On the Track of DNA Methylation: an Interview with Adrian Bird
Total Page:16
File Type:pdf, Size:1020Kb
Interview On the Track of DNA Methylation: An Interview with Adrian Bird Jane Gitschier* Departments of Medicine and Pediatrics, Institute for Human Genetics, University of California San Francisco, San Francisco, California, United States of America The day school let out for the summer, my daughter and I packed our bags for Britain, where we had lived for a few months in 2006. Annie was eager to reconnect with her friends there, and I had arranged to conduct three interviews. In desperation and with the clock ticking, I struggled to fit my bulky recorder into my wheelie when it dawned on me that the ‘‘talk app’’ on my daughter’s iphone should be up to the job. You can imagine the reluctance and skepticism on the part of my 15-year-old, but she managed to get into the spirit and acquiesced. First up on my schedule was Adrian Bird (Image 1), who holds the Buchanan Chair of Genetics at the University of Edinburgh and is also Director of the Wellcome Trust Centre for Cell Biology. Long before the word ‘‘epigenome’’ was coined, Bird began mapping the distribu- tion of DNA methylation (occurring at the cytosine of CpG dinucleotides) in the genomes of a variety of species. His work emerged just as agarose gels, restriction Image 1. Adrian Bird. enzymes, and Southern blots were being doi:10.1371/journal.pgen.1000667.g001 developed. Bird later spawned the idea of sun of Cambridge to find cold rain Yale] on gene amplification in frog CpG islands, pockets of DNA rich in penetrating the skylights at Edinburgh’s oocytes. When I went to Max’s in Zurich, unmethylated CpGs and frequently found Waverley Station. It felt very cozy to share he had a visitor named Ham Smith. in conjunction with the promoter regions the evening with them and their children, Gitschier: What year approximately of mammalian genes. Bird’s observation Tom and Annie: chatting, watching some was this? provided a roadmap for disease gene Twenty20 (an abbreviated form of cricket), Bird: 1973–1974. Ham was on sabbat- discovery for about 15 years, until human playing Uno, feeding the three guinea pigs, ical, and the first thing he did was to make genome draft sequences began to emerge. and experimenting with the iphone’s tape a restriction enzyme, HpaII—Haemophilus Bird’s laboratory then went on to app, which, to our delight, worked. parainfluenzae II. identify proteins that bound to methylated Gitschier: My first question is a two- I was making ribosomal RNA genes, DNA, one of which (MeCP2) was discov- part, integrated one. How did you get just for something to do really. We knew ered years later to be defective in Rett interested in methylation, and what was there was a difference between the ampli- Syndrome, a rare X-linked disorder in the state of the art at the time you started fied ribosomal RNA genes, which were which affected girls develop autism and a working on it? extrachromosomal in the oocyte, and the distinctive set of behaviors. This astonish- Bird: I first got interested when I was chromosomal ones, and that the difference ing turn of events propelled Bird to extend in Zurich doing a post-doc. was due to methylation. Don Brown and his studies on MeCP2 to a murine model Gitschier: Whom were you with there? Igor Dawid had shown that chromosomal for Rett Syndrome, ushering in new ideas Bird: Max Birnstiel. I had been in the rDNA had 5-methylcytosine and the about therapy for this devastating illness, States doing a post-doc [with Joe Gall at amplified didn’t. but still leaving open the question of MeCP2’s role in the brain. Citation: Gitschier J (2009) On the Track of DNA Methylation: An Interview with Adrian Bird. PLoS Genet 5(10): Bird and his wife Cathy Abbott, also a e1000667. doi:10.1371/journal.pgen.1000667 geneticist, invited me to spend the night Published October 16, 2009 prior to the interview with them (future Copyright: ß 2009 Jane Gitschier. This is an open-access article distributed under the terms of the Creative interviewees, take note!), and I was delight- Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, ed to do so. Still jet-lagged, I traveled by provided the original author and source are credited. train, leaving behind the uncharacteristic * E-mail: [email protected] PLoS Genetics | www.plosgenetics.org 1 October 2009 | Volume 5 | Issue 10 | e1000667 So I used HpaII to cut this amplified Bird: I’m not sure what I’m doing, to Bird: Yes, that was cited loads of times, [extrachromosomal] DNA, and it cut be honest! I knew I wanted to do because it was the mapping phase of the beautifully. But when I tried with purified something interesting, but I was just genome project where people wanted to chromosomal ribosomal DNA, it didn’t playing around more than anything else. map themselves into reality. cut. It was known that restriction enzymes Ham was playing around with the en- Gitschier: Well, people used them to were blocked by methylation in the zymes he knew about and Ed was doing find genes. They were like little flags that organisms from which they are derived; the technological things that he really liked said, ‘‘Hey over here, I’m a gene!’’ there is usually a restriction enzyme and a doing. So accidentally this gave rise to the Bird: Exactly. modification enzyme that matches it and idea that one could use restriction enzymes Gitschier: Did you coin the name that protects the genomic DNA of the host to map methylation in DNA. There was a ‘‘CpG islands’’? from its own destructive enzymes. So it conjunction of areas that were needed Bird: No I didn’t. We called them seemed that some of the methylation [in before one could exploit this properly, and ‘‘HTF islands’’ for ‘‘HpaII tiny frag- the chromosomal rDNA] might be mim- that didn’t happen until I got back to ments’’, but reviewers said, ‘‘What the icking the blockage that occurs in the Edinburgh. hell is HTF?’’ It was Marianne Frommer, Haemophilus parainfluenzae endogenous By 1975, I was back there in an MRC who was doing a sabbatical in my lab at enzyme. unit and mapping the methylated and the time the Cell paper was published, who Gitschier: Did Ham Smith know that nonmethylated sites in the ribosomal genes called them CpG islands—it made more HpaII didn’t cut methylated DNA before in Xenopus laevis. And it took absolutely sense. you did the experiment? ages to get that published. Gitschier: OK, let’s switch gears. At Bird: Probably, but he just made Gitschier: At some point you moved some point, you started to work on restriction enzymes in order to make away from frog oocytes. proteins that bound to methylated regions. himself at home. Restriction enzyme Bird: We looked at sea urchins— Bird: That arose by chance as well. technology was new, and agarose gels invertebrates—and found there was both The CpG islands provide an approximate had just been brought in. And another guy a methylated and an unmethylated frac- way to map methylation throughout the there was Ed Southern, who had just tion of the genome. The last things we genome, but at the time there was no way invented Southern blotting, so there was a came to were vertebrate cells and they to do that properly. So we kind of ran into bit of coincidence here. didn’t seem to have anything like what you a brick wall—what you really want to Gitschier: Was Ed on sabbatical? see in the invertebrates. When you digest- know is where the methyl groups are Bird: Yes, he was as well. I can’t ed the DNA with these methylation- throughout the genome. remember whether they were there at sensitive enzymes, nothing happened be- Gitschier: And you found out where exactly the same time. I think they cause most of the DNA is methylated. they weren’t. overlapped. Then we had the idea that maybe we Bird: So I decided to work on how the Gitschier: So, why is everybody com- could see a small fraction of unmethylated CpG islands might originate. We asked, ing to Max Birnstiel’s lab? DNA if we end-labeled it. That was the ‘‘Does something bind to the nonmethy- Bird: Max Birnstiel and Don Brown work of David Cooper who did a Ph.D. in lated sequence that might protect it from were hotly competing groups because both my lab. I can remember him doing the methylation?’’ Just by steric inhibition. We of them worked on ribosomal RNA genes, first end-labeling, because it made a made an oligonucleotide, and at the time it which you could purify by buoyant density horrible blob. Something that could be took Amersham about 4 months to do it. I centrifugation. Before you could clone artifactual, but it wasn’t, and we spent just made up a sequence full of CpGs that DNA, the only way to get hold of pure quite a lot of time showing that. I cloned were in restriction enzyme sites so we gene was A) because it was highly repeated mouse fragments derived from that blob, could test their methylation status, and and B) because its buoyant density was and that was our 1985 paper in Cell. And then we oligomerized them and methylat- different from the bulk [of the DNA]. So if then we restriction-mapped them and ed the sites using commercial enzymes, you ran enough cesium chloride gradients showed that they came from clusters of because you could buy HhaI and HpaII you could get it pure and then you could nonmethylated CpGs in the genome.