Research Interests

Evolution of gene networks in Arabidopsis

Overview: The organs of all organisms, from plant roots to human brains, are mosaics of different cell types. Thus, one fundamental task in organ development is assembling discrete, highly specialized cells. How is a specialized cell constructed at the genetic level and how do new cell types evolve? Taking a genomic approach to answer these questions, the analysis of gene regulation on a whole organ level confounds signals by mixing the genetic components of many different cell types. Thus, we have developed tools that examine gene activity within specialized cells in order to dissect the genetic pathways that lead to cellular specialization. Our research seeks to uncover the localized signals that regulate cellular differentiation. In a related question, we ask how an organism’s genetic machinery is manipulated (or invented) over time to create new cell types over evolution.

Navigating the “Redundome”:
High-throughput expression analysis has generated an explosion of data on gene regulation but a backlog of untested hypotheses about gene function. The bottleneck in functional analysis in model organisms is aggravated at least in part by functional redundancy at the genetic level. We are carrying out a phenotypic screen that uses cell-specific and other expression data to overcome redundancy and focus phenotypic analysis at the cell level for highly sensitive screening. For example, we have used several sources of data to generate a “formula” to predict genetic redundancy.
Our experimental approach asks how transcriptional regulators control differentiation events across and within cell types using a model organ in which cell types mature synchronously along a longitudinal gradient, the root of the model plant Arabidopsis thanliana. We first analyze the phenotypes of candidate regulatory genes involved in cellular differentiation. We also analyze potential differences in the targets of regulatory genes in different cell types using cell-specific expression profiling. We then examine the phenotypes of putative downstream cell-specific genes. Our goal is to identify the genes that control critical points in differentiation networks, their downstream genetic components, and the specific traits these gene cassettes control. For example, we are focusing on two cell types, xylem and columella, that sit at opposite ends of the stem cell niche, do not occupy the same tissue, yet exclusively share up-regulation of a set of genes. We are testing, by knockout analysis, whether this shared gene program represents the machinery that leads to one common trait between the two cell types, rapid programmed cell death.

Cellular Evolution

The evolution of gene expression can be tracked dynamically by extending cell type expression profiling techniques to other species. Our second major area of research builds on our methods that generate high-resolution expression maps in Arabidopsis. This entailsgenerating similar expression maps in other species. Rice, which is fully sequenced, is one of our targets. Subsequent analysis will focus on lower plants, including Selaginella and Physcomitrella, both of which are being sequenced. These comparative expression maps will provide 1) a proxy for ancestral expression patterns of duplicated genes prior to duplication. This will help assess the role of whether gene duplication leads to novel roles for duplicates by expansion or contraction of ancestral expression domains. 2) The profiles will help form specific hypotheses on the role of gene recruitment in the evolution of new cell types. For example, columella cells are unique to the root and, thus, hold important clues to the evolution of the root. We are testing the localization of columella specific transcripts in ancient plants that may or may not have a homologous root to Arabidopsis.