User:Kinkreet/What is the purpose of sleep?v6

For the survival of the organism
Can sleep serve, perhaps, the most important function of life? That is, to keep an organism alive? A landmark study by Allan Rechtschaffen, published in the journal Science in 1982, is one of the first to scientifically demonstrate the requirement of animals for sleep. The idea that sleep might be essential is not a novel concept, and researchers before have conducted many experiments to prove the fatal effect of prolonged sleep deprivation. But Rechtschaffen's study is the first to control for the physical stresses induced by the methods used.

A classic study
In his experiments, a pair of rats were placed on an elevated circular platform, which can be rotated; below this platform is a pool of water 3cm in depth. But although they share the same platform, they are physically separated by housing each of the pair in separate Plexiglas cages. Electroencephalography (EEG) and electromyography (EMG) are used to monitor the sleep stage of the rats. One of the rats are then assigned as the 'deprived' rat, and this rat will undergo total sleep deprivation; and the other will be the 'control' rat, accounting for the physical stresses induced by the methods used.

When the EEG and EMG indicates that the 'deprived' rat is at the onset of sleep, the circular platform will rotate in an random direction, and both the rats must respond to prevent it from falling off the platform and into the water. However, when the same indications are observed in the 'control' rat, it did not prompt the platform to rotate, and the control is allowed to sleep. By this design, although the sleep of the control rat will be disturbed, they will receive more sleep than the deprived rat. The control rat may sleep while the deprived rat is awake, but the deprived rat is always prevented from sleeping.

Thus, in practice, this experiment tested for the effect of severe sleep deprivation over moderate sleep deprivation. Since both rats share the same platform, their physical exertion would be about the same; furthermore, on average rats are forced to walk only 0.9 miles a day, much less than the 30 miles per day they would voluntarily run on a wheel.

Eight pairs of age-matched Sprague-Dawley male rats were used in this experiment. The results are resounding - three deprived rats died after 5, 13 and 33 days, and those remaining showed "obvious pathological signs". Contrast this with control rats, who had no fatalities during the experiment and only some showed mild pathologies. This study shows that sleep is essential and the lack of it can lead to early onset of pathologies and death.

Furthermore, although the experiments were designed to give total sleep deprivation, on average ~1.7 hours of sleep slipped in due to apparatus failures and other technical difficulties. One begs the question of the effect of true total sleep deprivation on animals, which one can reasonably predict will bring about more severe pathologies and death, at a shorter time frame.

Not so quick
It appears from the above study, we can already conclude that sleep is essential for the health, and ultimately, the survival of an organism. But unfortunately, it's not that simple. This is because there are animals which apparently do not sleep at all. Examples include the American bullfrog (Rana catesbiana), the Loggerhead sea turtle (Caretta careta), the yellow-footed tortoise (Chelonoidis denticulata), the American alligator (Alligator mississipiensis) and Dall's porpoise (Phocoenoides dalli). Thus, it can argued, if these animals do not need sleep and can still survive, then sleep is not essential for survival.

A possible criticism of these studies are the small sample size (ten or less), although the same can be said for the Rechtschaffen experiments. Even though these organisms have not been extensively studied, there are already reports that they may sleep after all, although in a different manner. For example, the Loggerhead sea turtle was observed to be in a sleep-like state of immobility and reduced responsiveness to stimuli.

Note the term "sleep-like state", one which is used a lot in sleep-studies, and it might be the reason why no conclusions are drawn as to whether sleep is essential or not. Studies which supports the essential function of sleep for survival are mostly conducted on mammals, and those that opposes this hypothesis are conducted on reptiles, amphibians and invertebrates. Clearly, we must establish a clearer definition of what we mean by sleep before we can carry this discussion further.

Lack of definition
In humans, sleep can be characterized in different ways - behaviourally, it is the reversible state of immobility and increased arousal threshold; electrophysiologically, it is the presence of characteristic waveforms in the brain during non-rapid eye movement (NREM) and rapid eye movement (REM) sleep, as well as the spatio-temporal distinction in brain activity.

For other animals, including invertebrates, a rough determination of sleep can be given by quiescence, and an increased arousal threshold.

Taking this to the extreme, researchers have even used the term 'sleep' in describing a "desiccated and dormant" state in maize (Arabidopsis thaliana), a state which helps the plant survive through unfavourable conditions. So in this sense, 'dormancy' might have been a more suitable word. But perhaps a more accurate definition of 'sleep' in plants would be a reduction in plant perception - the ability of plants to sense and respond to stimuli (similar to 'responsiveness' in animals).

This calls for sleep researchers to come together and decide on an agreed nomenclature that can span across phylogeny. If 'sleep' is simply immobility alongside reduced responsiveness, then how would this be different to a state of dormancy (such as hibernation, torpor, aestivation, cryptobiosis, and brumation)? If one researcher using one definition of sleep and argues for its essential function, and another using a different definition argues otherwise, there will never be an agreement, for they are not even talking about the same thing.

Interim conclusions
Since there is no clear-cut definition for sleep, we cannot say whether sleep is essential in all organisms. We can, however, assert that sleep is very important in maintaining health and long-term survival of humans, as apparent by our own experiences. We will now explore exploring the non-essential functions of sleep in mammals, and discover what it is that our body is unable to do during wake.

For learning and memory consolidation
One of the distinguishing abilities of humans compared to all other organisms on Earth, is our ability to carry out complex cognitive functions. This requires a process of learning and retaining memories. It has been shown in numerous studies that sleep is able to make memory more stable (easier to access and is retained longer) after sleep; there are even claims that sleep enhances learning by forming hidden links between memories to reveal insights. Thus, a possible function of sleep in humans, is to facilitate learning and memory.

Collectively, memory stabilization and enhancement constitute the principles of the process known as memory consolidation. This hypothesis relies on the notion that memories are not formed as a single event, but rather are processed through different stages. First, there is the acquisition phase, where the experience is encoded somehow into a form the brain can process and store. Then the memory is processed further, primarily through the process of consolidation. The process of consolidation is further subdivided into stabilization, where the memory is prevented from interference (from new experiences) or decay (or become forgotten), followed by enhancement, which takes the stabilized memories and improve its recollection and reveal hidden patterns, beyond the time of acquisition.

Definitions
To anyone that has read up on the study of sleep and memory, I'm sure you'll agree that the general consensus is that different stages of sleep, in different areas of the brain, are responsible for consolidating different types of memory. Thus, I will now introduce these differences.

First, sleep is not homogeneous - there are different types of sleep. In mammals, sleep is categorized into two broad categories - rapid eye movement (REM) and non-rapid eye movement (NREM) sleep. In humans, NREM sleep is further subdivided into NREM-1 to NREM-4, with the former representing the 'lightest' sleep (most easily aroused), and the latter the 'deepest' sleep. NREM-3 and NREM-4 are characterized by EEG readings of long, synchronous waves of large amplitude and low frequency (0.5 - 4 Hz); and REM sleep is characterized by desynchronous waves of higher frequencies. Human goes through a few alternating cycles of REM and NREM sleep per night, each lasting around 90 minutes. Normally, a night's sleep will consists of four REM stages, each one longer than the subsequent one. Although the length of each cycle remains roughly the same, NREM-3 and NREM-4 dominates in the early stages whereas NREM-2 and REM sleep dominate the latter periods of the night.

Like sleep, memory is not a homogeneous state, and is commonly categorized into declarative and procedural/non-declarative memory. Declarative memory describes memory which are can be consciously accessed (fact-based), such as personal experiences (episodic memory) and facts (semantic memory); procedural memory describes things which are unconscious, such as learning a skill, or understanding the grammar behind a language.

Consolidation of declarative memory have often reached mixed conclusions. While both types of memory have been shown to be consolidated after sleep,  the evidence for procedural memory consolidation is stronger.

Sleep improves procedural memory
If sleep is able to stabilize memory and enhance learning, then people who sleep after learning should perform better than those that don't, and this effect should last for a significant amount of time, beyond the physical strain and attention deficits induced by sleep deprivation.

Brawn et al. proposed to test this by testing the performance of subjects in playing first-person shooter games (FPS) after a night of sleep, or sleep-deprivation. Brawn chose subjects which is relatively unexposed to FPS and argues that for these subjects, learning the effect of each weapon, the hand-eye coordination skills etc., all requires learning and forming new procedural memories. In a FPS, players must incorporate and process a mixture of visual, auditory and spatial inputs, and produce outputs in the form of motor movement (pressing buttons on controller).

Subjects were first given two minutes to familiarize themselves with the controls of a FPS game called Unreal Tournament 2003, after which they were pre-tested for 7 minutes on a different map and the score was recorded (The score depends on how many times the player kills an enemy, and also on how many times they get killed). Subjects are then trained on a different game with similar controls and principles (Quake 3), in four different environments, each lasting 7 minutes each. A post-test is carried out after a certain amount of time after the last training session and the score of this post-test is recorded and compared to the pre-test score to get an indication of how well the memory is retained. The experiment is designed to test for the consolidation of procedural memory, and because the training and test are performed on different maps, and on different games, only the procedural memory of the skills acquired are carried forward, and not the declarative memory (e.g. the location of objects and stations).

Subjects were trained either in the morning (at 9am), or in the evening (at 9pm). When the subjects were tested straight after training (control group), their score improved by 8.63±|±1.62 and 7.71±|±1.23. When subjects trained in the morning are re-tested after 12 hours of wake (at 9pm), their score only improved by 4.28±|±1.08, showing a decay in memory during waking. However, a different group who was awake after training but had a night's sleep sees their score improve by 9.81±|±1.29, non-significantly higher than the control groups. It can be assumed that this group's memory was also decayed during the day, which means sleep is able to restore previously lost procedural memory, and possibly enhance it.

This studied also showed that memory that was consolidated during sleep is more stable. Subjects that were tested in the evening and allowed to sleep before being tested 12 hours later, showed enhancement of 10.34±|±1.55. A different group that were trained in the evening but was tested 24 hours later, showed a 9.95±|±1.86 improvement on the score, meaning the consolidated memory is not subject to the same level of decay as seen prior to consolidation.

This study can be deemed reliable due to a large sample size (results from 178 subjects were used) and the use of established protocols (based on a previous study by the same laboratory). However, its conclusions cannot be definitive because the sample are all from University of Chicago students with no more than 10 prior experiences of playing FPS games. This is a problem because 1) the age group of the subjects were confined to ~19.9 years old, and younger/older subjects might show different results, and 2) "no more than 10 prior experiences playing FPS games" means it is not a truly novel experience, and there were no way in ensuring the students were being truthful.

Although this study is by no means definitive, it supports other studies which have shown sleep-dependent enhancement in perceptual learning of synthetic speech, paired word association, rotary pursuit motor task, sequential finger tapping task, sequential motor task,, and visual texture discrimination tasks.

And the effect of consolidation might not be limited to the next morning or day. In a human study (n=25), subjects were trained on a pursuit task where only the horizontal trajectory is predictable. One half of the subjects were deprived of sleep the first night after training, whereas the other half were allowed to sleep. The subjects were then retested on the same task three days later. The three-day gap is to account for undesirable effects in attention-retention due to sleep deprivation, which is an undesirable factor in the experiment. Subjects which were allowed to sleep showed improvement whereas the sleep-deprived group showed no improvements. Using functional MRI (fMRI), they've also shown that brain activity in the superior temporal sulcus (STS) were more active in the sleeping group. This study shows that the sleep in the night immediately after memory acquisition is essential in memory consolidation, and this consolidation involves changes in the neural pathways that is retained after three days.

Active learning increases sleep propensity
If the purpose of sleep is to consolidate memory and facilitate learning, then there should be a homeostatic response whereby the more a person learn, the more sleep they'd require to consolidate that memory; indeed, evidence do support this. Studies on the fruit fly Drosophila melanogaster shows that flies which had a more intense or complex social experience requires more sleep after. This has also been shown in other animals, where animals who learned quickly had the biggest change in sleep structure. Human subjects undergoing an intensive foreign language learning increased the amount of REM sleep the night after; furthermore, the improvement of their learning correlated positively with the amount of REM sleep the previous night.

Sleep inspires insight
In the study by Brawn discussed above, subjects after a night of sleep improved their performance non-significantly beyond those that were tested immediately after training. Although this result is non-significant, what if sleep not only restores and consolidate previous memories, but actually form new associations and improve performance that is not possible immediately after training.

In a study, researchers took 66 subjects and asked them to carry out a Number Reduction Task. In this task, subjects are given an eight-digit string of numbers, which can only consist of three types of numbers - '1', '4' and '9'. They must process this string from left to right, taking adjacent pair of numbers and applying two simple rules. Rule 1 states if the pair of digits are the same (e.g. 1 and 1), the output is the same as the digit (e.g. 1), and rule 2 states if the pair of digits are different (e.g. 1 and 4), the output is the other number not in the pair (e.g. 9). The output from the first pair then pair up with the third digit, and this output is paired up with the next digit in the string until we reach the last output, which subject must submit as the final solution.

The important part of this task is that there is a hidden rule - that subjects do not need to complete the 8-digit string to get to the final solution, but the strings are designed so that the second output is always equal to the final solution. Subjects who realizes this hidden rule will be able to complete the task much faster, as they only need to do two iterations of the rules rather than seven.

Three task blocks (each consisted of 30 strings) were carried out. Subjects are then required to have 8 hours of nocturnal sleep, nocturnal wakefulness, or daytime wakefulness. After this 8 hour period, the subjects carried out the task again, this time lasting ten task blocks. The number of people that gained insight into the hidden rule (detected by their quickened elucidation of the final solution, as well as a questionnaire asking them about the hidden rule) is substantially higher in subjects who received 8 hours nocturnal sleep (59.1% as opposed to 22.7% in the wake groups).

To ensure this result is not merely due to an increase in attention, supplementary experiments were carried out where the tasks were carried out as 13 continuous task blocks, in the morning after sleep (when attention is high), or in the evening after daytime wakefulness (when attention is low). Both groups had 25% of its subjects gaining insight, much less than 59.1%. This shows that the gain of insight is independent on attention, and requires processing of previous experiences in the brain during sleep.

The fine prints
Previously, I've stated that there is no clear consensus for the benefits of sleep on declarative memory. This is perhaps due to researchers measuring improvement using different methods, or that different stages of sleep are involved in consolidating different memories, and individual researchers have been focusing on only one stage of sleep in their research. Recent research have attributed the consolidation of declarative memory to SWS, with little effect from REM sleep. Declarative memories appear to be unaffected by REM sleep, whereas procedural memories are. For non-declarative memory, there are some that attribute it to NREM-2 stage sleep, others have suggested SWS , and still others suggesting both are required. The general consensus, however, is that declarative memory are reinforced during SWS sleep, and procedural memory are processed primarily during REM sleep.

Thus, the benefit of sleep on memory consolidation should continue to be carried out with distinctions between different memories and different sleep stages, preferably with the subjects' performance measured using similar principles, to allow the field to come to a consensus.

Mechanism
So the evidence stacks up well for this hypothesis, but what is it about sleep that is so essential to consolidation, that it cannot occur during wake? The two prevailing models suggests that consolidation is achieved through replay of the memory during sleep, as well as downscaling the synaptic strength across the cortex.

Replay
To test whether memories are replayed during sleep, subjects trained on a serial reaction time task had their brain activity measured using positron emission tomography (PET) imaging, during the task as well during sleep. Comparing the two profiles, results showed that regions of the brain that was active during the tasks were significantly more active during REM sleep. The same group later found out that the extent of the replay correlated with the quality and strength of the memory developed, suggesting the similarities in brain activity during wake and REM sleep is due to the processing of these memories. Replay is also seen in songbirds at the single-cell level, where the firing patterns of neurons in the forebrain regions associated with learning vocalizations during singing, closely resembled that during sleep.

Synaptic homeostasis
During wake, we interact with the environment and acquire information which we process as memories, which are in the form of selective strengthening of synapses, and formation of new ones, a process known as synaptic potentiation. Synaptic potentiation means synapses are more frequently stimulated, more dendritic receptors form, and more neurotransmitters exported per stimulation. This process requires a lot of energy and space, and limits the number of further potentiation, leading to saturation and an inability to learn further. This hypothesis suggests that sleep functions to clear unused or irrelevant synapses, or their potentiation, to allow relevant associations to be reinforced, as well as to make way for new synapses to form and potentiate the next day. This is achieved through down-regulating the strength of all synapses of any particular neuron, desaturating neural connections, and ensure efficient use of space and energy, while preserving memory.

The synpatic homeostasis hypothesis is supported by behavioral studies in fruit flies, rodents and humans, which suggests that the more complex or intense an experience/training/social situation is, the more sleep that is required. EPSCs, which measures the depolarization of postsynaptic membrane potential after receiving a signal from the presynaptic neuron, showed that the frequencies and amplitude of these depolarizations increases after wake, and decreases during sleep. This supports the fact that synaptic strength is built up during the wake state, and is decreased during sleep - consistent with the hypothesis. Furthermore, the overall level of synaptic proteins in all major parts of the brain decreases gradually during sleep. In addition, this hypothesis is supported by anatomical studies which shows global increase in synaptic density in animals which are experiencing an rich environment.

The hypothesis also states that slow-wave sleep (NREM-3/4) are responsible for this 'synaptic downscaling'. SWS is dominant during the first half of a normal sleep cycle, and REM sleep dominates the latter half. Since REM sleep have been implied in memory consolidation of procedural memories, the process of clearing out irrelevant connections prior to them being consolidated makes sense.

Since reply is also known to occur during wake, replay might actually be a continuous process that occurs in parallel to synaptic homeostasis. In fact, it is possible that by replaying the memories, it is helping the downscaling effort by ensuring important connections are not attenuated. The role of sleep in enhancing this replay might be to silence out irrelevant memories that serves as interference, or it might make the process more efficient.

Learning and memory acquisition requires a brain that is plastic, that is it is able to acquire permanent structural and functional changes in neural pathways and synapses as a response to stimuli. So not only does sleep get rid the brain of irrelevant memories and strengthen relevant memories, it also maintain neuroplasticity, to ensure we are able to learn optimally the following day.

Sleep but not wake
Very well, replay and synaptic downscaling are observed in sleep and could be responsible for memory consolidation, but how come these processes cannot occur during wake - what makes sleep so special? The short answer is differences in environment.

There are high levels of noradrenaline (NA) and high phosphorylation of cAMP-responsive element-binding protein (CREB) during wake, which are known to lead to the up-regulation of long-term potential (LTP)-related genes such as Arc, Brain-derived neurotrophic factor (BDNF), nerve growth factor-induced gene A (NGFI-A), Homer, Neuronal activity-regulated pentraxin (Narp). These changes favours the storage of newly-acquired information by increasing the strength of the synapse.

During NREM sleep, the levels of NA, histamine, acetylcholine, serotonin, histamine and BDNF are diminished, thus likely disengaging the increase transcription of in LTP-related genes, as well as leading to dephosphorylation and subsequently internalization of α-N-3-hydroxy-5-methyl-4-isoxazolepropionate (AMPA) receptors, , and an up-regulation of molecules implicit in synaptic depression, such as calcineurin, protein phosphatase 1, meatbotropic glutamate receptor subunit 2, molecules involved in vesicle recycling (e.g. N-ethylmaleimide-sensitive factor (NSF)), and insulin.

Furthermore, using PET scans to image brain activity, it shows that after learning a motor skill, subsequent use of that skill employs different parts of the brain dependent on the time after the completion of practice. Brain activity shifts from the prefrontal regions of the cortex to the premotor, posterior parietal, and cerebellar cortex structures. It has been proposed that this shift is part of the consolidation process whereby the representation of that skill is changed into a more stable form. A study has shown that rat have neuronal replays, of periodic nature and of increasing activity, which occur simultaneously in both the cortex and the hippocampus. This means that consolidation might involve not only the part of the brain stimulated during wake, but also other regions as well. Performing replay during wake might not be possible because other parts of the brain might still be in active use.

New memories have shown to have a time period of approximately four to five hours where it is vulnerable to interference from more recent memories. Therefore, although stabilization is known to occur both during wake and sleep, the purpose of sleep might be to dim the sensory inputs and prevents newly-formed memories from interference, thus disallowing 'false' memories from being incorporated into the long term memory, ensuring the integrity of the memory.

Counter-arguments
A common fallacy is to argue that since sleep is required for learning and memory consolidation, and since humans are the most intelligent of all animals, humans should be the ones that sleep the most in the animal kingdom; this is not the case and thus the function of sleep cannot be for memory consolidation or learning. However, the amount of sleep depends on the need for sleep - if an organism has a lower ability to process the information that it receives, it might have to take longer for it to complete the consolidation process.

Yet another objection questions the methods used during research, and claims that sleep-deprivation not only deprives the subject of sleep, but also induces additional stress. Recently, researchers have addressed this issue and developed flies which can be induced to sleep using a temperature-gated nonspecific cation channel Transient receptor potential cation channel(UAS-TrpA1), and shown that this induced sleep (which the lack of do not provide additional stress) boosted memory of learning the flies undergone prior to sleep.

Interim conclusions
Humans are unique amongst the mammals largely because we have a large cortex. This cortex is involved in learning and memory, which requires permanent changes in the structure and function of the brain - a process known as brain plasticity. It is likely that any experiences we have during wake, which are perceived and processed in the brain, will lead to plastic changes in the brain. Not all of these experiences are useful, or indeed make any sense. Thus sleep is a homeostatic mechanism to prune away irrelevant connections, as well as to make sense of existing experiences, allowing us to learn from new experiences the following days.

For conserving energy
Hypothesis that suggests sleep serves to conserve energy have partly arisen through studies which shows that brain activity and energy consumption during sleep is lower than that during wake. For example, a study shows that the level of glucose metabolism in many regions of the brain decreases during NREM sleep compared to wake. However, a counter-argument for this hypothesis is that brain activity during REM sleep can vary between an intermediate state between NREM and wake, but can often be as, if not more, active during REM sleep than wake. Thus, there's no clear evidence that sleep uses less energy than wake.

Another avenue for this hypothesis is that the amount we, as well as small herbivores, sleep decreases during our lifetime. Newborns have a high surface area-to-volume ratio, meaning it is costly for the body to maintain body temperature, and being active will contribute to this cost even more. Thus, newborns will survive better if they are immobile and retire to a warm place (e.g. nest). Thus, sleep might serve an evolutionary function, to reduce activity and improve survival.

However, even if true, it is hard to distinguish whether sleep is the cause of the lowered metabolism, or is this reduction a consequence of other processes of sleep (e.g. synaptic downscaling).

A study that opposes the theory of energy conservation comes from the arctic ground squirrels (Spermophilus parryii). A study shows that these hibernating animals, as do others, periodically increases their body temperature during hibernation. Here, the researchers have observed "consistent" sleep during these periods, and have thus asserted that the purpose of euthermy is to sleep. Based on this assumption, they then go on to argue that because sleep is energetically expensive, the purpose of sleep cannot be for the purpose of conserving energy. However, a counter-argument may be that because euthermy is energetically demanding, the animal is prevented from depleting their energy supply by going into sleep, which conserves energy.

Interim conclusions
The major functions of sleep is unlikely to include energy conservation, although not discounting it as a regulating factor. Since we get sleepy even though we are not hungry, and we can get more sleepy after a big meal, discounts the hypothesis that sleep conserves energy. Furthermore, even if true, it does not explain why resting wakefulness cannot achieve the same effect, since resting wakefulness maintains a high level of consciousness while still maintaining energy by not moving. Retreating to a safe location and sleeping allowed the animal to be protected from predators, conserving energy during times when it is not hunting/grazing, might have been a function of sleep in the past, that is obsolete now, and serves only as an evolutionary ornament to remember our past.

Fighting pathogens
From experience, we know that when we are ill, through, say, a bacterial infection, we tend to be sleepier; it turns out, this might be our body's mechanism in preventing the infection from progressing. Indeed, many links have been found between sleep and immunity. The first sleep-inducing substances to be identified was muramyl peptide (initially called 'Factor S'), a component of the bacterial cell wall; it was found to increase in the cerebral spinal fluid during sleep deprivation. Chronic sleep loss has been observed to lead to increase in inflammatory mediators, which can act as immunomodulators; and these, in turn, have shown to alter sleep behaviour.

The best-studied components that links sleep and the immune system together are the cytokines IL-1β and TNFα. TNF mRNA levels and protein level in the cortex and hypothalamus of rats can increase by about two-fold and ten-fold, respectively, across the day, and this correlates positively with sleep propensity. IL-1 plasma levels are highest at the onset of sleep for humans, and IL-1 and TNF and their mRNA levels are highest at the NREM stage. Sleep deprived humans sees their circulating levels of IL-1β and TNFα protein,as well as levels of IL-1β mRNA increased. Administration of these cytokines can lead to symptoms of increased sleep propensity see after sleep deprivation including fatigue, poor cognition and increased sensitivity to pain. Thus, there seems to be a positive correlation between sleep propensity and the levels of IL-1β and TNFα.

Administration of IL-1β and TNFα leads to increase in NREM sleep and a suppression of REM sleep, and injection of TNF into the body cavity of mice induces 90 minutes more sleep within 9 hours of injection. Microinjection of TNF into the preoptic area (POA) increased NREM sleep, an effect which is reversed if TNF receptor fragments, which antagonises TNF, is injected instead. . showing a causative effect of the cytokines on altering sleep structure.

Antagonizing endogenous IL-1β and TNFα using inactivators, antibodies, soluble receptors, siRNA-knockdown animals, or removing cellular receptors, decreased the duration of NREM sleep in mice, including the increase in NREM sleep normally observed after sleep deprivation, excessive food intake, or acute elevation in temperature.

Therefore, it appears IL-1β and TNFα increases the amount of NREM sleep, but unclear on its affect on REM sleep. Of note here, although the cytokines increases NREM, the resulting NREM sleep is fragmented. But why does the body induces more sleep with IL-1β and TNFα? What is the purpose of the increased sleep duration? To answer this question, we'll have to look at what occurs when a bacterial infection is underway.

If we mimic bacterial infection by administering bacterial cell wall components such as lipid A, lipopolysaccharide and muramyl peptide, we see an antigenic response that results in the increased production and release of pro-inflammatory cytokines, including IL-1 and TNF. At the same time, NREM sleep is increased and REM sleep is predicably suppressed. Furthermore, when IL-1 or TNF are antagonized, the alteration in sleep is not observed.

Thus, this increase of NREM sleep mediated by IL-1 and/or TNF might be a response to infection. Since most infections occur outside the brain, the majority of the action of the immune system occurs outside the brain, and this takes up a large amount of energy, energy which is directed towards fighting infection. This can mean that the energy store in the brain is depleted more quickly during wakefulness (indicated by increased adenosine and depleted glycogen), and requires a long duration of sleep for this to be restored. Furthermore, if infection does occur in the brain, then increasing sleep might enhance the cellular maintenance tasks that is associated with NREM sleep, as will be discussed later. Impairing sleep might lower the effectiveness of the brain in regulating bodily processes, as supported by the observation that sleep deprived individuals receive little or no benefits in viral vaccination, possibly because the body is not mounting an effective immune response to it.

Mechanism
Most infections occur outside the brain, and thus most cytokines are produced in the peripheral immune system. To regulate something that is regulated inside the brain, the cytokines must be actively transported into the central nervous sysmtem (CNS) through the circumventricular organs (CVOs), which are structures of the brain that is highly vascularized. However, cytokines may also be synthesized in the brain from both neurons and glial cells. IL-1β and TNFα themselves induces their own production. This is done through activating nuclear factor κ β (NFκβ), a transcription factor that regulates IL-1β, TNFα and the adenosine A1 receptor (A1AR).

IL-1 and TNF are known to interact with the serotonin system, a wake-promoting neurotransmitter. This neurotransmitter are at its highest levels during wake, and decreases during NREM sleep, and is diminished in REM sleep, fitting the pattern seen for IL-1 and TNF. It is thought that serotonin carry out its wake-promoting functions by inhibiting sleep-promoting neurons in the anterior hypothalamus, preoptic area and basal forebrain. It is also thought that the neurons that releases serotonin also produce sleep-inducing factors such as IL-1, and these factors will accumulate and overwhelm the wake-promoting role of serotonergic cell bodies, eventually inducing sleep. In turn, IL-1 is thought to carry out its sleep-promoting effect through synthesis/release of growth hormone-releasing hormone, prostaglandin D2, adenosine and nitric oxide, as inhibiting any one of these systems prevented increase in NREM sleep as expected. The exact mechanism of these downstream effectors are less clear.

Noradrenergic neurons of the locus coeruleus and serotonergic neurons in the raphe nuclei firing is diminished, and the neurotransmitters they releases are thought to gate the entry into REM sleep by inhibiting neurons that promote REM sleep. Indeed, inhibiting serotonin allows REM sleep whereas administering it inhibited REM sleep.

Thus, from the mechanism discussed above, perhaps fighting infection does not induces more sleep, but instead it changes the structure of sleep to incorporate more NREM sleep. This further highlights the need for sleep research to break down different types of sleep and study them individually. It would also be interesting to see how the levels of immune modulators change during diurnal sleep (napping) to distinguish effects of circadian rhythm as opposed to sleep per se.

Cellular maintenance
It can be postulated that neurons, after a prolonged period of activity, or bacterial infection, will accumulate cellular stresses. As its biosynthetic machinery is unable to replenish the proteins and lipids (such as neurotransmitters) that are used up or damaged, quality control of protein folding through chaperones are decreased, leading to a higher level of misfolded proteins. These accumulate in the ER and causes ER stress. The temperature in the brain may also increase progressively during the waking period, which might lead to an increase in unfolding and entanglement of proteins. Furthermore, increased metabolic rates demands a higher activity in the mitochondria, and consequently more reactive oxygen species are produced and leaked into the cell, where it can damage the cell, unhindered due to the lack of functional regulatory proteins. Thus, sleep is essential to provide a time of inactivity, where these the energetic and biosynethetic demands are halted, allowing the cell to synthesize chaperones that will re-fold misfolded proteins or recycle them, allowing the cell to perform preventive cellular maintenance on minor problems each night, thus preventing major irreparable accumulated cellular damage.

An obvious question to ask is why does an organism need to undergo global, behavioural sleep while the problems are at the cellular level. Why can't local areas of the brain go to sleep and allow for the repair to occur? An argument might be that because the neurons are so interconnected, it is impossible to allow some neurons to rest while others are being excited. Furthermore, the high levels of modulates that promotes wake and neuronal activity might prevent the neurons from resting. Local sleep during wake might also impair our ability to perform cognitive tasks. This is seen in the fur seal, where if they are in water, they exhibit unihemispheric sleep to keep them moving and afloat in the water, whereas if it is on land it exhibits bihemispheric sleep; indicating that unihemispheric sleep is sub-optimal.

Thus, by this hypothesis, it'd be better off if there were no synaptic activity at all during sleep, as this would enhance the cellular maintenance task even further. But what we must remember is that sleep serves a multi-functional role, such as memory consolidation, which we have shown involves a replay of the memory traces in order to determine which memories are relevant. NREM sleep has been attributed to the process of cellular maintenance, and this precedes other non-vital processes such as memory enhancement, which are thought to occur during REM sleep, because cellular maintenance must take priority over these less important process, because if cellular maintenance is prevented, it can lead to irreparable cellular damage; and if widespread, organ damage.

Evidence that support this notion of cellular maintenance in response to high or prolonged synaptic activity relies on observation of a build up of damaging metabolites during wake, and its subsequent decline during sleep. For example, when neurons fire, ATP, a signalling molecule that denotes cellular stress, is released to the extracellular environment, and these can build up and binds to purine type 2 receptors, such as the P2X7 receptor, and this increases the level of IL-1β and TNFα release from glial cells. As seen above, this activate sleep-promoting neurons and increase sleep propensity. When ventricular extracellular ATP is antagonized, the amount of sleep decreases; conversely, when ATP is agonized, sleep is increased.

Recycling and destruction of damaged or misfolded proteins usually follows two pathways - through the proteasome, or through the process of autophagy (for bulky items). Performing a simple search on PubMed with the search terms 'autophagy' and 'sleep' returned twelve results, and for 'Proteasome' and 'sleep' returned twenty-one results, not all of which are relevant. This demonstrates the need for sleep research to shift from a systemic and behavioural focus, to a more cellular and molecular focus, looking at different regulators of different pathways change during wake/sleep, and then see how these pathways affect sleep. Therefore, needless to say, this logically-valid hypothesis remains a hypothesis, until further evidence is obtained.

Other hypotheses
Of course, these hypothesis to the function of sleep is by no means exhaustive, many other studies have shown an effect of sleep in allowing the body to synthesize macromolecules, to controlling the level of oxidative stress, to prevent aging, in regulating carbohydrate metabolism and endocrine function, in preventing diabetes, and preventing obesity However, I have chosen what I believe to be the best-characterized and those that make the most logical sense.

In studies linking sleep loss with obesity, many current studies show only correlation and not causation, and suggests that disrupted sleep decreases the leptin levels, but increases the ghrelin levels, which increases appetite and hunger. But a recent review raises the issue of a lack of experimental data; however, this might be difficult as obesity is a result of numerous factors and can emerge slowly. While an observational study does not provide evidence of causation, an experimental approach would be both impractical and unethical. Sleep loss has been associated with diabetes and cardiovascular diseases, where people that sleep less, are more likely to be diabetic and/or suffer fatal heart attacks. However, these links must be treated with caution, as one not necessarily causes the other, but rather, both might be effects from a common cause (e.g. disrupted lifestyle).

It was also paramount that the distinction between sleep homeostasis and circadian rhythm is established; and this is not observed in many studies supporting the other hypothesis (e.g. ). Although these two processes occur parallel to each other, the circadian component should be viewed as a regulation of the time to fall asleep according to ecology, and the homeostatic component as regulated by numerous systems. Both have an effect on sleep propensity. Rats which had their suprachiasmatic nuclei lesioned still exhibited both NREM and REM sleep, although the incidence and amplitude of slow waves are decreased. The benefit of a 60 - 90 minutes nap in memory consolidation is as effective as an 8-hour nocturnal sleep for learning a perceptual skill, and sleep during the day proved as effective as a night's sleep for a finger-tapping task. These studies further strengthen the assertion that sleep is independent of the circadian rhythm, but is affected (maybe part-regulated) by it.

Perspectives
Thus, more stringent studies discerning the different stages of sleep, and distinguishing sleep homeostasis and circadian rhythm, as well as conducting more interventional studies over observational ones, will be the next major step in progressing sleep research. Expanding from the orthodox methods and studying sleep in relation to other fields such as autophagy, metabolism and cancer, and use bioinformatics, -omics studies and biochemical assays to might allow us to identify universal regulators of sleep, and perhaps unify the different hypotheses, if one actually exists.

Moving forward, we must also be careful not to view sleep as a global event, but as a collective of locally regulated sleep. In a study with rodents, where rodents' local brain activity is measured using EEG, has shown that local areas of the brain go 'offline' after a long wake period, and more areas go 'offline' as the wake duration increases. Even though parts of their brain is 'offline', the rodent was still active, although their performance in a pellet reaching task is progressively impaired. Another study showed that repeatedly stimulating a single whisker in rats, leads to local changes specific to the somatosensory cortex to change from a wake-like state, into a sleep-like state.

Thus, sleep in the traditional sense should be viewed as a collective, but local, sleep in different areas of the brain. Only when enough areas are asleep do we close our eyes, become immobile, and fail to respond to stimuli below the arousal threshold.

Conclusion
The bottomline is this - the environment in the brain is different during sleep compared to wake, and this allows the brain to carry out processes that it is unable to during wake. The nature of these processes, however, are still up for debate. Indeed, the lack of sleep affects our mood, our health, our cognitive function, our appetite and numerous other aspects. It is no surprise that so many theories have emerged to explain the function of sleep, and now they seem to all starting fitting into each other.