This blog is dedicated to Yien Li and Shivaan, the most curious non-scientists I know.
(8 minute read)
I have just published my first first-author scientific paper and it is in the journal Nature. Pretty chuffed about that. You can find it online here and full-text & figures (for FREE) here.
But I realise that it’s not super accessible in terms of language. (It’s also not super accessible in terms of access… but that’s a story for another day. UPDATE: now accessible here!) The abstract is perhaps enough to throw you off:
“Man, this is gibberish! I can’t even understand the first sentence!!”
I know. Jargon is a bitch.
What do you do, Eamon?
I do “structural biology”. Structural biologists try to get three-dimensional pictures of proteins (and sometimes other stuff too).
Proteins are tiny machines.
You are made up of trillions of cells, which are very small bags filled with chemicals that just so happen to work together in such a way that the bag itself is considered to be “alive”. Amazing.
Some of those chemicals in the bag are the tiny protein machines – about 200,000 of which are working within a cell at any one point in time.
These tiny machines do all kinds of jobs: some carry oxygen and carbon dioxide molecules around your body to keep all of your cells alive; some change the electrical charge of cells in your brain to allow you to think; some detect individual particles of light as they enter your eye and enable you to see; some copy DNA so that your cells can divide and you can grow up. Pretty useful they are.
In fact, these machines are critically involved in pretty much every process that keeps you alive and functioning.
So we study them.
The problem is they are damn small.
They are so small that it is not possible to see one with even the most powerful light microscope. Not because the microscope’s magnification isn’t good enough, but because these machines are smaller than the wavelength of visible light.
Let me put this another way.
Imagine you are driving your car down a road full of potholes. These holes are of many different sizes but the only ones that you are worried about avoiding are the ones that are bigger than your tyres – you will definitely feel these ones (oompf!). All the smaller ones you won’t even notice because your tyres won’t fit into them. Your car tyres are like visible light and the smaller potholes are like proteins – they don’t affect each other. Oh, and you also happen to be wearing a blindfold, so the only way you know that a pothole is there is when you feel it by running over it. (In this analogy we want to hit the potholes because that means we can “see” them.)
If, however, you decided to cycle along this same road on your road-bike instead (still blindfolded), you would feel a lot more of the small holes thanks to your much narrower tyres.
We do the same thing to look at proteins: we use high energy X-rays – smaller tyres – which can feel the protein potholes. (This is in the case of one particular technique called “X-ray crystallography”. There are several other cool ways to see these proteins… that I won’t go into here.)
Okay, but you study something more specific, right?
For my PhD work, I’ve been trying to get a better picture of one important protein machine, called “Smoothened” (or “SMO” for short). Smoothened is a message carrier; it takes a message from one protein and passes it along to another protein. SMO sits in the wall of the cell and passes the message from the outside to the inside. If the message doesn’t get inside the wall then the cell cannot respond to it.
This particular message is very important because it tells the cell to grow and divide. This is essential when you are an embryo because you need to grow a lot (and in the right way) to eventually become an adult.
But if there is too much of this message when you are already an adult then it can cause an excessive growth of cells. We call this cancer. So understanding and potentially controlling this message, which is passed along by SMO, is very important.
We know that SMO carries the message but we don’t know precisely how.
So what have you discovered?
Well, the SMO protein machine has three different parts: one on the outside, one on the inside and one that spans the cell’s wall to connect up the other two parts (so it goes all the way through the wall).
Previous studies have looked at the outside part by itself or the wall-spanning part by itself. But we know that both these parts are important and we think that they interact with one another in carrying the message.
So what I did was to look at both of them together: outside part plus wall-spanning part (ignoring the inside part for the moment).
Sounds simple, right? The rub was that no one had actually done this before, solved these two parts together. It’s hard enough to get a picture of just one by itself. (In other words, my PhD project was bloody risky.)
But I gave it a go and I was very fortunate that it worked out.
Once we’d gotten the new three-dimensional picture of SMO, I found that these two parts of the machine do indeed interact very closely with one another. And what’s more, I found that a single molecule of cholesterol is very important to this interaction, sitting right in between the two parts. No one had ever seen this cholesterol before.
We (fourteen authors means I didn’t do it all myself!) did a bunch of different experiments to confirm that the cholesterol does actually sit in that position under normal circumstances (i.e. not just in the sample we shot with X-rays). For example, we altered the shape of the protein machine around the cholesterol to see if we could knock the cholesterol out of its place (we could) and also to see if this caused issues for message carrying (it did). This told us that cholesterol was indeed important for the signal and wasn’t just there for no reason (which sometimes happens in biology – not everything seems to have a purpose, unfortunately).
We also managed to get a picture of SMO that was stopped from working by an anti-cancer drug, called “vismodegib”. Think ‘spanner in the works’ – where the drug is the spanner and the protein is the works. The drug keeps the protein stuck and not working (i.e. not passing on the message).
This drug has been on the market since 2012 but until now no one had seen precisely how it works (this situation is disturbingly common – very often we know that drugs work without knowing precisely how they work).
This new picture of SMO with the drug explains why some people’s cancers do not go away after treatment with this drug. In these cases the place where the drug sits within SMO is a different shape from normal (i.e. it is mutated) so that the drug simply cannot fit in there and therefore it cannot stop the message.
Curiously, when this drug is present, cholesterol is not, even though they sit in different positions from one another.
Comparing these two different views of SMO also gives us some idea of how the signal might move from from the outside to the inside of the cell wall: the outside part can change position relative to the wall-spanning part and a protrusion from the wall-spanning part seems to act as a lever between the two. In this machine, the relative movement of the parts may be how the message is moved.
With this new information, perhaps we can devise better strategies to treat certain types of cancer. Particularly the nasty ones that don’t respond well to existing treatments. However, there is still much work to be done along this path before anything is likely to hit the clinic.
Cool! Nice one! But what is a ‘Wnt’? A ‘GPCR’?
‘Wnt’ and ‘Hedgehog’ in this context are simply groups of different protein machines that operate in a chain to carry a message from one place to another. These groups are referred to as ‘families’. The protein machines within these families often have cool/weird names – Sonic Hedgehog (yes, you read that correctly), Wnt (this is actually a mixture of two previous names), Smoothened, Frizzled, Patched, Suppressor of Fused, Dishevelled, … you get the idea. Most were named as a result of experiments originally done on fruit flies. So you can blame the fruit fly people. We certainly do.
‘GPCR’s are a very common type of wall-spanning protein machine (there are over 800 different ones in the human genome) – of which Smoothened is one. They essentially look like a bundle of seven sticks. Each of these sticks goes through the wall and usually there is a small space in the middle of the bundle for a drug (or something else) to sit. These are difficult proteins to work with but are generally good targets for drugs. The Nobel Prize in Chemistry in 2012 was given to the pioneering structural biologists who got the first pictures of them. Took them about 20 years to get there. No sweat.
So, I hope that clears a few things up! Let me know if there’s anything else specific you’d like me to explain in non-jargon. Thanks for reading!
