THE INTEGRITY PAPERS

Genre Group

ceptualinstitute.com

  Excerpted from

Scientists Meet at Rockefeller to Discuss Molecular Strategies in Biological Evolution

Author: Ricki Lewis
Date: July 20, 1998

In science, things often aren't as simple as they seem. This is certainly the case for the genetic code. Even as elegant experiments in the 1960s assigned DNA and messenger RNA (mRNA) base triplets to specific amino acids, researchers were wondering if a protein's blueprints were the sole meaning imparted by those long strings of A, T, G and C. But back then, they could do little more than wonder.

Today, with more than a dozen genomes sequenced, researchers can ask age-old questions as well as pose new ones about the meanings of DNA sequence organization, and begin to tease answers from data (R. Lewis, The Scientist, 12[1]:11, Jan. 5, 1998). That's why several dozen biochemists, geneticists, molecular biologists, and evolutionary biologists convened at Rockefeller University in New York City, June 27 to 29, for "Molecular Strategies in Biological Evolution," a conference sponsored by the New York Academy of Sciences. Talks revisited ideas of DNA's multiple roles, and explored how natural selection has shaped the ability of genomes to change, increasing variation.

"A genome contains a lot more information than we may have anticipated," said Lynn Caporale, a New York biotechnology consultant, who, with Werner Arber, a 1978 Nobel Prize winner, from the University of Basel, Switzerland, chaired the meeting. In 1984, long after the mechanics of protein synthesis had been worked out and geneticists had had a chance to ponder the subtleties of the genetic code, Caporale, then a professor of biochemistry at Georgetown University Medical Center, summarized the concept of a higher-order genetic code (L.H. Caporale, Molecular and Cellular Biochemistry, 64:5-13, 1984). Now that the Human Genome Project shows that probably less than 5 percent of our 3 billion DNA base pairs encode protein, Caporale's landmark paper is even more compelling. And the topics visited at the meeting fit in with functional genomics, the emerging field that is demonstrating that a genome is more than the sum of its parts.

Selecting Strategies for Adaptation

In the opening talk, "Chance Favors the Prepared Genome," Caporale painted a portrait of functionally flexible genomes able to combine pre-existing parts in novel ways. The enzymes that copy, move, and repair DNA are the key players in these shufflings. "Genomes evolve a balance between fidelity of replication, and exploration of new opportunities. Natural selection favors the evolution of strategies that increase the rate of adaptation," she said.

In the classic Darwinian sense, natural selection occurs because organisms with inherited traits that are beneficial in a particular environment preferentially survive long enough to reproduce, thereby gradually altering the genetic composition of a population. The talks emphasized that the very ability to generate variation is a selectable trait. And viewed under this evolutionary umbrella, genetic discoveries that seemed out of place as the details of gene function were worked out in the 1960s and 1970s now make sense.

"Evidence from genetic and biochemical studies reveals molecular mechanisms for a number of activities that would have been considered taboo mid-century," said James Shapiro, a professor of biochemistry and molecular biology at the University of Chicago. Examples include introns, jumping genes (transposable elements), and the overlapping genes of certain viruses. Intron-bearing genes can mix and match segments to encode new proteins; jumping genes cause mutations where they land and can rapidly introduce new DNA sequences, even shuttle them among species; and the rare overlapping genes derive maximal information from a single DNA sequence, perhaps the ultimate in molecular economy.

Speakers liberally evoked geographical metaphors, describing genomes as landscapes, terrain, scenery, and dynamic playgrounds. But the question-and-answer periods tempered any teleological tendencies--that is, the anti-Darwinian idea that evolution works with a purpose or goal, such as some ill-defined state of perfection. Use of the word "strategy," for example, literally implies forethought or intention. But the speakers and posters used the term to denote mechanisms that enhance the probability of survival to reproduce, the crux of natural selection. "Genomes don't 'know' what is right; they know efficient ways to explore," said Caporale.

Clues in Nonrandom Sequences

If the opposite of a goal-oriented genome is a totally random DNA sequence, then that description misses the mark, too. For genome sequences reveal distinct patterns, some unexpected, and many yet to be fathomed. Consider the matter of the third position of an mRNA triplet or codon. The third base is called "wobble" because it can vary, yet specify the same amino acid. For example, CGU, CGC, CGA, and CGG all encode arginine. If the wobble base occurs at random, as one might expect given that it makes no apparent difference in specifying the amino acid, then synonymous codons should be present in proteins in about equal frequency. But that isn't the case.

In Borrelia burgdorferi, the bacterium that causes Lyme disease, for example, a surface protein includes a five-amino-acid segment that is invariant at the protein and mRNA levels. "Many nucleotide sequences can encode the pentapeptide, but we only get a specific repeat. There must be some reason why," said Caporale. Added Giorgio Bernardi, of the Laboratory of Molecular Genetics at the Jacques Monod Institute in Paris, "there is a strange conservation in the third position. The so-called neutral third position may also carry a message, maybe more important than specifying the amino acid."

Bernardi presented an interesting hypothesis for the significance of nonrandom wobble base usage. He looked at enrichment for GC in 300+-kilobase genome segments called isochores. Such regions include many protein-encoding genes, which often have GC-rich stretches called CpG islands. An isochore called H3, which contains the highest concentration of GC base pairs, is found on most human chromosomes. "It has a distinct chromatin structure, and has the highest

rates of transcription and recombination," Bernardi said. So he calculated the percentages of G or C for the first, second, and third codon positions in H3 of several cold and warm-blooded vertebrates, including humans. "The percentage of GC in the third codon position is very different between warm and cold-blooded vertebrates. Some parts that are lightly enriched for GC in cold-blooded vertebrates have more GC in warm-blooded vertebrates," he said. Bernardi suggested that the excess GC increases the thermostability of DNA and RNA--which would be advantageous to a warm-blooded animal, and therefore selectable.

Another nonrandom genome pattern, whose function is better understood than the no-wobbly wobble, is a very variable DNA sequence flanked by highly conserved regions. This organization promotes rapid change. It is perhaps best known in the mammalian immune system, where gene pieces move and join, forming antibodies that recognize particular pathogens. Variable sequences sandwiched between conserved sections have since been discovered in several types of moveable genetic elements. "The immune system has been waving in our faces for 20 years what goes on in the genome," said Caporale.

"Cassette" genes in bacteria that confer resistance to antibiotics are another variation on this constant-variable-constant theme. A cassette gene includes a coding region plus a sequence called a recombination site that enables it to recognize and swap into a specific sequence called an integron. "The recombination sites are similar overall and have an imperfect inverted repeat. Their sequence is not as important as the symmetry, and they confer mobility on the gene cassette," said Ruth Hall, a molecular biologist at CSIRO Molecular Science in Sydney, Australia. Understanding how integrons work could lead to development of new, broader methods of combating antibiotic resistance, suggested Caporale.

Searching for Meaning in Repeats

A recurring theme at the meeting was the varied functions of repetitive DNA Sequence repeats comprise the bulk of the human genome, and in the early 1980s some geneticists arrogantly dismissed repeats as "junk." "That was an ugly phrase. So-called 'junk' DNA includes signals for RNA processing and DNA replication. It controls centromere function, telomeres in Drosophila melanogaster, and chromosome pairing in meiosis," said Shapiro.

Evidence shows that DNA repeats do indeed have meaning. For example, two "sibling" species of the fruit fly, Drosophila melanogaster and Drosophila simulans, look alike down to the tiniest bristle. "Yet you can easily distinguish them by their sequences of repetitive DNA," said Shapiro. And in humans, "triplet repeat" disorders such as Huntington's disease and fragile X syndrome result from mutant genes that have dozens of copies of a particular DNA triplet tacked onto them. Their discovery revealed an entirely new class of mutation. Perhaps understanding the normal function of these repeated sequences will reveal how in excess they cause symptoms that usually involve the brain.

Lively discussion toward the end of the meeting focused on short repeats called microsatellites, which occur in many gene complexes. Edward Trifonov, a structural biologist at the Weizmann Institute of Science in Rehovot, Israel, spoke of how microsatellites could modulate the mutability of protein-encoding genes, functioning in a manner analogous to tuning knobs on a stringed instrument. Also present was associate professor of anatomy David King of Southern Illinois University at Carbondale, who had independently used the same musical metaphor of "evolutionary tuning knobs" in a publication (D. King et al., Endeavor, 21[1]:36-40, 1997) and in a poster at the meeting. The idea is that changes in repeat numbers can control gene expression, allowing many nuances of a phenotype to emerge. This is typical of so-called quantitative traits, such as height, which show great variability. "Genes incorporating simple sequence repeats may be evolutionarily 'adjustable,' with the repetitive sequence providing site-specific control over the rate at which quantitative variation is introduced into a population," King said.

DNA repeats, moveable genes, and hidden messages in the genetic code may all begin to tell more complete stories as genome sequencing continues. The outpouring of genome information offers a tool of unprecedented power to identify and scrutinize the clues to evolution held in DNA sequences. As a result, the circa-1960s view of a gene as solely the instructions to build a protein is rapidly changing. "We have to look at genomes at the end of the 20th century from a different viewpoint than prevailed in the mid-20th century," concluded Shapiro.