Think Big, Dress Casual

Mike Snyder

Courtesy of Michael Marsland"We couldn't get that project funded for the life of us," says Yale University's Mike Snyder of the experiment that, in his opinion, launched the functional genomics era. It was the late 1980s, years before the dawn of DNA microarrays, and Snyder and his colleagues were proposing to use epitope-bearing transposons to tag every protein in yeast. With this collection, the scientists planned to track the positions of all 6,000 yeast proteins, information that would give them a handle on the proteins' potential functions."But funders just weren't interested," recalls Snyder. The emphasis at the time was on mapping and sequencing, not messing with function. "We kept resubmitting the same grant." The third time, it received a score that Snyder says couldn't be ignored. The researchers launched the project and haven't had to look back since.That Snyder prevailed, and that the transposon-tagging experiment took off, makes him a poster boy for the notion that "many genome-wide, proteome-wide, or other large-scale efforts succeed because of a singular vision," says Stan Fields, a Howard Hughes Medical Institute investigator at University of Washington, Seattle. "Mike's an imaginative guy who's not afraid to tackle highly ambitious projects."Snyder's ambition did not stop at the "localizome." Since nailing the subcellular distribution of more than 4,000 yeast proteins, Snyder and his colleagues have initiated projects to find all the coding sequences in yeast and humans and to map all the regulatory sequences in yeast and human DNA. The goal, says Snyder, is to "understand the function of every base and every protein in humans and yeast."That seems like a tall order. Fortunately, although the objectives are immense, the means are within reach, and they don't require an army. The transposon tagging, for example, was done largely by "one student working very industriously," says Snyder. The chromatin immunoprecipitation technique (ChIP/chip), which so many investigators use to identify and catalog regulatory sequences in DNA, was developed by "one talented student" in Snyder's lab working with "one talented postdoc" in Pat Brown's lab at Stanford University. The invention of the latter method was sparked by the observation that so many of Snyder's tagged proteins were found nuzzling up to chromosomes in the cell nucleus. "We wanted to know what they were binding to," he says.As a biologist, Snyder has spent a good portion of his career exploring the cell nucleus. In graduate school, he worked with Norman Davidson at the California Institute of Technology on gene organization and expression in Drosophila. As a postdoc, Snyder studied chromosome segregation with Ron Davis at Stanford, where he started to toy with transposon mutagenesis in yeast.Snyder still maintains an interest in problems that might be considered conventional biology. About half of his 30-person lab uses standard genetic, biochemical, and cell biological approaches to investigate cell division, cell-cycle control, and assembly of the microtubule networks that distribute chromosomes to daughter cells. The other half runs large-scale screens and cranks out massive data sets that require subsequent analysis and deconstruction.That balance works well, says Heng Zhu, a former postdoc who spearheaded Snyder's efforts to build a proteome chip, a microarray sporting nearly all the proteins in yeast. "Through these screens and assays, we generate a lot of data but we don't know what to make of them until the traditional biologists help us to look in depth," says Zhu. Those interactions, he says, always stimulate great discussion.The large-scale, discovery-based techniques are not meant to supplant hypothesis-driven research but to augment it, says Snyder: "The discoveries lead to new hypotheses." But the beauty of the unbiased approaches lies in the surprises. "In every project, we've found things we never would have predicted in a million years," he says.For example, in the course of the transposon tagging, Snyder identified a novel mitochondrial protein whose gene falls within an open-reading frame that encodes a ribosomal RNA – but on the opposite, noncoding strand of DNA. "What the heck is it doing there?" Snyder wonders. His lab is working to find out. More importantly, the results demonstrate that the genome might harbor quite a few surprises: genes buried inside introns or antisense DNA. Such genes, says Snyder, "are hard to find unless you use these random methods to look for them."Another unexpected finding is the discovery of a metabolic enzyme that appears to act as a transcriptional regulator. This unusual protein surfaced when Snyder washed labeled DNA over his proteome chip. The assay revealed many proteins (e.g., histones and transcription factors) known to bind to DNA, plus an enzyme involved in arginine biosynthesis. The results suggest that metabolic enzymes can directly regulate gene expression. Snyder wonders how many proteins might be "multifunctional in ways that we'd never anticipated."Although interesting observations result from these unbiased approaches, Snyder notes that "validating the data and turning them into an interesting biological story" can take time. For example, the team discovered the DNA-binding properties of Arg5,6 in 2000; the paper describing the result was published in Science just last month (Oct. 15).Getting the assays up and running also takes time, and talent. When Snyder first considered working on a proteome chip, conventional wisdom dictated that proteins couldn't be spotted on chips without becoming denatured and inactivated. "But I never liked that logic," says Snyder. And so he turned to Zhu, who joined the lab after hearing Snyder speak about the protein localization project at a Gordon Conference. "I realized immediately," Zhu says of the talk, that such large-scale, unbiased approaches were "the future of biology."So Zhu was primed to take on the challenge of the proteome chip. He worked out the proper conditions and then led the effort to produce the 6,000 different proteins the researchers hoped to display on the array.And when the chips were done, everyone wanted one. "When we published the first paper, people started writing, asking for arrays or suggesting collaborations," says Snyder. "We couldn't keep up with the demand." Snyder employs one person full-time to fulfill requests for cells and reagents, a job that he considers "one of the most important parts of these 'omics projects." He says that it makes sense to share. "The power of 10,000 labs working with a reagent is better than one." That generosity, says Fields, is part of what makes Snyder a "great citizen in the yeast community."To displace the demand for proteome chips, Snyder and his associates founded a company called Protometrix. In addition to producing and disbursing the yeast proteome chips, the company is developing a human proteome chip. Pharmaceutical companies could use such a chip to find the proteins to which their lead compounds bind, a method that Snyder says could dramatically accelerate the drug-discovery process.Although Snyder is no longer intimately involved with Protometrix, which was purchased by Invitrogen last spring, he, too, has turned his attention to human genomics. His lab raced Affymetrix to develop a microarray that contains all the non-repetitive pieces of human chromosome 22. The Affymetrix paper was published first, but the similarity of their results to Snyder's reassures him that he is doing something right. His lab uses the array to map transcription factor binding sites and to try to identify all the transcribed regions of the chromosome. "Half the stuff we found was not annotated," says Snyder, "which suggests that there's a lot more out there than people appreciate."The studies on chromosome 22 have made a big impact, says Steve Elledge, a Harvard University biologist who worked with Snyder when they were both post-docs in Davis's lab. And the leap from yeast to humans demonstrates Snyder's pioneer spirit. "Mike is kind of fearless in that way; he's willing to try new things," says Elledge, who reports that Snyder once dressed as a sandwich for Halloween – a costume that Elledge would subsequently sack out on after late nights in the lab or a little too much wine.If Snyder has been generous with his costume, as well as his wine and other reagents, he is equally giving of his time, support, and wisdom, says Zhu, who now heads his own lab at Johns Hopkins. "He always told me to think big, [and he] emphasized how important it is to stay ahead of the curve." He even coached Zhu on what to wear for his job interviews: "Don't dress too formally. Wear a sports jacket and nice pants, not a suit. And not a white shirt. Too formal."These days Snyder, too, is going casual. After stepping down as chair of his department, the pioneering researcher is taking a year-long sabbatical, in his own lab. He still heads Yale's Center for Genomics and Proteomics, but now gets to spend some time at the bench. "I just ran an experiment for the first time in 10 years," says Snyder. He was isolating polysomal RNA with a rotation student, trying to turn up some new transcribed regions, and "enjoying immensely" being back in the thick of things.Over the next year, Snyder hopes to master some new techniques, an accomplishment that will no doubt help him keep ahead of the curve. The only question remaining is: What will he wear?Karen Hopkin khopkin@the-scientist.com

Think Big, Dress Casual

Interested in reading more?

Become a Member of