January 2005--Proteomics is the firstborn child of the genomic revolution. And like most newborns, it's growing exponentially.
“At the beginning, proteomics had a very narrow definition,” says biochemist Heng Zhu, of Pharmacology and Molecular Sciences. “It simply meant using a 2D gel to reveal all the proteins in a cell or tissue or even organisms. You got a profile of proteins, so people called it proteomics. Now this word keeps growing. It's become an umbrella to cover many areas.”
Today “proteomics” encompasses a wide range of technologies and techniques tailored to the field, all aimed at shedding light on the structure and function of given proteins and on how they interact with other proteins, genes, cell lipids and drugs.
Because Hopkins has both steadily and wisely acquired the resources to make it a leader in this new field, the NIH awarded a $17 million grant to a team of researchers led by molecular geneticist Jef Boeke, director of Hopkins' High-Throughput Biology (HiT) Center. The center's goal is developing new technologies to examine protein interactions in systems ranging from yeast to human cells.
“One thing that really made our proposal stand out,” Boeke says, “was the technical expertise of the Hopkins team.”
Zhu, for example, developed his revolutionary protein chip technology in 1999, while still a postdoc at Yale. The chip, rows of microwells on a silicone sheet fitted over a microscope slide, permits rapid, high-throughput protein screening, making it possible to purify thousands of the molecules and study their activities with unprecedented speed and precision. Before the chip's invention, it could take years to single out a protein on the basis of its biochemical activity; now, thousands can be isolated in a few hours.
A caveat comes with what the chips and other technologies generate, however, stemming from the sheer volume of information and the way results can vary. Sophisticated data-wrangling is necessary before anyone can draw conclusions. Snapshots of a cell's protein activity, for example, can change sharply in a moment’s time. Slight differences in using chips can also make results less definitive. How many cells should you sample? How often? What statistics should you use? Bioinformatics has thus become a critical component of any proteomics project.
Biochemist Akhilesh Pandey discovered this in 1998, working in Denmark under researcher Matthias Mann. Pandey spent three years in the lab he calls “a mecca for proteomists” for its expertise in mass spectrometry, a standard way to identify proteins based on mass and electrical charge. In Denmark, he learned the value of “painstakingly going over data, trying to understand its significance,” and he's imported that approach to his own lab at Hopkins' Institute of Genetic Medicine.
“People like to say that I identified 1,000 proteins, 2,000 proteins,” he says, “but quality and reliability of data are much more important. I'd rather be known for 300 well-characterized proteins than 1,000 guesses.”
The difficulty of pinning down trends in techno-generated data disturbs some scientists (and grant reviewers) who characterize this type of work as “descriptive.” Accustomed to hypothesis-driven research answering a well-defined question, they don't see the value of accumulating masses of data on protein interactions.
“A lot of it is frankly tedious,” admits Boeke, “making these long lists of proteins and trying to glean information from them. But our bioinformatics specialists help spot patterns you might never have seen. That sometimes sparks a new hypothesis and opens a new avenue of investigation.”
Unlike the fairly stable genome, an organism's proteome—all of its proteins— constantly changes throughout its lifetime. It shifts, for example, as cells move from newborn to adolescent to senescent. And, of course, it changes with disease. Understanding protein dynamics will thus give researchers insights into fundamental biology and help them develop new targets for treating sickness and improving human health.
“This,” says Pandey, “is systems biology from the ground up.”
Jef Boeke of on microarrays 101 and scientific nomenclature