To a biologist, the number of proteins produced by an organism can seem almost as numerous as the stars in the sky.
Robin Rombach, Post-GazetteJonathan Minden's work has made it possible to identify proteins that differ from each other out of thousands that are the same. Minden is an associate professor of biological sciences at Carnegie Mellon University.
Click photo for larger image.
The human body, for instance, might make a million different varieties of protein, present in varying concentrations in each cell. But only a handful of those might be significant in diagnosing a cancer or in explaining how a cell functions.
So it makes sense that biologists would borrow some tricks from astronomers when it comes to analyzing an animal's complete set of proteins ---- what scientists call the proteome.
Jonathan Minden has employed one of those tricks in developing a tool for the new field of proteomics. It allows researchers to quickly identify specific proteins that differ among samples, while ignoring the vast number of proteins that don't.
It already has yielded surprising results that could force some rethinking of embryonic development.
In what has been called the first comprehensive proteomic analysis of a developing animal, Minden and his colleagues at Carnegie Mellon University looked for proteins that cause embryonic fruit fly cells to change shape during a developmental step called gastrulation. In this critical step, the embryo transforms from a hollow ball of all-purpose cells into a three-dimensional organism made of cells with specialized functions.
"We expected to find maybe five, 10 protein changes," Minden said, "but we found more than 100." The proteins were more than just the "cables and motors" necessary for shape changes, also including those for such functions as digestion, repair mechanisms and waste disposal.
"The cells seem to be primed for change," Minden said, noting quantities of all but one of these proteins seem to build up an hour before any reshaping occurs.
Every time the researchers tried to eliminate any of those protein changes, he added, the cell transformation process would stop dead in its tracks.
The findings, reported earlier this year in the journal Development, suggest that the process of gastrulation is far more complex than previously suspected.
"Gastrulation is still a big puzzle," said Ruth Lehmann, a developmental geneticist at New York University Medical School and the journal's editor. Minden's work likely will lead to new insights into the shape-changing mechanism, though it is too soon to say so conclusively, she added.
"This is all very new," Minden said. "Most people don't know what to make of it because it's coming out more complex than suspected."
The shape-changing mechanism is important not only in development, but in the functioning of the disease-fighting immune system. The immune cells known as B cells, Minden noted, change shape as they prepare to produce antibodies against a particular invader.
The study also was important, Lehmann said, because of its methodology. Minden used a tool he developed, called difference gel electrophoresis, that allows two or three protein samples to be directly compared. Though the samples might contain thousands of different proteins in various concentrations, the test quickly identifies only those proteins that differ between the samples.
Astronomers must take this same needle-in-a-haystack approach when looking for exploding stars, called supernovae. By observing the same area of sky at two different times, they can compare the two and eliminate all of the stars that remain unchanged; what remain are objects that might be supernovae and can be studied in more detail.
Minden accomplishes a similar feat for proteins by adapting a two-dimensional electrophoresis gel, a standard laboratory tool for separating protein samples based on electrical charge and molecular weight. The resulting patterns left in the gels, which are about as big as a letter-size sheet of paper, look something like spattered paint.
In difference gel electrophoresis, two protein samples are used, each labeled with a different color of fluorescent dye, typically red and blue. In the resulting gel, only areas where proteins in the samples differ in type or concentration will appear red or blue; those areas can be cut out of the gel for further analysis. Most areas of the gel, where the protein makeup of both samples is the same, will appear purple and can be ignored.
The technology was licensed and is marketed by Amersham Plc, a British maker of diagnostic and chemical agents that was recently acquired by General Electric.
The fruit fly gastrulation study, Lehmann said, "will set the stage for similar experiments to study other developmental processes, other organisms and even to apply this to the changes that occur when cells go from normal to malignant development."
Though the sequencing of the human genome and the genetic codes of a growing number of animals and plants has swamped researchers with information about thousands of genes, the potential data overload for protein researchers is even more overwhelming.
Not only are there perhaps 10 to 30 times more proteins than genes in any organism, but proteins "are a moving target," changing from cell to cell and from day to day, Minden said.
Making sense of this protein soup could potentially have big payoffs. Whereas genomic analysis might identify people who are susceptible to certain diseases, proteomic analysis could provide an early warning that disease processes are under way.
"Proteomics is more difficult," he said, "but it may be more informative in the long run."
First Published: April 19, 2004, 4:00 a.m.
