Stephen A. Goff

A High-Throughput Arabidopsis Reverse Genetics System

A collection of Arabidopsis lines with T-DNA insertions in known sites was generated to increase the efficiency of functional genomics. A high-throughput modified thermal asymetric interlaced (TAIL)-PCR protocol was developed and used to amplify DNA fragments flanking the T-DNA left borders from ∼100,000 transformed lines. A total of 85,108 TAIL-PCR products from 52,964 T-DNA lines were sequenced and compared with the Arabidopsis genome to determine the positions of T-DNAs in each line. Predicted T-DNA insertion sites, when mapped, showed a bias against predicted coding sequences. Predicted insertion mutations in genes of interest can be identified using Arabidopsis Gene Index name searches or by BLAST (Basic Local Alignment Search Tool) search. Insertions can be confirmed by simple PCR assays on individual lines. Predicted insertions were confirmed in 257 of 340 lines tested (76%). This resource has been named SAIL (Syngenta Arabidopsis Insertion Library) and is available to the scientific community at www.tmri.org.

A network of rice genes associated with stress response and seed development

We used a systematic approach to build a network of genes associated with developmental and stress responses in rice by identifying interaction domains for 200 proteins from stressed and developing tissues, by measuring the associated gene expression changes in different tissues exposed to a variety of environmental, biological, and chemical stress treatments, and by localizing the cognate genes to regions of stress-tolerance trait genetic loci. The integrated data set suggests that similar genes respond to environmental cues and stresses, and some may also regulate development. We demonstrate that the data can be used to correctly predict gene function in monocots and dicots. As a result, we have identified five genes that contribute to disease resistance in Arabidopsis.

Utility and distribution of conserved noncoding sequences in the grasses

Control of gene expression requires cis-acting regulatory DNA sequences. Historically these sequences have been difficult to identify. Conserved noncoding sequences (CNSs) have recently been identified in mammalian genes through cross-species genomic DNA comparisons, and some have been shown to be regulatory sequences. Using sequence alignment algorithms, we compared genomic noncoding DNA sequences of the liguleless1 (lg1) genes in two grasses, maize and rice, and found several CNSs in lg1. These CNSs are present in multiple grass species that represent phylogenetically disparate lineages. Six other maize/rice genes were compared and five contained CNSs. Based on nucleotide substitution rates, these CNSs exist because they have biological functions. Our analysis suggests that grass CNSs are smaller and far less frequent than those identified in mammalian genes and that mammalian gene regulation may be more complex than that of grasses. CNSs make excellent pan-grass PCR-based genetic mapping tools. They should be useful as characters in phylogenetic studies and as monitors of gene regulatory complexity.

Identification of rice (Oryza sativa) proteins linked to the cyclin-mediated regulation of the cell cycle.

Yeast two-hybrid assays were used to identify rice proteins interacting with two rice cyclins and other proteins potentially involved in cell cycling. The DNA sequences encoding 119 protein fragments identified were then compared by BLAST against proteins in GenBank. The proteins found include myosin-like proteins, transcription factors, kinesins, centromere proteins and undefined proteins. Based on interactions with cyclins and other elements required for cycling, we believe the undefined proteins may be involved in associated cycling processes. The identification of proteins involved in cell cycle regulation in rice may allow for the control of agronomic traits involving plant growth or development.