UDP-glucosyltransferase OsUGT75A promotes submergence tolerance during rice seed germination

Abstract

Submergence stress represents a major obstacle limiting the application of direct seeding in rice cultivation. Under flooding conditions, coleoptile elongation can function as an escape strategy that contributes to submergence tolerance during seed germination in rice; however, the underlying molecular bases have yet to be fully determined. Herein, we report that natural variation of rice coleoptile length subjected to submergence is determined by the glucosyltransferase encoding gene OsUGT75A. OsUGT75A regulates coleoptile length via decreasing free abscisic acid (ABA) and jasmonic acid (JA) levels by promoting glycosylation of these two phytohormones under submergence. Moreover, we find that OsUGT75A accelerates coleoptile length through mediating the interactions between JASMONATE ZIMDOMAIN (OsJAZ) and ABSCISIC ACID-INSENSITIVE (OsABI) proteins. Last, we reveal the origin of the haplotype that contributes to coleoptile length in response to submergence and transferring this haplotype to indica rice can enhance coleoptile length in submergence conditions. Thus, we propose that OsUGT75A is a useful target in breeding of rice varieties suitable for direct seeding cultivation.

Fig. 1. Identification on rice chromosome 11 of a QTL, qCL11, for coleoptile length under submergence.

a Manhattan plots for the whole population of rice accessions. The red arrow indicates the identified locus. b A Quantile–Quantile plot. c Identification of candidate genes in the qCL11 region. In a–c no data adjustments were made for GWAS with a threshold of 2.50 × 10⁻⁸ (0.01 significance level). d LOC_Os11g25990, the putative candidate gene for qCL11, was significantly induced during seed germination under anaerobic conditions using publicly available microarray data (http://www.genevestigator.com). Red, up-regulation; Blue, down-regulation. Values represent the log₂-fold gene changes. e, f Haplotypes of LOC_Os11g25990 identified in the coding sequence (CDS), 2-kb region upstream and 1-kb region downstream of the gene. IND indica, TEJ temperate japonica, TRJ tropical japonica. g Box-plots of coleoptile length in accessions containing the different haplotypes. n = 116/108 accessions. Center lines show the medians, box limits indicate the 25th and 75th percentiles, whiskers extend to the minimal and maximal values as determined by GraphPad Prism 8.0.1 software. h Chromosomal location of the introgressed Kasalath (with the Hap2 allele) segments in a near isogenic line (NIL) in a Koshihikari (with the Hap1 allele) background. i Representative images of Koshihikari (with the Hap1 allele) and NIL (with the Hap2 allele) under submergence (8 cm depth of water) for 7 days. Scale bars represent 10 mm. j Coleoptile length under submergence (8 cm depth of water) for 6 and 7 days. Data were presented as mean ± SD, n = 14. In g, j significant differences were determined by two-tailed Student’s t tests (*P < 0.05, **P < 0.01, ***P < 0.001). Source data are provided as a Source Data file.

Fig. 2. Characterization of the regulation of rice coleoptile length under submergence by OsUGT75A.

a Representative images of coleoptile length and b dynamic changes in coleoptile length among Nipponbare wild-type (WT) and Osugt75a mutant under submergence (8 cm depth of water). Data were presented as mean ± SD, n = 5. c Representative images of coleoptile length under submergence (8 cm depth of water) for 7 days and seedling establishment of direct seeding in soils under submergence (4 cm depth of water) for 12 days among WT and overexpressed OsUGT75A lines (OE-1 and OE-2). d Comparison of the relative expression levels of OsUGT75A in WT and overexpressed lines determined by qRT-PCR analysis. Data were presented as mean ± SD, n = 3 biologically independent samples. Expression is shown relative to that in the WT, the value of which was set to 1, with the OsActin gene being used as an internal control. e Comparison of the coleoptile lengths of WT and overexpressed lines. n = 23/22/26. Center lines show the medians, box limits indicate the 25th and 75th percentiles, whiskers extend to the minimal and maximal values as determined by GraphPad Prism 8.0.1 software. f The expression patterns of OsUGT75A in germinating seeds under normal and submerged conditions determined by qRT-PCR analysis. Data were presented as mean ± SD, n = 3 biologically independent samples. Expression is shown relative to that at 0 h under non-submerged conditions, the value of which was set to 1, with the OsActin gene being used as an internal control. g Histochemical staining for β-glucuronidase (GUS) activity in germinating seeds. Scale bars represent 10 mm. In b, d, e significant differences were determined by two-tailed Student’s t tests (*P < 0.05, ***P < 0.001). Source data are provided as a Source Data file.

Fig. 3. OsUGT75A regulates rice coleoptile length under submergence (8 cm depth of water) by influencing abscisic acid (ABA) and jasmonic acid (JA) levels.

a, c Representative images of the coleoptile length of WT and Osugt75a mutants in response to 5 and 10 µM ABA and/or JA treatments after 9 days of submergence. Scale bars represent 10 mm. b, d A comparison of coleoptile lengths in WT and Osugt75a mutants in response to control (H₂O), ABA, and JA treatments after 9 days of submergence. The contents of free ABA, JA, and JA-Ile in e–g the germinating seeds and in h–j the developing coleoptile in Nipponbare wild-type (WT) rice and Osugt75a mutants under submergence. In b, d data were presented as mean ± SD, n = 3 biologically independent experiments; different letters indicate significant differences (P = 0.05, one-way ANOVA). In e–j data were presented as mean ± SD, n = 3 biologically independent samples; significant differences were determined by two-tailed Student’s t-tests (*P < 0.05, **P < 0.01, ***P < 0.001). Source data are provided as a Source Data file.

Fig. 4. Glycosylation ABA and JA by OsUGT75A.

a The full-length GST-OsUGT75A fusion protein was quantified using bovine serum albumin protein standard (BSA). n = 3 independent experiments. The full-length GST-UGT75A fusion protein catalyzes the glycosylation of b ABA and c JA in vitro, but the GST tag could not. Arrow indicates the product ABA-Glu and putative JA-Glu. d JA-Glu formed in c was confirmed by MS/MS analysis in positive ionization mode. The molecular weight (M) of JA-Glu is 372.27. The reaction products had the ion peaks at m/z 395.2202 (M + Na⁺), m/z 210.1180 (M-H-Glu), and m/z 166.0911 (M-COOH-Glu). Quantification of absolute amounts of e the ABA-Glu and f the relative levels of JA-Glu in coleoptiles in Nipponbare wild-type (WT) and Osugt75a mutants under submergence (8 cm depth of water) for 7 days. Absolute amounts of ABA-Glu were measured by HPLC using the commercial available chemical standard. Relative levels of JA-Glu were determined by targeted HPLC. The level of JA-Glu is shown relative to that in WT, the value of which was set to 1. In e, f data were presented as mean ± SD, n = 3 biologically independent samples; significant differences were determined by two-tailed Student’s t tests (*P < 0.05, **P < 0.01, ***P < 0.001). Source data are provided as a Source Data file.

Fig. 5. In vitro and in vivo interaction assays between OsJAZs and OsABIs.

a–d Yeast-two-hybrid (Y2H) analysis revealed interactions between OsABI3/5 and OsJAZ6/7. Transformed yeast cells were grown on SD-Trp/-His/-Leu/-Ade medium. e–h In vitro pull-down assays revealed interactions between OsABI3/5 and OsJAZ6/7. n = 3 independent experiments. i Bimolecular fluorescence complementation analysis of the interactions between OsABI3/5 and OsJAZ6/7. n = 3 biologically independent experiments. Fluorescence, emitted as a consequence of complementation of the N-terminal region of the yellow fluorescence protein (YFP) fused to OsABI3/5 (OsABI3/5-nYFP) with the C-terminal region of YFP fused to OsJAZ6/7 (OsJAZ6/7-cYFP), was observed in the nuclei of rice protoplasts. No signal was observed in the negative controls. Bars represent 50 mm. Source data are provided as a Source Data file.

Fig. 6. OsUGT75A regulates coleoptile length involving the ABA and JA signaling pathways.

a Schematic diagrams of the reporters and effectors used in transient transactivation assays. b, c Transient dual-luciferase (LUC) reporter assays revealed that the OsABI3/5 activation of ABA response gene (OsEM1) expression could be repressed by OsJAZ6/7. Data were presented as mean ± SD, n = 4 biologically independent experiments. d Representative images of the coleoptile length of Nipponbare (NIP) or Zhonghua11 (ZH11) wild-type (WT), Osjaz6/7, and Osabi3 mutants under submergence (8 cm depth of water) for 7 days. Scale bars represent 10 mm. e A comparison of coleoptile lengths in WT, Osjaz6/7, and Osabi3 mutants under submergence. Data were presented as mean ± SD, n = 16. In b–e different letters indicate significant differences (P = 0.05, one-way ANOVA). Source data are provided as a Source Data file.

Fig. 7. Analysis of OsUGT75A haplotypes associated with coleoptile length under submergence.

a An OsUGT75A haplotype network. The size of circles is proportional to the number of samples for a given haplotype. IND indica, JAP japonica, TEJ temperate japonica, TRJ tropical japonica, Or Oryza rufipogon. b The diversity of elite haplotype 1 (Hap1) in rice accessions. c A comparison of elite Hap1 distribution and coleoptile lengths under submergence in the currently cultivated japonica and indica in China. d The development of a simple sequence repeat marker for OsUGT75A. n = 3 independent experiments. e Representative images of coleoptile length under submergence (8 cm depth of water) for 7 days and seedling establishment of direct seeding in soils under submergence (4 cm depth of water) for 14 days among indica Huanghuazhan wild-type (WT) and overexpressed OsUGT75A lines (OE-3 and OE-7). Scale bars represent 10 mm. f Comparison of the relative expression levels of OsUGT75A in WT and overexpressed lines determined by qRT-PCR analysis. Data were presented as mean ± SD, n = 3 biologically independent samples. Expression is shown relative to that in the WT, the value of which was set to 1, with the OsActin gene being used as an internal control. g Comparison of the coleoptile lengths of WT and overexpressed lines, n = 20. Center lines show the medians, box limits indicate the 25th and 75th percentiles, whiskers extend to the minimal and maximal values as determined by GraphPad Prism 8.0.1 software. In f, g significant differences were determined by two-tailed Student’s t tests (**P < 0.01, ***P < 0.001). Source data are provided as a Source Data file.

Fig. 8. Proposed working model forthe role of OsUGT75A in the regulation of rice seed germination under submergence.

a In wild-type plants, OsUGT75A catalyzes the glycosylation of abscisic acid (ABA) and jasmonic acid (JA) contributes to the decrease of free ABA and JA accumulation, which leads to the OsJAZ6/7 proteins interact directly with OsABI3/5 and repress the transcriptional activity of OsABI3/5 and subsequent regulate ABA signaling pathway. b In Osugt75a mutant plants, the lack of glucosyltransferase activity results in the accumulation of free ABA and JA, which leads to repressing the interaction between OsJAZ6/7 and OsABI3/5 proteins and activates the transcriptional activity of OsABI3/5 and subsequent regulate ABA signaling pathway. OsUGT75A thereby regulates seed germination under submergence involving the crosstalk of JA and ABA signaling pathways.