For the better part of this century, microbiologists have largely ignored evolutionary relationships among bacteria. But a revolution has occurred in microbiology with the advent of nucleic sequencing: Today, new phylogenetic relationships can be determined in far more detail and depth than was ever thought possible.
Carl R. Woese, 62, of the department of microbiology at the University of Illinois in Urbana, is widely considered the leader of this revolution. His 1987 review, "Bacterial evolution," published in Microbiological Reviews (51:221-71, 1987), comprehensively summarizes work done over the previous 10 years on the new phylogenetic categorization in microbiology. With more than 300 citations already (as tabulated from the Institute for Scientific Information's Science Citation Index), this review stands as the most cited microbiology paper of those published in the last three years (see accompanying chart).
Woese's paper discusses past failed attempts at developing a bacterial phylogeny. Stymied by technical difficulties, early micro- biologists gave up and discounted the area as unnecessary to the advancement of their field. But Woese argues in his review that without credible evolutionary guidelines, the theory that divides all organisms into either eukaryotes or prokaryotes rests on a shaky foundation. In his paper, he outlines the eubacterial and archaebacterial phylogenies that have been defined using sequencing data.
The importance of the new bacterial phylogeny goes far beyond a rewritten taxonomy, however. A detailed, accurate phylogenetic tree will help scientists understand bacteria not only in terms of their morphologies and biochemistries but also according to their relationships to other bacteria. This understanding will have a profound effect on the design and interpretation of experiments. Furthermore, sequencing information has extended the scope of evolutionary knowledge nearly tenfold. "It shows the evolutionist an intimacy between the evolution of the planet and the life forms thereon that he has never before experienced," says Woese.
Trained as a biophysicist, Woese characterizes his switch from biophysics to evolutionary problems as "an evolutionary process." His early studies of translation and the nature of ribosomal RNA eventually focused on the genetic code as an evolutionary problem. He began using nucleic sequencing technology, developed in the early 1960s, to determine ribosomal RNA sequences. Scientists consider these molecules to be the best historical chronometers because they occur in all organisms. Also, the different positions in their sequences change at different rates, making it possible to use these molecules to measure both close and the most distant phylogenetic relationships.
In the early days, reverse transcrip-tase had not yet been discovered, so it was not possible to determine complete ribosomal RNA sequences. By comparing ribosomal RNA fragments, however, researchers were able to identify phylogenetic groupings at many levels.
In 1977, Woese used this method, called oligonucleotide cataloging, to determine that the extreme halophiles, methanogens, and extreme thermophiles, now called the archaebacteria, are a distinct kingdom and not as closely related to the eubacteria (such as cyanobacteria or purple bacteria) as was previously thought (C.R. Woese et al., PNAS, 74:5088-90, 1977).
"Woese had to overcome a lot of initial resistance in getting the science community to accept that it is possible to establish phylogenetic trees for bacteria," microbiologist Roy A. Jensen of the University of Florida in Gainesville tells The Scientist. "Many scientists didn't trust the conclusions drawn from the sequence data."
Today most microbiologists have accepted nucleic sequencing as a valid tool to determine bacterial evolution. There is still a debate, however, over the best way to analyze sequence data to produce phylogenetic relationships. The two main and conflicting methods - those of Woese and of James A. Lake of the Molecular Biology Institute of the University of California at Los Angeles - were compared last year (see M. Gouy and W.-H. Li, Nature, 339:145-7, 1989), and evidence was found, by these reviewers, in support of Woese's arguments.
Still, the discrepancies produced by both Woese's and Lake's methodologies have yet to be reconciled. But until a verifiable analytic method is defined, Woese argues that "phylogenies derived from sequence analysis have to be accepted for what they minimally are: hypotheses, to be tested and either strengthened or rejected on the basis of other kinds of data."
Meanwhile, scientists are using the newly developed bacterial phylogenetics as a springboard for further experiments. For example, Jensen is studying the evolutionary biochemical pathways in organisms. When phylogenetic divergence of bacteria occurs relatively close in time, ascertaining the exact order of the tiny branching in a tree is difficult with nucleic sequencing. But once a credible phylogenetic tree is available, scientists can track a biochemical feature, such as gene fusions that produce multifunctional proteins, through history to fine-tune the branching order (Biosciences, 38:99-103, 1988). Scientists in other fields also use this methodology. For example, Norman Pace, of the Institute of Molecular and Cellular Biology at Indiana University in Bloomington, uses the new branching orders to study how related organisms interact on an ecological level.
Woese tells The Scientist that he is now concentrating on fleshing out the entire bacterial tree and identifying the branching subgroups. Ultimately, he hopes to unravel the identity of the universal ancestor, from which all organisms have descended. Before this will happen, he says, the entire bacterial genome must be identified. "Understanding the bacterial genome is necessary to open many doors throughout the entire field of biology and evolutionary study," says Woese. "Unfortunately, the thrust today is unraveling genomes in the medical arena, but this is like putting the cart before the horse. People who think that the medical genome project will have more impact are wasting their money."
Lisa Holland is a freelance science writer based in New York City.