DEVELOPMENT AND CHARACTERIZATION OF GENETIC RESOURCES FOR AGRONOMIC AND QUALITY TRAITS USING GENOMIC TOOLS
Title: Genetic diversity for grain mineral concentrations among diverse rice germplasm grown gnder aerobic and anaerobic field conditions
| Tarpley, Lee - |
| Salt, David - |
| Zhang, M - |
| Baxter, I - |
| Guerinot, M - |
| Punshon, Tracy - |
Submitted to: Rice Technical Working Group Meeting Proceedings
Publication Type: Proceedings
Publication Acceptance Date: January 15, 2010
Publication Date: February 22, 2010
Citation: Pinson, S.R., Tarpley, L., Salt, D.E., Zhang, M., Baxter, I., Guerinot, M.L., Punshon, T. 2010. Genetic diversity for grain mineral concentrations among diverse rice germplasm grown gnder aerobic and anaerobic field conditions. Rice Technical Working Group Meeting, Biloxi, MS, Feb. 22-25, 2010. CDROM.
Rice provides the major source of nutrition for a large proportion of the world’s population. Mineral nutrients such as Ca, Fe, and Zn play critical roles in human health, with over 3 billion people suffering from Fe and Zn deficiencies. Unfortunately for those who rely on rice for subsistence, rice grain is not a good source of these nutrients and can contain toxic elements such as As and Cd. As such, alterations in the mineral content of rice grain to either increase or decrease levels of various elements would impact human health. The first step toward breeding commercial rice lines with improved nutritional value is to identify germplasm having extreme nutritional traits.
The USDA Core Rice Collection is a subset of 1797 rice lines randomly selected from among the more than 17,000 accessions in GRIN. This Core subset contains rice accessions from 112 countries in 15 geographical regions, and was randomly selected in order to represent the wide genetic diversity contained within the larger set of rice lines contain in the USDA National Small Grains Collection (NSGC). We grew the 1700 O. sativa and O. glaberima members of the USDA Core Rice Collection in Beaumont, TX under both flooded and unflooded field conditions over two years (2007-2008), two replications per year. ICP-MS was used to analyze the harvested brown rice for variation in accumulation of 16 elements, namely Mg, P, K, S, Ca, Mn, Fe, Co, Ni, Cu, Zn, As, Rb, Sr, Mo, and Cd.
Because the soil redox state greatly affects the availability of soil nutrients, we grew the 1700 Core accessions under both flooded, and unflooded conditions, two replications per year, over two years. It was important to grow the accessions as close together as possible in order to minimize soil variance within the study. The accessions were drill-seeded into 5-seed hillplots, arranged in rows. Five hillplots were planted per row-grouping, with 61 cm between hillplots within each row, and 25 cm between rows. This relatively large distance between hillplots not only allowed us to walk between plots for management and harvest, but also minimized the variance of nutrient and sunlight availability between the plots inside versus on the ends of the field rows. Fifteen repeated check-plots paired with fifteen soil samples per paddy collected in a grid pattern. This allowed us to document that the impact of soil nutrient availability within each paddy caused minimal impact on grain element content. Twenty fully mature seed were selected per hillplot for ionomic analysis using ICP-MS. The standard rubber coating on huller rollers was found to contaminate seed samples with Zn during the hulling process. Therefore, we first replaced this rubber coating with a PU40 Polyurethane plastic. Standard coin envelopes with gummed flaps were used to contain the dehulled brown rice samples, after first verifying no elemental contamination of seed from their usage. Grain content of each of the 16 elements was averaged across replications and years.
Large (> 3x) variance in grain content was found for each of the 16 elements studied. Some element x element correlations were found to be significant and > 0.5, in particular P – K (r=0.56), P-Mg (r=0.66), and Zn –Ca (r=0.64). Both K and Mg were more directly correlated with P than with each other (r=0.35). We saw a wide range of grain concentration between the various elements. For example, the average grain concentration of Cd was 0.15 ppm, Fe averaged 15 ppm, and Ca averaged 100 ppm. In order to more easily compare elements with such diverse grain concentrations, we calculated Z-scores for each element, which reflect standard deviation from the population averages. In prior Arabidopsis studies, major-gene mutations generally caused Z-scores = 3. For future rice studies, we used a Z-score threshold of 3 for identifying accessions to be used as crossing parents to create segregating F2 progeny in which we will pursue gene mapping. The unflooded field condition revealed a wider range of grain element contents than did the flooded field condition. For all elements, the grain content histograms were skewed with significantly more accessions having > average content, than those having< average content. For all elements, several lines were found to have Z scores > 3, indicating they had extremely higher-than-average elemental concentration; but for several elements, no accessions were found to have Z scores < -3 (significantly low concentration). In several cases, the accessions identified as having high concentration of a particular element were found to come from geographically similar regions. For example, four of the 5 lines highest in Mo content all came from Malaysia. The common origin of the high-Mo accessions suggests that we were successful in identifying rice accessions containing a highly heritable gene underlying their high Mo grain content. It also suggests that we can study just one segregating progeny population, or combine data between the four related segregating progeny, in order to efficiently identify molecular markers linked to this high-Mo gene.