Viz Lab Summer Grant 2006

2008 Reports 2007 Reports 2006 Reports 2005 Reports 2004 Reports

Identification and analysis of genes expressed in canola flowers

Clay Carter and Brian Kram Biology

INTRODUCTION
Several members of the Brassicaceae plant family are cultivated as principal agricultural crops around the world; some of the more notable members include: canola, cabbage, cauliflower, broccoli, and various mustards. Brassica rapa serves as a major source for canola/rapeseed oil, which accounts for 13% of the oil produced annually by humans for consumption. Due to its largely self-incompatible nature, B. rapa is strongly dependent upon pollinator visitation (primarily honeybee) to achieve maximum crop yields. As a result, the volume of nectar produced and secreted from floral nectaries is often substantial -- up to 100 µL per day. Understanding the mechanisms regulating the complex suite of interactions, governing nectar production and secretion, could one day allow for the enhancement of highly desirable traits in agricultural crops. As a first step, we are performing large-scale studies of gene expression in floral nectaries (the organ that produces nectar), and are using VDIL software for high-throughput DNA sequence analysis.

OVERVIEW OF RESEARCH
Genetic control of nectary development, and the subsequent production/secretion of nectar, has largely remained a mystery. However, due to the advent of new technologies that allow us to sample genome-wide transcription of genes, we now have the potential to make great strides in understanding these molecular mechanisms. Nectar production and secretion are tightly regulated through gene expression, which can be triggered by a variety of stimuli. We have initiated large-scale DNA sequencing of genes expressed in Brassica nectaries to produce expressed sequence tags (ESTs). This is one way to identify key genes that are required for nectar production. In addition, our research involves the use of Affymetrix microarrays to analyze tissue-specific gene expression in the model plant Arabidopsis thaliana; this information will then be used to find homologues to the ESTs generated in Brassica rapa. Genes from Brassica and Arabidopsis share, on average, 87% sequence identity. As has been the case in past studies involving these genera, we expect the knowledge gained while studying the genetics of one system to be largely applicable to the other; that, in fact, is the underlying basis of our research. Following the identification of candidate genes involved in the regulation of gene expression in nectaries, various experiments will be conducted: 1) to describe how these genes affect de novo nectar production and 2) to define the encoded proteins. By the completion of this study, we hope to have identified several transcription factors and regulatory elements that modulate nectar production in plants.

APPLICATION OF VDIL SOFTWARE
Before sequencing of Brassica flower DNA can occur, it must first be attached to a piece of DNA of previously known sequence (vector DNA). After doing this, one can isolate the recombinant DNA and subject it to sequencing; in our case, samples were sequenced at the University of Minnesota Biomedical Genomics Center. A problem that occurs is that part of the data contains DNA sequence derived from the vector, which, if not edited out, can interfere with DNA sequence analysis. Another common problem is that DNA sequencing generally provides “good” results for only the first 600 nucleotides or so, with downstream sequence tending to be poor. This poor sequence data must also be removed prior to sequence analysis. Thus, the DNASTAR Lasergene software package, specifically Seqman, was used to remove vector sequence and trim off poor sequence data. SeqMan software allows for the designation of minimum length parameters and cutoffs for poor-quality trace data; resultantly, sequence data that do not meet these standards can automatically be discarded. Setting these guidelines prior to analysis of sequence output greatly reduced the time spent processing this information. After trimming off vector DNA and poor sequence data, Genequest software available at VDIL was used to identify what our DNA sequences encode. This software uses the Basic Local Alignment Search Tool (BLAST) to search for similar sequences found in an internationally searchable database. All together, VDIL software was used to analyze results from 576 different sequences. A brief summary of the general functions of these genes can be found in Figure 1: