The evolution of photosynthesis

Julian Hibberd is based in Cambridge and he’s been trying to unpick how the more advanced “modern” form of “C4” photosynthesis evolved. Chris Smith first met him 4 years ago when Julian published evidence that C4 photosynthesis had probably evolved not once but many times, a story he’s since been able to refine and build upon…

Julian - The ancestral form of photosynthesis is effectively unchanged since bacteria initially evolved the ability to fix carbon so probably, 3.6 billion years ago. The majority of land plants still use that system and that system relies on an enzyme which sounds unfortunately like a breakfast cereal called rubisco. rubisco is great. It fixes CO₂ and makes carbohydrates. But some of the time, it makes a mistake. Instead of taking CO₂, it takes oxygen. Because the concentration of oxygen in the atmosphere is now very high compared to that 3.6 billion years ago, those mistakes happen more often that waste energy generates a toxic compound. The C4 pathway is a system which effectively abolishes those mistakes by generating a little turbo-charger around this rubisco enzyme, such that more CO₂ is supplied to it.

Chris - So back in history then. The bacteria that evolve this in the first place were living in an environment which was dominated by carbon dioxide, hardly in the oxygen around. So they didn’t need an enzyme that really could discriminate between an oxygen molecule and a carbon dioxide molecule. And it’s only in the latter era now, this is more a problem so there is this pressure to adopt this turbo-charged form.

Julian - That’s exactly right, yes. Over evolutionary time and geological time, the CO₂ has come down because photosynthesis has been so successful, and we’re now in a completely reverse situation where CO₂ is relatively scarce and there's huge amounts of oxygen. And that’s led to this conflict between fixation of oxygen by rubisco and the fixation of CO₂.

Chris - How has it changed in order to make it a better discriminator between an oxygen molecule and a carbon dioxide molecule?

Julian - So in the C4 pathway, plants have developed a complex suite of traits which allows CO₂ to be pumped from one cell type to another cell type. And in that second cell type, we have the rubisco enzyme and that increases the concentration of CO₂ specifically around Rebisco which basically abolishes then the oxygenation reaction.

Chris - So it’s like a sieve.

Julian - It is a bit like a sieve, yeah. So that sieve involves a specific set of enzymes which initially fix the CO₂ into a different form. The CO₂ is then released at high concentrations around rubisco and that’s what's improving the efficiency of photosynthesis.

Chris - And if I look at all plants that do this, have they all got one common ancestor then or does it happen with the same common end-product, you get lots of CO₂ around your rubisco, but the way you get there is different in all these different plants?

Julian - There are multiple ancestors of C4 plants, so we estimate it’s evolved at least 60 times. And we think there's a general set of machinery and C3 plants which is sort of nascent and it allows C4 photosynthesis to develop. So, a lot of the regulatory structure associated with turning genes on or turning genes off, we have discovered more recently that quite a lot of that is sitting there in the C3 situation. It’s not being used exactly as it would be in a C4 leaf but in a way, it’s ready to be used.

Chris - And so, because it had that machinery, it’s a bit like me, walking into a kitchen and saying, “Well, that’s a knife and fork where we usually use that for eating but actually, I'm going to beat an egg with that.” And so, you're using a fork which is already made as a fork but you can do a different job with it, and it’s useful for both.

Julian - Yes, and I think that’s a great analogy. Actually, some of the work we’ve done more recently is a little bit more nuanced than that. Maybe it’s like, there's a fork sitting there and we would normally use it for our dessert. We’re going to carry on using it as a fork but maybe for a different course. Sounds terribly Cambridge, doesn’t it? But if we’re thinking about these forks and we say, “Well actually, this is a gene for a fork” and the ancestral C3 plant, those genes are very poorly expressed. In the C4 pathway, they're very highly expressed. And to our surprise, what we found is that in the ancestral C3 state, a lot of those genes are already regulated by cues which turn on photosynthetic genes. So such as light, they're actually sitting there as part of that photosynthetic architecture, regulatory architecture for photosynthesis. But they're not responding to it fully, so all you need to do is amplify that response as you go from C3 to C4 and hey presto! You have the full system working.

Chris - And is your ultimate goal then to understand photosynthesis, understand how plants work so that we can do it even better?

Julian - That’s one of the potential outcomes. There's a broad consensus now that despite millions or billions of years of evolution of photosynthesis, it’s not optimised for current agricultural practice. There's a broad consensus that there are multiple ways of improving photosynthesis. And so, if we understood it enough to be able to engineer it into a C3 crop such as rice or wheat, we should be able to dramatically improve crop yield, so up to 50 per cent improvement.