"Silent" mutations are not always silent

A mutation in a human gene that does not change the resulting amino acid can nevertheless change a protein's function, according to an online report from Science. The research marks the first time that the phenomenon has been confirmed in mammals.

"The habit we all have of disregarding nucleotide changes that don't change protein sequence may not be a good one," coauthor Michael Gottesman at the National Cancer Institute in Bethesda, Md., told The Scientist. "This may be a generalizable phenomenon that may lead to changes in function we haven't been thinking about."

Gottesman and his colleagues investigated MDR1, which encodes P-gp, a human membrane transport protein that helps cells pump out anticancer and other drugs. They focused on the C3435T, a single nucleotide polymorphism (SNP) that is synonymous or silent, encoding for the same amino acid as the gene's wild-type version.

The researchers first used viral vectors against cancerous HeLa cells to express wild-type P-gp and a number of versions bearing single polymorphisms at C1236T, G2677T or C3435T. They also experimented with specific haplotypes, or sets of polymorphisms, including combinations of the three single polymorphisms studied.

The team next examined how well these cells transported any of a variety of fluorescent molecules and tested how well P-gp inhibitors such as CsA, verapamil and rapamycin affected the protein's function.

Gottesman and his colleagues found that wild-type P-gps and variants with just one polymorphism apparently functioned the same. However, the haplotypes with C3435T proved more resistant against inhibition from CsA and verapamil, though not rapamycin, when they also included one or two other polymorphisms. The same experiments were conducted with epithelial cells of African green monkey kidney origin, monkey kidney cells and human T cell lines, yielding similar results.

"Observations regarding this polymorphism have been very confusing. Some people report it changes the transport of certain drugs, and other people find it has not much effect," Gottesman explained. "The fact that we need more than a single polymorphism to see this effect could explain some of the confusion."

C3435T-bearing haplotypes apparently led to differences in how P-gp folded. Conformation-sensitive antibody UIC2 bound significantly differently to haplotype P-gp than to wild-type, and haplotype P-gp was more susceptible to digestion by the enzyme trypsin.

The idea that synonymous mutations might lead to differently folded proteins was proposed by Ian Purvis at the University of Glasgow and his colleagues and, independently, by Anton Komar, now at Cleveland State University in Ohio.

"Many cases of silent SNPs and their possible link to diseases should be reexamined," Komar, who did not participate in this study, told The Scientist. "Also, one should be quite careful in constructing artificial genes for the purposes of gene therapy, for example, and pay careful attention to the choice of synonymous codons."

Gottesman and his colleagues speculate that synonymous mutations represent rare codons for which translation machinery is not optimized. The resulting interruption of the rate at which mRNAs are translated could affect how a protein is folded, they said. Recent experiments in prokaryotes suggest codon usage is not random.

Gottesman noted this idea was inspired in part by conversations with Randall Kincaid at Veritas Labs in Rockville, Md., who noted that replacing rare codons in malarial genes with rare bacterial codons could lead to improved expression of those genes in bacteria by better matching bacterial translation rates.

Gottesman readily acknowledged that evidence for the idea has not yet been produced. Still, "it's an interesting proposition," Wolfgang Sadee at Ohio State University in Columbus, who was not a coauthor, told The Scientist.

Sadee suggested experiments that measure translation rates in vitro might help. He added that recent work of his own suggests that C3435T might affect mRNA folding, which in turn might affect mRNA translation rate and subsequent protein folding.

Charles Q. Choi
cchoi@the-scientist.com

Links within this article:

C. Kimchi-Sarfaty et al. "A 'Silent' Polymorphism in the MDR1 Gene Changes Substrate Specificity." Science, published online ahead of print Dec. 21, 2006.
http://www.sciencemag.org

Michael Gottesman
http://ccr.cancer.gov/staff/staff.asp?profileid=5713

R. Lewis. "Race and the Clinic: Good Science?" The Scientist, Feb. 18, 2002.
http://www.the-scientist.com/article/display/12869

D. Secko. "Phase I of HapMap Complete." The Scientist, Oct. 26, 2005.
http://www.the-scientist.com/article/display/22811

I.J. Purvis et al. "The efficiency of folding of some proteins is increased by controlled rates of translation in vivo. A hypothesis," J. Mol. Biol. 193: 413-7. January 20, 1987.
http://www.the-scientist.com/pubmed/3298659

I.A. Krasheninnikov, A.A. Komar, I.A. Adzhubei. "Role of the rare codon clusters in defining the boundaries of polypeptide chain regions with identical secondary structures in the process of co-translational folding of proteins," Dokl. Akad. Nauk. SSSR 303: 995-9, 1988.
http://www.the-scientist.com/pubmed/3250842

Anton Komar
http://web.bges.csuohio.edu/faculty/komar.htm

F. Supek, K. Vlahovicek. "Comparison of codon usage measures and their applicability in prediction of microbial gene expressivity," BMC Bioinformatics 6: 182, July 2005.
http://www.the-scientist.com/pubmed/16029499

Wolfgang Sadee
http://medicine.osu.edu/pharmacology/1134.cfm

D. Wang et al. "Multidrug resistance polypeptide 1 (MDR1, ABCB1) variant 3435C>T affects mRNA stability," Pharmacogenetics and Genomics 15: 693-704. October 2005.
http://www.the-scientist.com/pubmed/16141795