Biography of Erin K. O'Shea

Having little background in biology has not hindered Erin K. O'Shea, professor of biochemistry and biophysics at the University of California at San Francisco (UCSF) and assistant investigator of the Howard Hughes Medical Institute. Trained primarily as a biophysical chemist, O'Shea has made her mark in several disciplines. As a graduate student, she made a significant contribution in her delineation of the physical structure of the leucine zipper motif (1, 2). Her later foray into yeast biology has yielded dozens of publications on signal transduction and proteomics and has established her as a leader in the field of cell and molecular biology.

For her many contributions to the fields of biochemistry, biophysics, and cell and molecular biology, O'Shea received the National Academy of Sciences (NAS) Award in Molecular Biology in 2001. Despite earning numerous awards throughout her career, O'Shea notes that the NAS award was the “biggest deal” for her, because her graduate advisor, Peter Kim [Massachusetts Institute of Technology (MIT), Cambridge, MA], and her postdoctoral advisor, Robert Tjian (University of California, Berkeley), had received the award in previous years. Also, by O'Shea's estimates, approximately 25% of winners of this award go on to win the Nobel Prize.

O'Shea was elected to the NAS in 2004 for her contributions to the understanding of signal transduction, regulation of protein movement into and out of the nucleus, and how phosphorylation controls protein activity.

The Coiled-Coil Leucine Zipper

After graduating with a bachelor's degree in biochemistry from Smith College (Northampton, MA) in 1988, O'Shea entered Kim's laboratory at MIT. A young scientist on the academic fast track, O'Shea finished graduate school in a mere two and a half years. After reading a paper in Science (3) that had proposed a structure for the leucine zipper—a region of repeating leucine residues—the eager graduate student set out to test it. O'Shea made synthetic peptides corresponding to the leucine regions of some transcription factors, such as the mammalian factors Fos and Jun and the budding yeast transcription factor Gcn4. She showed that they were α-helical and dimeric but that the orientation of the helices was not antiparallel, as had been proposed, but was parallel (1). She went on to solve the crystal structure of a leucine zipper two years later (2). With these discoveries, O'Shea made her mark on the world of protein structure: “The reason that this was such a big deal is, first, a lot of proteins have this coiled-coil region that's involved in dimerization—the leucine zipper defines a class of transcription factors that are one of the most common found in mammalian cells,” says O'Shea. The structure of a coiled coil had been independently proposed by Francis Crick and Linus Pauling based on their fiber diffraction studies of keratin in the 1950s. O'Shea's leucine zipper structure captured their attention: “Crick wrote me a letter, congratulating me on the high-resolution structure,” O'Shea notes when recalling how her paper was received by the protein structure community.

O'Shea is quick to credit Kim with putting her on the road to academic success. One of the most important aspects of research lies in asking the right questions and figuring out how to answer them. According to O'Shea, Kim had “tremendous intuition for picking interesting biological problems. Maybe even more important, he really encouraged me to do science.” O'Shea had intended to get a Ph.D. when joining Kim's laboratory, but she originally accepted a position in an M.D./Ph.D. program. “Then I started this work on the leucine zipper, right after I finished college,” she says. “I was struggling with whether I should go to medical school, and Kim really encouraged me to pursue a research career.” Kim told O'Shea that she was good enough to succeed in science. “I've never forgotten what he said. In times when things weren't going so well, it was something I could think about,” says O'Shea. “It's something that sticks with you.”

Figure 1.

Erin K. O'Shea. Photograph courtesy of Janet Yang and Brandon Toyama.

Nucleosomes and Gene Expression

O'Shea's productivity and outstanding performance during graduate school quickly earned her a highly coveted faculty position offer right out of graduate school. Although her faculty position at UCSF was assured, O'Shea was given some scientific free time. “They gave me two years to do whatever I wanted to do,” says O'Shea. She elected to spend that time working in the laboratories of Tjian at Berkeley and Ira Herskowitz at UCSF.

In Tjian's laboratory, O'Shea began to think about the focus of her research career. She decided the key question to address was how gene expression was regulated by chromatin. O'Shea knew the topic was not a natural extension of her previous protein structure work. “Tjian took me into his lab at a time when I really didn't know much biology at all, and that didn't seem to bother him,” she says. “When I decided I wanted to work on chromatin and gene expression, I am not sure he thought it was a great idea. But he listened to what I had to say about why I wanted to do it, and he supported me in every way possible.”

At that time, much of the transcription field was focused on studying transcription in purified in vitro systems, with purified components and a piece of plasmid DNA, says O'Shea. “I wanted to take it to the next level and study the DNA template in a more physiological context—with nucleosomes assembled,” she notes.

O'Shea's aspirations to study chromatin and gene expression in a physiological system led her to the laboratory of Herskowitz, a major figure in the field of yeast genetics. Herskowitz, who had determined the genetic pathways that allow budding yeast to change mating types, offered O'Shea the support she needed to make the jump from beakers to biology.

“I had never really worked with living organisms. Really, all I had done was physical chemistry—spectroscopy, peptide chemistry, and x-ray crystallography. I was changing fields,” recalls O'Shea. “From the beginning, Ira was so encouraging and never even thought twice about the fact that I was totally changing career tracks.”

At the time, the yeast gene PHO5 seemed like a straightforward model system to study chromatin and gene expression. When the gene is turned off, the nucleosomes sit in the regulatory promoter region of the gene; when the gene is turned on, in order for transcription to happen, the cells have to remove those nucleosomes.

But uncovering the mechanism behind nucleosome removal proved more difficult than O'Shea had expected: “It was such a hard problem,” she says. “I was bumbling and fumbling with an undergraduate, and we weren't making progress.”

So O'Shea turned to another attribute of the PHO5 gene: its control via a signal transduction pathway that responds to the nutrient phosphate. Researching the intricacies of this signal transduction pathway turned out to be the final shift in her career, and the phosphate system has been the axis of her research enterprise ever since. “Most of my research is centered around trying to understand at a very basic level how cells sense and respond to changes in their extracellular environment,” explains O'Shea. “I picked a model system, this PHO5 promoter. Then it turns out that PHO5 is one of these phosphate-regulated genes, and that's how I got to studying the signaling part of phosphate regulation.”

Using the budding yeast Saccharomyces cerevisiae, O'Shea set out to determine how the cell senses and responds to inorganic phosphate. Previous work by researchers in Japan had already identified many members of the pathway that regulated PHO5 expression (4). They had identified PHO2 and PHO4 as transcription factors—proteins that bind to the PHO5 regulatory region to turn it on. Two other pathway members, PHO80 and PHO85, seemed to collaborate to turn off PHO5 expression.

Despite earlier speculation to the contrary, O'Shea discovered that PHO80 and PHO85 actually belonged to the family of cyclin/cyclin-dependent kinase (cdk) proteins. “We found a cyclin/cdk complex that was involved in a process other than the cell division cycle; it was involved the signal transduction pathway that allows the cells to sense and respond to inorganic phosphate levels,” says O'Shea. When the cell has plenty of inorganic phosphate, the PHO80/85 complex turns off transcription of PHO5 by phosphorylating the transcription factor PHO4 (5).

“And that's noise, randomness... it's not predictable.”

Continuing her investigation of this pathway, she subsequently found that a gene, PHO81, codes for an enzyme inhibitor of the cyclin/cdk complex (6). PHO81, it turns out, has homology to a whole class of mammalian cdk inhibitors that regulate cell division. Not surprisingly, mutations in these mammalian genes have been linked to cancer.

O'Shea continued on, detailing how phosphorylation regulates the activity of this transcription factor. She determined that the transcription factor PHO4 is phosphorylated four times by the PHO80/85 complex (7). Three of the phosphates control the location of the transcription factor PHO4 within the cell (8). Two of those phosphates cause PHO4 to be exported from the nucleus (9), the third controls its import into the nucleus (10), and the fourth is involved in regulating DNA binding.

O'Shea notes that teasing out the contribution of each phosphorylation “was critically important, because this is how the cell controls the expression of dozens of genes, just by regulating this one transcription factor. Many transcription factors and regulatory proteins are multiply phosphorylated, and phosphorylation is important for their regulation,” she adds. “Our studies were really some of the first to demonstrate mechanistically how phosphorylation of a protein controls its import into the nucleus and export from the nucleus.”

Proteomics

“When I was appointed as a Hughes investigator four years ago, they challenged us to do something new,” says O'Shea. Together with her graduate school colleague, Jonathan Weissman, O'Shea decided to tackle proteomics. Their lofty goal: to determine the location and the abundance of all 6,200 proteins encoded by the yeast genome.

She and Weissman made two collections of yeast strains with 6,200 members each—one strain for each gene. They labeled the genes with GFP and analyzed the location of each of the protein products within the yeast cell (11). In a similar experiment, they tagged each gene with a different label to measure the abundance of each protein (12). Together, O'Shea and Weissman accomplished an almost inconceivable goal—they catalogued the location and amount of approximately 75% of the proteins encoded by the yeast genome.

Noise Happens

O'Shea has recently become more interested in things of a more abstract nature. One problem she is attempting to address is the role of noise or randomness in biological processes.

“When there are small numbers of molecules reacting with one another, you frequently get probabilistic effects. And that's noise, randomness... it's not predictable,” says O'Shea. “Many processes in biology, in particular those involving transcription, involve very small numbers of molecules and hence are subject to these probabilistic effects.”

Her laboratory now studies random fluctuations in transcription. “We measured these random fluctuations and tried to understand the origin of the randomness—how noise affects the process of gene expression, and how noise might be used or suppressed to influence phenotypic outcomes,” she says. The recent work suggests that the noise inherent in gene expression may be a result of chromatin remodeling. The removal of the nucleosomes from DNA, in preparation for transcription, is the source of much of the noise (13).

This noise, notes O'Shea, may have important biological and evolutionary implications. “It might be beneficial for cells to diversify phenotypically,” she says. “The noise that was generated to produce different amounts of proteins in different cells might allow a fraction of the cells to be poised to survive in an unpredictable environment.”

Although the chromatin/gene expression problem has been a challenge for her for nearly 10 years, O'Shea's team has made headway into addressing it: “Recently, we identified a class of molecules—inositol polyphosphates, small signaling molecules within the cell—that regulate that process (14). This class was one of the first small molecules found to regulate chromatin remodeling.” However, O'Shea concedes that it is still unclear whether it is a specific effect or something general.

Bridging Gaps and Making Connections

Although her initial journey into proteomics identified the cellular location and abundance of nearly all the yeast proteins, it was simply a snapshot under one condition that the yeast were experiencing at the time the GFP was visualized. With her Inaugural Article (15), published in this issue of PNAS, O'Shea is beginning to address the dynamic changes in the proteome. By screening the yeast strains that express the transcription factor-GFP fusions under different environmental conditions, she has identified a new transcription factor that senses environmental stress and translates that by regulating cell growth.

“The primary objective was to make new connections between a particular extracellular stress and changes in gene expression,” says O'Shea. Using the small molecule rapamycin, a drug that inhibits the highly conserved TOR signaling pathway, as the stressor, O'Shea was able to identify a transcription factor (Sfp1) that changes its location within the cell in response to stress. Sfp1 responds to the stressor by controlling the expression of the protein components of the ribosome. By controlling the function of the ribosome, this transcription factor regulates the amount of protein a cell makes, which in turn limits cell growth and size.

According to O'Shea, the transcription factor provides a link between the signaling pathways that are responsive to nutrients and stress and those that control cell growth and cell size through the control of ribosome production.

O'Shea hopes to link even broader connections to her work in the future: “This new work that we're doing in the general area of systems biology I hope will help move biology from being a very large collection of facts to reveal more fundamental rules and something about the logic of how cells are organized and how they do the remarkable things they do.”