September 19, 2013
Back in 2008 and 2011 there were huge spikes in the cost of food that caused riots in various parts of the world. These things were pretty bad and one of our favorite beast’s best products, ethanol, may have been at least partly to blame. In an attempt to deal with global warming, governments had created incentives that made it more lucrative to turn food into ethanol to power cars rather than keeping it as food to feed people. The law of unintended consequences reared its ugly head and caused food prices to rise high enough to be unaffordable by the very poor.
This situation arose because right now, pretty much the only commercially viable way to make ethanol is to use sugars like those found in sugar cane or starches like those found in corn. Ultimately this won’t be a problem once scientists learn to coax yeast or other microorganisms to make ethanol out of agricultural waste. Until then, though, one way to lessen the impact of ethanol production on food supplies might be to engineer a yeast strain that can more efficiently turn sugars into ethanol.
One of the most inefficient parts of yeast fermentation is that the silly thing converts anywhere from 4-10% of the sugars it gets into glycerol instead of ethanol. In a new study, Guadalupe-Medina and coworkers have engineered a strain of yeast that produces 60% less glycerol and 8% more ethanol than other commercial strains. If they can scale this up, it might help us feed both the world’s population and our cars.
It has been known for some time that yeast end up making glycerol during fermentation because of redox-cofactor balancing issues. In essence, the excess NADH that is made in fermentation reactions is reoxidized by converting part of the sugar into glycerol. One obvious way to get less glycerol would be to give the yeast some other way to reoxidize its NADH.
Guadalupe-Medina and coworkers decided to persuade yeast to use carbon dioxide instead of sugars. Not only would this make sugar use more efficient, but their particular plan would also convert that carbon dioxide into a precursor that could be shunted into the ethanol producing pathway. Theoretically the yeast should now increase its ethanol production both by wasting less sugar on glycerol and by turning carbon dioxide into ethanol. And it turns out that this idea actually worked in practice.
The first step was to introduce the Rubisco enzyme into the yeast. Rubisco (ribulose-1,5-bisphosphate carboxylase oxygenase) is really one of the key enzymes in life…it provides the foundation for almost all life on the planet by fixing carbon dioxide from the air into ribulose-1,5 phosphate. But that isn’t the important point here. No, the key point for this work is that in the process of doing this, the enzyme oxidizes NADH. By putting Rubisco in yeast, the yeast should now be able to reoxidize its NADH without making useless glycerol.
Of course this is easier said than done! Rubisco is multi-subunit in most beasts and persnickety to boot. But with a bit of work, they managed to get Saccharomyces cerevisiae to express a working copy of Rubisco.
So they would only have to introduce a single gene, the authors used the single subunit enzyme from T. dentrificans. As expected, this gene alone was not enough. They knew from previous work that Rubisco would not work in yeast without the help of a couple of E. coli chaperones, groEL and groES. When they expressed all three genes at the same time, they got Rubisco to fix carbon dioxide in Saccharomyces cerevisiae.
The next step was to introduce the enzyme phosphoribulokinase (PRK) so that the ribulose-1,5 phosphate could be converted into 3-phosphoglycerate, a precursor in the ethanol pathway. Luckily this was much easier than Rubisco and worked on the first try. They had now engineered a Frankenyeast that should be able to make more ethanol and less glycerol.
When they tested the new strain, Guadalupe-Medina and coworkers found they had indeed engineered a more efficient yeast. In anaerobic chemostat conditions, this yeast made 68% less glycerol and 11% more ethanol than the usual commercial strain. They obtained similar results, 60% less glycerol and 8% more ethanol, in batch fermentations. They had succeeded in improving an already awesome beast.
If this strain works on an industrial scale and if commercial producers all used this strain instead of the ones they currently use, the authors calculate we could get an extra 5 billion liters of ethanol added to the 110 billion we are already making. That might just be enough to tide us over until scientists come up with a way to make ethanol commercially from non-food sources.
by D. Barry Starr, Ph.D., Director of Outreach Activities, Stanford Genetics
Categories: Research Spotlight