2014 ASHS Annual Conference

In the 21st century, cancer is expected to surpass heart disease as the number one cause of death in the United States. With the current cancer treatment methods only showing limited efficacy in decreasing cancer-related deaths, some believe that in the short term the prevention of carcinogenesis may be the more prudent objective.  Several dietary factors have been shown to increase one’s risk of developing cancer, including saturated and trans-fats, but other compounds have been shown to be associated with a decreased cancer risk.  Among these compounds are the glucosinolates (GSs), which are found in many common crops from the Brassica genus including broccoli, cauliflower, cabbage, and kale.  Even though there is evidence that glucosinolates, and more specifically their bioactive hydrolysis products, have a negative effect on carcinogenesis, there is still much to learn about these compounds.  Many genes involved in GS and hydrolysis product formation have been identified, but little is known about the underlying genetic factors leading to variability in these compounds, and subsequently, chemopreventive bioactivity.  In this study we have attempted to determine how GS and hydrolysis product profiles differ between species within the Brassica genus and how this relates to cancer chemopreventive activity (as measured by quinone reductase (QR) activity in murine liver cells).  Our data confirm previous reports that GS profiles vary greatly between species, although much of this variation may be attributed to the tissue that is harvested for a given crop.  We also observed significant variation in GS profiles between subspecies within a given Brassica species.  Large variations in QR activity were seen between species  as well as between subspecies within a species, although correlations between individual GSs and QR activity were inconsistent. We have used multiple regression analysis to determine how well QR can be explained using GSs and hydrolysis products as predictor variables, both for the genus as a whole and for individual species.  We have used the same multiple regression techniques to determine the predictive ability of transcript abundance data of key GS biosynthetic genes for estimating final GS levels.  Also, using GS profiles and transcript abundance of key hydrolysis genes as predictor variables, we have attempted to forecast the final hydrolysis product profiles.  Finally, we have used information from GS and hydrolysis product analyses to choose individuals for sequence analysis who likely show allelic differences for key GS biosynthesis and hydrolysis genes.