Univariate and multivariate QTL analyses reveal covariance among elements in the rice ionome that is affected by production environment and tissue type.

Authors: LIU, HUAN - Nanjing Agricultural University; LONG, SU-XIAN - Nanjing Agricultural University; Pinson, Shannon; TANG, ZHONG - Nanjing Agricultural University; GUERINOT, MARY LOU - Dartmouth College; SALT, DAVID - University Of Nottingham; ZHAO, FANG-JIE - Nanjing Agricultural University; HUANG, XIN-YUAN - Nanjing Agricultural University

Submitted to: Frontiers in Genetics
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 1/6/2021
Publication Date: 1/25/2021
Citation: Liu, H., Long, S., Pinson, S.R., Tang, Z., Guerinot, M., Salt, D.E., Zhao, F., Huang, X. 2021. Univariate and multivariate QTL analyses reveal covariance among elements in the rice ionome that is affected by production environment and tissue type. Frontiers in Genetics. https://doi.org/10.3389/fgene.2021.638555.
DOI: https://doi.org/10.3389/fgene.2021.638555

Interpretive Summary: To improve human health by alleviating malnutrition, there is widespread interest in developing crop varieties that accumulate more nutritive elements in their edible grains (e.g., calcium or iron) while limiting the accumulation of toxic elements such as arsenic and cadmium. Identifying and molecularly-tagging genes for desired traits provides breeders with knowledge and tools with which to speed the variety development process. In a study published in 2014, we used a mapping population derived from a cross between the US variety ‘Lemont’ and ‘TeQing’, a rice variety from China, to identify 134 chromosomal regions containing genes affecting the accumulation of 16 individual elements in rice grains. While three genes that notably increase grain concentrations of copper, molybdenum, and magnesium were cloned on the basis of that study, the remaining 131 genes reported have not been studied further, largely due to the use of only 176 markers that mapped the genes to large rather than narrow or precise regions. Recent advances in molecular technology have reduced the cost of developing thousands of new markers based on DNA sequencing. We sequenced the two parents and the 257 cross-progeny in our mapping population, and created an enhanced genetic map containing more than 3000 markers. Armed with this enlarged genetic map, along with new statistical procedures that allow us to consider the elements not just as individual traits, but as sets of meaningfully related and prioritized traits, we reanalyzed the grain data collected and used previously. One unique aspect of this grain dataset is that it includes element analyses of grains grown in both flooded paddies and unflooded fields. This is of particular relevance because we know that the availability of nutrients for plant uptake changes greatly when soils are flooded. For our new study, we also collected data on element concentrations in grains, roots and shoots of rice plants grown under carefully controlled greenhouse conditions. From our modernized genetic analyses we have learned that the concentrations and ratios of the elements differ more between roots, shoots, and grains than they do between grains produced with versus without a flood. This suggests that plant mechanisms, such as tissue-to-tissue transport, affect grain element concentrations more than uptake differences due to soil flooding. We also discovered several genes that affect not just one, but multiple grain elements. While some of these multi-element genes were also detected based on their impact on individual elements, some were not detectable until we used principal component analysis, which essentially calculates new multi-element “traits” that measure relatedness among the 16 elements. To prioritize further effort, we looked at all of our elemental and relational QTLs and found some genomic regions containing clusters of as many as 15 or 20 QTLs. Looking at annotated genes in databases, in two regions where we identified large QTL clusters we found nearby annotated genes previously shown to affect plant uptake or grain accumulation of one element (e.g. the copper gene cloned after our 2014 publication), but now suggested to affect grain accumulation of multiple elements. The improved mapping precision accomplished with our enlarged marker map, along with the new knowledge created on relationships among the elements in grains, and in roots and shoots as well, will provide guidance for continued characterization of genes useful for improving the nutritive value of rice and other grains.

Technical Abstract: As one of the most important staple food crops, rice not only provides more than one fifth of daily calories for half of the world’s human population but is also a major dietary source of both essential mineral nutrients and toxic elements. Production of nutritious and safe rice with grains enriched in essential mineral nutrients and reduced in toxic elements is essential for maintenance human health. Identification of the quantitative trait loci (QTLs) or genes that control the concentrations of mineral nutrients and toxic trace metals (the ionome) in rice will facilitate development of nutritionally improved rice varieties. QTL analyses have traditionally considered each element separately without considering their interrelatedness. In this study, we performed principal component analysis (PCA) for the concentrations of 16 elements in grains, shoots and roots of a population of rice recombinant inbred lines grown in 7 years under 4 different conditions. PCA revealed distinct ionomic patterns in different tissues or in a single tissue under different growth environments. Taking advantage of a high density linkage map we derived from whole genome resequencing, univariate and multivariate QTL analyses were performed using the elemental concentrations or principal component scores as traits, respectively. Identified were 167 unique elemental QTLs based on analyses of individual elemental concentrations as separate traits, 53 QTLs controlling covariance among elemental concentrations within a single environment or tissue (PC-QTLs), and 152 QTLs which determined covariation among elements across environments/tissues (aPC-QTLs). Compared to the elemental QTL mapping intervals, 25 of 53 PC-QTLs and 99 of 167 aPC-QTLs co-localized with at least one elemental QTL, resulting in 28 novel PC-QTLs and 68 novel aPC-QTLs. Fifteen QTL clusters with elemental QTLs, PC-QTLs and aPC-QTLs co-localized were identified, and their candidate genes isolated, including a heavy metal P-type ATPase gene OsHMA4 for a copper QTL cluster and a cadmium and manganese transporter gene OsNRAMP5 for a manganese QTL cluster. The identification of both elemental QTLs and PC QTLs in this study will facilitate the cloning of underlying causal genes and the dissection of the complex regulation of the ionome in rice.