Hacking the Genome

In pondering genome structure and function, evolutionary geneticist Laurence Hurst has arrived at some unanticipated conclusions about how natural selection has molded our DNA.

Jun 1, 2012
Karen Hopkin

Laurence D. Hurst, Laboratory of Evolutionary Genetics and Genomics, Department of Biology and Biochemistry, University of Bath NICK MORRISH PHOTOGRAPHY

These days Laurence Hurst pores over data sets. As a kid, he pored over bones. “You find plenty of dead sheep on the moors down in Cornwall,” he says. “I’d go out collecting these stinky sheep bones and bury them in the garden. I was utterly fascinated by the idea that every bone had a name.”

He was also mesmerized by microbes. “I remember the shock I felt the first time I looked in a microscope and saw a Euglena swimming away. There was this whole other world sitting there. That was pretty mind-blowing. To this day I have a love of protists, in no small part because I’ve always gravitated toward simple problems to answer big questions.”

The biggest question of all: Is the genome like an exquisitely engineered Swiss watch, in which carefully crafted parts fit together perfectly and every feature is optimized to function flawlessly? Or, as Hurst puts it, “is it just some cheap Mickey Mouse watch that’ll tell you the time, but its components are poor-quality and it includes lots of crap that’s frankly unnecessary?”

Hurst would like to know. “Looking at the genome, the question comes up again and again. And the reason I get very excited about it is, we really don’t know the answer. But now, for the first time, we’re drowning in the data we can use to address the issue. We just have to be clever about it.”

So far, Hurst—who in 1997 became a full professor at the University of Bath at the age of 31—has made clever use of the data to address why there are two sexes, why the genetic code was no accident, why gene order matters, and why RNAs with no function are functionally important. Here he contemplates new ways to teach evolution, the joys of throwing a javelin, and where to put the coffeepot.

HURST HUSTLES

The other Cambridge. After completing his undergraduate studies at the University of Cambridge in 1987, Hurst landed a fellowship that allowed him to spend a year at Harvard. “I loved the rush of new ideas, new ways to look at evolution. You go to a given university and they teach you their view of the world. This was first time I’d ever got to see there are other places that have completely other views—not just different, but really disagreeing with what I’d been taught as an undergraduate. I found that really exciting.”

“It was originally asserted that gene order in mammals is random. This was before anybody even had a complete eukaryotic genome, which was theoretical hubris taken to the nth degree.

Battle of the sexes. As a graduate student working with Bill Hamilton and Alan Grafen at the University of Oxford, Hurst tackled the age-old question: Why are there two sexes? “Imagine you’re a protist swimming in a pond. Since you can’t mate with anybody who’s the same as you, the best solution would be to have as many mating types as possible—to boost the odds of success. Two is the worst solution because you’ll only succeed half the time. Yet two is what protists have got.” Why? “The answer we came up with has to do with coordinating the uniparental inheritance of organelles.” In many organisms, for example, mitochondria are passed down from the mother. Having more than two sexes could introduce confusion about whose organelles are going to make it to the next generation, a situation that could be deleterious. The study—published in the Proceedings of the Royal Society in 1992—got written up in Science. “Only later did I realize that having a two-page article about you in Science as a graduate student is breathtakingly unusual.”

One in a million. When scientists first started sequencing collections of human proteins, they noticed that in situations where an amino acid had been changed, it was often replaced by one that was chemically similar. “There could be two explanations,” says Hurst. “One is that if substitutions are not chemically similar, then selection simply gets rid of them and you don’t see them in the population. The other is that it’s a property of the genetic code—that most mutations will produce amino acids that are chemically similar.” To find out, Hurst fired up a computer and scrambled the genetic code—keeping start and stop codons the same, but jumbling the rest of the codons so, for example, UUU might no longer code for phenylalanine. For each of these “alternative codes,” he assessed the effect of mistranslation—whether reading one of the three base pairs in a codon incorrectly would dramatically change the chemical character of the amino acid it encodes. “We simulated a million different genetic codes, and it turned out not one was as good as the real genetic code in terms of its ability to minimize the effect of translation errors.” The resulting paper, “The genetic code is one in a million,” appeared in the Journal of Molecular Evolution in 1998.

You say potato. Until recently, evolutionary biologists asserted that synonymous mutations—changes in nucleotide sequence that do not change the amino acid sequence of an encoded protein—were selectively neutral. In other words, they don’t benefit an organism, but they aren’t particularly harmful, either. Hurst found otherwise. “In mammals, the ends of our exons are defined by splicing enhancer motifs. These motifs are 6 to 8 nucleotides long and the more of these motifs you have in an exon, the more accurate splicing is. Now, if we suppose that splicing is done for a good reason, and if you don’t do it properly it’s going to be bad for you, it’s obvious that a mutation that changes a splicing enhancer motif to something that is not a splicing enhancer motif is going to be deleterious”—even if that mutation is ‘synonymous’ in the sense that if a protein were produced from that exon, it would still have the correct amino acid sequence. “And that’s what we found. Working with sequence data, we estimated that about 9 or 10 percent of mutations in mammals could disrupt splicing enhancer motifs, and be deleterious, while at the same time being synonymous”—which means they could be acted on by selection. “The paper was initially rejected because the editor said, ‘It is well known that, in mammals, synonymous mutations are not under selection.’ To this day I think I still have that e-mail.”

Keeping up with the Joneses. “It was originally asserted that gene order in mammals is random. This was before anybody even had a complete eukaryotic genome, which was theoretical hubris taken to the nth degree.” But Hurst suspected that gene order might matter. He had noticed that in the yeast genome, many pairs of neighboring genes were transcribed from a single, bidirectional promoter. When one was turned on, so was the other. And he wondered: Could their positioning be a happy accident? To find out, Hurst and colleagues compared the genomes of Saccharomyces cerevisiae and Candida albicans. They asked: If two genes are coexpressed in S. cerevisiae, are they more likely to be neighbors? And will neighbors in S. cerevisiae have counterparts that also cozy up in C. albicans? “What we got was an unambiguous, lovely, delicious signal showing that genes that sit next door to each other and are highly coexpressed in yeast tend to stay next door to each other and stay highly coexpressed” in C. albicans—suggesting that their proximity is purposeful.

“I keep the coffee machine in with my graduate students. Sometimes you get students who are a little reluctant to chat. Grabbing a coffee is a good excuse to go talk with them about this, that, and the other.”

CUTting-edge research. “Recently we got the idea that bidirectional promoters are actually little devices that ensure consistently open chromatin. And we found a really cute explanation for how this works.” That cute explanation involves cryptic unstable transcripts, or CUTs. “They’re small RNAs, found in yeast, which are made at high levels and destroyed the second that they’re made. It’s a classic example of: Is this a Mickey Mouse watch and you’re just generating rubbish? Or is it a Swiss watch, the mechanism of which we don’t yet understand? In this case, I’m fairly certain we’re looking at a Swiss watch and I think we now understand what’s going on. Imagine you’ve got an essential gene that really has to be upregulated when it’s needed. When a transcription factor comes along, it would be a bit of a disaster if that promoter is stuck away in closed chromatin.” To solve that problem, Hurst discovered, yeast tend to put essential genes next to genes that encode CUTs. “By continually making these CUTs, you keep that chromatin in an open configuration, which promotes expression of the essential gene next door”—work published in Genome Biology and Evolution in 2011.

First things first. Hurst recently launched an experimental project to determine the best way to teach evolution to teens. “The hypothesis I’m working on came from something I read in Science, which said that if you look across cultures, it turns out there’s a good correlation between people’s understanding of evolution and their understanding of genetics. It could be that if you understand genetics—you know about alleles and you understand that alleles can change in frequency—then evolution is just the next logical next step. So the hypothesis we’re going to test is: if we teach genetics first, then kids will get evolution better than if we teach evolution first and genetics second. If that works, we should be able to put together some materials we can take to schools to help the teachers teach.”

HURST HOLDS FORTH

Eureka! “Darwin spends the first chapter of The Origin of Species discussing pigeons. He knew he was about to overturn a world view, yet he doesn’t start off with a great big theological debate. He starts off talking about pigeons. Fantastic! The first few chapters of The Origin of Species are actually the most impressive bit of writing.  The way Darwin establishes the argument in those first four chapters is just wonderful: there’s heritability, there’s variation, there’s a process of selection, there’s mortality. So by the end of Chapter 4, you end up going, ‘Ooh, I’ve got a really good idea! I know how evolution works!’ Of course it’s Darwin’s idea, but he’s put it into your brain in a most fantastic fashion.”

History in the making. “There’s an energy you get around the Harvard campus which I never see in Oxford and Cambridge or the British university system as a whole. To some extent we may be inhibited in Britain because our traditions are just so old and so deep, it’s almost as if change is unimaginable. Everything is so rooted in history, you can’t see how the world could be any different. Whereas at Harvard, I got the distinct impression that they wanted to be making history, rather than sitting on top of history.”

Cheap, fast, and fun. “I think with the $1,000 genome we can now start to do playful science. Say you want to know which cell type gives rise to which other cell type. There’s two ways to do it. With the classic developmental approach, you have to track the damn things as they divide. Now, you just sequence them all up in a given individual, and because in every cell division you get mutations, you can construct not a phylogenetic tree but an ontogenetic tree. That ontogenetic tree will tell you the series of cell divisions that gave rise to every cell in the body. Wouldn’t that be fantastic?”

Insider info. Hurst revels in grant review panels. “You can learn a lot from listening to your colleagues talk about grants. You’ve got the world’s experts telling you what the problems might be with a particular analysis or why a certain data set is not going to do what you think it will do. How valuable is that? I sit on a lot of editorial boards, as well. It’s a very good way to keep up with the zeitgeist.”

Coffee talk. “I keep the coffee machine in with my graduate students. Sometimes you get students who are a little reluctant to chat. Grabbing a coffee is a good excuse to go talk with them about this, that, and the other, and see how things are going. That would be my advice to new supervisors: put your coffeepot in with your graduate students.”

HURST HANGS LOOSE

Hurst the hurler. “At age 13, I was given a javelin—as was the rest of my class. They told me to throw it, and discovered that I could throw it twice as far as anybody else. It was very good when I got bored [with studying] for my exams as a schoolkid. We lived in a house that had an open field next to it, and I would take out the javelin to blow off some steam.”

Lip service. Hurst and his daughter play trumpet in her primary-school orchestra. “I dropped trumpet at the age of 14 because I never really had the lips for it. But when I heard that the orchestra encourages parents to come and play, I bought a trumpet and discovered that I could remember the fingering. At the moment we’re doing the James Bond theme. So most evenings you’ll find me just tootling away.”

Out of (i)touch. “The reason I don’t have an iPhone is that I know I’d be eternally checking my e-mail if I did. I love being on vacation with my family, completely cut off from the rest of the world. During those 2 weeks, you can e-mail me as much as you like—I will be utterly oblivious to it. There’s more to life than science.”