Editorial: “Time for a new food revolution“
There’s plenty of room for improvement in the ways plants turn sunlight into food – the technology has been around for a billion years
FOOD prices high right now. And although they are unlikely to stay that way, the long-term outlook is clear. From an increasingly rich population in Asia demanding more meat to the weather growing ever wilder, there are many reasons to think . So the focus is once again turning to ways to boost crop yields, just like 50 years ago, and a few biologists have grand plans for achieving this. Instead of tinkering around with the body, like conventional breeders, they want to upgrade the engine.
Advertisement
Plants pirated the machinery they use for photosynthesis from bacteria more than a billion years ago. The same machinery is found in every single plant today, from tiny insect-eating sundews to colossal redwood trees – and it has barely changed. For all the architectural complexity that plants have evolved, they are still powered by the same engine they have had from the beginning. That’s a bit like building a modern aircraft carrier and powering it with a Victorian steam engine.
By contrast, while photosynthetic bacteria don’t look like they have changed much, some have made big improvements under the hood. They can convert carbon dioxide into food far more efficiently than most plants, and many are also able to make their own nitrogen fertiliser. If crop plants could be upgraded with just some of the improved machinery found in modern bacteria, agriculture could be revolutionised once again.
The story begins about one-and-a-half billion years ago, when photosynthetic bacteria were enslaved by a more complex cell. The descendants of those bacteria lost their ability to live independently and evolved into the cellular solar power stations known as chloroplasts. There is one amoeba whose “chloroplasts” evolved from a modern cyanobacterium around 60 million years ago, but the chloroplasts in all algae and plants derive from a single ancient cyanobacterium.
Why does this matter? Think of the countless cyanobacteria living in the sea. If a mutation enables one cyanobacterium to photosynthesise more efficiently, it will grow and reproduce faster, and its descendants could come to dominate a population within weeks. The rampant gene-swapping among simple cells means other kinds of bacteria could acquire this trait too.
Now suppose that same mutation occurs in a chloroplast in a plant. It might not be beneficial in plants, as what is good for chloroplasts can be bad for the host cell. Even if the mutation is beneficial, the odds are the chloroplast is in a leaf that will fall to the ground and die. And even if the mutation occurs in a cell that eventually gives rise to a new plant, the much slower reproduction rate of plants means it will take many decades for the mutation to spread through a population.
These differing rates of evolution perhaps explain why, as levels of carbon dioxide in the atmosphere fell over the past billion years, cyanobacteria evolved an elegant means to adapt to this change, while plants only managed a costly compromise.
Primordial air
Plants need CO2 for making food via photosynthesis. They add this CO2 to another molecule, using an enzyme called rubisco, and by doing this over and over again, plants acquire the carbon needed to make sugars, proteins and fats. When photosynthesis evolved around 2 billion years ago, the Earth’s atmosphere contained lots of CO2 and no oxygen. It did not matter that rubisco is in fact happy to grab either CO2 or oxygen – and when it grabs oxygen it destroys food instead of making it. As CO2 levels fell and oxygen levels rose, though, this unwanted reaction became ever more common, making photosynthesis less and less efficient.
As rubisco’s wayward tendencies became a serious handicap, cyanobacteria evolved a way to increase CO2 levels within their cells up to a thousandfold, recreating the ancient CO2-rich atmosphere rubisco evolved in. They do this by putting the enzyme inside tiny internal compartments called carboxysomes.
Unlike the membrane-bound organelles found in plant and animal cells, these microcompartments are made out of proteins and have regular geometrical shapes, like the shells of viruses. The beauty of carboxysomes is that while it is difficult for CO2 to escape from them, bicarbonate ions – formed when CO2 reacts with water – can diffuse in. Inside them is another enzyme, carbonic anhydrase, that converts bicarbonate back into CO2.
CO2 levels in carboxysomes are boosted still further by transport proteins in the outer membranes of the cyanobacteria. These actively pump bicarbonate ions and CO2 into the cell. Most likely, cyanobacteria evolved these mechanisms just 400 to 350 million years ago. The amount of CO2 in the atmosphere plummeted during this time, as plants spread across the land.
Green plants took another direction. They evolved a slightly different form of rubisco that is less likely to grab oxygen. The catch is that it also works far more slowly, so plants have to pack their chloroplasts with vast quantities of the enzyme to keep photosynthesis ticking along at a reasonable rate. A quarter of the nitrogen plants need is used just to make rubisco. “Crop plants may have missed the boat in terms of getting some of the advanced carbon-dioxide-concentrating mechanisms that cyanobacteria have,” says Dean Price, a plant molecular biologist at Australian National University in Canberra.
Some plants have evolved something similar, though, although it is not as efficient. In the past 20 million years or so, as levels of CO2 fell to new lows, a few found a way to concentrate CO2 using a process called C4 photosynthesis. Two important crops, maize and sorghum, are C4 plants, and a big effort is now under way to transfer this trait to other crop plants such as wheat and rice (New Scientist, 14 September 2010, p 40).
Price and his colleagues are taking a different approach. They are now working to upgrade the chloroplasts of crop plants with cyanobacterial innovations. The quickest way to do this is just to add some of the bicarbonate pumps to chloroplasts. In plants, CO2 merely diffuses through the chloroplast membranes, so active chloroplasts can have 20 per cent less CO2 than the rest of the cell.
Two of the pumps, SbtA and BicA, are each encoded by a single gene and so should be relatively easy to transfer. “We think it’s doable,” Price says. “It’s just a matter of having enough funds to try enough options.”
If Price and his team succeed – and he thinks they will within three years – that single small upgrade could boost photosynthetic rates by between 15 and 25 per cent, he calculates. “These are the sorts of numbers that make plant breeders quite interested,” he says. “They’re interested in 3 to 5 per cent, so if we can get 15 per cent, that would make a big difference.”
Grand goal
The grander goal is to engineer the whole carbon-concentrating mechanism of modern cyanobacteria into chloroplasts. This would involve adding at least eight or nine genes to chloroplasts: about five for the proteins that form the carboxysomes themselves, plus those for transport proteins and carbonic anhydrase. They might also have to add the bacterial version of rubisco and turn off carbonic anhydrase elsewhere in the chloroplast, to prevent it from turning hard-won bicarbonate into CO2 before it gets into carboxysomes.
“There’s a lot that has to be just right for this to work,” Price says. “Some of the proteins are required in very small amounts and others in very large amounts. It’s a question of getting all those balances right.”
In theory, though, it is simpler than converting plants to C4 photosynthesis, which might require tweaks to hundreds of genes. Price’s team is starting by adding carboxysomes to the lab favourite, the Escherichia coli bacterium, which is easy to work with. Once they have done that, they will try to do the same with another lab standard, tobacco. Only then will they be ready to try it in crop plants. “It could happen very quickly or it could take a long time,” Price says.
So far, nobody has built a carboxysome from scratch. But Martin Warren, a synthetic biologist at the University of Kent, UK, and his colleagues have assembled another kind of protein microcompartment from scratch in E. coli. They have even managed to get the cell to deliver a fluorescent jellyfish protein to their construct and plan to use their tailor-made microcompartment to carry out industrially important reactions (see “Chambers of secrets”). Their success strongly suggests that building a carboxysome is also feasible, says Warren. “I don’t see it as being problematic. If anything, it should be slightly easier, because there are fewer subunits,” he says.
Some people are not so sure, however. Perhaps the reason green plants have never evolved anything like carboxysomes, says Andreas Weber, a plant biochemist at Heinrich Heine University in Duesseldorf, Germany, is that they do not work in plants for reasons we may be about to discover.
Weber is optimistic about the prospects for another kind of upgrade – changing how plants obtain nitrogen. The element is vital for making protein and thus for growth, but most plants can only get nitrogen by absorbing whatever nitrogen compounds happen to be in the soil. This is why farmers usually have to add expensive nitrogen fertilisers, which damage the environment in many ways, including by increasing emissions of a potent greenhouse gas called nitrous oxide.
If crops could get nitrogen directly from the atmosphere instead, like some groups of bacteria and cyanobacteria, the benefits would be huge. Of course some plants, notably beans and peas, already exploit nitrogen-fixing bacteria by housing them in nodules in their roots. Can’t we just get these symbiotic bacteria to associate with a wider range of crops? The problem is, no one is sure how to achieve this. “It would take a long time to understand all the genes necessary to have a symbiotic association with a higher plant,” says James Golden, a molecular microbiologist at the University of California at San Diego.
What is understood, though, is how bacteria fix atmospheric nitrogen, so genetic engineers are now considering adding the genes involved directly to plants. The chloroplast is the obvious place to put them. For one thing, cyanobacterial nitrogen-fixing genes might work best in this setting. “You’re essentially putting bacterial genes in a bacterial system,” says Eric Triplett, a microbiologist at the University of Florida in Gainesville.
Juggling act
Better still, chloroplasts already produce several enzymes closely related to those used in nitrogen fixation. Borrowing these components could greatly reduce the number of genes that need to be added to just eight or so. “That’s still a lot, but it’s less than the 20 we used to think,” says Triplett.
The catch, and it’s a big one, is that the key enzyme in the nitrogen-fixation reaction – nitrogenase – is destroyed by oxygen, yet photosynthesis produces oxygen. The problem is so serious that in some filament-forming cyanobacteria, one in every 10 cells or so turns off photosynthesis and specialises in fixing nitrogen – a rare step towards multicellularity in bacteria.
However, a few bacteria, such as Azotobacter, perform a more delicate juggling act. They fix nitrogen at night and during the day they chemically convert their nitrogenase to an inactive form that is not destroyed as oxygen levels rise during the day. It might be possible to attach a molecular “timer” to nitrogen-fixing genes that could then be inserted into a chloroplast, Golden says. Alternatively, the genes could be activated only in root tissues where photosynthesis does not occur.
The possibility of creating crop plants that produce their own fertiliser in this way is more than just idle speculation. At least one major company is already trying to engineer nitrogen-fixing enzymes into crop plants, says Ray Dixon, a molecular biologist at the John Innes Centre in Norwich, UK, though he cannot reveal any details.
The billion-year upgrade is a few years away at best, but if it happens, the pay-off could be as great as that of the Green Revolution in the 1960s. “These are seriously ambitious schemes,” says Dixon. “You’ve got to try them.”
Chambers of secrets
The textbooks will tell you that bacteria are not much more than bags of chemicals. In fact, many contain protein-walled microcompartments.
By concentrating chemicals in these reaction chambers, the speed of certain processes – such as capturing CO2 during photosynthesis (see main story) – can be greatly increased. Microcompartments can also protect cells by keeping toxic by-products contained until they can be converted into harmless forms. They are essentially tiny factories within cells – and their potential is making big industries pay attention. Researchers have already begun creating bacteria with customised microcompartments designed to carry out novel reactions.
For example, Todd Yeates, a structural biologist at the University of California at Los Angeles, is trying to encapsulate all the enzymes required to produce butanol, a potential biofuel. This should help keep intermediate products in the pathway from leaching away and so increase the reaction’s efficiency, he says. Martin Warren at the University of Kent has already achieved this with another simple biofuel, though he is unwilling to give further details until the work has been published.
Our imagination is the only limit on the possibilities, Warren says. “I don’t think people have realised just how easy and manipulatable these things are.”
