AUSTRALIAN FRONTIERS OF SCIENCE, 2008

The Shine Dome, Canberra, 21-22 February

Mitochondrial proteomics in plants: From big lists to a functional understanding of an energy organelle
by Professor Harvey Millar

Harvey Millar has a degree and PhD in plant biochemistry and molecular biology from the Australian National University. He has previously held fellowships in the UK and Australia and is currently an ARC Australian Professorial Fellow at University of Western Australia. He has been awarded the Peter Goldacre Medal from the Australian Society for Plant Scientists, the Premier's Prize for Early Career Achievement in Science for Western Australia, and the Science Minister’s Prize for Australian Life Scientist of the Year. His research is focused on understanding the role mitochondria play in the primary carbon and nitrogen metabolism of plants and their response to oxidative stress.

What I want to talk to you about is mitochondria, and in a plant context this is always a bit interesting, because in primary school we are all told, very early on, that animals do respiration and plants, well, they do the other thing. They do photosynthesis. For many people, then, it is somewhat unnatural to start talking about respiration and mitochondria in plants, but indeed they do have mitochondria and they do respire.

Today I want to talk to you about mitochondria in the context of something that we call plant energy biology, which is to understand these organelles in a plant perspective, in an integration of a number of organelles of the cell.

So what on earth is plant energy?

(Click on image for a larger version)

You can look at this in a number of different ways, but here are some scale ideas to give us an idea of what we are talking about when we say 'plant energy'.

There are about 100 gigatonnes of CO₂ fixed per year into carbohydrate by land plants. That is equivalent to about 4000 exajoules, 1018 joules, of energy. To look at this activity from a perspective of energy: the annual plant energy 'harvest' from the sun into carbohydrates really dwarfs our annual fossil fuel energy usage. But we can't access this energy, because plants have it. They have it, and they kind of want to keep it, thanks very much.

The leaf area index on this slide is here to show that a lot of this activity of land plants is, obviously, focused around the equator but indeed it operates on most of the continents on the planet.

Another way of looking at it is to bring it down from those crazy numbers of exajoules and gigatonnes to think about what is happening each day.

Here is an interesting aspect of plant energy biology. There are about 10 million tonnes of leaf starch which is being made each day by photosynthesis. This is only a proportion of the carbon that the plants take in, but they make starch so they make the equivalent of flour, if you like in their leaves. If you calculate this, you realise that they make about a kilogram for every man, woman and child on the planet, every day. But do they give it away to the people who are starving? No. At night they break it down and use it themselves for respiration. So they have a midnight snack of 10 million tonnes, just to get them through the night.

If we take it down to a minute, we see that forests around the world are making sugars at about 100,000 tonnes a minute from sunlight, and they are breaking down about half of those sugars every minute, through respiration, to make ATP (adenosine triphosphate) for plant growth. And the other half, roughly, is being used for plant growth, actually biomass accumulation if they lose it all, they won't grow.

(Click on image for a larger version)

We tend to interact with plant energy and these concepts from the perspective of the major crops that we grow. We think about canola and the oil, we think about sugarcane and the sugar, we think about various grain crops and the starch that they make.

(Click on image for a larger version)

But underlying these major products is a very complex series of biochemical reactions which determine how much of this is made, when it is made, how that responds to environmental conditions, and indeed how that process is regulated inside the plant for its own purposes.

While we can attempt to look at crop plants in various ways, a lot of the science that I want to talk to you about is very difficult to do in crop plants, simply because our tools are too weak to use in these systems.

(Click on image for a larger version)

So a lot of scientists in plant biology now use our equivalent of the lab rat. This slide illustrates our 'lab rat', a plant called Arabidopsis thaliana.

(Click on image for a larger version)

Even though there are various aspects of this plant which don't really replicate what is happening in crops, when you come down to the basic biochemistry, the basic pathways, the basic energy biology, you find it is really very conserved and very consistent in all plants. So we can use this model to understand energy biology in plants.

If we really want to understand what is happening in energy biology, we can't simply look at a field and figure it out; we actually have to start looking at what is happening inside the cells. We have to go subcellular.

(Click on image for a larger version)

If we look inside cells and this is an artist's impression (I could show you a real one, but it wouldn't be as pretty) what we see is that eukaryotic cells are filled with sub-compartments. Those compartments contain about 30,000 different proteins, in the case of the plants. Each compartment is its own little 'mini-factory', and the proteins are the machines in those factories.

(Click on image for a larger version)

When we think about plant energy biology, there are basically three contenders in this process. The first is the chloroplast, the thing that makes the plant a plant and not an animal: we have chloroplasts as well as mitochondria in plants, whereas as animals we only have mitochondria. Scientists predict that there are something like 3500 discrete protein types that are working in this substructure to do photosynthesis. What are they doing?

They are doing a variety of things, but one of the main things we think about is that they use carbon dioxide and water and, of course, light to make oxygen and to make sugars.

(Click on image for a larger version)

Another sub-compartment, which very few people know much about, is called the peroxisome. These are strange little structures. They are also organelles. They are basically involved in degrading fats and degrading oxygen radicals, and there are probably about 500 proteins involved in that process.

(Click on image for a larger version)

Then there are mitochondria. We predict that there are something like 1500 proteins that are actually dedicated to this work in mitochondria. They use oxygen and sugars or organic acids, and they make ATP, amongst other things.

(Click on image for a larger version)

One of the unique aspects of plant biology and plant mitochondria is that the mitochondria in plants do things which are different from what is done by the mitochondria that you and I have. For one thing, they are really involved in the photosynthesis process. They recycle compounds that are required for photosynthesis to continue. So if you stop mitochondria in a plant, you stop photosynthesis: it will slowly grind to a halt.

Mitochondria in plants are also really interesting because they make vitamins. (These are called vitamins because our mitochondria don't make them and thus they are essential components we need to gain from our diet)). Things like folate, biotin and ascorbic acid are all made inside mitochondria in plants.

A lot of work has been done on the final aspect, looking at novel bypasses of the respiratory chain which give plant mitochondria more flexibility. One classic example is rotenone, which is an inhibitor of the first complex in the electron transport chain. Plant mitochondria don't need complex I they can bypass it in various ways and so they are resistant, to a large degree, to rotenone. The other example is cyanide, a classic respiratory poison that we all know kills humans really well. Plants are resistant to cyanide. They can survive with cyanide; they even can make cyanide when they want to.

(Click on image for a larger version)

We think about these three organelles together, not separately not thinking about mitochondria just by themselves, but thinking about these three organelles as a kind of system that is happening inside the cell.

As we have already been told, the nucleus is where all the genes live. And the genes have to be made into proteins inside the cell and sent to these different organelles, hence the numbers shown here: 1500, 3500, 500 proteins.

The organelles also talk to each other, and they talk to each other through things like fatty acids and lipids, amino acids, vitamins and sugars. So these commodities which are the major products of plants, the major things that we want to harvest from plants, are also the way that these three organelles communicate and talk to each other, because they are substrates and products in biochemical reactions, and they influence the function of each other and of course they influence what is happening in the nucleus.

So that is the sort of dynamic that we work with which looks great as a little diagram.

(Click on image for a larger version)

The only problem is that, even though we think there are about 500 proteins sent to peroxisomes, 3500 sent to chloroplasts and 1500 sent to mitochondria, we really don't know which ones. We have an idea of some of them we know some of the major ones but it is probably a conservative estimate to suggest that, for something like 60 per cent of the proteins that go to the organelles, we don't actually know which ones they are in the nucleus. We have 30,000 to choose from, and they are a bit cryptic as to where they are planning to go, when we just look at their sequence.

If you want to actually figure out what proteins go to which organelles and how the system works, and what all those proteins are really doing there, you need to go and do something in the lab.

(Click on image for a larger version)

What we do in plant mitochondria and this is true for many people working on mitochondria around the world is to become a mitochondriologist, somebody who studies mitochondria. It is not the organism that you work on any more, but you select these things and isolate them and purify them so that you can work with them in a tube.

We basically use size and density to do this. The different parts of the cell have different sizes and densities. If you take a tissue and put it in a blender, you get all these little bits and pieces, and they all have relationships to each other soluble proteins, ribosomes, starch grains, the nuclei, various parts of the cell they all have a density and they all have some sort of size. And so you can use centrifugation to, fairly crudely, purify a particular component based on its density and its size.

If we do this, we find that by using these density gradients, running at about 40,000 g for an hour, we can separate mitochondria away from other structures in the cell, but they are not terribly pure from plants, because mitochondria is still only a small proportion of the total cellular composition.

(Click on image for a larger version)

What we have been doing recently is to try to add a third dimension to that process, accepting that there are differences in density and that there are differences in size, but saying, 'Well, what other parameter can we use, which is truly independent of density and size, to help us in this purification process and add a third dimension to it?'

The one that we use is surface charge, the actual charge on the outside of an organelle. An organelle in aqueous solutions is not neutrally charged; it has a net charge. If we put it in a chamber for what is called free-flow electrophoresis, we can run a little spot of organelles through this process, through a laminar flow; we can apply a few hundred volts across this; and they will deviate, they will move towards the positive electrode. So we can use this as a way of separating organelles, based on their surface charge, but keeping them intact so that we can do biology on them in the end.

(Click on image for a larger version)

Then you have to say, 'Well, what did we actually get? We have done this but it is just little opaque solutions of things. What did we actually find?'

So we have done various enzymology I was going to show you that, but it's a bit boring so I thought I would show you a picture instead to show you that if you use transmission electron microscopy, you can actually go from a preparation which has these mitochondrial structures in it but also a whole variety of other stuff, other parts of the cell which couldn't be separated by that process. But if you use size, density and surface charge separations, you can get preparations which are really predominantly just the mitochondrial structures.

This lets us ask the question, 'Okay, we've got them. We've got them in a tube. We know what they are. We can even make them work and do various things with them. But what are they made of? What are the proteins that were sent by the cell to that location?'

(Click on image for a larger version)

What we do is develop two-dimensional maps. These are not small any more. What you see here is probably about 20 cm x 20 cm, so it is not a small thing, it is something you can have on your desk and have a look at. It is a two-dimensional separation of proteins which are then stained so that we can actually see the proteins on this backdrop.

What this allows us to do is to have a picture an abstract picture, but a picture of all of the composition of the mitochondria in terms of its proteins, each spot being a different protein. And the abundance of that is shown by the intensity of the colour.

We think there are about 1500 proteins, but unfortunately we can only see about 600 little spots on these gels. And when we try to identify them, using mass spectrometry which I will explain in a minute we can probably only identify about 150 different things. That is really about the abundance of them: some things are very lowly abundant, some don't appear in this kind of separation. We have other kinds of separation, but ultimately all these gel-based separation techniques that we use have intrinsic problems of sensitivity.

(Click on image for a larger version)

An alternative way is to avoid all these matrixes, gels and so on that you have to put proteins into and then try to get them out again. Instead we just take an organelle and directly attack it with proteases, so we chop it up into all of the different peptides that could be derived from all of those proteins. You take this complex mixture of peptides you have got a solution, and 60,000 different peptide types and you can separate that across HPLC (high performance liquid chromatography) gradients, you can fire that into a mass spectrometer and then, when you find a peptide in there, you can break it. In a collision-induced dissociation of that peptide, you break it up, and that produces a fingerprint for you, like the one shown here. These are all the products that derived from physically smashing a peptide into bits. And you can use this fingerprint, basically, to identify the protein.

(Click on image for a larger version)

To do that, we try to match the information to what is actually in genomes. So for this kind of biology you need the genome. If you don't have a genome, you can't effectively figure out what the proteins are. So the genomics and genome sequencing have been very helpful for this.

When you come to looking at proteins and where they are encoded in genomes, it is worthwhile knowing that although people in genomics and people who work on genomes basically say that if you have got the genome sequence you know what's going on, sometimes they fail to tell you that in fact every piece of DNA that you have can be read six different ways. And until you have read it all six different ways and you have got some way of assessing which of those different ways of reading it was the most valid, or in fact what was actually produced, it is difficult to be absolutely sure.

We are getting very sophisticated in our bioinformatics, in being able to predict what should be the gene product of a certain piece of DNA, but the ultimate proof is the fact that these things turn into proteins.

(Click on image for a larger version)

So we have techniques now where we can take these little fingerprints and link them to particular frames of the genome.

(Click on image for a larger version)

That means we can say, 'Okay, that little part there is actually making a protein, and this is the frame it is in.'

(Click on image for a larger version)

We can link that back to the genome and say, 'That must be a gene, then, because that is where that peptide must have come from. It only maps at one place in the genome.'

In this way we can, obviously, find what the genes are that encode these things. But we can also discover new protein encoding genes, as we are doing in plants, which are not annotated because they are just a bit odd, they break the rules a little bit, and it is hard to figure them out by other mechanisms.

(Click on image for a larger version)

Of course, we are not the only people doing this. The animal guys are at it, and the yeast guys as well, and so there is a whole variety now of comparative mitochondrial proteomics going on. People taking these lists of proteins they have physically found in mitochondria and asking the question, 'Okay, if this is what a mitochondrion looks like in yeast, and this is what a mitochondrion looks like in human heart, or in mouse' and we have done the same thing in plants 'what is a mitochondrion, then?'

When you do that, you find that the conserved functions, the things that are found in all mitochondria, are the things you would expect. They are the really basic machinery of doing respiration. But many of the regulatory functions and many of the ancillary functions and other pathways are actually distinct. So what we find is that mitochondria in different organisms are a place in the cell to put stuff. They have a basic role, and then it is a place where other stuff can be put if it is useful to put it there.

Maybe it is a nice micro-environment for a particular biochemical pathway to happen; maybe it is just convenient; maybe you need to keep something away from something else that it is not very friendly to, and so you put it in a particular location. And that seems to be what is happening with mitochondria.

(Click on image for a larger version)

But even though it is the case that there are big differences when you start comparing humans and plants, and people think, 'Well yes, we'd kind of expect that. They're pretty different things when you look at them,' there are still large changes in what a mitochondrion is in different parts of a plant.

If you think about a stem of a plant, or a leaf of a plant, or a flower of a plant, or in fact the roots down there in the soil, you realise they all have a different role. Mitochondria have a different composition in those different locations, because they are doing something different. And we can use techniques to start probing that.

(Click on image for a larger version)

We can use fluorescent dyes and fluorescent dyes are used very broadly in biological sciences now, for comparing two samples. You have different coloured fluorescent dyes, which means you can treat two samples separately with a different fluorescent dye, and when you mix them together and do whatever crazy stuff you do with them, you can then ultimately, using lasers, separate away which components came from which sample, because of the fluorescent dye that is attached to them.

So we do this with mitochondria. The image at the left arose from comparing leaf mitochondria and stem mitochondria. The leaf is shown in red, and the stem is shown in green. The yellow indicates when they were the same. So whenever you see green, that means it is only in the stem mitochondria; if you see red, it is only or predominantly in the leaf mitochondria; when it is yellow, it is equally abundant.

You can see the very big differences in the actual mitochondrial composition of mitochondria from these different types, because they are doing different jobs.

We can also use transcript information so, the abundance: how much are these genes turned on in a plant? We can use this information, and start looking at a whole variety of different tissues and see the way in which particular genes that make proteins that go to mitochondria turn on and off in different tissues.

We can start building what we call transcriptional networks, which is a way of trying to understand how a plant builds a mitochondrion. What is its process? What order does it do things in? What things are linked, and what things aren't linked? When you start understanding those linkages, you start understanding the dynamics of building biochemical pathways.

(Click on image for a larger version)

The last thing I want to mention, briefly, is environmental stress and understanding that.

We have known for a long time that the respiratory process is quite sensitive to environmental stress. Cold, heat, drought, salinity, waterlogging, all these phenomena occur and they have the problem that when they do occur, respiration is damaged in various ways.

So we have spent a lot of time trying to understand mitochondrial dysfunction. How do you break a mitochondrion when the conditions change when it suddenly gets a bit colder than it should do, or a bit hotter than it should do, or you have a drought situation?

(Click on image for a larger version)

What we are trying to do here is to understand mitochondrial adaptation to environmental stress. The questions we are asking are using these big approaches that are predominant now in plant biology and, indeed, in all biology, of trying to measure as many things as you can in a biological situation, trying to measure as many of the different molecules as possible at the different levels that biology is occurring at, so that we can try to understand, when we put it all together, what the whole picture is trying to tell us.

So we can look at transcripts, we can look at the genes. We have technology that allows us to study all 30,000 genes at once, so we can look at all 30,000 to see how they are changing.

At the protein level we can ask the question: how is the abundance of proteins changing during a stress event? There we are probably, at the moment, studying thousands. We can't really study them all; there are technical limitations at the moment.

Lastly, how do the changes in enzyme function alter the metabolites, the small molecules that run around in plants and accumulate, and that we call 'food'? All these things change, and we can measure them, but we can at the moment probably measure and identify only a few hundred of these, when we know that there could be thousands. So there are still great limitations at the protein and metabolite levels, in really understanding how stress is impacting on the biology that we can see at a macro-level we can see the plant isn't very happy, we can see yields go down. But to understand what the sites are, how that damage is occurring, what metabolites are changing that is still a challenge.

(Click on image for a larger version)

So we want to integrate all that data. We want to start integrating it, building physical and kinetic models. We are doing that to a limited degree, and as we build up more information about what is in there we can make our models better our kinetic models and our physical models.

When we integrate that we can start asking biological questions through these data and start visualising and trying to comprehend from these different levels of data what is happening at the gene level to what is happening at the protein level, what is happening with metabolites and then what actually happens to the plant phenomics. We can start to understand that process.

(Click on image for a larger version)

Why do we want to do that? Well, it is because we want to start engineering novel plant lines with altered attributes by specifically changing plant energy functions. Shown here are plant varieties which are currently in Perth, where we have altered mitochondrial and other functions which affect growth, improve drought tolerance and alter flowering time. If you start tinkering in the engine-room of plants, you can start having quite dramatic impacts on what is happening to the plant and how it responds to environments.

(Click on image for a larger version)

This slide acknowledges people who have been involved in this work, and some international collaborators, as well as all the other people in plant energy biology, because this is really an Australia-wide effort to try to understand plants at this fundamental level.