As the photosynthetic factory of the plant cell, the chloroplast contains its own complement of genes distinct from the comparably sized mitochondrial genome in the energy center of the cell or the much larger genome in the cell nucleus.

"The chloroplast genome can be more informative in some ways than the complete nuclear genome, and easier to analyze than plant mitochondrial DNA," said Brent Mishler, professor of integrative biology at the University of California, Berkeley, and director of the Jepson and University Herbaria.

Mishler is one of nine principal investigators on a new project, supported by $3 million over five years from the National Science Foundation, to isolate and sequence chloroplast and mitochondrial genomes from 50 to 100 representative plants, drawing on the expertise of the U.S. Department of Energy's Joint Genome Institute (JGI) in Walnut Creek, Calif. The grant was among the largest of seven collaborative projects funded last month by NSF's "Assembling the Tree of Life" program.

The biologists will compare chloroplast genomes, as well as mitochondrial genomes and nuclear genes, along with morphological characteristics to determine plant relationships among the more ancient plant groups such as the mosses, algae and ferns. Their work will complement that of on-going projects looking at other branches of the green plant family tree, such as the well-studied seed plants.

"The whole nuclear genome is enormous and it's very difficult technically to get the same portions of a genome out from a lot of different organisms," said co-PI Jeffrey L. Boore, a scientist at the Lawrence Berkeley National Laboratory, head of the evolutionary genomics laboratory at JGI and an adjunct associate professor of integrative biology at UC Berkeley. "But with organelles, either mitochondria or chloroplasts, we can pull out this bit of DNA that is physically separate from the nuclear genome and get this collection of homologous genes. So we get a pretty good collection of genes for one price."

Both chloroplasts and mitochondria originated more than a billion years ago, when bacteria colonized early single-celled organisms, establishing a symbiotic relationship that has allowed plant cells to get energy from sunlight and both plant and animal cells to produce energy efficiently.

Among the questions Mishler, Boore and their colleagues want to answer are, how many times has land been colonized from the sea by green algae, where did plants acquire the adaptations essential to life on land, and how many times did multicellular plants evolve?

To date, only two entire plant genomes have been sequenced - a plant called Arabadopsis thalianafrom the mustard family and rice - and JGI is at work on a third, the poplar tree.

"Data from the already sequenced genomes have not yet been analyzed comparatively," said Mishler, who specializes in the study of mosses and other bryophytes. "Only about 15 green plant chloroplast genomes have been sequenced, and even fewer mitochondrial genomes - about 10 - so our project will be a big step forward."

The group plans to test various ways of comparing genomes to elicit evolutionary relationships. In particular, they want to find the best methods to use for groups of organisms that have a long evolutionary history.

"We're going to test theories and methods for analyzing genes comparatively," said Mishler, who was one of the leaders of the "Deep Green" initiative that several years ago reported the first draft of the tree of life for green plants. "Right now we don't know the best ways to analyze the DNA once we have it."

Typically, biologists look at the same gene in many different species and document the sequence changes that accumulate over time. Assuming a roughly constant rate of change as a result of random mutations, scientists can estimate the time since two lineages split from one another to evolve separately.

Boore said, however, that comparing DNA sequences directly may not be the best method, because the same mutation could show up more than once, throwing into doubt any conclusions about plants being from the same lineage. He has had great success looking at gene rearrangements within the mitochondrial genomes of animals - at genes that switch places, flip or duplicate.

"When gene order rearrangements define some specific evolutionary branching, we've judged that those are very, very powerful characters because they are very unlikely to rearrange in the same way in two different lineages," Boore said. "We feel that when we find gene rearrangements, we are confident that that part of the tree is well resolved."

Following mitochondrial gene rearrangements over time, he and his team several years ago established convincingly that the myriapods - millipedes and centipedes - are not the ancestor of modern insects, as most people assumed. Rather, these many-segmented creatures emerged from the ocean earlier than insects. Crustaceans - crabs and lobsters - are more closely related to insects than are millipedes and centipedes, he said.

Boore, Mishler and other members of the collaboration hope to find chloroplast as well as plant mitochondrial genes that change slowly over time, and thus would be suitable for assessing long-term evolutionary change, as well as fast-mutating genes suitable for studying more recent evolution.

Other principal investigators on the grant are research botanist Alan R. Smith of UC Berkeley's University Herbarium, Charles O'Kelly of the Bigelow Laboratory for Ocean Sciences in Maine, Paul G. Wolf of Utah State University, Karen Renzaglia of Southern Illinois University, Dina Mandoli and Richard Olmstead of the University of Washington and Michael Donoghue of Yale University. Mandoli plans to construct bacterial artificial chromosome (BAC) nuclear genome libraries for about 50 plants representing deep-branching lineages of green plants, for use by other researchers.

Mishler will host a Tree of Life symposium on the UC Berkeley campus next February to discuss the structure of the green plant family tree and how it relates to other family trees, such as those of arthropods (insects and spiders) and vertebrate animals.