Elizabeth A. Kellogg

University of Missouri-St. Louis

It has been clear to me for some time that there needs to be a new synthesis between molecular genetics and phylogenetics. We are rapidly approaching a time when the broad outlines of plant phylogeny are in place, which gives us the basic framework within which to ask what happened in evolutionary time. I have been approaching this question using the grass family as a model system. Work in my lab has been largely phylogenetic, but has increasingly involved studies of developmental morphology, histology, gene expression, and molecular evolution of developmentally important genes. In this work we have been trying to describe at the molecular level the changes that have occurred at deep nodes in the phylogeny, events in the long-distant past. In the process of this work, I have found deep divisions between the scientific cultures of those who study phylogenetics and those who study genomics, divisions that require far more interdisciplinary training than is commonly done. It is generally true that geneticists are interested in genetic struictures and functions that are conserved across large groups of organisms, whereas phylogeneticists are interested in aspects that vary. This immediately means that the genes and functions they are interested in studying are quite different. There is the need for many additional tools and procedures for screening candidates genes to focus on those that are conserved, and thus of general interest, and those that are variable and thus of potential evolutionary importance. Such tools would benefit the entire community of those studying phylogenomics. It is not clear how any one lab could proceed alone in what needs to be a community endeavor.

Projects in my lab at the moment include a collaborative phylogeny of the entire Gramineae, based on 8 different sets of data, which produces a strongly supported phylogeny for the family. We are also in the process of developing a molecular phylogeny for the subfamily Panicoideae, a group of about 3000 species that is the size of many angiosperm families. These phylogenies form the basis for a number of developmental studies. Much diversity in the grasses, as in the angiosperms as a whole, is diversity of flowers and inflorescences. We have extensive data on the developmental morphology of flowers and inflorescences in the panicoid grasses. This shows that floral structure proceeds in a unique pathway involving cell death in the gynoecium of some flowers to produce unisexual flowers. We have also shown that variation in inflorescence structures is apparently due to modest changes in the timing of development, affecting the length of time primary and secondary meristems produce lateral primordia.
In collaboration with colleagues from the John Innes Centre in England, we are mapping inflorescence varation in a segregating F2 population in a cross of Setaria italica x viridis. This should allow us to determine how many genes are controlling each of the inflorescence characters that differentiate the parents, and by correlation with maps of other grass species, should let us identify candidate genes responsible. Studies of the gene expression of candidate genes have been hampered by lack of appropriate antibodies for immunolocalization, and by the difficulty of cloning candidate genes from many disparate taxa for use as probes for in situ hybridization. We are beginning to do in situ hybridization of Tasselseed2 in the gynoecium of the panicoids. We are also investigating expression of leafy in other panicoids to determine if its expression is conserved across the group. We will begin work on other developmental genes in the fall.

It is difficult to find any single biologist who has skills in developmental morphology, Southern blotting, Northen blotting, immunolocalization, RT-PCR, in situ hybridization, moleuclar phylogenetics, and comparative biology. The necessity of acquiring all the technical and analytical skills creates a hurdle that few graduate students or post-docs are willing to jump. In general, they opt for projects that are less demanding and more routine. The proposed RCN will I hope give us ways to address the technical and analytical issues. For example, if members of the RCN community decided to pool resources and make joint applications for funding, we could perhaps develop a set of antibodies to critical genes, much as the animal community did with the antibodies to distalless and engrailed. A set of immunolocalization experiments would provide expression data on many genes for many taxa, and would help to define the problems. this is not somehting that any one lab would do, however. As another example, RCN workshops might devise ways to use microarray technology as a way to screen genes for those of particular evolutionary interest.