Metabolomics

'Omics' words tend to get a large amount of bad press in biology. Starting fairly sensibly with genomics, the group expanded to include such things as proteomics, metabolomics, transcriptomics and then when a bit crazy, with seemingly every branch of biology (and possibly even a couple of physicists) wanting to find an 'omics' to work on. The 'Tree of Life' blog even started giving out awards to some of the more hilarious ones.

Omics words are like the cool silent kids in black leather jackets with a hint of drug-taking about them. One of them is fine, a couple of them can be fun but if you get too many it just starts getting messy and looks a bit gratuitous. And eventually people realise there's nothing particularly amazing about them and they become Public Enemy Number One. Which is a pity as many of them are actually quite sensible and can be very useful.

As far as I've been lead to believe 'ome' means 'complete set of '. The genome is the set of all the genes in a cell, proteome the set of all proteins, transcriptome the set of all mRNA transcripts made from the DNA and the metabolome is the set of all metabolic reactions taking place in the cell. 'Omics' is the study of 'omes'.

On the face of it, metabolomics looks to be a mindnumbingly insane task. To study and document every metabolic reaction happening in a cell, to create a model of it, and then use that model to predict how levels of substances change in response to changing conditions seems almost impossible. Just to give some idea of the task, here's a quick diagram of a couple of metabolic pathways involving manipulation of carbon chains:

Diagram taken from the SYSFYS project carried out by the University of Helinski Computer Department. This picture is one of two reasons I am currently studying biochemistry.

Each little dot on the diagram above represents a metabolite, and that diagram doesn't include enzymes, or the things that affect enzymes, or any method of regulation. It doesn't include quantitative analysis of the flux through every pathway, and how different concentrations of metabolites or regulators affect that flux. All of that information is tied up in metabolomics.

One of the first things that's noticeable about that diagram (other than that it looks like the London Tube Map as designed by Tim Burton) is that all the branches appear to be interconnected, everything is joined together. They become a lot more interconnected when you start considering regulation of each step, as many of the metabolic enzymes are regulated by similar compounds (ATP, for example). And what that means is that a change in the levels of one metabolite can have an unprecedented effect on the levels of another. More importantly, anything that accidentally gets missed out of the diagram could cause the model to work incorrectly. To put it in (slightly) more mathematical terms: there are a lot of parameters floating around.

How do you even start studying something like that?

Analysis can be split into two broad categories, open and closed. Open analysis involves taking a sample, and looking for metabolites. It's primarily used to find novel entities, and is rather open ended, in that you start without much of an idea of what you're going to find. 'Looking for metabolites' is done by pretty much any method used to detect proteins; mostly NMR spectroscopy, Liquid and gas chromatography, various Mass Specs and chromatography methods which would take up a whole blog post on their own (which I can write, if anyone's interested, it will be good revision).

Closed analysis focuses on a specific molecule (or molecules) and tries to find out as much as possible about them; what they interact with, what interaction rates are, how it's reactions are controlled, etc. This can be a lot more sensitive than open analysis, and you start with a clear idea of what you're searching for. Apparently it's better for producing papers as well.

By looking for different proteins, and then examining them in detail, a picture can be gradually built up of the metabolic pathways and their interactions. While I'm sure this has many uses in humans (for medical purposes) one of the applications I've been most exposed to is (surprise, surprise) in bacteria, where an understanding of existing metabolic pathways can be used to enhance synthetic ones. By playing around with the enzymes and fluxes of pathways involved in (say) a certain antibiotic precursor, you can encourage bacteria to be far more productive in antibiotic synthesis.

Like all things, metabolomics is at it's best when combined with other methods to give a fuller picture. The information gained from both metabolomics and transcriptomics was used in the reference below to find a key transcriptional compound, Stearyl-CoA desaturase, involved in fatty liver production. Fatty liver is formed from lipid accumulation in the liver, caused by orotic acid supplementation in rats, and excessive drinking in humans. The metabolic diagram below shows the effects of the orotic acid addition (green denotes an increase in a substance, and red a decrease) which for anyone who is not instantly able to pick out glycerol substrates (like me) simply shows just how much work is involved in metabolomics, and how complicated it can get.Despite my fascination with metabolomics and the pretty diagrams they produce, I don't think it's an area I would really go into. Nevertheless it's produced some very fascinating results, with some very worthwhile applications for many different scientific disciplines.

---

Griffin, J. (2004). An integrated reverse functional genomic and metabolic approach to understanding orotic acid-induced fatty liver Physiological Genomics, 17 (2), 140-149 DOI: 10.1152/physiolgenomics.00158.2003

---

Follow me on Twitter!

Exploring protein interactions: yeast two hybrid systems

>Proteins are one of the key molecules inside cells; involved in signalling, intracellular transport, metabolism and gene control. They rarely work alone, most of the proteins in the cell are part of large complex networks consisting of many interacting proteins. Various techniques exist in order to find these interactions, and one of the most common is the use of yeast two hybrid systems.

Yeast is apparently quite a nice organism to work with (I've never worked with it myself, I must say, apart from a few practicals in second year, where I almost set my lab-partners hand on fire). The genome is fully annotated, the organism is well characterised, and yeast grows and responds quite fast, so experiments shouldn't take too long.

The yeast-two hybrid system is based around molecules called transcription factors, which are normally used by the cell to active gene expression. In yeast, there are several transcription factors which consist of two separate molecules, which need to be in close proximity in order for a gene to be expressed. If you attach an experimental protein to one half of the transcription factor, and another experimental protein (that you think interacts with the first) to the other half you can test for interactions. If the two proteins do interact then the transcription factors will be brought close together and the gene downstream of them (which acts as a reporter gene) will be expressed:The diagram above (taken from the reference below) shows this process. The blue jigsaw-shaped 'X' and 'Y' proteins are the experimental proteins, being tested for interactions and the yellow shapes are the two parts of the transcription factor. The big white cloud is the polymerase, which begins the process of turning the DNA into protein. The reporter gene can be set to code for a vital compound such as histidine; stick the whole system into a histidine deficient mutant and you have a marker system to see if the proteins interact. If they do, the histidine is produced and the cells can grow, if they don't, the cells die.

One of the most useful things about this technique is that it can be automated, and used to scan whole libraries of proteins to see if they interact. By using a matrix, each protein X (the 'bait' protein) can be given a defined position and then systematically exposed to a number of different protein Y (the 'prey' protein). If you're taking Y from a clonal library, and have a sufficiently intelligent robot, the whole procedure can be carried out with minimal human input.

The yeast two-hybrid system has been invaluable for determining many important protein-protein interactions however there are some problems with it. Firstly, this is a yeast two component system, and most protein complexes consist of many interacting proteins, certainly more than two! Secondly, this whole system relies on two soluble proteins interacting in the nucleus (where the DNA is) and so doesn't work for membrane bound protein interactions.

In view of this, several modifications have been made to the original methodology to make it more useful for trapping a wider range of protein interactions. It's been expanded into the three component system, which identifies proteins that interact with (or inhibit) both the the bait and the prey. Using another natural yeast system (the G protein system) has allowed transmembrane proteins to be identified as well:
In this system protein Y contains a binding site for a subunit of the G protein, while X is a membrane-spanning protein. G proteins are membrane bound proteins that activate transcription factors inside the nucleus. If the two proteins interact then the G protein subunit bound to the Y is sequestered away from the rest of the complex, and the G protein signal cannot be transmitted. There is therefore no signal to the transcription factors, meaning reporter genes (in this case genes for pigmentation rather than for death) will be turned off.

Even with these modifications there still are problems with the yeast two hybrid system. There is very little quantitative analysis involved to see how strongly the protein is binding, and the discovery of new interactions will always be limited by the choice of proteins to screen. However one of the largest advantages of yeast two hybrid systems is that it carries out protein binding analysis inside the cell, in proper cellular conditions, unlike affinity binding which, while it can identify much larger protein complexes, involves taking proteins out of the cell and handling them in vitro.

Yeast two hybrid systems are therefore still one of the main practical methods used for determining protein interactions. The field of systems biology is a fascinating one, and while attempting to catalogue the whole spectrum of cellular interactions may seem like a daunting task ('interactomics' for the funding people) working towards it will only reveal more and more useful information about the complex and fascinating networks of proteins within the cell.

[btw: The reference below is a great source of information about the many different modifications made to the basic yeast two hybrid system, thoroughly recommended for anyone interested in protein interactions or systems biology]

---

Brückner A, Polge C, Lentze N, Auerbach D, & Schlattner U (2009). Yeast two-hybrid, a powerful tool for systems biology. International journal of molecular sciences, 10 (6), 2763-88 PMID: 19582228

Lab Rat guide to Fourier Transformations

Protein crystallography is one of those mysterious things that I always feel I should know more about. It starts with protein crystals, fair enough, and then heads rapidly out beyond the Magical Event Horizon leaving me with a picture of set of small fuzzy dots which mysteriously resolve themselves into an electron density map. Somewhere in all this someone will mention the dreaded Fourier transformation, at which point I know all hope is lost and a might as well stop listening.

With the help of a good lecture and an utterly outstanding website though, I'm starting to get the hang of it. X-ray crystallography works by shooting X-rays at a protein crystal. As the crystal is in a regular lattice structure, the X-rays are scattered in a regular way. Each scattered ray can be characterised by the amplitude and the phase, as shown below:Each dot on the resulting image (which usually looks something like the figure shown on the right) corresponds to a scattered X-ray. The position of the dot gives (by way of a set of clever equations) the amplitude of the wavelength. However in order to calculate the position of an atom you need both the amplitude and the phase, and you can't get the phase from just the position of the dot. And here is where the dreaded Fourier transformation comes in, to help give information about the phase.

The Fourier transformation is an equation, which gives reciprocal information about a molecule, for example if the molecule is a single small point, the Fourier transformation (i.e what it looks like after plugging its positional coordinates into the Fourier equation) will be a large fuzzy blob, as shown below (all pictures from now on are taken - with permission - from Kevin Cowtan's Book of Fourier) . Molecule is on the left and its Fourier transformation on the right:
For the more mathematically minded, what the Fourier transformation actually does is take a function and express it as the sum of a set of different sine and cosine waves. Apparently this can be done for any continuous function. By using this transformation you can get not only the amplitude, but also the phase of any given molecule inside an atom. You can also make the equation go backwards as well, turning the red fuzzy blob on the left back into the individual point. As most molecules contain many, many atoms there are various tricks you have to do in order to make this easier for large molecules, but first a quick proof of this concept using my favourite pictures on the Cowtan website.

Here is a picture of a duck:
And here is the Fourier transformation of the duck, with the amplitude represented by the brightness of the colour and the phase represented by the actual colour:
In order to show that the Fourier transformation really does show the phase, this transformation is mixed with the Fourier transformation of a picture of a cat. Rather than mix them both equally, the amplitude (brightness) of the duck transformation is mixed with the phase (actual colour) of the cat to give the following Fourier transformation:Performing the Fourier equation on this blob (to turn it back into an animal again) gives the following result, hopefully not unsurprising if I've managed to explain this alright:
It's a cat! A rather blotchy cat, true, with a fairly trippy background, but nevertheless the Fourier transformation has faithfully reproduced the phases, rather than the amplitudes.

So how does this help when looking at molecules? It turns out that the fuzzy black-and-gray-dots picture that is the end result of X-ray crystallography (shown above and reproduced here on the right) is the Fourier transform of the atomic electron clouds inside the protein crystal. That picture is like the fuzzy coloured blobs that the duck and cat images came out of. In the same way that those turned into animals, this picture can be turned into the approximate shape of the electron clouds surrounding a molecule. For low resolution images this can show the secondary structure of a protein, the positions of alpha helices and beta sheets and a general idea of protein shape. For high resolution images, individual amino-acid residues can be seen, allowing a much more detailed view of the structure to be generated.

It isn't always perfect. Sometimes you do get the equivalent of a blobby cat with a trippy background and have to play around with homologous comparisons and allowed bond-angles to get a meaningful structure. There are plenty of strategies that exist to help you get a better image as well, particuarly for larger molecules which need more help resolving phases. From what I've heard though, once you've actually got the crystal, the rest seems like childs play in comparison. I know people who have spent their whole PhD's, and longer, just trying to isolate and concentrate a single protein crystal...

Making Mutants

Nowadays there are many different techniques for looking at gene and protein structure and functions. You can make protein crystal structures, you can see what substrates the protein binds too, you can do various chemical assays to open the protein up and see what it looks like inside. The most classic way however is the scientific equivalent of hitting it until it stops working, and then seeing what you've damaged. This technique, in a slightly more sophisticated wording, seems to be the cornerstone of much of biochemistry, and probably developmental biology as well.

In most cases the 'hitting' is a lot less random than I've probably made it sound. Say you have a stretch of DNA that binds to a protein, and you want to know which parts of the DNA actually physically bind to the protein. The best way to do this is to get the sequence of interest and change the base pairs (that make up the DNA sequence) very specifically, to see which changes stop the protein binding.

Starting with the hypothetical DNA sequence AATATAT. In order to find which bases bind to your protein, you need to make a few very specific point mutations. The way most of the labs around me do this is with a kit from Stratagene called QuikChange (R). Your DNA sequence is likely to be inside a plasmid (small circular piece of DNA) so you design a small primer with the change you want to make, i.e AAGATAT. Adding this to the plasmid, along with some polymerase to expand the DNA, and some nucleotides to expand it with and you get a perfect copy of the plasmid, perfect except for the small difference of the T-G mutation.

QuikChange provides plasmids for you to put your DNA sequence in, and these plasmids have been methylated; some of the DNA base-pairs will have methyl groups attached. To get rid of this original plasmid (after all, you only want your mutated copy, not the un-mutated original) you use a restriction enzyme (DpnI) that literally chops up methylated DNA, leaving nothing behind but your mutated sequence.

Then you add your protein, and see how it binds. If the binding is still just as strong, then that clearly wasn't an important residue. If the binding is weaker, or if less of the protein binds, then that might have been one of the important ones.

(You can of course just use PCR with mutated primers to create single mutants. But you do run a risk of introducing other accidental mutations through the PCR process. And when it doesn't work it's incredibly irritating. I did some research over the summer which proved conclusively that it is possible for PCR mutagenesis to not work for a continuous period of over two months)