Science & Technology Story
Biologists find that a gene's location affects evolutionary change
Posted October 18, 2010; 02:35 p.m.
A gene's location on a chromosome plays a significant role in shaping how an organism's traits vary and evolve, according to findings by scientists at Princeton University and New York University.
The research, which appears in the latest issue of the journal Science, suggests that evolution is less a function of what a physical trait is and more a result of where the genes that affect that trait reside in the genome.
"We usually think of natural selection as acting directly on the genes that influence a specific trait," said Leonid Kruglyak, one of the co-authors on the paper and the William R. Harman '63 and Mary-Love Harman Professor in Genomics at Princeton. "This paper shows that there is an indirect evolutionary force that can override this direct effect -- selection acting on neighboring genes that just happen to travel together on a chromosome with the genes that influence the trait."
Physical traits found in nature, such as height or eye color, vary genetically among individuals. While these traits may differ significantly across a population, only a few processes can explain what causes this variation -- namely, mutation, natural selection and chance.
In the study, the researchers sought to understand, in greater detail, why traits differ in their amount of variation. But they also wanted to determine the parts of the genome that vary and how this affects expression of these physical traits.
To do this, the researchers analyzed the genome of the worm C. elegans, the first animal species whose genome was completely sequenced. It is viewed as a model organism for studying genetics. In their analysis, the researchers measured approximately 16,000 traits in C. elegans. The traits were measures of how much each gene was being activated in the worms' cells.
The researchers began by asking if some traits were more likely than others to be susceptible to mutation, with some physical features more likely than others to vary. Different levels of mutation indeed explained some of the results. The findings also revealed significant differences in the range of variation due to natural selection. Those traits that are vital to the health of the organism, such as the activity of genes required for the embryo to develop, were much less likely to vary than were those of less significance to its survival, such as the activity of genes required to smell specific odors.
The results, however, left most of the pattern of variation in physical traits unexplained -- some important factor was missing, the researchers said.
To search for the missing explanation, the researchers considered the makeup of C. elegans' chromosomes -- specifically, where along its chromosomes its various genes resided.
Chromosomes hold thousands of genes, with some situated in the middle of their linear structure and others at either end. In their analysis, the researchers found that genes located in the middle of a chromosome were less likely to contribute to genetic variation of traits than were genes found at the ends.
"As we wrote in the paper, genes that influence a trait can vary less due to their own effects than due to the genomic company they keep," said Kruglyak, who also is a professor of ecology and evolutionary biology and the Lewis-Sigler Institute for Integrative Genomics at Princeton.
The biologists considered why location was a factor in the variation of physical traits. Using a mathematical model, the researchers were able to show that genes located near large amounts of other genes are evolutionarily tied to their genomic neighbors. Specifically, natural selection, in which variation among vital genes is eliminated, also removes the differences in neighboring genes, regardless of their significance. In C. elegans, genes in the centers of chromosomes are tied to more neighbors than are genes near the ends of the chromosomes.
In addition to Kruglyak, the research was conducted by Matthew Rockman, an assistant professor at New York University's Department of Biology and Center for Genomics and Systems Biology, and Sonja Skrovanek, formerly a research associate at Princeton's Lewis-Sigler Institute and the Howard Hughes Medical Institute.
The study was supported by grants from the National Institutes of Health.