Spring 2015 /Daring Minds II/
 

A CRISPR Cut

Jennifer Doudna ’85 didn't set out to revolutionize Genetic Engineering, but that may be exactly what she did.

PC_Doudna

THE ACTION AT scientific conferences mostly happens in conference rooms and hotel bars, but sometimes the players break out to see the sights. That’s what Jennifer Doudna was doing at a conference put on by the American Society of Microbiology in spring 2011. She attended a session on a type of bacterial genetic sequence called a clustered, regularly interspaced, short palindromic repeat—CRISPR, for short. They seemed to have something to do with an immune system for bacteria, though to be honest, Doudna, a biochemist at UC Berkeley who specializes in the three-dimensional structure of genetic material, thought they were “a boutique area of science” at best.

That night, Doudna ran into someone else who was working on the same problem. In Germany, Emmanuelle Charpentier’s lab was studying what most people call “flesh-eating bacteria,” a species of Streptococcus, and Charpentier’s team had found something important in one of their pet bug’s CRISPRs—it made a protein called Cas9. But she needed help to understand all of the moving parts. So Charpentier asked Doudna if she wanted to team up. Doudna said yes. Typical conference stuff.

Over the next few months, Doudna’s postdocs in California worked with Charpentier’s teams in France and Germany. But what they started to figure out began, slowly, to look a lot bigger than just an immune system for flesh-eating bacteria. CRISPR/Cas9 made a complex structure of protein and genetic material that looked like it could cut DNA—which is to say, genes—but it was precisely targeted, almost as simple as putting a cursor between two letters on a computer screen and clicking delete. “There were other techniques in the literature, but they were difficult,” Doudna says. “This is the kind of technique that, in principle, anybody who knows anything about molecular biology will be able to do.”

Doudna and Charpentier wrote a paper for Science, one of the world’s premiere scientific journals. When it came out, in summer 2012, the scientific community went nuts. By the end of 2013, hundreds of papers from labs all over the world had confirmed that, yes, not only was CRISPR a quick-and-easy way to edit a genome as easily as Word edits a magazine article, but it worked in just about every living thing—yeast, zebrafish, mice, stem cells, in-vitro tissue cultures, and even cells from human beings. Most gene-editing techniques work in theory, but in practice require wrestling to the ground complicated, ornery techniques that often fail. But with CRISPR: You want that gene over there? You got it. Companies formed seemingly overnight to turn CRISPR into medicines, research tools, and maybe even profits. Doudna’s lab was at the center of a shift that could be every bit as significant as being able to sequence the genome.

CRISPR1a

THE KIND OF CELLS that you and I have encode information in the form of deoxyribonucleic acid—DNA, a long backbone of two spiraling strands bridged by “base pairs,” the famous A, T, C, and G that comprise the genetic sequence. But DNA isn’t the only genetic material. When it’s time to make proteins, cells unspool lengths of DNA from their tightly-packed chromosomes and make a cheap copy called ribonucleic acid, or RNA. It’s this RNA that other machinery in the cell reads—the sequence of A, C, G, and U (replacing the T) represent amino acids, and amino acids put together are the proteins of which we are mostly made. It’s a cool system.

RNA, though, is kind of weird. Because in addition to containing information, it can also form structures that do jobs. In fact, the biological machine that reads RNA and outputs proteins, called a ribosome, is itself made of RNA. In this particular corner of molecular biology, the map is also the terrain. This dual personality is behind the “RNA world” theory, the idea that RNA’s ability to both carry and initiate the code of life means it gave rise to all life on earth.

It’s also what compelled Jennifer Doudna, freshly graduated from Pomona, to get her PhD. Growing up in Oahu she knew she wanted to be a biochemist; a set of seminars she attended in high school had sealed that deal. She studied biology as an undergraduate, worked in real labs, and pointed herself at grad school in Boston as soon as it was time to apply. She ended up getting in the laboratory of Jack Szostak, who came up with the RNA world idea.

Doudna decided that she wanted to understand those RNA structures—figuring out the structure of ribosomes and other so-called catalytic RNA. Basically that meant trying to get them to crystalize and then x-ray them. It was, Doudna says, a methodological challenge that was “every bit as cool as I could have imagined.” Eventually she ended up a professor at Yale, and she solved a few of those structures. Doudna was earning a reputation as an ace in a field without many practitioners. “She’s careful and diligent in pursuing all the leads without cutting corners,” says George Church, a Harvard Medical School geneticist who remembers Doudna’s student days there. “And she has a good knack for picking the right topics.”

The West Coast lured her back; in 2010, Doudna took a job at UC Berkeley. “I had always considered myself a basic scientist,” she says. “But you want to feel like your work is going to help solve human problems at some level.” She started working on diseases caused by RNA mutations, and on a technique called RNA interference, or RNAi. Basically it uses small molecules to interrupt the translation of RNA into proteins, to try to fix problems before they start. And to make it work, you have to understand the structural characteristics of RNA.

At about the same time, some food researchers in Copenhagen were learning something new about yogurt. Turning milk into yogurt requires specialized bacteria, but every so often those bacteria get sick—just like people, they get attacked by viruses trying to hijack their cellular machinery. The Danish team found that bacteria exposed in advance to the viruses, called bacteriophages, became immune. It was like vaccination, but for microbes.

Durell_140916_PC_doudna_0155

HOW’D IT WORK? In the late 1980s scientists found long, repeating sequences in bacterial DNA that were the same back-to-front—palindromes, in other words. And between the palindromes: nonsense. At least, that’s what they thought at the time. But the genetic gibberish turned out to be quoted from bacteriophage DNA. Put all that together and you got RNA structures that could target specific DNA sequences in a virus, and a protein that would chop that DNA up, destroying it. It was, in other words, an immune system. The Danish yogurt makers had hit upon a rudimentary way of programming it to hit specific viral targets.

That’s why Charpentier started studying it. “I’m interested in how bacteria cause diseases, and how they can become resistant to antibiotics,” she says. “Initially the goal was to look for this class of small RNAs to find one with a nice regulatory function. Coming to CRISPR was in a way a bit by chance.” It wasn’t crazy to imagine that she’d find a useful enzyme in her work—most of the DNA- and RNA-cutting enzymes used in labs were isolated from organisms found in nature.

The thing was, though, that even the most advanced techniques for cutting-and-pasting DNA and RNA were really tough to use. The two best approaches, “zinc-finger nucleases” and “TALEN,” required the creation of a new, bespoke protein every time, coded to the specific sequence a researcher wanted to cut. “Zinc finger nucleases were originally priced at about $25,000 each. You could do it yourself for a little less, but it extracted a corresponding amount of flesh,” says Church. “TALENs looked easier, but they were particularly hard to engineer biologically. It lasted for about a year, a year and a half, as a fad.”

The RNAi Doudna was working with turned out to have similar problems. Usually you try to engineer a bacterial or insect cell to make protein or RNA. However, you then have to purify the protein or RNA that you want out of all the gunk you don’t. Those methods took a whole skill set, and not everyone had it.

In the late 2000s, though, postdocs in Doudna’s lab were starting to get really good results in experimenting with CRISPR. It was easier to do and didn’t need custom-made proteins. Doudna’s and Charpentier’s two labs together realized that in the case of CRISPR/Cas9, the same protein was doing the cutting every time. The only thing that changed were two adjacent structures made of RNA—and you could engineer their function into one, a short, synthetic stretch called a guide RNA that was really easy to make. “We figured out how to program Cas9 to cut any sequence in DNA just by changing the guide RNA,” says Doudna. When she and one of her students realized what they had, sitting in her office at Cal, “we looked at each other and said, this could be an incredible tool for genome engineering.”

On the other side of the world, Charpentier was just as stunned. “All the other tools, each time you want to target DNA at a specific site you have to engineer a new protein. This requires time, and it’s not easy for someone who isn’t used to it,” Charpentier says. “With Cas9, anyone can use the tool. It’s cheap, it’s fast, it’s efficient, and it works in any size organism. It’s revolutionizing biology.”

So what’s CRISPR actually going to be for? That’s still being determined, at labs all around the world—and a handful of companies that spun up in the dizzy aftermath of the Doudna-Charpentier paper. “There’s a distinction between CRISPR-Cas9 as a therapeutic tool, using it to correct mutations in cells that would then be reimplanted in a patient—or delivering Cas9 directly,” says Charpentier. “But the other possibility is more indirect. That’s using it as a tool in labs for development, to help screen drugs or to understand a disease by using it to create models of the disease in animals.” In other words, you could use the technique as a medicine, to correct a mutation directly, or use it to induce a mutation you wanted to study in an animal to test other possible drugs.

Charpentier herself is one of the founders of a company, Crispr Therapeutics, based in Basel, that’s planning to focus on making treatments for people. The company’s CEO, Rodger Novak, is a longtime drug development exec, and the company is well-funded, but even Novak acknowledges what they’re doing won’t be easy. “What biotech pharma always struggles with is the biology of the target. In many instances we don’t know until late-stage development, pivotal human trials, if the target we’re using is the target we need.” Novak says. “The other challenge is delivery. If you go after the liver or the lungs or the brain, very different requirements apply.” He says he’s cautiously confident.

Meanwhile on the other side of the Atlantic, Doudna had teamed up with a few other CRISPR pioneers, as well as George Church, to start her own company: Editas, based in Cambridge. Doudna remains a “co-founder,” but is no longer associated with the company. Church and Doudna got $43 million from a handful of well-known venture investors to spin the company up, aiming, he says, at “a large number of genetic diseases, both common and rare, especially those that might require the removal or editing of DNA, rather than just the addition.” Other therapies in trials are better than CRISPR-Cas9 at adding DNA, inserting a gene, says Church. CRISPR-Cas9 is much better at cutting—harkening back to its original function in the flesh-eating Streptococcus Charpentier studied.

Back in Doudna’s lab at Berkeley, though, her team is still trying to answer some fundamental questions. No one doubts that CRISPR works, but some researchers still worry about whether they can target it narrowly enough to work as a therapeutic. But Doudna would like to know how its targeting system works at all—with just 20 bases of RNA it can somehow home in on any sequence of DNA. No one knows how. “We’d love to figure that out,” Doudna says. And no one knows how it acquires the “spacer” sequences, the genetic information between the palindromes. That would be the secret to how CRISPR works as an immune system in bacteria, and for now it’s a mystery.

As the excitement around CRISPR has continued to grow, no one seems more surprised than Doudna herself. “I was working away in my lab on a bacterial immune system. Genome engineering wasn’t on my radar,” she says. “If you had asked me in 2007 if the CRISPR system was going to be useful, I don’t know what I would have answered.” Today, though, Doudna is a little more sure: If it continues to work as we expect, it’s going to change everything.

 

EDITOR’S NOTE: In November, Doudna and Charpentier were awarded the 2015 Breakthrough Prize for Life Sciences, each receiving a stipend of $3 million. Their discovery was also featured as one of 10 “World Changing Ideas” on the cover of Scientific American.