Mapping and validation of quantitative trait loci associated with concentrations of 16 elements in unmilled rice grain

Min Zhang,Lee Tarpley,Elena Yakubova,Xin-Yuan Huang,Ivan Baxter,Shannon R M Pinson,David E Salt,Brett Lahner,Mary Lou Guerinot

doi:10.1007/s00122-013-2207-5

Abstract

Key Message QTLs controlling the concentrations elements in rice grain were identified in two mapping populations. The QTLs were clustered such that most genomic regions were associated with more than one element. In this study, quantitative trait loci (QTLs) affecting the concentrations of 16 elements in whole, unmilled rice (Oryza sativa L.) grain were identified. Two rice mapping populations, the ‘Lemont’ × ‘TeQing’ recombinant inbred lines (LT-RILs), and the TeQing-into-Lemont backcross introgression lines (TILs) were used. To increase opportunity to detect and characterize QTLs, the TILs were grown under two contrasting field conditions, flooded and irrigated-but-unflooded. Correlations between the individual elements and between each element with grain shape, plant height, and time of heading were also studied. Transgressive segregation was observed among the LT-RILs for all elements. The 134 QTLs identified as associated with the grain concentrations of individual elements were found clustered into 39 genomic regions, 34 of which were found associated with grain element concentration in more than one population and/or flooding treatment. More QTLs were found significant among flooded TILs (92) than among unflooded TILs (47) or among flooded LT-RILs (40). Twenty-seven of the 40 QTLs identified among the LT-RILs were associated with the same element among the TILs. At least one QTL per element was validated in two or more population/environments. Nearly all of the grain element loci were linked to QTLs affecting additional elements, supporting the concept of element networks within plants. Several of the grain element QTLs co-located with QTLs for grain shape, plant height, and days to heading; but did not always differ for grain elemental concentration as predicted by those traits alone. A number of interesting patterns were found, including a strong Mg–P–K complex.Electronic supplementary materialThe online version of this article (doi:10.1007/s00122-013-2207-5) contains supplementary material, which is available to authorized users.

Highlights

As one of the most important staple crops, rice provides more than 40 % of the daily calories for the world’s population (Parengam et al 2010)
The first population was a set of ‘Lemont’ × ‘TeQing’ recombinant inbred lines (LT-RILs) similar in genetic structure to the mapping population analyzed by Norton et al (2010), while the second population was a companion set of TeQing-intoLemont backcross introgression lines (TILs) similar in genetic structure to the chromosome segment substitution lines (CSSLs) observed by Ishikawa et al (2005)
The staggered-planting dates employed in the 2002, 2003, and 2006 LT-RIL studies successfully narrowed the period of heading time and decreased the variance in day length, light intensity, and air and soil temperatures during grain-fill among the LT-RILs

Summary

Introduction

As one of the most important staple crops, rice provides more than 40 % of the daily calories for the world’s population (Parengam et al 2010). Four rice studies have reported QTLs associated with accumulation of nutritional and potentially toxic elements in rice grain (Ishikawa et al 2005; Lu et al 2008; Norton et al 2010; Stangoulis et al 2007) The earliest of these rice studies (Ishikawa et al 2005) utilized a set of chromosome segment substitution lines (CSSLs) and identified three QTLs associated with Cd concentration. Norton et al (2010) used atomic absorption spectrometry to evaluate concentrations of Ca and Mg in rice grain and leaves, plus inductively coupled plasma mass spectrometry (ICP-MS) to evaluate an additional 15 elements to identify ionomic QTLs segregating among an F6 indica × japonica mapping population. The first population was a set of ‘Lemont’ × ‘TeQing’ (japonica × indica) recombinant inbred lines (LT-RILs) similar in genetic structure to the mapping population analyzed by Norton et al (2010), while the second population was a companion set of TeQing-intoLemont backcross introgression lines (TILs) similar in genetic structure to the CSSLs observed by Ishikawa et al (2005)

Methods

Results

Conclusion