Search our site

Custom Search

Friday, 10 July 2015

How do we fall asleep?

If there’s one thing everyone has in common, it’s sleep. Regardless of age, gender, race or religion, without enough sleep we fail to function normally. As you’ve probably experienced, a lack of sleep can be detrimental for mood and focus. Lose enough of it and you are more prone to serious health issues like heart disease, diabetes and even death. The amount of sleep we’ve had seems to be one of the things we think about (read: regret) daily, but what about how it happens? This article aims to introduce sleep as a fascinatingly interwoven structure of processes, not just a simple currency of life, exploring the inner workings of the brain that might be responsible for how it happens. The reasons behind how we sleep are distinct from the potential reasons behind why we sleep, which we covered in a previous article (and video).

So, what is sleep? The definition of most verbs (for example, running[1]) describes how it happens. However anywhere you look, sleep is not currently defined in this way. Instead, its physiological effects are outlined. Breathing slows, most muscles are relaxed and the eyes go through varying periods of rapid movement, called REM. Sleep is defined in this way because we don’t know how it happens.

Photograph 'Sleep Like A Baby' ©peasap (CC BY)
We can fall asleep before we've even been born, but how do we do it? Photograph ©peasap via Flickr (CC-BY)

Electroencephalogram with alpha, theta and delta waves present during the stages of non-REM and REM sleep.
Using a method called electroencephalography (EEG), brain waves can be measured during the stages of sleep. The short, frequent, alpha waves seen whilst awake gradually change to less rapid theta waveforms. The largest, slowest waves are visible in part of stage three and for all of the deepest sleep, stage four. These are known as delta waves. Image © Sleepdex, used with permission.
From observing sleep, we do know what happens once you drift off. It is a cycle of five stages over the course of an hour and a half to two hours (on average), where your brain moves between ‘deep sleep’ and lighter, REM sleep. Stage one - where you are gradually falling asleep, moving in and out of it - generally only occurs once. The other four stages repeat themselves until you wake up. These are established by the varying levels of brainwaves you produce whilst sleeping. From stages two to four, the latter being deepest sleep, brainwaves gradually slow down. Once we reach REM sleep however, our brains start working at similar rates to when we’re awake! This increase in activity is translated into dreams as well as increases in blood pressure and heart and breathing rates.

The stages of sleep that we know about present the brain as a whole, using brainwaves produced through synchronicity of electrical signals sparked across the brain. Although we understand how the brain transitions between these stages of sleep, we don't understand how the brain transitions from wakefulness to sleeping. Research and experimental data is increasingly promoting a theory that it is instead a self-organised process where only certain sections of the brain need to be in a ‘sleep-like’ state to send us to sleep [2].

These sections are called cortical columns; columns of nerve cells, or neurons, that are the structural basis of the brain’s cortex (outer layer). Though their function is debated, they are a pathway of neurons used to send and receive signals in the central nervous system (in this case, the brain). Mice’s whiskers are an example of how such a network acts, where one cortical column complements a single whisker. The information from that whisker is sent to the brain via the cortical column, and the mouse gathers knowledge of its surroundings.There are different types of neurons, some of which may be vital for our bodily functions.

3-D Model of thalamocortical circuits between VPM axon, dendrite and soma columns.
A 3-D model of a cortical barrel column; each colour represents a different nerve cell type within the column. The middle and right columns show the extensions of the nerve cells, dendrites and axons respectively. Dendrites are responsible for delivering information to the cell, whilst axons carry it away. Image ©Marcel Oberlaender et al.[6]

Cortical columns are an example of a neuronal network, which are present throughout the brain. If these neuronal networks can be shown to ‘sleep’ individually and independently, it may be an indicator of the brain’s behaviour as a whole, through separate units.

Neuronal networks have a second key component; neuroglia, or just glia, that protect, insulate and support the neurons. Importantly, they also play a role in modulating synaptic transmission - how one neuron sends a signal to the next - by releasing calcium (Ca2+) in response to ATP, adenosine triphosphate - a molecule that is the main energy source in all cells.

So how does this relate to sleep? ATP is responsible for triggering the release of two molecules from the glia that researchers have found are fundamental for sleep; tumour necrosis factor (TNF) and interleukin-1 (IL1). These are cytokines, proteins used for cell communication. The way ATP induces TNF and IL1 release is similar to how these cytokines work themselves, through receptors. Receptors are found on the outside of cell membranes and they ‘receive’ the certain types of molecule it is built for. When a molecule is ‘received’ and binds to this receptor, it causes a response on the other side of the cell membrane. For example imagine a James Bond style eye scanner only lets your eye open a door. The scanner (receptor) analyses your eye (molecule), and if it is correct it elicits a response - in this case a door opening before you meet your gadget expert, Q. In more relevant terms, ATP binds to a receptor on the glia; TNF and IL1 are released.

Though we don’t know exactly how TNF and IL1 interact with the brain to send it into the first stages of sleep, there are some things that are now generally accepted. Firstly, that TNF can induce the sleep-like state in cortical columns. The columns could be described as switches. Individual cortical column states are capable of influencing organism behaviour, so if enough cortical columns are in the sleep-like state, or switched off, in up to 90% of cases the organism itself will sleep (and vice versa when it’s awake) [3]. So by analysing cellular levels of TNF, and looking at which ‘switches’ are off and on, a link has been established. Additionally, the longer a cortical column is in a wake-like state the more likely it is that it will enter a sleep-like state. In other words, the greater its activity, the more ‘tired’ it is. IL1 has been shown to increase the power of delta waves in the non-REM stages of sleep [4]. Furthermore, when inhibitors of IL1 are introduced the power of these waves decreases, resulting in a lower ‘intensity’ of sleep.

If cortical columns have been subjected to a sustained workload they will still be asleep the next day - people can almost literally be ‘half-asleep’(!) - but we wake up anyway. Maybe if enough of these columns are in the wake-like state for us to function, we do. Sleep then would not be dependent on the whole brain, but on enough of the neuronal networks in a sleep state to induce it. As a result, scientists are now beginning to think differently of the long standing agreement that sleep is a top-down, cognitive engagement of the entire brain.

Another family of proteins, the kinases, may also be involved in how we fall asleep. Cyclin-dependent kinase 1 (Cdk1) is responsible for suppressing sleep and promoting wakefulness, but a combination of proteins (Taranis and Cyclin A) have been found to inhibit Cdk1, and therefore induce sleep [5]. At the moment there is no evidence of this in humans as Taranis is unique to the fly that was studied, so we don’t know if this could be another mechanism involved in our sleeping habits. There is a protein similar to Taranis in humans, but it has not yet been investigated.


This animation shows the varying activity of someone’s brainwaves when they are falling asleep and waking up. Each line represents a single brainwave, with the colours distinguishing different ‘nodes’ or areas of the brain that are active. Additionally, the taller the loop is the greater the strength of the brainwave. Whilst awake, the alpha waves visible are tall and rapidly changing. However once the patient (who is being sedated) starts to fall asleep, there is a clear change in not only the size of the brainwaves, but also their location. New areas of the brain are being engaged at low levels with new connections being formed. Finally, as they begin to reawaken the brainwaves revert to the original patterns. Video credit: Srivas Chennu, University of Cambridge


Despite rapid progression in the study of sleep recently, a large amount of questions still surround it. It is still very difficult to tell when a tissue (like the brain’s cortex) enters a sleep-like state. Changes in the measures currently used to determine sleep can be subtle. Even in the current ‘gold-standard’ of measuring sleep, the EEG, there are discrepancies. For example a hyperventilating teenager can produce the delta waves characteristic of our deepest sleep stage, despite being fully awake. For now, the solution to how we sleep is only something we can dream about.

This article was written by Joshua Fleming, a biological sciences student from the University of Leicester conducting a summer internship in science writing at TWDK.

References
why don't all references have links?

[1] Dictionary.com,. 2015. 'The Definition Of Run'. (web)
[2] Jewett, Kathryn A., Ping Taishi, Parijat Sengupta, Sandip Roy, Christopher J. Davis, and James M. Krueger. 2015. 'Tumor Necrosis Factor Enhances The Sleep-Like State And Electrical Stimulation Induces A Wake-Like State In Co-Cultures Of Neurons And Glia'. Eur J Neurosci. doi:10.1111/ejn.12968.
[3] Rector, David M., Irina A. Topchiy, Kathleen M. Carter, and Manuel J. Rojas. 2005. 'Local Functional State Differences Between Rat Cortical Columns'. Brain Research 1047 (1): 45-55. doi:10.1016/j.brainres.2005.04.002.
[4] Yasuda, T, H Yoshida, F Garcia-Garcia, D Kay, and JM Krueger. 2005. 'Interleukin-1Beta Has A Role In Cerebral Cortical State-Dependent Electroencephalographic Slow-Wave Activity'. Sleep 28 (2): 177-184
[5] Afonso, Dinis J.S., Die Liu, Daniel R. Machado, Huihui Pan, James E.C. Jepson, Dragana Rogulja, and Kyunghee Koh. 2015. 'TARANIS Functions With Cyclin A And Cdk1 In A Novel Arousal Center To Control Sleep In Drosophila'. Current Biology. doi:10.1016/j.cub.2015.05.037.
[6] Marcel Oberlaender et al., ‘Cell Type–Specific Three-Dimensional Structure of Thalamocortical Circuits in a Column of Rat Vibrissal Cortex’ Cereb. Cortex 2012;22:2375-2391 doi: 10.1093/cercor/bhr317