Gregor Mendel Professor of Genetics and Medicine Harvard Medical School
Geneticist, Brigham and Women’s Hospital Investigator, Howard Hughes Medical Institute

When he began a postdoc in Ronald Davis’s laboratory at Stanford University in 1984, Stephen Elledge wanted to develop new ways to knock out and mutate specific genes in mammals. His first experimental results contained a serendipitous artifact that laid the foundation for a scientific career studying how eukaryotic cells deal with damage to their DNA.

As a start to designing those gene-targeting tools, Elledge, now a professor of genetics at Harvard Medical School, began by trying to clone the mammalian homolog of recA, a bacterial gene required for DNA repair via recombination. Because there was no mammalian recA homolog, Elledge attempted to clone the Saccharomyces cerevisiae (baker’s yeast) homolog using a novel method that included an antibody step. The...

“DNA was cool itself, but the fact that you could take it apart and put it back together and test ideas on genes—that totally blew my mind. I decided I wanted to do that.”

Elledge had found that Rnr2 protein levels increased when yeast cells were grown in the presence of agents that damaged DNA. He mentioned this to David Stillman, who was at Stanford to interview for a faculty position, and who studied cell cycle regulation of proteins as a postdoc in Kim Nasmyth’s lab at the MRC Laboratory of Molecular Biology in the U.K. Stillman pointed out that ribonucleotide reductase was cell cycle regulated—rather than remaining stable, the RNA and protein levels fluctuate throughout the cell cycle. Elledge decided RNR2 was worth another look. He found that RNR2 RNA levels increased dramatically, even more than the protein levels, upon exposure of cells to DNA damage and that mutations in RNR2 resulted in hypersensitivity to DNA damage.

“I thought, wow, this is gigantic induction. Then I thought, there must be a sensory pathway that recognizes the DNA damage that’s going on in the cell,” says Elledge. Studying RNR2’s regulatory elements, he found those that were necessary to induce the production of higher protein levels in response to DNA damage and identified factors that bind these DNA elements to mediate the response of RNR2 to DNA damage.

Elledge’s idea that eukaryotic cells sense the progress of DNA replication and transform that information into a DNA-damage response was new. While most molecular biologists thought signaling pathways worked by sensing signals extrinsic to the cell and relaying the information to the nucleus, Elledge was proposing an internal signaling pathway that senses cell-intrinsic events. Those results led him to study how cells monitor roadblocks to replication and DNA damage, such as nicks and double-stranded breaks, and how the cell handles that information.

Here, Elledge, talks about how he fell in love with chemistry, how the crux of his graduate thesis was based on a misunderstanding, and why his life partner had to be a scientist.

Emerging Elledge

It’s all matter. Elledge was born in Paris, Illinois, and lived much of his childhood with his paternal grandmother. His family was not “academic,” says Elledge. He attributes his own interest in science partly to the Science Research Associates (SRA) reading program: reading-level-rated pamphlets on different subjects, including science. “This was during the Space Race era, so there was a real effort by the U.S. government to get kids interested in science.” The final levels—bronze, silver, and gold—were physics and chemistry subjects on matter, atoms, and subatomic particles. “The one on how matter was built out of smaller building blocks really got my attention,” says Elledge. “The idea that you could explain everything from smaller and smaller components really appealed to me. I remember sitting on this couch at my grandmother’s that had fraying edges, where you could peel back layer after layer until you got to the wood, and thinking that this couch is built just like all matter.”

A thing of beauty. By middle school, Elledge was checking out chemistry books from the library. In high school, he excelled in math and chemistry classes. He was on the chemistry team and participated in an interstate contest sponsored by the American Chemical Society, taking first place in the exam competition. “I just loved chemistry: the way the periodic chart self-assembles and predicts the properties of different elements, and the idea that there is a physical reality behind everything and everything has an explanation. I thought that was so beautiful,” says Elledge.

Biology eye-opener. In 1974, Elledge entered the University of Illinois at Urbana–Champaign on a full scholarship and majored in chemistry. Thinking he would work in the chemical industry, he all but ignored biology. When his pre-med roommate told him he should pay attention to this “DNA makes RNA makes protein and that it’s really cool, I just said, ‘Yeah, yeah,’ and ignored him.” says Elledge. But learning about recombinant DNA in a senior-year biochemistry course opened his eyes to biology. “DNA was cool itself, but the fact that you could take it apart and put it back together and test ideas on genes—that totally blew my mind. I decided I wanted to do that,” he says. Elledge applied to graduate school in biology, but was nervous about getting in because he had no lab research experience. He decided on the Massachusetts Institute of Technology (MIT), based on advice that it was the best place with plenty of good potential advisors.

Effective Elledge

Genetically inclined. Elledge entered MIT in 1978 as a biology graduate student. To compensate for his lack of biology knowledge, he overloaded on catch-up courses, taking 13 over three semesters. Despite thinking he would do enzymology research, Elledge joined Graham Walker’s bacterial genetics lab, drawn to Walker, “who was a really nice person,” and to the lab, thanks to a paper he had read about the RecA protein, a bacterial protease that’s essential for DNA repair. In Walker’s lab, Elledge worked on the umuC (unmutable C) gene that, when mutated, resulted in strains that couldn’t produce genomic mutants even when the bacteria were grown in the presence of mutagens. Elledge initially cloned the umuC gene using a technique he developed himself because the standard plasmid library of E. coli DNA to complement the mutant phenotype didn’t work with umu gene mutants. “I discussed a strategy of how to do it with Graham, but when I told him how I did it, he asked how I ever thought of that, because the method I had used was not the one he suggested. I had misunderstood him, and it turned out that the best idea of my thesis was one that no one actually had!” Elledge showed that umuC was really two genes, umuC and umuD, and they worked together to promote error-prone repair.

Genetic tools. Besides working on the DNA-damage signaling pathway, Elledge also focused on creating new laboratory methods. As a graduate student, he had already designed novel bacteriophage lambda cloning vectors. As a postdoc in Davis’s lab, Elledge designed multifunctional lambda phage vectors that could be converted to plasmids for expression in yeast and E. coli. When Elledge started his own laboratory at the Baylor College of Medicine in Houston in 1989, one of his first experiments was to create a human cDNA library using his phage vectors. Elledge did an experiment—a repeat of one he heard Paul Nurse describe in a talk—to find human genes by complementing a yeast cell-division-cycle (CDC) mutant, CDC28. Elledge not only identified the same gene as Nurse, but also the CDK2 gene, required for eukaryotic cells to proceed to the S phase of the cell cycle.

Revealing how cells cope with DNA damage. In 1993, Elledge and his first graduate student, Zheng Zhou, identified Dun1, a yeast kinase they showed was directly involved in the signal transduction pathway that controls the DNA-damage response. The work was the first to demonstrate that the damage response in eukaryotes is regulated by phosphorylation, providing evidence for Elledge’s hypothesis that an intracellular signaling pathway monitors genomic integrity. Elledge’s lab, together with that of biochemist Wade Harper, then also at Baylor, discovered cyclin-dependent kinase (CDK) inhibitors, including p21, which modulate the activity of the CDK-cyclin complexes that control transition to the S phase of the cell cycle. His lab continued to identify components of both the yeast and mammalian DNA-damage signaling pathway, components of which also interacted with the cell cycle machinery.

In 1995, Elledge’s graduate student Jim Allen found yeast mutations that result in cell death if the cells are exposed to DNA damage. The mutations were in the RAD53 gene, which encodes a protein kinase that acts upstream of DUN1, and revealed that a kinase signaling pathway is responsible for maintaining genomic integrity—alerting the cell that there is DNA damage and arresting the cell cycle until that damage is fixed. His lab also identified two yeast genes, MEC1 and TEL1, which encode kinases that act upstream of Rad53. Upon DNA damage, Mec1 and Tel1 can transduce the DNA damage signal to Rad53, which ultimately results in arrest of the cell cycle. The team also discovered a role for the DNA damage response in regulation of BRCA1, a tumor suppressor gene that can lead to breast cancer when mutated.

Start signal. In 2003, Elledge moved his lab to Harvard University, where he continues to study DNA-damage signaling. That year, he and Lee Zou identified one of the ways that the pathway senses a damaged DNA replication fork—the accumulation of single-stranded DNA (ssDNA) coated with the replication protein A (RPA). “This is what happens at the top of the pathway that leads to all of the signal transduction: when there are stopped or stalled replication forks, you get longer stretches of ssDNA,” says Elledge.

State of the cell. “Many biologists used to think that DNA damage was just about arresting the cell cycle, but the fact is that the DNA-damage pathway regulates about 5 percent of the genome, so it’s really a global controller that throws many switches on and off,” says Elledge. “These switches have to do with repair at the right time and the right place and with communication to other cells, to the immune system. It’s remarkable that the cell can figure out whether and how its DNA is damaged and can then do something about it.” To understand how broadly the DNA-damage-signaling pathway extends within the function of a eukaryotic cell, Elledge’s lab set out to identify all the substrates of the kinases within the pathway. In 2007, they demonstrated the extent of the influence of the DNA-damage response on cell function—beyond mediating the cell cycle—by showing that two of the upstream kinases that mediate the response modify nearly 1,000 proteins, including ones involved in repair, but also in senescence, apoptosis, and metabolism. Many of these proteins, including the DNA-damage response and cell-cycle ones, are mutated in cancers.

Leading Edge Elledge

Can’t stop, won’t stop. In addition to working on DNA-damage response, Elledge’s lab is also developing tools to understand how the immune system is wired, including how immune cells and antibodies recognize their epitopes. His lab’s first stab at studying HIV, a genetic screen, identified more than 250 proteins HIV needs for its life cycle in its human host, and they have performed similar screens for hepatitis C virus and influenza A. Elledge’s lab also recently developed a blood test that can provide a personalized history of an individual’s exposure to viruses by identifying the immune system’s memory of the viral exposure, using antibodies in the blood. Of 600 individuals studied, the study found, the average person had been exposed to 10 viral species over his or her lifetime. Now, Elledge wants to study immunology.

Lesson learned. “I was really nervous about doing research in graduate school because I had no experience, and I learned a really valuable lesson right away. Someone gave me a plasmid and told me its concentration in the sample. I was supposed to transform it into E. coli. I tried and tried and thought my method and plates were bad. I finally figured out that the person told me the wrong plasmid concentration by a factor of 1,000. That’s when I realized that you can’t trust anyone else’s reagents, a really valuable lesson I tell my graduate students all the time.”

Partner in science. Elledge is married to Harvard geneticist Mitzi Kuroda. “I had to marry a scientist because no one else could put up with my passion for science unless they really understood it themselves, and I think that was a huge part of my success—I was able to follow my passion. I love and want to talk about it all the time, and to have a partner who shares that same passion is great,” he says. 

Greatest Hits

  • Discovered that mRNA levels of the ribonucleotide reductase enzyme, which helps make DNA nucleotides, increased dramatically in response to DNA damage, resulting in the proposal that a signal transduction pathway senses the rate of DNA replication and adjusts DNA synthesis and repair for accurate genome synthesis
  • Identified the CDK2 gene that encodes the cyclin-dependent kinase 2 enzyme, and together with Wade Harper established how Cdk2 protein kinase is activated and functions to control the transition from the G1 to the S phase of the cell cycle
  • Devised cloning technologies, including the first hybrid plasmid and bacteriophage vector and derivatives that could be expressed in either E. coli or S. cerevisiae and used for the two-hybrid system
  • Identified DUN1, a gene that encodes a classic kinase activated by DNA damage, providing evidence for an intracellular signaling pathway activated directly by DNA damage
  • Showed that the DNA-damage signaling pathway communicates with and influences many cellular functions beyond DNA repair, including senescence, apoptosis, and metabolism

Interested in reading more?

Magaizne Cover

Become a Member of

Receive full access to digital editions of The Scientist, as well as TS Digest, feature stories, more than 35 years of archives, and much more!