Watching genes at work12 February 2001
PHILIP BALL
Now that the human genome is mapped, it is time, says Eric Lander of the Whitehead Institute for Biomedical Research in Cambridge, Massachusetts, "to take global views of biological processes: simultaneous readouts of all components".
The technology to do this is DNA arrays or 'gene chips'. Barely a decade old, these devices consist of ranks of single-stranded DNA molecules, synthesized in the laboratory and attached at one end to a wafer of silicon or glass like so many hairs. Roughly, each strand represents a single gene.
The DNA in our cells is double-stranded: each strand bound to a complementary partner. The strands fit together like the two halves of a zip. In the same way, the short DNA strands on a gene chip bind extremely selectively to their complementary strands. These complementary DNA (cDNA) fragments can be reconstructed from a cell in which the corresponding gene has been activated (so that it is producing its respective protein molecule).
A DNA microarray shows which genes were active in the cell.
The cDNA molecules are tagged with fluorescent molecular labels. So the DNA strand on the gene chip 'lights up' when it binds to its partner, showing which genes were active in the cell at that instant. This allows researchers to monitor the expression of every gene in a cell's genome simultaneously -- even when the function of those genes is not known.
But DNA arrays tell only part of the story, and even that can be ambiguous. It can be hard, for example, to obtain repeatable results from these experiments. More fundamentally, the arrays monitor the cell's production of RNA, the molecules that provide the template for protein synthesis ('expression'). Unfortunately there is not necessarily a direct relationship between RNA expression and protein expression. Yet it is the amount of proteins that really counts -- these are the molecules that do all the work.
And some genes encode more than one kind of protein, as proteins are generally modified chemically after they have been expressed from RNA. There may be as many as three times more kinds of protein than gene in human cells.
The time is right to launch a Human Proteome Project
With this in mind, some researchers are starting to develop arrays for directly measuring the amounts of different proteins in a cell. These protein arrays will be a basic tool for the science of 'proteomics', which focuses on proteins rather than genes. But proteins are harder to track. One approach is to manufacture tailored antibodies, which recognize and stick to each protein in the cell, and then tether the antibodies to a chip.
Proteomics is the next frontier for data-gathering biologists, says Ian Humphery-Smith, a biochemist at the University of Utrecht. "The time is right to launch a Human Proteome Project."
