Fig. 1. Phylogenetic tree and protein structure of JMJ30 in selected monocotyledons and Arabidopsis. A, Phylogenetic analysis of JMJ30 proteins from wheat (Triticum aestivum), barley (Hordeum vulgare), purple false brome (Brachypodium distachyon), rice (Oryza sativa), sorghum (Sorghum bicolor), foxtail millet (Setaria italica), and Arabidopsis (Arabidopsis thaliana). The phylogenetic tree was constructed using the neighbor-joining method based on JMJ30 protein sequences, with 1 000 bootstrap replicates in MEGA11. The values at the nodes indicate the percentage of the bootstrap confidence level. B, Protein domain architecture of JMJ30 in different species. Gray boxes represent variable regions, while blue boxes denote conserved JmjC domains.

Fig. 2. Expression pattern and subcellular localization of OsJMJ712. A, Expression levels of OsJMJ712 in various rice tissues. The bar chart display expression values, and the heatmap shows gene expression as transcripts per million (TPM) values normalized by Z-score transformation. RNA-seq data were obtained from the Rice Genome Annotation Project database (RGAP, https://rice.uga.edu/expression.shtml). Transcript abundances were quantified as TPM. DAP, Days after pollination. B and C, Rhythmic expression patterns of OsJMJ712 under long-day (LD) transferred to constant light (LL) (B) and short-day (SD) transferred to constant dark (DD) (C) conditions. Fourteen-day-old seedlings grown under LD or SD conditions were transferred to LL or DD, respectively. Samples were collected every 6 h. The Actin gene was used as an internal control. Data are mean ± SD (n = 3). Black bars represent the dark period and light bars indicate the light period. ZT, Zeitgeber time; ZT0 is defined as the time of lights on. D, Subcellular localization of OsJMJ712-GFP. 35S:OsJMJ712-GFP was transformed into rice protoplasts and visualized by fluorescence microscopy. 35S:RPL1-CFP was used as a nuclear marker. Bright-field and merged images are shown. Scale bars, 10 μm. GFP, Green fluorescent protein; CFP, Cyan fluorescent protein.

Fig. 3. Generation of OsJMJ712 gene-editing mutants and their heading date phenotypes under long-day (LD) and short-day (SD) conditions. A, Schematic of OsJMJ712 gene structure and CRISPR/Cas9-targeted mutation sites. sgRNA targets and protospacer adjacent motif (PAM) sequences are highlighted in blue and gray shading in the wild type (WT), respectively. Red nucleotides indicate insertions; red dashes indicate deletions. Sequence gap lengths are shown in parentheses. B, Partial amino acid sequences of WT and osjmj712 mutants. Red letters highlight predicted altered amino acids. Asterisks (*) indicate stop codons. C, Flowering phenotype of WT and osjmj712 mutants under LD conditions. Scale bar, 10 cm. D and E, Statistical analysis of heading date for WT and osjmj712 mutants under LD (D) and SD (E) conditions. Data are mean ± SD (n = 12). Significant differences were determined by a two-tailed unpaired Student’s t-test (*, P ≤ 0.05; **, P ≤ 0.01).

Fig. 4. Diurnal expression patterns of key flowering-related genes in wild type (WT) and osjmj712 mutants under long-day (LD) and short-day (SD) conditions. Black bars represent the dark period and light bars indicate the light period. ZT, Zeitgeber time; ZT0 is defined as the time of lights on. The Actin gene was used as an internal control. Data are mean ± SD (n = 3). Significant differences were determined by a two-tailed unpaired Student’s t-test (*, P ≤ 0.05; **, P ≤ 0.01).

Fig. 5. OsJMJ712 is an H3K36me3 demethylase. Assay of histone H3 lysine methylation levels in vivo. Western blot analysis of histones isolated from wild type (WT) and osjmj712 mutants under long-day (LD, A) and short-day (SD, B) conditions using the indicated methylation-specific antibodies listed on the left. Anti-H3 was used as a loading control.

Fig. 6. H3K36me3 levels at Ehd1 and RFT1 loci in osjmj712 mutant and wild type (WT). A, Schematic diagrams of Ehd1 and RFT1 gene structures and the fragments detected by chromatin immunoprecipitation-quantitative PCR (ChIP-qPCR). Numbers 1‒4 represent the regions analyzed. B, ChIP-qPCR analysis of H3K36me3 enrichment at the Ehd1 and RFT1 loci. The Actin gene was used as an internal control for normalization. Numbers 1‒4 represent regions 1‒4 in A. IgG was used as a negative control. Data are mean ± SD (n = 3). Asterisks indicate significant differences between osjmj712 mutant and WT (Student’s t-test, *, P ≤ 0.05; **, P ≤ 0.01).

Fig. 7. Rhythmic expression of OsLHY in osjmj712 mutant and wild type (WT). A and B, Diurnal expression patterns of OsLHY in WT and osjmj712 mutant under long-day (LD, A) and short-day (SD, B) conditions. C and D, Free-running rhythms of OsLHY expression in WT and osjmj712 mutant under constant light (LL, C) and dark (DD, D) conditions. Black bars represent the dark period and light bars indicate the light period. ZT, Zeitgeber time; ZT0 is defined as the time of lights on. The Actin gene was used as an internal control. Data are mean ± SD (n = 3). Significant differences were determined by a two-tailed unpaired Student’s t-test (*, P ≤ 0.05; **, P ≤ 0.01).

Fig. 8. OsLHY directly binds to OsJMJ712 promoter. A, Transcriptional activity of the OsJMJ712 promoter was assessed by dual-luciferase assays in Nicotiana benthamiana leaves. REN, Renilla luciferase; LUC, Firefly luciferase. Data are mean ± SD (n = 5). Significant differences were determined by Student’s t-test (**, P ≤ 0.01). B, Interaction between OsLHY and OsJMJ712 promoter detected by yeast one-hybrid (Y1H) assays. pHis2‐pOsJMJ712 and OsLHY‐Rec2 were co-transformed into yeast cells, and transformants were grown on SD/‐Trp/‐Leu/‐His (SD/-TLH) medium and SD/‐TLH medium supplemented with 50 mmol/L 3‐amino‐1,2,4‐triazole (3‐AT). C, Binding of OsLHY to evening element (EE) sequences in OsJMJ712 promoter detected by electrophoretic mobility shift assay (EMSA). EE fragments were 5ʹ-biotin-labelled as the Bio‐probe. Unlabelled probe was used as competitor. Purified GST‐OsLHY fusion protein was incubated with the designed probes. The red arrowhead indicates the protein‐DNA complex, and the black arrowhead indicates the free probe.

Fig. 9. Proposed model for the role of OsJMJ712 in regulating flowering in rice. OsJMJ712 is a component of the circadian clock, and its expression is directly suppressed by OsLHY. Loss of function of OsJMJ712 increases OsLHY expression, indicating a potential negative transcriptional feedback loop. Acting as an H3K36me3 demethylase, OsJMJ712 removes H3K36me3 marks at the Ehd1 and RFT1 loci, thereby repressing their transcription and ultimately delaying flowering.