Genetics: Gene Players May Tip Balance to Diabetes Collaboration: Harvard, MIT Announce Institute to Advance Genomic Medicine Imaging: Technique Tracks Tumor Escape into Lymph Nodes Drug Technology: Molecules from Novel Genetic Code Aimed at Drug Discovery Outreach: Four Directions Fetes 10th Anniversary Resistance to Food Poisoning Seems Factor in Worm Longevity Immune Cells Deal Death Blow to Damaged Neurons Molecular Timekeeper Keeps Up Speed Through Precision   Proceedings of the HMS Faculty Council Spiegelman Wins Bristol-Myers Squibb Award for Metabolic Research HMS and Harvard Pilgrim Launch Center for Child Health Care Zelen Leadership Award Presented HMS Faculty Member Named Kirsch Investigator HMS Appointments to Full Professorships In Memoriam: Francis Wolfort Howard Blume Honors and Advances   Resident Ridicules Nurse's Body Size   Group Gives Enabling Support   Front Page

DRUG TECHNOLOGY Molecules from Novel Genetic Code Aimed at Drug Discovery On the 50th anniversary of the discovery of DNA's structure, a team of researchers at Brigham and Women's Hospital and Columbia University has shown that the genetic code is not carved in stone. They have essentially rewritten the code, producing novel peptidelike substances that may provide a valuable technology for discovering new drugs and clarifying the process of genetic translation. Anthony Forster and Stephen Blacklow (l to r) modified the process of translation to create short peptide chains with unnatural amino acids incorporated at several sites. They hope to use this technology for finding new drugs and studying translation. (Photo by Jeff Cleary) In translation, tRNA adaptor molecules decode genetic messages into peptides and proteins. HMS and Columbia University researchers have managed to modify some steps in this process to produce peptide variants, dubbed peptidomimetics. The researchers were able to chemically produce their own tRNAs that could read a genetic code of their choosing. They then created modified amino acids, which they attached to their engineered tRNAs for incorporation into the peptidomimetics. Toppling Dogma "We proposed to change the central dogma of biology in the test tube," said Anthony Forster, HMS instructor in pathology and lead author of the study. Accepted by biologists everywhere, the dogma describes the genetic information path that passes from DNA to RNA to protein. "We reasoned that if we understood enough about translation, and we did the experiment in the right way, we ought to be able to show DNA makes RNA makes peptidomimetics." "We proposed to change the central dogma of biology in the test tube." --Anthony Forster The idea of changing the genetic code and putting unnatural amino acids into peptides is not new. For years researchers have been able to create proteins containing a single unnatural amino acid by decoding a single stop codon with a suppressor tRNA. But Forster and fellow researcher Stephen Blacklow, HMS associate professor of pathology, figured out a way to incorporate several unnatural amino acids sequentially in the growing peptide using any group of codons. "The technology was here in 1989, but you really had to have a reason to do it and be motivated to put it all together," said Forster. The reason Forster and Blacklow wanted to make these highly specific peptidomimetics was to provide a potential new route to drug discovery. They intend to do this by creating a library of many different mimetic sequences, which they can then unleash on potential targets. Peptide mimics that bind the desired target can be repeatedly recovered and amplified. Ultimately the "winners" predominate and can be identified by their associated gene sequence. The advantages of using peptidomimetics instead of natural peptides in looking for drugs are that they are more resistant to degradation by proteases and may be able to cross membranes. The variety of mimics that researchers can produce may be another benefit. "The utility is not in the yield of product, it's in the diversity of product," said Forster. And libraries that are vastly larger than those used in industry may produce molecules with higher affinities earlier on in the drug discovery process. From Scratch In order to make the peptidomimetics, the scientists, whose work appears in the May 27 Proceedings of the National Academy of Sciences, had to recreate the process of translation. To do this, they purified ribosomes and translation factors from E. coli. Then they synthesized tRNAs that recognize sense codons and attached unnatural amino acid derivatives synthesized by Virginia Cornish's lab at Columbia. In normal translation, proof-reading tRNA synthetases ensure that the appropriate amino acid is bound to its corresponding tRNA. The researchers removed these synthetases and other interfering factors from their translation system so only the amino acids they chose would be attached to their engineered tRNAs. The researchers also arbitrarily chose codons that they assigned to the unnatural amino acid-charged tRNAs for translation. "This is rewriting the genetic code," said Forster. By creating their own code, Forster and his colleagues could bypass nature's normal process of making proteins and generate their own proteinlike mimics that contain the unnatural amino acids of their choosing, ordered as they want them. "Optimizing the efficiency of the translation system proved more challenging than expected," said Blacklow. But the researchers succeeded in producing a highly defined system that is already helping to elucidate some of the remaining mysteries of translation. --Gaia Remerowski