Since Theodor Schwann first peered inside animal cells in the 1830s, curious biologists have sought to identify cellular components and comprehend how they work together. But until recently, light microscopes could not distinguish objects much smaller than a mitochondrion—an organelle about one-fiftieth the diameter of a typical cell. Electron microscopes have long helped scientists like Grigorieff make out smaller objects, such as molecular machines and small organelles—but only in dead, chemically fixed cells. To better understand such objects, biologists tried to obtain sharper images of them and see them in action. But until the last decade, they had little luck.

Now that's changing. "Every 20 years or so there's a big technical advance that really changes the way biomedical research is done," says Gerald M. Rubin, vice president of HHMI and director of Janelia Farm. In the late 1950s, x-ray crystallography allowed biologists to see the atomic structure of proteins, which carry out most of the work of the cell; in the 1970s, cloning and DNA sequencing led to new insights into evolution, gene regulation, and the biochemical workings of individual proteins.

At Janelia Farm, HHMI has focused a good deal of its efforts on developing new microscopy methods and new computing methods to analyze images. Nationally, the organization is investing tens of millions of dollars each year in microscopy. That's because imaging, Rubin says, "is the most important technology that we need now."

The human eye can readily distinguish objects as small as 100 micrometers across, about the width of a human hair. A typical human cell is about 10 micrometers in diameter and therefore invisible to the naked eye. As of the early 1990s, even state-of-the-art light microscopes could distinguish only objects larger than 0.2 micrometer in diameter—half the wavelength of blue light—which meant that smaller but important structures like ribosomes and spliceosomes were impossible to see in living cells. Biologists' vision was restricted by the light microscope's resolution—its ability to create a sharp image by distinguishing between two adjacent points. Optics dogma dictated that resolution was limited to about half the shortest wavelength of light used, so many scientists thought it could get no better.

At Janelia Farm, Nikolaus Grigorieff says he'll try to push single-particle cryoelectron microscopy methods to "routinely get to such high resolution that you can build an atomic model."

Cellular structures were also difficult to view because they're usually transparent, making it hard to distinguish them from the watery cytoplasm in which they sit. Phase-contrast microscopes made that easier by using interfering light waves to distinguish cellular structures from background. And biologists developed a plethora of chemical stains and fluorescent antibodies that bound to specific cellular structures, making them visible. But cells usually had to be killed first, and scientists then had to surmise what the structures did when the cells were alive.

To see molecular machines and small organelles, scientists used transmission electron microscopes, which utilize electromagnetic coils to focus electron beams instead of glass lenses to focus light. But electron beams destroy biological tissue, so researchers could only see into dead, chemically fixed cells. And electron microscopists like Grigorieff who wanted to determine the atomic structures of protein complexes could not do it. So biologists filled textbooks with descriptions of what they could see, saying little about what they couldn't.

	PAGE 2 OF 6	Go Back | Continue