Tuesday, December 4, 2012

The human machine: circuits and wires

The previous post in this series can be found: here.

In the first post of this 'human machine' series, I explained how 'energy' (that abstract entity) is processed and used by our bodies in order to converted the chemical energy in our food into the work energy required to keep us ticking over nicely. I discussed in this how we are all actually powered by electrical circuits that buzz along in the internal membranes of our cell's power stations, the mitochondria. Better yet, not only are we powered by currents of electrons, familiar to us as standard electricity, but also by currents of protons, and so are actually working off energy being extracted from two forms of electrochemical potential. We're pretty sophisticated machines!

The work energy generated by these processes is used in myriad ways, but one very important one is the creation of another electrical current that is the foundation of everything you've ever done and every thought you've ever had: the neuronal action potential. This is the electrical signals that run along the neurons in your brain and body in general, constantly relaying information back and forth throughout the whole complex machine. Without it we would be like plants, with one part of our bodies completely unaware of what's happening to the rest of it, and animal life as it is familiar to us would be entirely impossible. Most people have, I expect, heard of the notion of electrical signals running throughout our bodies (it's why the machines built the Matrix, right?), but few will actually know what that means. In today's post I'm going to be talking about what neuronal signals actually are, and so explain why being hit by lightning is a bad thing but being defibrillated (like in ER) can be a good thing.

Neurons - shooting the messenger

Your body is an intricately wired piece of equipment - neurons connect every part to every other via the nervous system. In my last post in this series, I discussed how cells talk to one another; well the role of the nervous system is to allow tissues to talk to one another. This is often dictated by the central nervous system, made up of the brain and spinal chord, which is obviously vital for any concious decisions you might want to relay to the rest of your body, yet the majority of the signals coming from the brain are completely subconscious - regulating the conditions of your body, for example. The peripheral nervous system, made up of all of the non-CNS neurons, tends to be less involved in the decision making, and more acting as a simple relay of information; either sensory signals working their way back to the brain, or important instructions being handed down from the CNS to the peripheral tissues like muscles. 

This entire network is dependent on a single family of cells: neurons. These are a highly-specialised type of cell that have evolved solely to pass information along from one cell to the next. This specialisation is immediately evident in their structure. Neurons are a hell of a lot longer than most cells, most of their length coming from an extended segment called the axon, or nerve fibre. Most neurons have just one axon but some, particularly in the brain, can have several. At one end of the axon is the cell body that contains the vast majority of the organelles of the cell - i.e. the bits that do the work - mitochondria, nucleus, endoplasmic reticulum etc. Dendrites, from the Greek for 'fingers', extend from the cell body and contact other neurons or cells. At the end of the axon furthest from the cell body is the axon terminal; simply the end of the cell. Both the dendrites and the axon terminus form structures known as synapses, which is really just the point at which a neuron comes into contact with another cell. 

The humble neuron - simple but effective.

The axon itself is usually surrounded by a sleeve of other cells called the myelin sheath (made up of either Schwann cells or oligodendrocytes, but let's not go into that here!) that act as insulation for the electrical signal. And that's basically it - nothing all that fancy in the structure, it's very to-the-point. If I had to design an information-relaying cell I can't think of a better shape for it than a long, thin and simple design with synapses at both ends to allow information in and out. So, while elegant, the structure of the cell itself is not that astounding. It is instead once you start to look at the mechanism by which information travels along these cells that you really start to see the brilliance in its engineering.

Travelling down the wire

A firing neuron is a busy thing. The popular perception is that electricity travels down neurons much like it does in a copper wire, with electrons moving down an electrical potential from one end to the next. This is not the case in neurons. For a start, one end of the cell has no more electrons than the others, so there is no electrical potential (which just means a gradient of electron concentration) for them to move down. Then you'd have the problem of how to move the electrons - they can't move freely along the cell membrane because it's made of fats that don't conduct electricity well, and to pass it through the water-filled centre of the cell would likely do a lot of damage (errant electrons are dangerous things)! Instead evolution has come up with an ingenious way around this by developing a mechanism to skim an electrical signal along the outside of the neuron. This is done not by moving free electrons, but instead by moving charged particles called ions. Atoms have no charge when they contain the same number of negatively charged electrons as positively charged protons, but this can change if they either gain or lose electrons, thereby becoming either negatively or positively charged ions, respectively. The main ions that take part in neuronal signal transduction are sodium (Na+) and potassium (K+), both of which have a single positive charge per ion. It's the redistribution of these ions relative to a resting neuron that constitutes the electrical signal in a firing neuron.

The process by which this occurs is called the 'action potential'. In a resting neuron, the number of negatively and positively charged ions on either side of the cell membrane is not equal, and so the two sides of the membrane have different overall charges. In quantifiable terms, the inside of the cell is 70mV more negatively charged than the outside because there are more negative ions (like chloride) on the inside and more positive ions (like Na+) on the outside. This difference is maintained by the fact that the cell membrane is semi-impermeable to ions and so they cannot redistribute spontaneously, and by active ionic pumps that sit in the membrane and make sure everything is where it should be. The action potential is essentially a pulsed disruption of this equilibrium such that for a very brief moment the outside of the cell becomes more negatively charged than the inside. This effect zips along the surface of the cell very quickly, almost like a Mexican wave (or just 'The Wave' as I'm given to understand it's called in the US!) of charge displacement.

So what causes this change? Well our key players, as I mentioned earlier, are Na+ and K+. In the resting neuron, there is more Na+ outside the cell than inside, whereas there is more K+ inside the cell than outside. At the start of an action potential, channels open in the cell membrane that allow Na+ (and only Na+) to enter the cell because of simple diffusion down their chemical gradient. The absolute numbers of Na+ ions that enter the cell aren't very large, but they have the significant effect of reversing the polarity of the membrane such that the outside is now more negatively charged to the tune of around 40mV. This change in polarity causes separate K+ channels to open and so allow K+ to exit the cell, which thereby cancels out the exterior negative charge and brings the cell back to a resting membrane potential. After the action potential the distribution of K+ and Na+ is restored to normal by an Na+/K+ pump, which uses our old friend ATP to power the exchange of one exterior K+ for one interior Na+ until the balance is restored. The fact that this pump has to work pretty much constantly in active neurons makes them incredibly energy-hungry cells, which is why your brain uses 20-25% of your total calorific consumption - I hope you're putting it to good use!

The neuronal action potential.

And that's it, basically. This depolarisation and subsequent repolarisation of the membrane starts at one end of the cell and then shoots along the membrane to the other where it reaches a synapse with another cell and so either passes the signal along to the next neuron via the release of neurotransmitters, or induces whatever the desired response is from another type of cell (e.g. causing muscle contraction). The firing of the action potential on one part of the membrane partially depolarises the next bit, in turn causing it to fire and set of the next bit etc. This can be disrupted by outside sources of charge, principally the negative charge that comes with electrical current. This is the principle behind a defibrillator; if your heart is beating erratically the electrical pulse from the defibrillator overwhelms the effect of the action potential to very briefly inhibit neuronal firing in the heart so that it can restart with a normal rhythm (it doesn't, contrary to popular belief, restart a heart that has stopped beating). A similar thing happens if you are struck by lightning - aside from the burns caused by the heat of the strike, the action potentials in your neurons are disrupted and so this can cause brain damage or stop the heart from beating.

The moving parts of the machine

Whilst it's very interesting to know how the events of this cycle result in neuronal transmission, as a molecular biologist I am most interested in the engineering behind the whole thing. It's a fairly simple process but it requires some absolutely astounding biological machines to make it work. Principally what I am talking about are the two channels mentioned above: the Na+ channel and the K+ channel. These are remarkable bits of equipment - they have to be specific only to their designated ion and also be sensitive to the membrane potential such that they are open or closed at the correct stages of the action potential. For this reason, they are known broadly as 'voltage-gated ion channels'. These channels have been the subject of intense research for half a century and their architecture and inner workings are now quite well understood. I'm going to give a brief overview of how the K+ channel works to give you a sense of just how unbelievably well evolution has engineered these things to work, and really just how sophisticated a machine you are!

The human voltage-gated K+ channel - a thing of beauty! So much so that it has inspired art.

Let's start with the first point I raised: how does the K+ channel stay specific to K+ ions? The channel is, like all channels, a protein - a string of amino acids linked together in sequence and then folded up to produce the final product. Different amino acids have different chemistries, some even have negative and positive charges. That is stage one of the solution - since K+ is positively charged, lining the inside of the channel with negatively charged residues will not only attract K+ in, but also repel negatively charged ions, thereby giving some specificity. However, Na+ is also positively charged, so how to stop it from going through? This where the engineering gets really clever! The K+ channel has a segment in the centre of its pore called the 'P loop', which is made up of 5 amino acids. 4 of these P loops some together in the full channel to made a section known as the 'selectivity filter', which, as the name suggests, gives selectivity to the channel for K+ over Na+. It does this by forming bonds with K+ ions passing through the channel to stabilise their overall charge and so make the whole process energetically favourable. Na+ is smaller than K+ and so is not able to efficiently form these bonds with the selectivity filter, thereby leaving Na+ passage as an energetically unfavourable process. 
The selectivity filter of the K+ channel. K+ ions passing through it are shown in red and purple; they fit perfectly in the spaces created between the P loops, whereas Na+ would be too small to fit optimally.
So, on to the next problem: how are they able to open and close in response to changes in membrane voltage? We are still working towards a full understanding of voltage-gating in these channels but nonetheless understand the general principles. Both this and the Na+ channel can exist in 3 distinct states: deactivated (closed), active (open), and inactivated (closed). The channel is capable of switching between the deactivated and active states by subtle changes in the positions of amino acids on its intracellular and extracellular sides. These amino acids alter their charge as the membrane voltage changes, which in turn causes them to shift position and pull the rest of the protein with them. In the deactivated state they are positioned such that the 5 amino acids of the selectivity filter collapse to block the central pore of the channel, whereas in the active state they help to stabilise the conformation shown above. The third state ('inactivated') is also impermeable to K+ ions but instead occurs after the channel has been active for some time. This state cannot easily revert back to the active form and so stops K+ passage entirely. This state is reached when a ball-like segment of the channel becomes correctly charged so that is will insert itself into the central pore and block it like a plug. In order for the channel to become open again the ball must be removed, which only happens after the action potential is fully finished. Unsurprisingly, this is known as 'ball and chain' inactivation.

The three states of K+ channels.

The intricacy of these proteins is quite beautiful, and it is worth remembering that they do their jobs over time-scales that are imperceptibly small to us; a neuron can fire 100 times per second quite easily, which means these channels are opening and closing on the order of nanoseconds. I think it's always important to remember that things are often more complicated than they are typically portrayed, and the complex processes governing such a ubiquitous event as neuronal transmission is a good example of this. I hope I've given you a better sense of what those vague 'electrical impulses' actually are, and that you can see just how unbelievable our construction really is!

The next post in this series can be found here.


  1. Hi, could you please provide references for the figures you have shown? In particular I'm looking for a reference for the action potential diagram. Thanks

    1. Hi there,

      Certainly, apologies for not giving these in the text. I got the action potential figure from a American high school's website (http://kvhs.nbed.nb.ca/gallant/biology/action_potential_generation.html) but I believe the original was published by Benjamin Cummings through Pearson (http://www.pearsoned.co.uk/Imprints/BenjaminCummings/).

      The K+ channel figure is taken from the OPM database (ID=2r9r; http://opm.phar.umich.edu/protein.php?search=2r9r) and the selectivity filter is from the PDB (ID=1k4c; http://www.rcsb.org/pdb/explore/explore.do?structureId=1k4c).

      I'm afraid I can't find the source of the ball and chain figure, but I'm pretty sure I got it from another university/school website, sorry about that!