Scientists had long assumed that any genetic mutation that does not alter a protein sequence should have no impact on human health. But recent research has shown that such synonymous DNA changes can trigger disease in a number of ways. Alla Katsnelson talks to scientists and biotech companies who are speaking up about ‘silent’ mutations.
It all started with an expression problem. Michael Gottesman and his lab members at the US National Cancer Institute in Bethesda, Maryland were studying a membrane protein involved in drug metabolism called P-glycoprotein to understand why some people develop resistance to chemotherapy during cancer treatment. But when the scientists tried to express large quantities of the protein in bacterial cells, they hit a wall.
“It was a real mess,” Gottesman recalls. “We couldn’t do it.”
The genetic code is read in triplets called codons, 64 of them representing just 20 amino acids. That means there is more than one codon for each amino acid, and different organisms preferentially use certain codons to make translation faster. One standard trick for boosting the expression of human genes in other organisms is to swap around nucleotides to get the DNA triplets most often used by the host’s cellular machinery. But a colleague of Gottesman’s suggested a different tack: as proteins elongate, the translation process needs to slow down and speed up to achieve proper folding, and perhaps the distribution of frequent and rare codons might control that rhythm.
The idea got Gottesman thinking about a niggling problem. The gene that encodes P-glycoprotein, called multidrug resistance 1 (MDR1), has about 50 single nucleotide polymorphisms, a handful of which are located in the coding region but at a position where they don’t affect the protein’s amino acid sequence. One, for example, in exon 26 of this 209-kilobase-long gene switches an ATC codon to ATT, both of which encode the amino acid isoleucine. Scientists routinely assume that such ‘silent’ or ‘synonymous’ mutations don’t affect the protein’s function, but clinical data clearly showed that people carrying these mutations metabolize drugs differently. “We were trying to think of how it could be that these synonymous mutations caused these changes,” Gottesman says. Maybe, he thought, they were meddling with the rhythm, thereby changing the protein produced.
The researchers then expressed the ATT codon along with two other naturally occurring polymorphisms and saw that the expression levels of messenger RNA (mRNA) and protein remained the same, but the protein’s activity was altered. Just as Gottesman had hypothesized, the evidence pointed to a shape shift in the resultant P-glycoprotein, caused by altered timing of translation.
The findings, published online in Science in late 2006, weren’t the only report of this type of mutation at work. A paper published at the same time in the same journal reported that synonymous polymorphisms in a gene encoding a protein called catechol-O-methyltransferase, which modulates responsiveness to pain, affected the loops and turns that make up the structure of the gene’s corresponding mRNA, and, with it, the level of protein expression. The two studies were the first published examples of human genes in which naturally occurring mutations produced proteins with an unchanged amino acid sequence but clearly different functional effects on disease. And, in the five years since, the idea that such mutations can have dramatic and far-reaching effects is beginning to take off.
(Click here to continue reading.)