Back in ancient Greece, in the 5th century B.C., Hippocrates recommended using the bark of the willow tree to treat pain. A few thousand years later in the 18th century, chemists figured out they could extract a painkiller from willow bark called Salicin. And then in the 19th century, they realized that instead of extracting the painkiller, they could make it themselves in test tubes and improve it, as salicin and it's relative salicylic acid were foul tasting and caused stomach bleeding. This eventually led chemists to the creation of aspirin.
"They went from studying the nature of molecules they found in the environment to actually trying to make them," says Jeff Boeke, a professor at Johns Hopkins University. "It's something that's happening in biology right now, it's referred to actually as synthetic biology."
Manipulating yeast DNA
He and other synthetic biologists are starting to do with living creatures what chemists did with chemicals 100 years ago. In Boeke's case he works a lot with yeast, and he decided he wanted to whip up not aspirin or synthetic sugar, but a much more complex chemical: DNA. The Venter Institute did this last year with a bacterial genome, but the yeast genome is much larger and more complex.
"Yeast cells are more like human cells," says Boeke. "They're more advanced in some sense, they have chromosomes that are more like human chromosomes."
So Boeke took aim at reproducing one of those chromosomes -- about 1 percent of the total yeast genome. While he was at it, he thought, why not make some improvements. Using software his team developed, and enzymes and markers to snip, his team basically edited Mother Nature's blueprint for the yeast.
"What a lot of people call Junk DNA, which are repetitive sequences that I believe are the spice of the genome, but I also believe in yeast at least, we will be able to remove every single scrap of those sequences and have no impact on the life of the yeast."
Spinning the wheel of genes
Repetitive Junk DNA -- also known as transposons -- can move around and cause all kinds of tomfoolery in the genome that would mess with biologists' future engineering plans. So Boeke's team cut it out. But they did something else, too. They engineered a scramble switch in the yeast's DNA to cause massive, random, mutation on demand among the yeast's non-critical genes.
"You could consider the yeast genome a deck of 5,000 cards," says Boeke. "We can shuffle the deck and we could make 1,000 different shuffled decks and see how those different decks behave. One deck might be a winning hand for poker, another might be a winning hand for Pinochle or some other game."
So every time they press the mutate button (it's not really a button -- they send a chemical signal using estrogen) they get thousands of complex mutations. Ideally, that could yield some really useful yeast. For industries playing the games of biofuels, vaccines, or medicines - all of which depend on yeast - Boeke says it could produce some real winners.
Stepping into the world of designer genomes
Jim Collins, a professor of biomedical engineering at Boston University, says Boeke's work editing down the genome is an interesting step for the field. And he says it's one more step toward discovering what's called the minimal genome, which is the smallest number of genes needed to have a functioning living cell. It's a simple, stable platform that will allow bioengineers to build on as they move into the brave new world of designer genomes.
But while this is a major step for synthetic biology, it's also a sign of how far the field still has to go. Peter Enyeart, a graduate student at the University of Texas at Austin, helped review Boeke's paper for the journal, "Nature." He says the scramble technology is useful because it speeds up something far more effective than anything humans can do yet: evolution.
"You make a whole bunch of variants and iteratively select out the ones that work best -- you basically evolve them," says Enyeart. "If you just try to intelligently design it, if it works at all it probably won't work very well. Evolution remains our best option for designing code which makes sense since that's how nature does it."
So as advanced as this work is, Boeke says it's still early.
"It's really only a stepping stone to where we really want to get, which is to have someone to sit down at a computer terminal and say 'OK, I'd like to design a plant to grow into a very specific shape so that when it dries up it will be a container that we can use, it will be biodegradable, it will have the following characteristics." Boeke says.
Biologists have no idea how to do that right now. It would require processing power far in excess of what currently exists -- even today's supercomputers have difficulty predicting how a protein folds, let alone how 12 million base pairs interact to produce a living creature. But designing more productive yeast could hold far more promise than better bread.
[Music: "Scramble" by California Guitar Trio from Pathways]