The big picture

July 14, 2004

Tags: Young LabGenetics + Genomics

CAMBRIDGE, Mass. — Like many researchers who have their hands in a number of academic and industry collaborations, Richard Young’s calendar is filled with meetings with lab members and collaborators, conference calls, and lectures. To make this triple-booked schedule worse, many of his appointments take him out of town. Travel is one area, however, where he has learned to tame the chaos.

Four years ago, when this wayfaring started to pick up, Young took matters into his own hands and got his pilot’s license. Now, no longer at the mercy of commercial airline schedules, he can fly to and from Washington, D.C., or Chicago and make it home for dinner with his wife and 7-year-old daughter.

“It has its drawbacks,” he observes. “I can’t sit back and break out my laptop or just read a book. I need to stay focused on flying the aircraft and trying not to have too much fun.” An important aspect of flying, Young says, is maintaining situational awareness—knowing precisely where the aircraft is in three dimensions and projecting where it needs to be at any time throughout the rest of the flight.

There’s a phrase for what happens when pilots lose the situational awareness that Young describes: “falling behind the plane.” While communicating with air traffic controllers or changing flight plans due to poor weather, the pilot may lose his sense of where the plane is and where it needs to be. The technicalities of how to fly and land a plane may have been mastered, but when a pilot loses his grasp on the big picture, Young notes, “the fun ends right there.”

It’s here that biologists might learn something from the world of aviation.

In 1996, while attending a conference at Cold Spring Harbor Laboratory, Young began to get the uneasy feeling that the world of biology was, in a sense, “falling behind the plane.” Sitting in the lab’s Grace Auditorium, he and a few dozen researchers presented findings about the individual genes they’d been studying. They debated the details of how these few genes are regulated, but there was no talk of the big picture: how all genes work together to produce living cells and organisms.

The problem, in Young’s opinion, was that biology was stuck in the one-gene-at-a-time approach; genes were studied in isolation from the larger context of the genome. It was like trying to understand how to fly an airplane by studying the engine in great detail while knowing nothing of the principles that make precise flight possible. If scientists didn’t step back to look at all an organism’s genes at once, to see the individual parts in the context of the whole, then all this genetic research would never get off the ground.

That was eight years ago. Today, the 50-year-old scientist, like many of his fellow biologists, has long eschewed the one-gene-at-a-time approach. “Biology is undergoing a revolution right now,” he says. “Everything from how we train undergraduates to how we conduct science is becoming more and more genome-based.”

Young’s gaze in this revolution is focused on the complex networks through which genes and proteins communicate. In particular, Young is studying the more than 2,000 proteins called transcription factors that switch genes on and off in humans. Dozens have been linked to any number of diseases. Young’s plan is to locate them all and figure out which genes they control. Drafting such a map could change the face of drug development and help doctors pinpoint an individual’s risk for diabetes, hypertension, and other health problems with a simple analysis of their genetic profile. While the possibilities of such work are limitless, scientists are constrained by the limitations of conventional technology. Creating this intricate protein map would take centuries using available tools.

Young doesn’t have centuries. If he is to map the interactions of genes, proteins, and disease in his lifetime, he must find a new way to do what, until now, wasn’t possible.

The science of everything

Young is by no means alone in his aim to tackle the biological big picture through a multidisciplinary approach. Research centers and programs combining the work of biologists, engineers, chemists, and computer scientists are cropping up around the country, part of an emerging field called systems biology. Although the term was first coined in the 1960s, it has become increasingly popular in recent years.

At its most basic level, systems biology is an examination of cellular life as an integrated system rather than as individual molecules. Systems biologists typically take the data from multiple experiments and use computer algorithms to weave the parts into a whole, almost like re-creating an atlas of the U.S. by analyzing bits and pieces of maps from individual states. One technological advance that has made this possible—and which also is central to Young’s current research—is the microarray chip, a quarter-sized slice of either glass or silicon that can contain up to 100,000 gene fragments. These chips provide snapshots of the genome at work, showing, for example, which genes are turned on or off at any given time.

“When you describe systems biology, it’s almost like saying you’re trying to understand the science of everything,” says Wendell Lim, associate professor of cellular and molecular pharmacology at University of California, San Francisco. “In many ways it’s a much more vague term than genomics.”

The phrase “science of everything” certainly seems fitting when one looks at the human genome as a whole. The genome, two copies of which fit inside a single cell, is composed of about 3 billion nucleotides—the DNA building blocks represented by the letters A, C, T, and G. However, the typical human gene averages anywhere from just a few hundred to a few thousand nucleotides. For Young to study groups of genes and proteins within the context of the whole genome is like a marine biologist studying a particular species of fish with an eye on the entire population of the Atlantic.

In many ways, though, the genome is an entire population. Thirty thousand genes produce hundreds of thousands, perhaps even millions, of proteins. These genes and proteins all communicate with each other through intricate networks responsible for carrying out the cell’s work. Young and many other scientists believe that a person’s medical future is embedded in these communication networks. And he’s determined to find out exactly who is talking to whom, and more importantly, what they’re saying.

From yeast to humans

To eavesdrop on this conversation Young turned first to common baker’s yeast. It might seem strange that the desire to uncover the deepest biological mysteries of human life would lead a scientist to one of the main ingredients of beer and pizza dough. But yeast is a proven testing ground for biologists, the perfect context in which to hone scientific exploratory techniques before moving to the bedlam of human cells.

The human genome contains vast terrains of DNA that don’t serve any known purpose. In fact, this so-called “junk DNA” makes up over 90 percent of our genome. To complicate matters further, one gene can produce dozens, sometimes even thousands, of different proteins. The yeast genome, by contrast, is simple, neat, orderly—and small. There are only 6,000 genes in all, and each gene produces only one protein. For these reasons, yeast was the logical starting point.

Young’s particular approach for viewing the entirety of the genome has a self-evident logic to it. If you want to find out what any “organization” is all about, find out who’s running it. With the genome—yeast or human—that part is easy. The whole show is run by transcription factors, proteins that bind to genes and act as control switches, flipping the genes on and off. Transcription factors give the orders; genes follow them. So, then, to understand how the genome runs, there was no better place to begin than by first locating all the transcription factors and finding out what they’re telling the genes to do. “The only problem,” says Young, “was that with traditional laboratory tools, this would take over a hundred years to pull off—even in yeast.”

In the late 1990s, microarray technology hit the biology scene, at last providing scientists with a method to quickly analyze genes en masse. But to make sense of the reams of data microarrays provided, some intense computational power was required. Young put together a team that included, among others, postdoctoral associate Duncan Odom and David Gifford, a computer scientist at Massachusetts Institute of Technology where Young also is a professor of biology. Young’s lab assembled the micro-array chips; Gifford constructed the algorithms to crunch the data.

In October 2002, the research team reported in the journal Science that they had discovered the binding points of 106 of yeast’s 200 transcription factors. The technology they developed—which reduced to a matter of months what would take centuries using conventional methods—enabled the scientists to create a map of sorts, a schematic that diagramed how transcription factors and genes in yeast communicate with each other. It was a complex picture, since one transcription factor can bind to and regulate multiple genes. In addition, transcription factors send signals to each other, as well as communicating with genes—a system that Young calls a regulatory network. For the first time, scientists had a working set of operating instructions for an entire genome.

“But this was simply a proof of context experiment,” says Young. “Doing this in yeast just proved that the tools worked.”

The real test, he says, was to try the system out on the human genome. Transcription factors are known to play key roles in many common diseases, but no one had yet developed a process for hunting them down and identifying all their points of operation. If Young’s technique worked in human tissues, the payoff could be immense.

Tuning into the networks

For human genes, Young had two choices. He could use readily available and plentiful lines of cultured cells, but most of these cell lines have, over generations, developed genetic abnormalities that might compromise study results. Far more challenging, but equally rewarding, would be to use donor-grade human tissue samples, the same quality used in transplant procedures. This would be as close to a living, breathing body as the researchers could get.

Acquiring these tissues is no easy task. It requires persistence, patience, and a willingness to respond quickly to a call from a donor center. Last year, one of those first calls came through on Odom’s cell phone during a weekly lab meeting. Odom checked his caller ID: It was the Joslin Diabetes Center. Staff there had pancreatic tissue samples for his research, and he needed to pick them up fast. He quietly ducked out of the meeting, loaded some plastic test tubes into a cryogenic carrier, threw on his coat, and ran for the subway.

When he arrived at Joslin, he transferred the tissue into the test tubes, filled the tubes with formaldehyde, and capped them with brightly colored lids. The process, Odom recalls, was “pretty anti-climactic.” The end result, however, could be a windfall of insight into human disease.

Odom spent the next week extracting all the genetic information from these tissue samples and applying it to a new suite of microarray promoter chips. An algorithm developed by Gifford, a modified version of the one used in the yeast research, interpreted the data and displayed the complex networks that these genes and proteins form. Odom examined the networks, looking at four transcription factors associated with type 2 diabetes.

His task was to take the donated pancreatic tissue from Joslin Diabetes Center and liver tissue that he received from another donor program at the University of Pittsburgh, comb the entire genome in both types of samples, locate every single point to which each of these transcription factors bind, identify each gene that they control, and learn how these transcription factors communicate both with these genes and with each other.

“It’s an extremely complex and deeply integrated network,” says Odom, “one that orchestrates the creation and maintenance of the pancreas and other human organs.” If a transcription factor is damaged, the cell may end up producing the wrong amounts of any number of proteins. The entire network can be thrown off balance, causing a change in insulin release that could lead to diabetes.

The researchers were successful in their hunt, and in February, they reported the location of every genome binding point of these four transcription factors—again completing in months what in the past would have taken centuries. The team discovered that one of the transcription factors regulates nearly half of the 3,000 genes necessary to make both a pancreas and a liver. In a world where scientists tend to examine individual genes and proteins to find the molecular causes of disease, pinpointing this one transcription factor could yield a wealth of genetic information. Perhaps, Young suggests, researchers might be able to develop medications that modify the activities of mutated forms of this transcription factor. Doing so would correct the activity of 1,500 genes and possibly even prevent type 2 diabetes in at-risk individuals.

The study, which was published in Science, in addition to uncovering some compelling basic biology about type 2 diabetes, demonstrated that the technology works in human tissue, signaling a new phase in human genomic research.

“Before, we were just looking at conditions one gene at a time,” says Graeme Bell, molecular biologist at University of Chicago and coauthor of the latest Science paper. “Now we can see the whole playing field, and more importantly, we can see the players.”

Making a list

Now that the scientists have a good model to follow, Young plans to use it to study all the tissues and organ systems in which gene regulators are involved in disease. And that is why he’s compiling a list—a very long list—of every known transcription factor related to diseases and conditions such as cancer, hypertension, birth defects, neurological disorders, and obesity, among others. His goal is to map every gene-protein communication network that each of these transcription factors regulates and do this in every human tissue.

“In the end,” says Bell, “these efforts will provide a detail of understanding of the regulation of gene expression that will open doors and possibly lead to a whole new approach to many of the most common diseases.”

Still only a few months old, this most recent Science paper has caused a surge of interest in the therapeutic value of identifying these gene-protein networks. Scientists such as former Whitehead Fellow Trey Ideker are developing technologies to analyze the new information about gene-protein interactions that researchers like Young are discovering.

“In the next five to 10 years, we’re going to see the study of these networks dominating the scene in biology,” says Ideker, who now is an assistant professor of bioengineering at University of California, San Diego. “I’d say that right now, this field is where the Human Genome Project was in 1985.”

Of course, back then sequencing the human genome was still something of a dream. Today, the completed project, essentially a massive “parts list” in which the chemical building blocks of DNA are laid out in a linear line, often has been compared to dissecting a Boeing 747 and placing every last nut and bolt on the ground next to each other. Although we can see the entire body of the aircraft from the inside out, there’s nothing to indicate what goes where—let alone how the machine stays airborne.

By understanding the complexities of how genes and proteins interact, Young and others in the field are developing an instruction manual scientists can use to shape the information from the genome project into a map of human disease. In a sense, they’re figuring out how to build an airplane, a daunting task even for the most seasoned pilot. The key for Young, it seems, will be to recall the lessons of aviation that have guided him from the runway, to 20,000 feet above ground, and back down again: Keep your eyes on the horizon. Know where your plane is and where it should be at all times. And never lose sight of the big picture.

Written by David Cameron.


Communications and Public Affairs
Phone: 617-258-6851


New genomics tool boosts diabetes research

Whitehead Institute is a world-renowned non-profit research institution dedicated to improving human health through basic biomedical research.
Wholly independent in its governance, finances, and research programs, Whitehead shares a close affiliation with Massachusetts Institute of Technology
through its faculty, who hold joint MIT appointments.

© Whitehead Institute for Biomedical Research              455 Main Street          Cambridge, MA 02142