Of SNPs and Smells

The language of a DNA sequence is more than meets the eye. A string of A, C, T, and G can indeed encode a string of amino acids. At a higher level, however, nuances of sequence may impart evolutionary information, because variants accumulate over time. Single nucleotide polymorphisms--SNPs--for example, are scattered throughout the human genome on average about one every 500 to 1,000 DNA bases. But they are not distributed evenly, and their clustering or paucity may hold clues to the past. A team of researchers from the Crown Human Genome Center at the Weizmann Institute of Science in Rehovot, Israel, is analyzing SNP patterns to paint a portrait of the evolution of the human sense of smell. 1 This sense derives from an inch-square patch of epithelium high in the nasal cavity. Here, some 500 types of olfactory receptor proteins bind different combinations of odorant molecules. The bending sends signals to the brain's olfactory bulb, which passes them to the cerebral cortex, ultimately enabling us to distinguish a sizzling pizza from a smelly foot. Humans sniff the world with about 12 million cells bearing olfactory receptors on waving cilia; bloodhounds have 4 billion such cells. Yoav Gilad, left, and Doron Lancet The sense of smell was once crucial to human survival, and to an extent still is, despite reliance on such stand-ins as smoke detectors. "Some say we have lost our sense of smell. But our 500 intact, coding olfactory receptor genes are still enough to provide a good olfactory system. We can still avoid fire and smoke, look for mates, protect ourselves from food poisoning, and enjoy our food and drink. We do use our sense of smell," explains Doron Lancet, director of the Crown Human Genome Center at the Weizmann Institute. And like most inherited traits, sharpness of the sense of smell varies. "We can test for certain odorants, and individuals have highly disparate thresholds. No stimulation from an odorant field at all is called specific anosmia," Lancet adds. Olfactory receptor (OR) genes form clusters of intact, intronless coding genes, as well as pseudogenes. As nonfunctional stretches of DNA whose sequences are similar to those of functional genes, pseudogenes are thought to be akin to ghosts of genes past--they are well-studied in the beta globin gene cluster. The 500 human olfactory receptor genes are shadowed by about the same number of silent pseudogenes. And because pseudogenes impart no phenotype, variant bases such as SNPs would be expected to accumulate because they are not removed through natural selection. The researchers found evidence for this, and more--the coding genes are, at the same time, being preferentially retained. Dror Sharon Secrets in SNP Patterns Lancet's group has been investigating human olfaction for many years, and the pace has accelerated with the influx of human genome data. But it was the discovery of one unusual gene variant that spearheaded the current work. "Two years ago, Dror Sharon, who was a postdoc and is now at Harvard Medical School, made the seminal discovery, an unusual polymorphic variant that we call the 'ancient allele.' Sharon found that it had many SNP differences compared to the most common allele," relates Lancet. Next, graduate student Yoav Gilad picked up the trail. He hypothesized that the unusual ancient allele with many SNPs was actually a pseudogene, and possibly even represents a common ancestor between human and chimp. Gilad then extended the analysis by comparing SNP sequences and frequencies among seven intact coding OR genes, five pseudogenes, and seven nearby introns in a cluster on chromosome 17, as well as four other well-studied genes as controls. And by several measures, the researchers found that the pseudogenes and introns not only had more SNPs than the OR genes, but they had many more haplotypes (combinations of SNP sites). Put another way, the data revealed that the OR genes are remarkably like one another, compared to the differences among the other types of sequences. For example, a measurement called the "average nucleotide diversity" is 0.11 percent for pseudogenes, 0.10 percent for introns, yet only 0.03 percent for the intact OR genes. "We found that 'Gilad's rule' held--most of the pseudogenes have many SNPs and several alleles, while the coding regions of genes had the typical behavior of about one SNP per 1,000 bases, not more. The pseudogenes, in contrast, had two to seven per 1,000, with a bias toward all SNPs looking like a common ancestor. It was an interesting irregularity," recalls Lancet. So they brought in population geneticist Michael Nachman of the University of Arizona, Tucson, and Karl Skorecki of the Rambam Medical Center and Rappaport Faculty of Medicine at Technion-Israel Institute of Technology. "We conducted many statistical tests and further experiments. And when the white smoke rose, it indicated that it wasn't negative or purifying selection, but a weak but discernible positive selection. There aren't too many cases of that known for humans," Lancet adds. The Bigger Picture DNA sequences are the stuff of the 21st and 20th centuries, but it was the work of a 19th century mind that provides the perspective for this new glimpse into the workings of the olfactory genes. Although Charles Darwin knew nothing of DNA sequences, his principle of natural selection underlies the ebb and flow of changes within genomes. He wrote in his famed On the Origin of Species in 1859, " ... I have called this principle, by which each slight variation, if useful, is preserved, by the term Natural Selection ...."2 More specifically, whether a change is perpetuated depends on whether it confers on its bearer a reproductive advantage. Natural selection is usually thought of as a negative process, a weeding out of genetic variants that counter reproductive success. But selection can be positive too, and this is what the team realized could explain what they were seeing in the SNP patterns of the OR genes compared to the pseudogenes. Sharon explains the difference. "In a negative selection process, new mutations might have a disadvantage, and thus will be eliminated from the population. In a positive selection process, new mutations might have an advantage, and thus will be fixed relatively fast in the population." Mark Seielstad, a research associate at the Harvard School of Public Health who wrote an article accompanying the Weizmann research paper, 3 points out that it isn't only the number of variant DNA nucleotides, but their natures that are intriguing in this multigene family. "The strongest evidence for something unusual is that in the OR genes, the number of mutations [that] change the amino acid sequence is elevated relative to the number of synonymous changes, which do not change the amino acid sequences. Negative selection acting to eliminate deleterious variation would tend to get rid of the amino acid altering mutations, because these are the ones that are likely to result in a protein with reduced function, while allowing some synonymous variation to enter the population, because these have little or no effect on the function of the protein." Lancet thinks that the positive selection "signature" revealed in the SNP patterns can be explained by evolution acting on entire multigene families, rather than on individual genes. Olfaction may result from a complex interplay and redundancy in gene function. "Each odorant molecule can be detected by more than one olfactory receptor. Each gene is dispensable. We call this the 'shifting terrain scenario,'" he says. DNA sequences that encode amino acids that have no effect on the function of on olfactory receptor can stay, and advantageous mutations become fixed in the genome. Concludes Lancet, "Positive selection often acts on genes that are not essential. If a gene duplicates, the duplicate can undergo very strong positive selection. If there is a repertoire of genes, some may be dispensable, and the genome can continue to explore new terrain." Ricki Lewis (rickilewis@nasw.org) is a contributing editor for The Scientist. References 1. Y. Gilad et al., "Dichotomy of single-nucleotide polymorphism haplotypes in olfactory receptor genes and pseudogenes," Nature Genetics, 26:221-4, October 2000. 2. C. Darwin. On the Origin of Species, New York and Toronto, A Mentor Book, p. 74, 1859. 3. M. Seielstad, "Whiffs of selection," Nature Genetics, 26:131-2, October 2000.

Of SNPs and Smells

Become a Member of

Keywords

Meet the Author

Ricki Lewis

Published In

October 2000

Related Research Resources

A Nanoparticle Delivery System for Gene Therapy

Research Resources

Podcasts

Webinars

Videos

Infographics

eBooks

Enhancing Therapeutic Antibody Discovery with Cross-Platform Workflows

Considerations for Cell-Based Assays in Immuno-Oncology Research

From Water Bears to Grizzly Bears: Unusual Animal Models

Sex Differences in Neurological Research

Products

Product News

Scaling Lentiviral Vector Manufacturing for Optimal Productivity

Generating High-Quality mRNA for In Vivo Delivery with lipid nanoparticles

Tecan introduces Veya: bringing digital, scalable automation to labs worldwide

A Visual Guide to Lentiviral Gene Delivery