Issue Date: January 22, 2007
Silent No Longer
The more scientists study the genetic code, the more it reads like poetry. In a poem, every word, every line break, even every syllable can carry more than a literal meaning. So too can the molecular letters, syllables, and words of the genetic code carry more biologically relevant meanings than they appear to at first.
Now, a cadre of researchers is discovering intriguing depths of meaning in "synonyms" in the genetic code—very short wordlike sequences, or codons, that translate into exactly the same amino acids during the construction of a protein. Scientists are finding that synonymous codons influence the temporal pattern by which a messenger RNA (mRNA) molecule bearing genetic specifications from a cell's nucleus is translated by machinelike ribosomes into protein molecules.
These punctuations in the RNA-to-protein translation process have unexpected consequences: They can change the timing by which nascent proteins fold as they elongate and peel away from ribosomes. This means that two stretches of mRNA that differ only in synonymous codons can translate into two proteins that have identical amino acid sequences but different three-dimensional shapes. Such differences can convey important, even grave, biological and medical meanings. It's akin to the way the same hand can fold into an affirming thumbs-up gesture or into a shape involving the middle finger that conveys another sentiment altogether.
"We know that one individual given drug A will have to sleep for three days, but another taking the same drug will suffer no such effect," notes Michael M. Gottesman, chief of the Laboratory of Cell Biology at the National Cancer Institute (NCI) in Bethesda, Md. He now thinks that such individual differences in response to drug treatments and in susceptibility to diseases could correspond to different synonymous codons that lead to differently folded protein products. Most researchers have assumed that this type of genetic variation is too subtle to matter much. In fact, an often-used moniker for the variation is "silent polymorphism." Nonsilent polymorphisms are those variations in a gene's code that do lead to amino acid changes.
Last month, Gottesman and coworkers reported results of their investigation of a silent polymorphism that isn't so silent (Science, DOI: 10.1126/science.1135308). They found it in the gene that codes for P-glycoprotein (P-gp), a protein that takes residence in cell membranes, where it pumps drug molecules out of the cell. By purging the cell of drugs, this protein renders about half of human cancers resistant to a diversity of drugs.
Gottesman's group discovered that a silent polymorphism sometimes found in this gene gives rise to a version of P-gp that is less effective at expelling drugs from cells than the "wild type" of the protein. The researchers conjecture that the altered protein function derives from a synonymous codon's effects on the timing of translation and folding as the P-gp protein is being made and as it insinuates itself into a cell's membrane. In their studies, the researchers expressed the gene with and without the silent polymorphism in cultured human carcinoma cells, an AIDS-related human cell line, and two lines of cells derived from monkey kidney.
"The beauty of the paper is that it is based on natural examples," that is, living cells, comments Anton Komar of Cleveland State University. He was one of the first scientists to suggest, in the late 1980s, that silent polymorphisms in genes might have important biological consequences. Previously, Komar and others had found evidence that synonymous codons might affect protein folding, but those studies were done in cell-free test-tube preparations. "Nobody paid attention," Komar recalls. The consensus view, he points out, has long been that only those polymorphisms that translate into amino acid substitutions in the associated proteins were biologically or medically significant. To Komar, Gottesman's findings ought to change that view.
"Looking closely at silent polymorphisms could become a vast project now," Komar says. "We have the whole genome in hand."
Gottesman was attracted to research into silent polymorphisms three years ago during a discussion with Randall Kincaid, a former immunology lab head at the National Institutes of Health in Bethesda, who now runs Veritas, a biotech company in nearby Rockville. Kincaid mentioned a malaria vaccine project that required him to produce loads of a human protein in a microbial host. The protein, however, kept folding up and aggregating into unusable clumps. Kincaid told Gottesman that to circumvent this protein-folding headache, his team used a genetic engineering technique that involves exchanging some of the codons in the human gene with synonymous codons that are more prevalent in the microbial host used to manufacture the protein in bulk.
During that 2003 discussion, Gottesman says, "a light bulb went off in my head." For Gottesman, Kincaid's protein-folding headache sounded like a potential answer to a mystery he and his colleagues had been encountering in their research on P-gp. Listening to Kincaid, Gottesman wondered if the differences in folding that his team had observed stemmed from the silent polymorphisms found in the gene for P-gp.
Silent polymorphisms are among a more general class known as single-nucleotide polymorphisms, or SNPs (pronounced "snips"). SNPs consist of one nucleotide letter substituting for another. In the mRNA transcribed from a gene, every string of three nucleotides constitutes a codon that corresponds to and is ultimately translated into one of 20 amino acids.
For example, the mRNA codon designated UUU (uracil-uracil-uracil) encodes the amino acid phenylalanine, whereas the codon UUA (uracil-uracil-adenine) encodes leucine. Because a leucine replaces a phenylalanine, the polymorphism is nonsilent in this case, and the codons are nonsynonymous. On the other hand, the mRNA codons GGU, GGC, GGA, and GGG all encode glycine. That makes them synonymous codons, and their protein constructs all have the same amino acid sequence.
Gottesman's group traced one particular silent SNP in the gene for P-gp—in which a GGC codon changes into GGT—to altered protein activity. Both codons correspond to glycine. Using several analytical methods, the researchers concluded that the folding, final shape, and function of P-gp indeed are influenced by silent SNPs.
"These results may not only change our thinking about mechanisms of drug resistance, but may also cause us to reassess our whole understanding of SNPs in general and what role they play in disease," states NCI Director John E. Niederhuber in a press release.
Komar conjectures that synonymous codons might affect protein folding by tweaking the timing of that folding. In cells, he notes, the concentrations of amino acid-toting transfer RNA (tRNA) molecules, each of which corresponds to a specific mRNA codon, roughly mirror the overall frequencies at which the codons appear.
During protein translation, the mRNA codons sequentially specify which tRNA must come into the ribosome complex to deliver the next amino acid to be stitched onto the growing protein. A polymorphism that substitutes an infrequent codon for a relatively common but synonymous codon ought to result in a delay in translation because there is less of the corresponding amino acid-bearing tRNA around, Komar says. Because of the momentary pause, the growing protein could fold in a different way than if the pause were absent.
The details of the altered folding kinetics remain largely unknown, but recent work by Luda Diatchenko of the University of North Carolina and her colleagues has opened up one route of investigation into those matters (Science 2006, 314, 1930). Like Gottesman's group, they found that different synonymous codons in a gene can lead to changes in the production of its protein product. The gene Diatchenko's team studied encodes a neurotransmitter-degrading enzyme called human catechol-O-methyltransferase, or COMT. This enzyme is central to the regulation of pain perception. The COMT gene exists in three common variants, each one consisting of both silent and nonsilent codon changes.
Depending on which variant a person has, he or she is likely to have low, average, or high pain sensitivity. The researchers found that differences in COMT production derive far more from differences in synonymous codons in the COMT gene than in nonsynonymous ones that lead to amino acid changes.
Moreover, Diatchenko and her colleagues were able to relate those codon and clinical differences to the presence or absence of a specific stabilizing loop structure in the mRNA molecules encoding the enzyme. The mRNAs that were more stable yielded COMT activities up to 25 times higher than that associated with the least stable mRNA. The researchers surmise that these stability differences influence either the rate at which the mRNA molecules are degraded or at which they can be translated into protein. Because the more stable mRNAs produce more of the neurotransmitter-degrading enzyme, they ultimately correspond to less pain sensitivity.
"We need to give much more weight to synonymous changes," Diatchenko concludes. "Now that we know that the difference in COMT expression depends on the secondary structure of mRNA, we can think of targeting this mechanism" to alleviate such conditions as persistent pain, she says.
Confirming that the genetic code has built into it "colons or commas" that influence the kinetics of protein synthesis and folding, Komar notes, is a reminder that the code has yet to be fully decrypted. It's a molecular poem whose deconstruction must continue. The question now for Komar and others is whether they've identified a previously hidden stratum of meaning in the genetic code that will significantly help account for the differences that make individuals unique, in illness and in health.
- Chemical & Engineering News
- ISSN 0009-2347
- Copyright © American Chemical Society