Image: M. YUSUPOV et al.; copyright Science

PROTEIN FACTORIES in all cells, known as ribosomes, are revealing their secrets. New detailed images are showing how their two subunits (blue and gray) interact with each other and with tRNAs (red, orange, yellow).

Scientists in the genome race did not cross a finish line draped with the usual 10-foot-long yellow ribbon but rather one marked by a string of genetic code stretching on for miles. From there researchers have started on a new course, hurrying to learn how this expanse of DNA instructs our cells to assemble proteins. Central to the process are ribosomes. These tiny organelles crank out thousands of proteins needed to sustain life with remarkable efficiency. And two papers in Friday's issue of Science offer a wealth of new information about how they workand how antibiotics can foul their performance.

"As biologists we are fascinated by these results because of their fundamental importance in understanding how the genetic code gets translated into proteins," says Venki Ramakrishnan, head of the U.K. Medical Research Council Laboratory of Molecular Biology and lead author of one of the papers. "However, pharmaceutical and biotech companies are keenly interested because this research not only helps us to understand how many known antibiotics work but also helps us to understand the basis of certain kinds of resistance. This will hopefully allow us to design new antibiotics in the future that can overcome the growing worldwide problem of resistance."

The new results are the latest in a fast-paced series of advances biologists have made in creating ever-sharper images of the ribosome and its two main parts: a large subunit called 50S and a smaller subunit, 30S. These two lumpswhich together measure some 200 angstroms wideare linked by a series of bridges. Between them lie three distinct docking stations of sorts. Transfer RNA (tRNA), carrying the building blocks of proteins, or amino acids, move from one docking station to the next, as if on a conveyor belt. In the process they are compared with genetic instructions from the cell nucleus, or messenger RNA (mRNA). When the two match, the tRNA drops its load and the ribosome catalyzes a reaction fusing the amino acid onto a growing protein chain (see sidebar)

In 1999, Harry Noller of the University of California at Santa Cruz and his colleagues used x-ray crystallography in a novel way to produce pictures of the entire ribosome at a resolution of 7.8 angstroms. Then in August of last year, Peter Moore, Thomas Steitz and their colleagues at Yale University published an account of the ribosome's 50S subunit in even greater detail. Another group from the Max Planck Institute and the Weizman Institute, headed by Ada Yonath, has worked on the 30S subunit. Now Ramakrishnan's team reports on the structure of the 30S subunit bound to pieces of mRNA and tRNA resolved to 3.1 to 3.3 angstroms in the presence and absence of antibiotic. And Noller's team has come back with images of the entire ribosomealso bound to both tRNA and mRNAat a resolution of 5.5 angstroms.

Ramakrishnan's work reveals for the first time the exact four parts the ribosome uses to make sure that the three base pairs on a tRNA (an anticodon) match up properly with three on an mRNA (a codon). This codon-anticodon link is crucial to guarantee that the proper protein gets made. The team found that two different residuesA1492 and A1493 change shape and probe the geometry of codon-anticodon interaction to check that the first two base pairs are matched. The third pairing, they observed, is afforded more freedom, in support of Francis Crick's wobble hypothesis. This flexibility helps explain how a single anticodon can bind to more than one codon.

To examine the effects of antibiotics, the scientists worked with ribosomes from the bacteria Thermus thermophilus. Bacterial and human ribosomes are sufficiently different that antibiotics disrupt the actions of the former but not the latter. "Although these antibiotics were discovered several decades ago," Ramakrishnan says, "we haven't understood in detail how they work." The scientists found that the antibiotic paromomycin induced some of the same structural changes in A1492 and A1493 as did matching mRNA and tRNA base pairs. Thus, they conclude that the antibiotic makes it easier for the bacterial ribosomes to accept mismatched codon-anticodon pairs, leading to the production of many incorrect proteins and the bacteria's death.

Noller's team uncovered just as many important insights. They, too, produced images of T. thermophilus ribosomes, which are slightly smaller than those in higher organisms and so easier to resolve using x-ray crystallography. Husband and wife Marat Yusupov and Gulnara Yusupova worked on creating near-perfect crystals for generating sharper images. The group, which also included Jamie Cate of the Whitehead Institute, found that higher resolutions were attainable when tRNA and mRNA were bound to the ribosomeperhaps because they helped to stabilize it.

In fact, much of what the researchers found provides evidence of just how dynamic the ribosome actually is. For one thing, they were able to see in clear detail the great distances tRNAs must travelover 20 to 50 angstromsalong the ribosome's conveyor belt. They noted that a prominent kink in tRNA reaches out to ribosomal proteins in each of the three docking stations. And some ribosomal proteins also extend thin tails down into the docking stations to hold on to the tRNA backbone. Their analysis further revealed that the ribosome's two subunits must themselves move in order for the tRNA to complete its journey.

How the two subunits move in relation to one another remains to be seen, but it undoubtedly involves the numerous molecular bridges that stitch the two parts together and come in contact with the tRNA. In Noller's last 7.8-angstrom image, they were able to see 10 of these connections. The new, sharper picture shows another two. Scientists speculate that these bridges may also help the two subunits communicate, relaying the status of protein production.

When Noller's last work was published in 1999, he commented that "what we have at present are a few snapshots, and ultimately what we would like is a movie of the ribosome in action." Now scientists have captured a few vital scenes. "Both the Noller and Ramakrishnan manuscripts are landmark contributions to the understanding of ribosome structure and function," Albert E. Dahlberg of Brown University writes in an accompanying article in the same issue of Science. "Our appetites have been whetted, and we now look forward to seeing crystal structures representative of different conformations of the dynamic ribosome." Frame by frame, the film will be completed.