Covariation among root biomass, shoot biomass, and tiller number in three rice populations

Authors: Barnaby, Jinyoung; Pinson, Shannon; CHUN, JAEBUHM - Rural Development Administration - Korea; BUI, LIEM - Cuu Long Delta Rice Research Institute

Submitted to: Crop Science
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 3/3/2019
Publication Date: 5/16/2019
Citation: Barnaby, J.Y., Pinson, S.R., Chun, J., Bui, L.T. 2019. Covariation among root biomass, shoot biomass, and tiller number in three rice populations. Crop Science. 59:1516-1530. https://doi.org/10.2135/cropsci2018.09.0595.
DOI: https://doi.org/10.2135/cropsci2018.09.0595

Interpretive Summary: RBreeding for improved rice root systems could improve crop yields by increasing drought tolerance and uptake of nutrients, while also protecting the environment by reducing excess nutrients in runoff or and/or reducing methane emissions , which are also affected by roots. Collection and measurement of roots is laborious, preventing direct selection for root biomass (RB) in large breeding populations. Tiller and root development are physiologically connected in rice where the root system is composed predominantly of nodal roots. Each tiller and its associated roots emerge from the same node on an originating stem. We tested the hypothesis that genes affecting tiller number (TN) also have sufficiently large effect on shoot biomass (SB) and RB and that selections made on TN data can be used to impart improvements in RB and SB. We evaluated the association between TN, RB, and SB in three recombinant inbred line (RIL) rice mapping populations by first selecting subsets that were divergent for TN, then evaluating if those contrasting RILs exhibited corresponding shifts in SB, RB, plant height (Ht), and the calculated traits SB/RB, RB/TN, and SB/TN. Traits were measured in 6-week old potted plants to allow control of environmental factors known to affect TN. Selection for high versus low TN resulted in corresponding shifts in RB and SB in all three populations, suggesting linkage or sharing of genes between TN, RB, and SB. Ability to select for RB with fast, nondestructive TN data proved robust across the three populations having variable genetic backgrounds and TN loci. Gene interactions were indicated with TN selections causing larger shifts in RB and SB among a sd1 (semidwarf) subgroup, than among their siblings possessing Sd1. Trait relationships and quantitative trait loci (QTLs) were studied further using 62 RILs from one of the mapping populations. One of the six TN-QTLs co-located with a RB-QTL and a SB-QTL; suggesting that the TN-RB-SB association is due to some but not all TN-QTLs.

Technical Abstract: Breeding for improved rice root systems could improve crop yields by increasing uptake of nutrients and drought tolerance, while also protecting the environment by reducing excess nutrients in runoff or reducing methane emissions, which are also affected by roots. Collection and measurement of roots is laborious, preventing direct selection for root biomass (RB) in large breeding populations. Tiller and root development are physiologically connected in rice where the root system is composed predominantly of nodal roots. Each tiller and its associated roots emerge from the same node on an originating stem. We tested the hypothesis that genes affecting tiller number (TN) also have sufficiently large effect on shoot biomass (SB) and RB and that selections made on TN data can be used to impart improvements in RB and SB. We evaluated the association between TN, RB, and SB in three recombinant inbred line (RIL) rice mapping populations by first selecting subsets that were divergent for TN, then evaluating if those contrasting RILs exhibited corresponding shifts in SB, RB, plant height (Ht), and the calculated traits SB/RB, RB/TN, and SB/TN. Traits were measured in 6-week old potted plants to allow control of environmental factors known to affect TN. Selection for high versus low TN resulted in corresponding shifts in RB and SB in all three populations, suggesting linkage or sharing of genes between TN, RB, and SB. Ability to select for RB with fast, nondestructive TN data proved robust across the three populations having variable genetic backgrounds and TN loci. Gene interactions were indicated with TN selections causing larger shifts in RB and SB among a sd1 (semidwarf) subgroup, than among their siblings possessing Sd1. Trait relationships and quantitative trait loci (QTLs) were studied further using 62 RILs from one of the mapping populations. One of the six TN-QTLs co-located with a RB-QTL and a SB-QTL; suggesting that the TN-RB-SB association is due to some but not all TN-QTLs.