal firing was sparse and
the average discharge rate of single cells was less than expected from animal studies.
The latency of firing at up-state onset across all layers was in the range of 10 ms,
suggesting close inter-laminar coupling at up-state onset.
Here we provide strong direct experimental evidences that slow wave activity in
humans is characterized by hyperpolarizing currents associated with suppressed cell
firing, alternating with high levels of oscillatory synaptic activity associated with
increased cell firing. Our results emphasize the major involvement of supragranular
layers in the genesis of slow wave activity.
87
Csercsa Richárd
BIBLIOGRAPHY
Achermann, P. and Borbely, A.A. (1997) Low-frequency (< 1 Hz) oscillations in the
human sleep electroencephalogram. Neuroscience, 81, 213-222.
Amzica, F. and Steriade, M. (1995) Disconnection of intracortical synaptic linkages
disrupts synchronization of a slow oscillation. J Neurosci, 15, 4658-4677.
Amzica, F. and Steriade, M. (1998) Cellular substrates and laminar profile of sleep Kcomplex. Neuroscience, 82, 671-686.
Andrews-Hanna, J.R., Reidler, J.S., Sepulcre, J., Poulin, R., Buckner, R.L. (2010)
Functional-anatomic fractionation of the brain's default network. Neuron, 65,
550-562.
Aserinsky, E. and Kleitman, N. (1953) Regularly occurring periods of eye motility, and
concomitant phenomena, during sleep. Science, 118, 273-274.
Atkinson, R.C. and Shiffrin, R.M. (1968) Human memory: a proposed system and its
control processes. In Spence, K.W. (ed) The psychology of learning and
motivation: II. Academic Press, Oxford.
Axmacher, N., Elger, C.E., Fell, J. (2008) Ripples in the medial temporal lobe are
relevant for human memory consolidation. Brain, 131, 1806-1817.
Barinka, F., Druga, R., Marusic, P., Krsek, P., Zamecnik, J. (2010) Calretinin
immunoreactivity in focal cortical dysplasias and in non-malformed epileptic
cortex. Epilepsy Res, 88, 76-86.
Bazhenov, M., Timofeev, I., Steriade, M., Sejnowski, T.J. (2002) Model of
thalamocortical slow-wave sleep oscillations and transitions to activated States.
J Neurosci, 22, 8691-8704.
Blake, H. and Gerard, R. (1937) Brain potentials during sleep. Am J Physiol, 119, 692703.
Bódizs, R. (2003) Az alvás és jelenségköre. In Pléh, C., Kovács, G., Gulyás, B. (eds)
Kognitív idegtudomány. Osiris Kiadó, Budapest, pp. 601-618.
88
Laminar analysis of the slow wave actvity in humans
Borbely, A.A. (1982) A two process model of sleep regulation. Hum Neurobiol, 1, 195204.
Born, J., Rasch, B., Gais, S. (2006) Sleep to remember. Neuroscientist, 12, 410-424.
Bremer, F. (1936) Nouvelles recherches sur le mécanisme du sommeil. C R Soc Biol,
460-464.
Buzsaki, G. (1989) Two-stage model of memory trace formation: a role for "noisy"
brain states. Neuroscience, 31, 551-570.
Buzsaki, G. (1998) Memory consolidation during sleep: a neurophysiological
perspective. J Sleep Res, 7 Suppl 1, 17-23.
Buzsáki, G. (2006) Rhythms of the brain. Oxford University Press.
Campbell, I.G. and Feinberg, I. (2009) Longitudinal trajectories of non-rapid eye
movement delta and theta EEG as indicators of adolescent brain maturation.
Proc Natl Acad Sci U S A, 106, 5177-5180.
Cash, S.S., Halgren, E., Dehghani, N., Rossetti, A.O., Thesen, T., Wang, C., Devinsky,
O., Kuzniecky, R., Doyle, W., Madsen, J.R., Bromfield, E., Eross, L., Halasz, P.,
Karmos, G., Csercsa, R., Wittner, L., Ulbert, I. (2009) The human K-complex
represents an isolated cortical down-state. Science, 324, 1084-1087.
Chalupa, L.M. and Werner, J.S. (2001) The Visual Neurosciences. The MIT Press.
Chauvette, S., Volgushev, M., Timofeev, I. (2010) Origin of Active States in Local
Neocortical Networks during Slow Sleep Oscillation. Cereb Cortex.
Chemelli, R.M., Willie, J.T., Sinton, C.M., Elmquist, J.K., Scammell, T., Lee, C.,
Richardson, J.A., Williams, S.C., Xiong, Y., Kisanuki, Y., Fitch, T.E., Nakazato,
M., Hammer, R.E., Saper, C.B., Yanagisawa, M. (1999) Narcolepsy in orexin
knockout mice: molecular genetics of sleep regulation. Cell, 98, 437-451.
Chrobak, J.J. and Buzsaki, G. (1996) High-frequency oscillations in the output networks
of the hippocampal-entorhinal axis of the freely behaving rat. J Neurosci, 16,
3056-3066.
89
Csercsa Richárd
Cirelli, C. and Tononi, G. (2008) Is sleep essential? PLoS Biol, 6, e216.
Clemens, Z., Fabo, D., Halasz, P. (2005) Overnight verbal memory retention correlates
with the number of sleep spindles. Neuroscience, 132, 529-535.
Clemens, Z., Molle, M., Eross, L., Barsi, P., Halasz, P., Born, J. (2007) Temporal
coupling of parahippocampal ripples, sleep spindles and slow oscillations in
humans. Brain, 130, 2868-2878.
Cohen, N.J. and Squire, L.R. (1980) Preserved learning and retention of patternanalyzing skill in amnesia: dissociation of knowing how and knowing that.
Science, 210, 207-210.
Colrain, I.M. (2005) The K-complex: a 7-decade history. Sleep, 28, 255-273.
Compte, A., Sanchez-Vives, M.V., McCormick, D.A., Wang, X.J. (2003) Cellular and
network mechanisms of slow oscillatory activity (<1 Hz) and wave propagations
in a cortical network model. J Neurophysiol, 89, 2707-2725.
Contreras, D. and Steriade, M. (1995) Cellular basis of EEG slow rhythms: a study of
dynamic corticothalamic relationships. J Neurosci, 15, 604-622.
Contreras, D., Timofeev, I., Steriade, M. (1996) Mechanisms of long-lasting
hyperpolarizations underlying slow sleep oscillations in cat corticothalamic
networks. J Physiol, 494 ( Pt 1), 251-264.
Crunelli, V. and Hughes, S.W. (2010) The slow (<1 Hz) rhythm of non-REM sleep: a
dialogue between three cardinal oscillators. Nat Neurosci, 13, 9-17.
Csicsvari, J., Hirase, H., Czurko, A., Buzsaki, G. (1998) Reliability and state
dependence of pyramidal cell-interneuron synapses in the hippocampus: an
ensemble approach in the behaving rat. Neuron, 21, 179-189.
Daan, S., Barnes, B.M., Strijkstra, A.M. (1991) Warming up for sleep? Ground squirrels
sleep during arousals from hibernation. Neurosci Lett, 128, 265-268.
Dang-Vu, T.T., Schabus, M., Desseilles, M., Albouy, G., Boly, M., Darsaud, A., Gais,
S., Rauchs, G., Sterpenich, V., Vandewalle, G., Carrier, J., Moonen, G., Balteau,
E., Degueldre, C., Luxen, A., Phillips, C., Maquet, P. (2008) Spontaneous neural
90
Laminar analysis of the slow wave actvity in humans
activity during human slow wave sleep. Proc Natl Acad Sci U S A, 105, 1516015165.
Dantzker, J.L. and Callaway, E.M. (2000) Laminar sources of synaptic input to cortical
inhibitory interneurons and pyramidal neurons. Nat Neurosci, 3, 701-707.
Dayan, P. and Abbott, L.F. (2001) Theoretical Neuroscience: Computational and
Mathematical Modeling of Neural Systems. The MIT Press, Cambridge,
Massachusetts.
De Gennaro, L., Marzano, C., Fratello, F., Moroni, F., Pellicciari, M.C., Ferlazzo, F.,
Costa, S., Couyoumdjian, A., Curcio, G., Sforza, E., Malafosse, A., Finelli,
L.A., Pasqualetti, P., Ferrara, M., Bertini, M., Rossini, P.M. (2008) The
electroencephalographic fingerprint of sleep is genetically determined: a twin
study. Ann Neurol, 64, 455-460.
Delorme, A. and Makeig, S. (2004) EEGLAB: an open source toolbox for analysis of
single-trial EEG dynamics including independent component analysis. J
Neurosci Methods, 134, 9-21.
Destexhe, A., Contreras, D., Steriade, M. (1999) Spatiotemporal analysis of local field
potentials and unit discharges in cat cerebral cortex during natural wake and
sleep states. J Neurosci, 19, 4595-4608.
Destexhe, A., Rudolph, M., Pare, D. (2003) The high-conductance state of neocortical
neurons in vivo. Nat Rev Neurosci, 4, 739-751.
Destexhe, A., Hughes, S.W., Rudolph, M., Crunelli, V. (2007) Are corticothalamic 'up'
states fragments of wakefulness? Trends Neurosci, 30, 334-342.
Détári, L. (2008) Biológiai ritmusok, Lecture notes.
Diekelmann, S., Wilhelm, I., Born, J. (2009) The whats and whens of sleep-dependent
memory consolidation. Sleep Med Rev, 13, 309-321.
Diekelmann, S. and Born, J. (2010) The memory function of sleep. Nat Rev Neurosci,
11, 114-126.
Donlea, J.M., Ramanan, N., Shaw, P.J. (2009) Use-dependent plasticity in clock
neurons regulates sleep need in Drosophila. Science, 324, 105-108.
91
Csercsa Richárd
Elston, G.N. (2003) Cortex, cognition and the cell: new insights into the pyramidal
neuron and prefrontal function. Cereb Cortex, 13, 1124-1138.
Elston, G.N., Benavides-Piccione, R., Elston, A., Zietsch, B., Defelipe, J., Manger, P.,
Casagrande, V., Kaas, J.H. (2006) Specializations of the granular prefrontal
cortex of primates: implications for cognitive processing. Anat Rec A Discov
Mol Cell Evol Biol, 288, 26-35.
Entz, L., Fabo, D., Eross, L., Halasz, P., Wittner, L., Karmos, G., Ulbert, I. (2008)
Cortical electrical stimulation induced slow oscillation in different vigilance
states in humans. FENS Forum. Geneva, Switzerland.
Entz, L., Fabo, D., Eross, L., Halasz, P., Wittner, L., Csercsa, R., Dombovari, B.,
Karmos, G., Ulbert, I. (2009) Detection of slow oscillation elicited by cortical
electrical stimulation in different vigilance states in humans. Society for
Neuroscience. Chicago, USA.
Eriksson, K.S., Sergeeva, O., Brown, R.E., Haas, H.L. (2001) Orexin/hypocretin excites
the histaminergic neurons of the tuberomammillary nucleus. J Neurosci, 21,
9273-9279.
Euston, D.R., Tatsuno, M., McNaughton, B.L. (2007) Fast-forward playback of recent
memory sequences in prefrontal cortex during sleep. Science, 318, 1147-1150.
Fabo, D., Magloczky, Z., Wittner, L., Pek, A., Eross, L., Czirjak, S., Vajda, J., Solyom,
A., Rasonyi, G., Szucs, A., Kelemen, A., Juhos, V., Grand, L., Dombovari, B.,
Halasz, P., Freund, T.F., Halgren, E., Karmos, G., Ulbert, I. (2008) Properties of
in vivo interictal spike generation in the human subiculum. Brain, 131, 485-499.
Faraguna, U., Vyazovskiy, V.V., Nelson, A.B., Tononi, G., Cirelli, C. (2008) A causal
role for brain-derived neurotrophic factor in the homeostatic regulation of sleep.
J Neurosci, 28, 4088-4095.
Ferezou, I., Haiss, F., Gentet, L.J., Aronoff, R., Weber, B., Petersen, C.C. (2007)
Spatiotemporal dynamics of cortical sensorimotor integration in behaving mice.
Neuron, 56, 907-923.
Fonseca, M. and Soriano, E. (1995) Calretinin-immunoreactive neurons in the normal
human temporal cortex and in Alzheimer's disease. Brain Res, 691, 83-91.
92
Laminar analysis of the slow wave actvity in humans
Fort, P., Bassetti, C.L., Luppi, P.H. (2009) Alternating vigilance states: new insights
regarding neuronal networks and mechanisms. Eur J Neurosci, 29, 1741-1753.
Frank, M.G. and Benington, J.H. (2006) The role of sleep in memory consolidation and
brain plasticity: dream or reality? Neuroscientist, 12, 477-488.
Freeman, J.A. and Nicholson, C. (1975) Experimental optimization of current sourcedensity technique for anuran cerebellum. J Neurophysiol, 38, 369-382.
Gabbott, P.L., Jays, P.R., Bacon, S.J. (1997) Calretinin neurons in human medial
prefrontal cortex (areas 24a,b,c, 32', and 25). J Comp Neurol, 381, 389-410.
Gais, S., Plihal, W., Wagner, U., Born, J. (2000) Early sleep triggers memory for early
visual discrimination skills. Nat Neurosci, 3, 1335-1339.
Gais, S., Molle, M., Helms, K., Born, J. (2002) Learning-dependent increases in sleep
spindle density. J Neurosci, 22, 6830-6834.
Gao, L., Meng, X., Ye, C., Zhang, H., Liu, C., Dan, Y., Poo, M.M., He, J., Zhang, X.
(2009) Entrainment of slow oscillations of auditory thalamic neurons by
repetitive sound stimuli. J Neurosci, 29, 6013-6021.
Gilestro, G.F., Tononi, G., Cirelli, C. (2009) Widespread changes in synaptic markers as
a function of sleep and wakefulness in Drosophila. Science, 324, 109-112.
Girardeau, G., Benchenane, K., Wiener, S.I., Buzsaki, G., Zugaro, M.B. (2009)
Selective suppression of hippocampal ripples impairs spatial memory. Nat
Neurosci, 12, 1222-1223.
Gonzalez-Burgos, G., Barrionuevo, G., Lewis, D.A. (2000) Horizontal synaptic
connections in monkey prefrontal cortex: an in vitro electrophysiological study.
Cereb Cortex, 10, 82-92.
Greicius, M.D., Krasnow, B., Reiss, A.L., Menon, V. (2003) Functional connectivity in
the resting brain: a network analysis of the default mode hypothesis. Proc Natl
Acad Sci U S A, 100, 253-258.
Grenier, F., Timofeev, I., Steriade, M. (2003) Neocortical very fast oscillations (ripples,
80-200 Hz) during seizures: intracellular correlates. J Neurophysiol, 89, 841852.
93
Csercsa Richárd
Hahn, T.T., Sakmann, B., Mehta, M.R. (2006) Phase-locking of hippocampal
interneurons' membrane potential to neocortical up-down states. Nat Neurosci,
9, 1359-1361.
Halasz, P. (2005) K-complex, a reactive EEG graphoelement of NREM sleep: an old
chap in a new garment. Sleep Med Rev, 9, 391-412.
Halgren, E., Wang, C., Schomer, D.L., Knake, S., Marinkovic, K., Wu, J., Ulbert, I.
(2006) Processing stages underlying word recognition in the anteroventral
temporal lobe. Neuroimage, 30, 1401-1413.
Harris, K.D., Henze, D.A., Csicsvari, J., Hirase, H., Buzsaki, G. (2000) Accuracy of
tetrode spike separation as determined by simultaneous intracellular and
extracellular measurements. J Neurophysiol, 84, 401-414.
He, J. (2003) Slow oscillation in non-lemniscal auditory thalamus. J Neurosci, 23,
8281-8290.
Hebb, D.O. (1949) Organization of Behavior: a Neuropsychological Theory. John
Wiley & Sons.
Heitler, J.W. (2006) DataView v5: software for the display and analysis of digital
signals in neurophysiology.
Herculano-Houzel, S., Collins, C.E., Wong, P., Kaas, J.H. (2007) Cellular scaling rules
for primate brains. Proc Natl Acad Sci U S A, 104, 3562-3567.
Hill, S. and Tononi, G. (2005) Modeling sleep and wakefulness in the thalamocortical
system. J Neurophysiol, 93, 1671-1698.
Hoffman, K.L. and McNaughton, B.L. (2002) Coordinated reactivation of distributed
memory traces in primate neocortex. Science, 297, 2070-2073.
Hromadka, T., Deweese, M.R., Zador, A.M. (2008) Sparse representation of sounds in
the unanesthetized auditory cortex. PLoS Biol, 6, e16.
Huber, R., Ghilardi, M.F., Massimini, M., Tononi, G. (2004) Local sleep and learning.
Nature, 430, 78-81.
94
Laminar analysis of the slow wave actvity in humans
Huber, R., Ghilardi, M.F., Massimini, M., Ferrarelli, F., Riedner, B.A., Peterson, M.J.,
Tononi, G. (2006) Arm immobilization causes cortical plastic changes and
locally decreases sleep slow wave activity. Nat Neurosci, 9, 1169-1176.
Iber, C., Ancoli-Israel, S., Chesson, A., Quan, S.F. (2007) The AASM Manual for the
Scoring of Sleep and Associated Events: Rules, Terminology and Technical
Specifications. American Academy of Sleep Medicine, Westchester.
Isomura, Y., Sirota, A., Ozen, S., Montgomery, S., Mizuseki, K., Henze, D.A., Buzsaki,
G. (2006) Integration and segregation of activity in entorhinal-hippocampal
subregions by neocortical slow oscillations. Neuron, 52, 871-882.
Jenkins, J.G. and Dallenbach, K.M. (1924) Obliviscence during sleep and waking.
American Journal of Psychology, 35, 605-612.
Ji, D. and Wilson, M.A. (2007) Coordinated memory replay in the visual cortex and
hippocampus during sleep. Nat Neurosci, 10, 100-107.
Kali, S. and Dayan, P. (2004) Off-line replay maintains declarative memories in a
model of hippocampal-neocortical interactions. Nat Neurosci, 7, 286-294.
Kattler, H., Dijk, D.J., Borbely, A.A. (1994) Effect of unilateral somatosensory
stimulation prior to sleep on the sleep EEG in humans. J Sleep Res, 3, 159-164.
Keller, C.J., Cash, S.S., Narayanan, S., Wang, C., Kuzniecky, R., Carlson, C.,
Devinsky, O., Thesen, T., Doyle, W., Sassaroli, A., Boas, D.A., Ulbert, I.,
Halgren, E. (2009) Intracranial microprobe for evaluating neuro-hemodynamic
coupling in unanesthetized human neocortex. J Neurosci Methods, 179, 208218.
Knake, S., Wang, C.M., Ulbert, I., Schomer, D.L., Halgren, E. (2007) Specific increase
of human entorhinal population synaptic and neuronal activity during retrieval.
Neuroimage, 37, 618-622.
Krueger, J.M. and Obal, F. (1993) A neuronal group theory of sleep function. J Sleep
Res, 2, 63-69.
Krueger, J.M. (2008) The role of cytokines in sleep regulation. Curr Pharm Des, 14,
3408-3416.
95
Csercsa Richárd
Krueger, J.M., Rector, D.M., Roy, S., Van Dongen, H.P., Belenky, G., Panksepp, J.
(2008) Sleep as a fundamental property of neuronal assemblies. Nat Rev
Neurosci, 9, 910-919.
Kurth, S., Ringli, M., Geiger, A., LeBourgeois, M., Jenni, O.G., Huber, R. (2010)
Mapping of cortical activity in the first two decades of life: a high-density sleep
electroencephalogram study. J Neurosci, 30, 13211-13219.
Lakatos, P., Shah, A.S., Knuth, K.H., Ulbert, I., Karmos, G., Schroeder, C.E. (2005) An
oscillatory hierarchy controlling neuronal excitability and stimulus processing in
the auditory cortex. J Neurophysiol, 94, 1904-1911.
Lee, A.K. and Wilson, M.A. (2002) Memory of sequential experience in the
hippocampus during slow wave sleep. Neuron, 36, 1183-1194.
Li, C.Y., Poo, M.M., Dan, Y. (2009) Burst spiking of a single cortical neuron modifies
global brain state. Science, 324, 643-646.
Liu, Z.W., Faraguna, U., Cirelli, C., Tononi, G., Gao, X.B. (2010) Direct evidence for
wake-related increases and sleep-related decreases in synaptic strength in rodent
cortex. J Neurosci, 30, 8671-8675.
Lomo, T. (1966) Frequency potentiation of excitatory synaptic activity in the dentate
area of the hippocampal formation. Acta Physiologica Scandinavia, 68, 128.
Loomis, A.L., Harvey, E.N., Hobart, G. (1938) Distribution of disturbance-patterns in
the human electroencephalogram, with special reference to sleep. J
Neurophysiol, 1, 413-430.
Lopez-Bendito, G., Shigemoto, R., Kulik, A., Paulsen, O., Fairen, A., Lujan, R. (2002)
Expression and distribution of metabotropic GABA receptor subtypes
GABABR1 and GABABR2 during rat neocortical development. Eur J Neurosci,
15, 1766-1778.
Luczak, A., Bartho, P., Marguet, S.L., Buzsaki, G., Harris, K.D. (2007) Sequential
structure of neocortical spontaneous activity in vivo. Proc Natl Acad Sci U S A,
104, 347-352.
Lynch, M.A. (2004) Long-term potentiation and memory. Physiol Rev, 84, 87-136.
96
Laminar analysis of the slow wave actvity in humans
Mahowald, M.W. and Schenck, C.H. (2005) Insights from studying human sleep
disorders. Nature, 437, 1279-1285.
Mann, E.O., Kohl, M.M., Paulsen, O. (2009) Distinct roles of GABA(A) and GABA(B)
receptors in balancing and terminating persistent cortical activity. J Neurosci,
29, 7513-7518.
Mao, B.Q., Hamzei-Sichani, F., Aronov, D., Froemke, R.C., Yuste, R. (2001) Dynamics
of spontaneous activity in neocortical slices. Neuron, 32, 883-898.
Maquet, P. (2001) The role of sleep in learning and memory. Science, 294, 1048-1052.
Margrie, T.W., Brecht, M., Sakmann, B. (2002) In vivo, low-resistance, whole-cell
recordings from neurons in the anaesthetized and awake mammalian brain.
Pflugers Arch, 444, 491-498.
Marr, D. (1970) A theory for cerebral neocortex. Proc R Soc Lond B Biol Sci, 176, 161234.
Marr, D. (1971) Simple memory: a theory for archicortex. Philos Trans R Soc Lond B
Biol Sci, 262, 23-81.
Marshall, L., Helgadottir, H., Molle, M., Born, J. (2006) Boosting slow oscillations
during sleep potentiates memory. Nature, 444, 610-613.
Marshall, L. and Born, J. (2007) The contribution of sleep to hippocampus-dependent
memory consolidation. Trends Cogn Sci, 11, 442-450.
Massimini, M., Huber, R., Ferrarelli, F., Hill, S., Tononi, G. (2004) The sleep slow
oscillation as a traveling wave. J Neurosci, 24, 6862-6870.
Massimini, M., Ferrarelli, F., Huber, R., Esser, S.K., Singh, H., Tononi, G. (2005)
Breakdown of cortical effective connectivity during sleep. Science, 309, 22282232.
Massimini, M., Ferrarelli, F., Esser, S.K., Riedner, B.A., Huber, R., Murphy, M.,
Peterson, M.J., Tononi, G. (2007) Triggering sleep slow waves by transcranial
magnetic stimulation. Proc Natl Acad Sci U S A, 104, 8496-8501.
97
Csercsa Richárd
Maviel, T., Durkin, T.P., Menzaghi, F., Bontempi, B. (2004) Sites of neocortical
reorganization critical for remote spatial memory. Science, 305, 96-99.
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.
Psychol Rev, 102, 419-457.
Meskenaite, V. (1997) Calretinin-immunoreactive local circuit neurons in area 17 of the
cynomolgus monkey, Macaca fascicularis. J Comp Neurol, 379, 113-132.
Mohajerani, M.H., McVea, D.A., Fingas, M., Murphy, T.H. (2010) Mirrored bilateral
slow-wave cortical activity within local circuits revealed by fast bihemispheric
voltage-sensitive dye imaging in anesthetized and awake mice. J Neurosci, 30,
3745-3751.
Molle, M., Marshall, L., Gais, S., Born, J. (2002) Grouping of spindle activity during
slow oscillations in human non-rapid eye movement sleep. J Neurosci, 22,
10941-10947.
Molle, M., Marshall, L., Gais, S., Born, J. (2004) Learning increases human
electroencephalographic coherence during subsequent slow sleep oscillations.
Proc Natl Acad Sci U S A, 101, 13963-13968.
Molle, M., Yeshenko, O., Marshall, L., Sara, S.J., Born, J. (2006) Hippocampal sharp
wave-ripples linked to slow oscillations in rat slow-wave sleep. J Neurophysiol,
96, 62-70.
Moroni, F., Nobili, L., Curcio, G., De Carli, F., Fratello, F., Marzano, C., De Gennaro,
L., Ferrillo, F., Cossu, M., Francione, S., Lo Russo, G., Bertini, M., Ferrara, M.
(2007) Sleep in the human hippocampus: a stereo-EEG study. PLoS One, 2,
e867.
Moruzzi, G. and Magoun, H.W. (1949) Brain stem reticular formation and activation of
the EEG. Electroencephalogr Clin Neurophysiol, 1, 455-473.
Mukhametov, L.M., Supin, A.Y., Polyakova, I.G. (1977) Interhemispheric asymmetry
of the electroencephalographic sleep patterns in dolphins. Brain Res, 134, 581584.
98
Laminar analysis of the slow wave actvity in humans
Murphy, M., Riedner, B.A., Huber, R., Massimini, M., Ferrarelli, F., Tononi, G. (2009)
Source modeling sleep slow waves. Proc Natl Acad Sci U S A, 106, 1608-1613.
Nadasdy, Z., Hirase, H., Czurko, A., Csicsvari, J., Buzsaki, G. (1999) Replay and time
compression of recurring spike sequences in the hippocampus. J Neurosci, 19,
9497-9507.
Nadel, L., Samsonovich, A., Ryan, L., Moscovitch, M. (2000) Multiple trace theory of
human memory: computational, neuroimaging, and neuropsychological results.
Hippocampus, 10, 352-368.
Nauta, W.J. (1946) Hypothalamic regulation of sleep in rats; an experimental study. J
Neurophysiol, 9, 285-316.
Nemeth, D., Janacsek, K., Londe, Z., Ullman, M.T., Howard, D.V., Howard, J.H., Jr.
(2010) Sleep has no critical role in implicit motor sequence learning in young
and old adults. Exp Brain Res, 201, 351-358.
Nevian, T. and Sakmann, B. (2006) Spine Ca2+ signaling in spike-timing-dependent
plasticity. J Neurosci, 26, 11001-11013.
Nicholson, C. and Freeman, J.A. (1975) Theory of current source-density analysis and
determination of conductivity tensor for anuran cerebellum. J Neurophysiol, 38,
356-368.
Nishino, S., Ripley, B., Overeem, S., Lammers, G.J., Mignot, E. (2000) Hypocretin
(orexin) deficiency in human narcolepsy. Lancet, 355, 39-40.
Olcese, U., Esser, S.K., Tononi, G. (2010) Sleep and synaptic renormalization: A
computational study. J Neurophysiol.
Ozen, S., Sirota, A., Belluscio, M.A., Anastassiou, C.A., Stark, E., Koch, C., Buzsaki,
G. (2010) Transcranial electric stimulation entrains cortical neuronal populations
in rats. J Neurosci, 30, 11476-11485.
Pace-Schott, E.F. and Hobson, J.A. (2002) The neurobiology of sleep: genetics, cellular
physiology and subcortical networks. Nat Rev Neurosci, 3, 591-605.
99
Csercsa Richárd
Pappenheimer, J.R., Miller, T.B., Goodrich, C.A. (1967) Sleep-promoting effects of
cerebrospinal fluid from sleep-deprived goats. Proc Natl Acad Sci U S A, 58,
513-517.
Pastalkova, E., Itskov, V., Amarasingham, A., Buzsaki, G. (2008) Internally generated
cell assembly sequences in the rat hippocampus. Science, 321, 1322-1327.
Petersen, C.C., Malenka, R.C., Nicoll, R.A., Hopfield, J.J. (1998) All-or-none
potentiation at CA3-CA1 synapses. Proc Natl Acad Sci U S A, 95, 4732-4737.
Petersen, C.C., Hahn, T.T., Mehta, M., Grinvald, A., Sakmann, B. (2003) Interaction of
sensory responses with spontaneous depolarization in layer 2/3 barrel cortex.
Proc Natl Acad Sci U S A, 100, 13638-13643.
Peyrache, A., Khamassi, M., Benchenane, K., Wiener, S.I., Battaglia, F.P. (2009)
Replay of rule-learning related neural patterns in the prefrontal cortex during
sleep. Nat Neurosci, 12, 919-926.
Quiroga, R.Q., Reddy, L., Kreiman, G., Koch, C., Fried, I. (2005) Invariant visual
representation by single neurons in the human brain. Nature, 435, 1102-1107.
Raichle, M.E., MacLeod, A.M., Snyder, A.Z., Powers, W.J., Gusnard, D.A., Shulman,
G.L. (2001) A default mode of brain function. Proc Natl Acad Sci U S A, 98,
676-682.
Raichle, M.E. (2006) Neuroscience. The brain's dark energy. Science, 314, 1249-1250.
Raichle, M.E. and Snyder, A.Z. (2007) A default mode of brain function: a brief history
of an evolving idea. Neuroimage, 37, 1083-1090; discussion 1097-1089.
Rasch, B., Buchel, C., Gais, S., Born, J. (2007) Odor cues during slow-wave sleep
prompt declarative memory consolidation. Science, 315, 1426-1429.
Rasch, B., Pommer, J., Diekelmann, S., Born, J. (2009) Pharmacological REM sleep
suppression paradoxically improves rather than impairs skill memory. Nat
Neurosci, 12, 396-397.
Ravagnati, L., Halgren, E., Babb, T.L., Crandall, P.H. (1979) Activity of human
hippocampal formation and amygdala neurons during sleep. Sleep, 2, 161-173.
100
Laminar analysis of the slow wave actvity in humans
Rector, D.M., Topchiy, I.A., Carter, K.M., Rojas, M.J. (2005) Local functional state
differences between rat cortical columns. Brain Res, 1047, 45-55.
Riedner, B.A., Vyazovskiy, V.V., Huber, R., Massimini, M., Esser, S., Murphy, M.,
Tononi, G. (2007) Sleep homeostasis and cortical synchronization: III. A highdensity EEG study of sleep slow waves in humans. Sleep, 30, 1643-1657.
Ros, H., Sachdev, R.N., Yu, Y., Sestan, N., McCormick, D.A. (2009) Neocortical
networks entrain neuronal circuits in cerebellar cortex. J Neurosci, 29, 1030910320.
Rosanova, M. and Ulrich, D. (2005) Pattern-specific associative long-term potentiation
induced by a sleep spindle-related spike train. J Neurosci, 25, 9398-9405.
Sakata, S. and Harris, K.D. (2009) Laminar structure of spontaneous and sensoryevoked population activity in auditory cortex. Neuron, 64, 404-418.
Sanchez-Vives, M.V. and McCormick, D.A. (2000) Cellular and network mechanisms
of rhythmic recurrent activity in neocortex. Nat Neurosci, 3, 1027-1034.
Saper, C.B., Chou, T.C., Scammell, T.E. (2001) The sleep switch: hypothalamic control
of sleep and wakefulness. Trends Neurosci, 24, 726-731.
Schwark, H.D. and Li, J. (2000) Distribution of neurons immunoreactive for calciumbinding proteins varies across areas of cat primary somatosensory cortex. Brain
Res Bull, 51, 379-385.
Scoville, W.B. and Milner, B. (1957) Loss of recent memory after bilateral hippocampal
lesions. J Neurol Neurosurg Psychiatry, 20, 11-21.
Sherin, J.E., Shiromani, P.J., McCarley, R.W., Saper, C.B. (1996) Activation of
ventrolateral preoptic neurons during sleep. Science, 271, 216-219.
Siapas, A.G. and Wilson, M.A. (1998) Coordinated interactions between hippocampal
ripples and cortical spindles during slow-wave sleep. Neuron, 21, 1123-1128.
Siegel, J. (2002) The Neural Control of Sleep and Waking. Springer-Verlag, New York.
101
Csercsa Richárd
Siegel, J.M. (2001) The REM sleep-memory consolidation hypothesis. Science, 294,
1058-1063.
Siegel, J.M. (2005) Clues to the functions of mammalian sleep. Nature, 437, 1264-1271.
Siegel, J.M. (2008) Do all animals sleep? Trends Neurosci, 31, 208-213.
Somogyi, P., Kisvarday, Z.F., Martin, K.A., Whitteridge, D. (1983) Synaptic
connections of morphologically identified and physiologically characterized
large basket cells in the striate cortex of cat. Neuroscience, 10, 261-294.
Song, S., Howard, J.H., Jr., Howard, D.V. (2007) Sleep does not benefit probabilistic
motor sequence learning. J Neurosci, 27, 12475-12483.
Squire, L.R. (1992) Memory and the hippocampus: a synthesis from findings with rats,
monkeys, and humans. Psychol Rev, 99, 195-231.
Staba, R.J., Wilson, C.L., Bragin, A., Fried, I., Engel, J., Jr. (2002) Sleep states
differentiate single neuron activity recorded from human epileptic hippocampus,
entorhinal cortex, and subiculum. J Neurosci, 22, 5694-5704.
Staba, R.J., Wilson, C.L., Fried, I., Engel, J., Jr. (2002) Single neuron burst firing in the
human hippocampus during sleep. Hippocampus, 12, 724-734.
Steinvorth, S., Wang, C., Ulbert, I., Schomer, D., Halgren, E. (2010) Human entorhinal
gamma and theta oscillations selective for remote autobiographical memory.
Hippocampus, 20, 166-173.
Stellwagen, D. and Malenka, R.C. (2006) Synaptic scaling mediated by glial TNFalpha. Nature, 440, 1054-1059.
Steriade, M., Nunez, A., Amzica, F. (1993a) A novel slow (< 1 Hz) oscillation of
neocortical neurons in vivo: depolarizing and hyperpolarizing components. J
Neurosci, 13, 3252-3265.
Steriade, M., Nunez, A., Amzica, F. (1993b) Intracellular analysis of relations between
the slow (< 1 Hz) neocortical oscillation and other sleep rhythms of the
electroencephalogram. J Neurosci, 13, 3266-3283.
102
Laminar analysis of the slow wave actvity in humans
Steriade, M. and Amzica, F. (1996) Intracortical and corticothalamic coherency of fast
spontaneous oscillations. Proc Natl Acad Sci U S A, 93, 2533-2538.
Steriade, M., Timofeev, I., Grenier, F. (2001) Natural waking and sleep states: a view
from inside neocortical neurons. J Neurophysiol, 85, 1969-1985.
Steriade, M. and Timofeev, I. (2003) Neuronal plasticity in thalamocortical networks
during sleep and waking oscillations. Neuron, 37, 563-576.
Steriade, M. (2006) Grouping of brain rhythms in corticothalamic systems.
Neuroscience, 137, 1087-1106.
Stickgold, R. (2005) Sleep-dependent memory consolidation. Nature, 437, 1272-1278.
Takashima, A., Petersson, K.M., Rutters, F., Tendolkar, I., Jensen, O., Zwarts, M.J.,
McNaughton, B.L., Fernandez, G. (2006) Declarative memory consolidation in
humans: a prospective functional magnetic resonance imaging study. Proc Natl
Acad Sci U S A, 103, 756-761.
Tamas, G., Somogyi, P., Buhl, E.H. (1998) Differentially interconnected networks of
GABAergic interneurons in the visual cortex of the cat. J Neurosci, 18, 42554270.
Tamas, G., Lorincz, A., Simon, A., Szabadics, J. (2003) Identified sources and targets of
slow inhibition in the neocortex. Science, 299, 1902-1905.
Tenke, C.E., Schroeder, C.E., Arezzo, J.C., Vaughan, H.G., Jr. (1993) Interpretation of
high-resolution current source density profiles: a simulation of sublaminar
contributions to the visual evoked potential. Exp Brain Res, 94, 183-192.
Timofeev, I. and Steriade, M. (1996) Low-frequency rhythms in the thalamus of intactcortex and decorticated cats. J Neurophysiol, 76, 4152-4168.
Timofeev, I., Grenier, F., Bazhenov, M., Sejnowski, T.J., Steriade, M. (2000) Origin of
slow cortical oscillations in deafferented cortical slabs. Cereb Cortex, 10, 11851199.
Timofeev, I., Grenier, F., Steriade, M. (2001) Disfacilitation and active inhibition in the
neocortex during the natural sleep-wake cycle: an intracellular study. Proc Natl
Acad Sci U S A, 98, 1924-1929.
103
Csercsa Richárd
Tononi, G. and Cirelli, C. (2006) Sleep function and synaptic homeostasis. Sleep Med
Rev, 10, 49-62.
Triarhou, L.C. (2006) The percipient observations of Constantin von Economo on
encephalitis lethargica and sleep disruption and their lasting impact on
contemporary sleep research. Brain Res Bull, 69, 244-258.
Tseng, K.Y., Kasanetz, F., Kargieman, L., Riquelme, L.A., Murer, M.G. (2001) Cortical
slow oscillatory activity is reflected in the membrane potential and spike trains
of striatal neurons in rats with chronic nigrostriatal lesions. J Neurosci, 21,
6430-6439.
Tucker, M.A. and Fishbein, W. (2009) The impact of sleep duration and subject
intelligence on declarative and motor memory performance: how much is
enough? J Sleep Res, 18, 304-312.
Turner, D.A., Li, X.G., Pyapali, G.K., Ylinen, A., Buzsaki, G. (1995) Morphometric
and electrical properties of reconstructed hippocampal CA3 neurons recorded in
vivo. J Comp Neurol, 356, 580-594.
Ulbert, I., Halgren, E., Heit, G., Karmos, G. (2001) Multiple microelectrode-recording
system for human intracortical applications. J Neurosci Methods, 106, 69-79.
Ulbert, I., Karmos, G., Heit, G., Halgren, E. (2001) Early discrimination of coherent
versus incoherent motion by multiunit and synaptic activity in human putative
MT+. Hum Brain Mapp, 13, 226-238.
Ulbert, I., Heit, G., Madsen, J., Karmos, G., Halgren, E. (2004) Laminar analysis of
human neocortical interictal spike generation and propagation: current source
density and multiunit analysis in vivo. Epilepsia, 45 Suppl 4, 48-56.
Ulbert, I., Magloczky, Z., Eross, L., Czirjak, S., Vajda, J., Bognar, L., Toth, S., Szabo,
Z., Halasz, P., Fabo, D., Halgren, E., Freund, T.F., Karmos, G. (2004) In vivo
laminar electrophysiology co-registered with histology in the hippocampus of
patients with temporal lobe epilepsy. Exp Neurol, 187, 310-318.
Vanhatalo, S., Palva, J.M., Holmes, M.D., Miller, J.W., Voipio, J., Kaila, K. (2004)
Infraslow oscillations modulate excitability and interictal epileptic activity in the
human cortex during sleep. Proc Natl Acad Sci U S A, 101, 5053-5057.
104
Laminar analysis of the slow wave actvity in humans
Velluti, R.A. (2008) The Auditory System in Sleep. Elsevier Academic Press.
Vertes, R.P. and Siegel, J.M. (2005) Time for the sleep community to take a critical
look at the purported role of sleep in memory processing. Sleep, 28, 1228-1229;
discussion 1230-1223.
Volgushev, M., Chauvette, S., Mukovski, M., Timofeev, I. (2006) Precise long-range
synchronization of activity and silence in neocortical neurons during slow-wave
oscillations [corrected]. J Neurosci, 26, 5665-5672.
von Economo, C. (1929) Schlaftheorie. Ergebn. Physiol. (München), 312-339.
Vyazovskiy, V., Borbely, A.A., Tobler, I. (2000) Unilateral vibrissae stimulation during
waking induces interhemispheric EEG asymmetry during subsequent sleep in
the rat. J Sleep Res, 9, 367-371.
Vyazovskiy, V.V., Cirelli, C., Pfister-Genskow, M., Faraguna, U., Tononi, G. (2008)
Molecular and electrophysiological evidence for net synaptic potentiation in
wake and depression in sleep. Nat Neurosci, 11, 200-208.
Vyazovskiy, V.V., Faraguna, U., Cirelli, C., Tononi, G. (2009) Triggering slow waves
during NREM sleep in the rat by intracortical electrical stimulation: effects of
sleep/wake history and background activity. J Neurophysiol, 101, 1921-1931.
Wagner, T., Axmacher, N., Lehnertz, K., Elger, C.E., Fell, J. (2010) Sleep-dependent
directional coupling between human neocortex and hippocampus. Cortex, 46,
256-263.
Walker, M.P. and Stickgold, R. (2006) Sleep, memory, and plasticity. Annu Rev
Psychol, 57, 139-166.
Wang, C., Ulbert, I., Schomer, D.L., Marinkovic, K., Halgren, E. (2005) Responses of
human anterior cingulate cortex microdomains to error detection, conflict
monitoring, stimulus-response mapping, familiarity, and orienting. J Neurosci,
25, 604-613.
Waters, J. and Helmchen, F. (2006) Background synaptic activity is sparse in neocortex.
J Neurosci, 26, 8267-8277.
105
Csercsa Richárd
Whittington, M.A., Traub, R.D., Jefferys, J.G. (1995) Synchronized oscillations in
interneuron networks driven by metabotropic glutamate receptor activation.
Nature, 373, 612-615.
Wierzynski, C.M., Lubenov, E.V., Gu, M., Siapas, A.G. (2009) State-dependent spiketiming relationships between hippocampal and prefrontal circuits during sleep.
Neuron, 61, 587-596.
Wilson, M.A. and McNaughton, B.L. (1994) Reactivation of hippocampal ensemble
memories during sleep. Science, 265, 676-679.
Wittner, L., Henze, D.A., Zaborszky, L., Buzsaki, G. (2006) Hippocampal CA3
pyramidal cells selectively innervate aspiny interneurons. Eur J Neurosci, 24,
1286-1298.
Wittner, L., Huberfeld, G., Clemenceau, S., Eross, L., Dezamis, E., Entz, L., Ulbert, I.,
Baulac, M., Freund, T.F., Magloczky, Z., Miles, R. (2009) The epileptic human
hippocampal cornu ammonis 2 region generates spontaneous interictal-like
activity in vitro. Brain, 132, 3032-3046.
Wyrwicka, W., Sterman, M.B., Clemente, C.D. (1962) Conditioning of induced
electroencephalographic sleep patterns in the cat. Science, 137, 616-618.
Yoshimura, Y., Dantzker, J.L., Callaway, E.M. (2005) Excitatory cortical neurons form
fine-scale functional networks. Nature, 433, 868-873.
Zola-Morgan, S.M. and Squire, L.R. (1990) The primate hippocampal formation:
evidence for a time-limited role in memory storage. Science, 250, 288-290.
106
Laminar analysis of the slow wave actvity in humans
PUBLICATIONS
Relevant to thesis
’Laminar analysis of the slow wave activity in humans’,
Csercsa R., Dombovári B., Fabó D., Wittner L., Erőss L., Entz L., Sólyom A., Rásonyi
Gy., Szűcs A., Kelemen A., Jakus R., Juhos V., Grand L., Magony A., Halász P.,
Freund T. F., Maglóczky Zs., Cash S. S., Papp L., Karmos Gy., Halgren E., Ulbert I.,
Brain, 133(9):2514-5., 2010., IF: 9.490
’The human K-complex represents an isolated cortical down-state',
Cash S. S., Halgren E., Dehghani N., Rossetti A., Thesen T., Wang C., Devinsky O.,
Kuzniecky R., Doyle W., Madsen J. R., Bromfield E., Erőss L., Halász P., Karmos Gy.,
Csercsa R., Wittner L., Ulbert I.,
Science, 324(5930):1084-7., 2009., IF: 29.747
Non-relevant to thesis
’Short and long term biocompatibility of NeuroProbes silicon probes’,
Grand L., Wittner L., Herwik S., Göthelid E., Ruther P., Oscarsson S., Neves H.,
Dombovári B., Csercsa R., Karmos G., Ulbert I.,
J Neurosci Methods, 189(2):216-29., 2010., IF: 2.295
’Control and data acquisition software for high-density CMOS-based microprobe
arrays implementing electronic depth control’,
Seidl K., Torfs T., De Mazière P. A., Van Dijck G., Csercsa R., Dombovári B.,
Nurcahyo Y., Ramirez H., Van Hulle M. M., Orban G. A., Paul O., Ulbert I., Neves H.,
Ruther P.,
Biomedizinische Technik, 55(3):183-91., 2010., IF: 0.525
107
Csercsa Richárd
ACKNOWLEDGEMENTS
First, I would like to thank my supervisor, Dr. István Ulbert for his guidance
throughout my undergraduate and graduate years, for his inexhaustable enthusiasm and
support in all aspects of work.
I am much obliged to Professor György Karmos for his outstanding lectures that
directed my interest to neuroscience, and for his sponsorship.
I thank Balázs Dombovári, László Grand, and Andor Magony for their friendly
support professionally as well as personally. I also thank Richárd Fiáth, Domonkos
Horváth, Bálint Kerekes, Lucia Wittner, and all of my co-workers at the Institute for
Psychology for the good atmosphere and inspiring collaboration. I learned a lot from
their interesting experiments and the interdisciplinary topics of our conversations.
I owe Dr. Loránd Erőss, Dr. László Entz, and Dr. Dániel Fabó for their
invaluable cooperation and for providing the data for my work, and Dr. Zsófia
Maglóczky for the histology analysis.
I would like to express my thanks to Péter Kottra for his assistance in the lab.
I thank Dr. István Czigler, the director of the Institute for Psychology of the
Hungarian Academy of Sciences for providing the facilities for my work.
I am grateful to Dr. Patrick Ruther, Professor Oliver Paul, and Karsten Seidl of
the Laboratory for Microsystem Materials of the Institute for Mikrosystemtechnik
(IMTEK) in Freiburg, Germany, for having me at their institute and giving me the
opportunity to get acquainted with electrode design and fabrication.
Last but not least, I would like to thank my family for their encouragement and
support.
108