Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2025 Jun 24;48(10):7392–7407. doi: 10.1111/pce.70042

Rice Zinc‐Finger Homeodomain 1 (OsZHD1) Finetunes Flowering Time by Modulating the Photoperiodic Response

Yehyun Yim 1, Hyeryung Yoon 1,2, Sung‐Hwan Cho 1,2, Lae‐Hyeon Cho 3, Kiyoon Kang 4,✉, Nam‐Chon Paek 1,2,✉
PMCID: PMC12415411  PMID: 40556134

ABSTRACT

Plants exhibit photoperiod sensitivity (PS), which influences flowering time in response to changes in daylength, allowing cereal crops to optimise seed production at different latitudes. Rice is a facultative short‐day (SD) plant, flowering early under SD and late under long‐day (LD) conditions. This study identifies the rice (Oryza sativa) Zinc‐finger Homeodomain 1 (OsZHD1) as a new photoperiod regulator that modulates flowering time in response to daylength. Compared to the parental japonica cultivar, oszhd1 knockout mutants exhibited enhanced PS, resulting in flowering that occurred earlier under SD and later under LD. Conversely, transgenic rice overexpressing OsZHD1 showed a reduction in PS, resulting in delayed flowering under SD and accelerated flowering under LD. RT‐qPCR analysis revealed that OsZHD1 plays a role in the Ehd1‐Hd3a/RFT1 pathway for flowering induction. We found that OsZHD1 interacts with Hd1 and OsPRR37, but not with Ghd7 or DTH8. In addition, OsZHD1 binds directly to the Ehd1 promoter, potentially downregulating Ehd1 in SD and upregulating it under LD to reduce PS. SNP analysis showed that rice varieties with the japonica‐type OsZHD1a allele flowered earlier than those with the indica‐type OsZHD1b allele. Taken together, these results highlight a critical role of OsZHD1 in finetuning flowering time at different latitudes.

Keywords: flowering time, natural variation, OsZHD1, Rice

Summary statement

  • •

    Rice is a photoperiod‐sensitive plant that flowers early in short days and late in long days.

  • •

    OsZHD1 acts to reduce photoperiod sensitivity in flowering. It appears that natural variation of OsZHD1 alleles contributes to optimal flowering time for rice cultivation at different latitudes.

1. Introduction

Flowering represents a pivotal stage in the life cycle of sessile plants, making the transition from vegetative to reproductive growth (Cho et al. 2017; Srikanth and Schmid 2011). Flowering time, referred to as heading date, is controlled by a complex interaction between environmental factors, such as light quality, daylength, and temperature, and internal mechanisms, such as photoreceptor‐mediated light signalling, flowering up‐ and down‐regulators, and positive and negative phytohormone regulation (Andrés and Coupland 2012; Song et al. 2015; Shi et al. 2023). Among these factors, photoperiod during the growing season plays a pivotal role as it is responsible for sensing and responding to ambient changes in daylength (Hayama and Coupland 2004; Yano et al. 2001). Based on their photoperiod sensitivity (PS), plants are classified into long‐day (LD), short‐day (SD), and day‐neutral types.

Cultivated rice (Oryza sativa) is a facultative SD plant that flowers early in SD and late in LD. It is derived from the wild rice Oryza rufipogon, which is native to tropical and subtropical Asia (Garris et al. 2005). O. rufipogon exhibits strong PS, with flowering being triggered in response to SD conditions. During the domestication process, natural and artificial selection on alleles controlling floral regulatory pathways allowed for the reduction of PS in cultivated rice (Khush 1997; Zong et al. 2021). As a result, the cultivation area of rice varieties has expanded to higher latitudes, ranging from 40° S to 53° N (Koo et al. 2013; Shrestha et al. 2014; Li et al. 2015). This expansion highlights the adaptive flexibility of rice, allowing it to grow in a wide range of environments beyond its native tropical and subtropical habitats (Osnato 2023).

Rice flowering is primarily regulated by the florigen genes Heading Date 3a (Hd3a) and RICE FLOWERING LOCUS T1 (RFT1), which are homologous to Arabidopsis FT. While rice requires both Hd3a and RFT1 to induce flowering under SD, only RFT1 is sufficient to induce flowering under LD (Komiya et al. 2008; Komiya et al. 2009). Florigen genes are expressed in leaf vascular tissues and transported to the shoot apical meristem (SAM) (Komiya et al. 2008; Komiya et al. 2009; Tamaki et al. 2007). Subsequently, the florigen proteins activate the floral meristem identity genes OsMADS14 and OsMADS15, which are essential for flower development (Taoka et al. 2011). Transcription of the florigen genes is regulated by the Heading date 1 (Hd1), a transcription factor, and Early heading date 1 (Ehd1), a transcriptional regulator that integrate photoperiod signals to regulate flowering time. The conserved rice flowering regulator Hd1, which is homologous to the Arabidopsis CONSTANS (CO), regulates rice flowering through the Hd1‐Ehd1‐Hd3a regulatory module (Yano et al. 2000). Ehd1 is a rice‐specific regulator that acts as upstream of florigen genes and mediates the Ghd7 (for grain number, plant height, and heading date 7)‐Ehd1‐Hd3a/RFT1 regulatory module (Shim and Jang 2020; Yano et al. 2000; Doi et al. 2004; Zhao et al. 2015). Hd1 promotes rice flowering under SD and represses it under LD (Hayama et al. 2003; Yano et al. 2001). The rice genome encodes several flowering repressors in response to photoperiod. Under SD, OsGI (Hayama et al. 2003), Ehd2 (Matsubara et al. 2008), Ehd4 (Gao et al. 2013), OsMADS50 (Lee et al. 2004) and OsMADS51 (Kim et al. 2007) acts as positive regulators of Ehd1, while OsCOL4 (Lee et al. 2010) and OsCOL10 (Tan et al. 2016) function to downregulate Ehd1. Under LD, Ghd7 (Xue et al. 2008), Days to heading 8 (DTH8; Wei et al. 2010), and Pseudo‐Response Regulator 37 (OsPRR37; Koo et al. 2013) function as LD‐specific flowering repressors by downregulating Ehd1 expression. Ghd7 is a CCT domain‐containing TF that acts as a primary target of phytochrome signalling pathways that control flowering in response to daylength (Xue et al. 2008). DTH8 encodes a putative HAP3 subunit of the NF‐YB transcription factor, a protein complex known for its role in regulating gene expression in response to environmental cues (Wei et al. 2010). OsPRR37, a pseudo‐response regulator, has an N‐terminal pseudo‐receiver (PR) domain and a C‐terminal CCT domain, which are involved in photoperiodic signalling (Koo et al. 2013). Hd1, Ghd7, OsPRR37, and DTH8 form an intricate regulatory network by interacting with each other and collectively downregulating the Ehd1‐Hd3a/RFT1 pathway (Cai et al. 2019; Du et al. 2017; Gao et al. 2014; Koo et al. 2013; Li et al. 2015; Sun et al. 2022; Zong et al. 2021). These core photoperiod regulators delay flowering under non‐inductive LD conditions, making rice a facultative SD plant.

Zinc finger‐homeodomain (ZF‐HD) TFs consist of a zinc finger (ZF) domain at the N‐terminus and a homeodomain (HD) at the C‐terminus (Windhövel et al. 2001). The HD domain serves as a DNA‐binding domain that regulates the expression of downstream genes. The ZF domain contributes to both DNA binding and interaction with regulatory proteins (Ariel et al. 2007; Ciftci‐Yilmaz and Mittler 2008). ZF‐HD TFs have been identified in a wide range of plant species and have been shown to play crucial roles in processes such as photosynthesis, vegetative growth, stress resistance, and flower development (Park et al. 2007; Seo et al. 2011; Street et al. 2008; Tran et al. 2007; Xu et al. 2014; Zhao et al. 2018). Among the 11 ZF‐HD genes in the rice genome, the functions of OsZHD1 and OsZHD2 have been characterised. Overexpression of OsZHD1 leads to abaxial leaf curling by altering the distribution of bulliform cells. Thus, transgenic rice plants overexpressing OsZHD1 exhibit abaxially curled and drooping leaves (Xu et al. 2014). In addition, OsZHD1 and its homologue OsZHD2 promote root growth by upregulating ethylene and auxin biosynthesis genes. The functional redundancy of OsZHD1 with OsZHD2 results in defective root development in the oszhd1 oszhd2 double mutant (Yoon et al. 2020). OsZHD1 and OsZHD2 are also involved in the floral meristem transition at the SAM (Yoon et al. 2023).

In this study, we identified a potential role for OsZHD1 as a newovel photoperiod regulator that affects flowering time in response to daylength. Compared to the parental japonica rice lines, we found that null mutation of OsZHD1 activated the Ehd1‐Hd3a/RFT1 pathway under SD and repressed it under LD, resulting in early flowering under SD and delayed flowering under LD. Overexpression of OsZHD1 reduced PS, resulting in late flowering under SD and early flowering under LD. OsZHD1 regulated the expression of Ehd1 by directly binding to its promoter region. Among the core photoperiod regulators, OsZHD1 interacts with Hd1 and OsPRR37, but not with Ghd7 and DTH8. It is speculated that natural variation in OsZHD1 has allowed for the regional expansion of rice cultivation to higher latitudes. Taken together, we propose that OsZHD1 functions to reduce PS to adjust flowering time, which contributes to ensuring an adequate vegetative growth period for reproductive success at different latitudes.

2. Materials and Methods

2.1. Plant Materials and Growth Conditions

The rice japonica cultivars “Dongjin” (DJ) and “Hwayoung” (HY) and two independent T‐DNA insertional knockout mutant lines, oszhd1‐1 and oszhd1‐2, were used in this study. The oszhd1‐1 T‐DNA insertion mutant (PFG_3A‐60050), generated in the DJ background, was obtained from the Crop Biotech Institute of Kyung Hee University, South Korea (Jeon et al. 2000). The oszhd1‐2 T‐DNA insertion mutant (PFG_1C‐10508), generated in the HY background, was described in Yoon et al. (2023). Rice plants used for flowering time observations were grown in the experimental field under natural long‐day (NLD) conditions (≥ 14 h of daylight) in Suwon, South Korea (37° N latitude). Plants used for diurnal and weekly expression analysis of flowering time genes were grown in chambers under either controlled SD (10‐h light at 30°C/14‐h dark at 24°C) or LD conditions (14.5‐h light at 30°C/9.5‐h dark at 24°C). Flowering time was recorded as the number of days after germination to the day of first panicle emergence from the main culm for each individual plant.

2.2. Vector Construction and Rice Transformation

To generate OsZHD1 overexpression plants in the DJ background, the coding sequence (CDS) of OsZHD1 (LOC_Os09g29130) was amplified and ligated into the TA cloning vector pCR8/GW/TOPO (Invitrogen, USA). It was then inserted into the pMDC32 Gateway compatible binary vector carrying the 35S promoter through an LR recombination reaction (Curtis and Grossniklaus 2003). The 35S:OsZHD1 construct was introduced into Agrobacterium tumefaciens strain LBA4404 by the freeze‐thaw method (An et al. 1989) and subsequently transformed into calli derived from mature embryos of DJ seeds. Transformed calli were selected on 1/2 Murashige and Skoog medium containing 40 mg L− 1 hygromycin, and the selected lines, OsZHD1‐OE1 and OsZHD1‐OE2, were confirmed by RT‐qPCR analysis. Homozygous plants of the T2 generation were used for functional analysis. Gene‐specific primers used in this study are listed in Table S1.

2.3. RNA Extraction and Quantification of Gene Expression

Total RNA was extracted from rice leaf blades using the MG Total RNA Extraction Kit (MGmed, Seoul, South Korea) according to the manufacturer's instructions. First‐strand cDNAs were synthesised from 2 μg of total RNA using M‐MLV reverse transcriptase (Promega, Madison, WI, USA), 10 mM dNTPs, 5× reaction buffer, and oligo(dT)15 primers. The cDNAs were diluted with nuclease‐free water to a final volume of 100 μL. Real‐time quantitative PCR (qPCR) analysis was performed on a LightCycler 480 (Roche, Switzerland). Each reaction contained 2 μL of diluted first‐strand cDNAs, 10 μL of 2× GoTaq qPCR Master Mix (Promega, Madison, Wisconsin, USA), and 1 μL of 10 pM gene‐specific primers, in a final volume of 20 μL. PCR cycling conditions for all reactions were as follows: 95°C for 2 min, followed by 45 cycles of UBQ95°C for 5 s, 59°C for 15 s, and 72°C for 10 s. OsUBQ5 (LOC_Os01g22490) was used as an internal control (Jain et al. 2006).

2.4. Subcellular Localisation

To analyze the subcellular localisation of OsZHD1 in rice protoplasts, the OsZHD1 CDS was amplified from the first‐stranded cDNAs using gene‐specific primers and cloned into the pGA3651‐GFP vector containing the maize Ubiquitin (Ubi) promoter and the green fluorescent protein gene (GFP). The pGA3651‐GFP (control) and pGA3651‐OsZHD1 plasmids were then transformed into rice protoplasts using polyethylene glycol (PEG) solution [40% (w/v) PEG 4000, 100 mM CaCl2, and 2 mM MES]. After incubation for 14–16 h at 25°C in the dark, GFP signals were detected using a fluorescence microscope (Zeiss Axioskop 2, Carl Zeiss, Germany). GFP fluorescence was excited at 488 nm and detected at emission wavelengths of 505–530 nm.

2.5. Yeast Two‐Hybrid (Y2H) and Three‐Hybrid (Y3H) Assays

Y2H assays were performed using the Matchmaker GAL4 system (Clontech, CA, USA). The full‐length or truncated CDSs of OsZHD1 were cloned into the bait vector pGBKT7. Similarly, the full‐length or truncated CDSs of Hd1, OsPRR37, DTH8, and Ghd7 were individually ligated into the prey vector pGADT7. Each bait‐prey plasmid pair was co‐transformed into the yeast strain AH109 using the polyethylene glycol/lithium acetate (PEG/LiAc) method. Transformants were selected on the synthetic defined (sd)/‐Leu/‐Trp medium and subsequently screened for protein‐protein interactions on the sd/‐Leu/‐Trp/‐His medium supplemented with 3‐amino‐1,2,4‐triazole (3‐AT). For Y3H assays, the full‐length CDS of Hd1 was cloned into multicloning site I (MCSI) of the pBridge vector (Clontech, CA, USA), and that of OsZHD1 was cloned into MCSII of the same vector. The full‐length CDSs of Ghd7 and DTH8 were cloned individually into the prey vector pGADT7. The respective plasmid combinations were co‐transformed into the yeast strain AH109 and the transformants were sequentially selected on the sd/‐Leu/‐Trp, sd/‐Leu/‐Trp/‐His/+3‐AT, and sd/‐Leu/‐Trp/‐Met medium to assess ternary protein‐protein interactions.

2.6. Bimolecular Fluorescence Complementation (BIFC) Assays

The BiFC assays were performed according to the protocol described by Cho et al. (2022). The CDSs of OsZHD1, Hd1, OsPRR37, DTH8, and Ghd7 were amplified using the DJ cDNAs and cloned into the pAM‐PAT‐35S:YFPn (YN) and pAM‐PAT‐35S:YFPc (YC) vectors to generate OsZHD1‐YN, OsZHD1‐YC, Hd1‐YN, OsPRR37‐YN, DTH8‐YC, and Ghd7‐YN. The constructs were separately transformed into the protoplasts isolated from the Oc cells using a 40% PEG solution. After 12‐h incubation in the dark at 25°C, YFP signals were detected at excitation and emission wavelengths of 488 nm and 505‐530 nm, respectively, using an AxioSkop 2 microscope (Carl Zeiss, Germany). DAPI was used as a nuclear localisation marker. Empty vectors, pAM‐PAT‐35S:YFPn (YN) and pAM‐PAT‐35S:YFPc (YC), were used as negative controls.

2.7. Luciferase Complementation Imaging (LCI) Assays

The LCI assays were performed as previously described by Cho et al. (2022). The CDSs of OsZHD1, Hd1, OsPRR37 were amplified and cloned into the pCAMBIA 1300‐cLUC and pCAMBIA 1300‐nLUC vectors. Each resulting vector was transformed into Agrobacterium tumefaciens strain GV3101. The Cultured cells were suspended in an infiltration buffer (10 mM MES, 10 mM MgCl2, and 150 mM acetosyringone) and incubated for 2 h in the dark at room temperature. For transient expression, the abaxial sides of Nicotiana benthamiana leaves were infiltrated with equal volumes of Agrobacterium cells carrying the construct vectors. After 2 days, the infiltrated abaxial side of the leaves were sprayed with 1 mM d‐luciferin (Goldbio, Olivette, USA) and imaged using a chemiluminescence imaging system (FUSION FX7; Vilber, Germany). ImageJ (ver 1.51) was used to calculate relative luminescence unit (RLU) intensities. Empty vectors, pCAMBIA 1300‐cLUC and pCAMBIA 1300‐nLUC, were used as negative controls.

2.8. Yeast One‐Hybrid (Y1H) Assays

The OsZHD1 CDS was cloned into the prey vector pB42AD for Y1H assays. The promoter region of Ehd1 (−2068 to +19 bp relative to the ATG start codon) was divided into 4 fragments, which were amplified by genomic PCR and subsequently inserted into the pLaczi2μ vector as bait. These plasmid constructs were co‐transformed into the yeast strain EGY48 using the PEG/LiAc method. Transformants were selected on sd/‐Trp/‐Ura medium, and positive colonies were plated on X‐Gal (5‐bromo‐4‐chloro‐3‐indolyl‐beta‐d‐galacto‐pyranoside) containing the sd/‐Trp/‐Ura medium for blue colour development.

2.9. Chromatin Immunoprecipitation (ChIP) Assays

For ChIP assays, the pUbi:MYC and pUbi:OsZHD1‐MYC constructs were transfected into rice protoplasts as previously described (Saleh et al. 2008; Zhang et al. 2011). Briefly, the transfected protoplasts were cross‐linked with 1% (v/v) formaldehyde in W5 solution [154 mM NaCl, 125 mM CaCl2, 5 mM KCl, and 2 mM MES (pH 5.8)] for 30 min under vacuum. Subsequently, nuclei were subsequently isolated, lysed, and chromatin complexes were purified and sonicated as described previously (Saleh et al. 2008). DNA fragmentation was performed using a VCX500 ultrasonic processor (Sonics, CT, USA). Immunoprecipitation was performed using an anti‐MYC monoclonal antibody (cat. no. 9B11; Cell signalling Technology, Massachusetts, USA) and protein A agarose beads (Merk Millipore, USA). DNA recovered from the agarose beads was purified using a QIAquick Purification Kit (Qiagen, Weinbunt, Germany). qPCR analysis was conducted on a LightCycler 2.0 instrument (Roche Diagnostics, Basel, Switzerland) under the following conditions: 95°C for 5 s, 59°C for 15 s and 72°C for 10 s. ACTIN (LOC_Os03g50885) was used as an internal control.

2.10. Dual Luciferase (LUC) Assays in Rice Protoplasts

For the dual LUC assays, the 2.5 kb promoter region of Ehd1 upstream of the start codon was amplified and inserted into the pJD301 vector to generate the reporter plasmid. The full‐length CDS of OsZHD1 was cloned into the pGA3817 vector carrying a 6X MYC tag to generate the effector plasmid. The effector, reporter and renilla luciferase (REN) plasmids were co‐transformed into rice protoplasts and incubated in darkness at 25°C for 14 h. The protoplasts were then collected and lysed in a passive lysis buffer provided with the Dual‐Luciferase Reporter Gene Assay Kit (PR‐E1910, Promega, USA). LUC activity was measured using a GloMax 20/20 luminometer (Promega, USA) according to the manufacturer's instructions. REN driven by the Ubi promoter was used as an internal control. Relative LUC activity was calculated as the ratio of LUC/REN for each sample.

2.11. Analysis of Single‐Nucleotide Polymorphism (SNP)

To investigate the natural variation of OsZHD1 in cultivated rice and wild rice (Oryza rufipogon), SNPs in the OsZHD1 CDS were identified using publicly available databases. The SNPs of O. sativa were analyzed using data from the 3 K Rice Genome Project available through the International Rice Genebank Collection Information System (IRGCIS) database (Mansueto et al. 2017), while the SNPs of O. rufipogon were analyzed using data from OryzaGenome Release 2.0 provided by the National Institute of Genetics under the National Bioresource Project (Huang et al. 2012; https://shigen.nig.ac.jp/rice/oryzabase/). The japonica cultivar Nipponbare allele was used as the reference genome in this SNP analysis. Flowering time data associated with the SNPs were obtained from the IRGCIS database.

2.12. Statistical Analysis

Statistical analyses were performed using the GraphPad Prism version 10 for Windows (GraphPad Software). A two‐tailed Student's t‐test was used for pairwise comparisons, and one‐way ANOVA followed by Tukey's HSD test was used for comparisons among three or more groups. p < 0.05 were considered statistically significant. Data are presented as the mean ± standard deviation of at least three biological replicates. Additional details, including statistical methods and sample sizes, are provided in the figure legends.

3. Results

3.1. Expression Patterns of OsZHD1

To investigate the spatial expression of OsZHD1, the rice organ tissues were collected from the japonica rice ‘Dongjin’ (DJ) at 63 days after sowing (DAS, tillering stage) and 118 DAS (panicle initiation stage) grown in the rice paddy field under NLD. RT‐qPCR analysis revealed that OsZHD1 transcripts were detected predominantly in the leaf tissues, including leaf sheath, leaf blade, and flag leaf (Figure 1a). In addition, OsZHD1 expression was diurnally rhythmic, peaking at dawn (ZT1, zeitgeber time 1) under SD and LD (Figure 1b,c). We further investigated the expression levels of OsZHD1 in fully expanded second leaves of DJ until flowering under SD and LD. It showed that OsZHD1 expression was highest at 3 weeks after sowing (WAS) under SD and at 7 WAS under LD (Figure 1d,e).

Figure 1.

Figure 1

Open in a new tab

Expression profiles of OsZHD1. (a) OsZHD1 expression was measured in various tissues of the japonica rice cultivar ‘Dongjin (DJ)’ grown in the rice paddy field under NLD (≥ 14 h of daylight) conditions. Vegetative and reproductive stages of DJ plants were grown for 63 DAS and 118 DAS, respectively. Total RNA was isolated from the DJ plant tissues at ZT1 (1 h after dawn). RT, root; LS, leaf sheath; LB, leaf blade; TB, tiller base; FL, flag leaf; IN, internode; and PN, panicle; NO, node. (b, c) Diurnal expression patterns of OsZHD1 were measured every 3 h in the fully developed second leaves of DJ plants grown for 21 DAS under SD (10‐h light/14‐h dark) (b) and for 56 DAS under LD (14.5‐h light/9.5‐h dark) conditions. (c). Black and grey boxes indicate the dark period. (d, e) Weekly expression patterns OsZHD1 were measured in the fully developed second leaves of DJ plants. Leaf samples were collected at ZT1 every week starting from 2 WAS under SD (d) or every 2 weeks starting from 3 WAS under LD (e). Relative transcript levels were determined by RT‐qPCR and normalised to OsUBQ5 (LOC_Os01g22490). Means and standard deviations were taken from 30 plants. NLD, natural long day; SD, short day; LD, long day; ZT, zeitgeber time; DAS, days after sowing; WAS, weeks after sowing.

Previous studies have shown that OsZHD1 localises to the nucleus (Xu et al. 2014), and forms homodimers (Figueiredo et al. 2012). We also found that OsZHD1‐GFP fusion proteins localised exclusively to the nucleus in rice protoplasts (Figure S1a). Furthermore, split LUC assays in the N. benthamiana leaves co‐expressing OsZHD1N and OsZHD1C (N‐ and C‐terminal LUC fusions of OsZHD1) showed a strong luciferase signal (Figure S1b), confirming the homodimerization of OsZHD1.

3.2. OsZHD1 Acts to Reduce PS in Rice Flowering Time Regulation

To determine the possible function of OsZHD1 in flowering time regulation, we obtained two independent T‐DNA insertion mutants (PFG_3A‐60050; oszhd1‐1 and PFG_1C‐03234; oszhd1‐2) from the RiceGE database (http://signal.salk.edu/cgibin/RiceGE). We confirmed that the T‐DNA fragments in oszhd1‐1 and oszhd1‐2 mutants were inserted into the OsZHD1 genomic DNA of their respective parental japonica rice, DJ and ‘Hwayoung’ (HY) (Figure 2a, Figure S2a). Next, we examined the effect of T‐DNA insertion on OsZHD1 expression in the rice plants at 60 DAS grown under NLD. Semi‐quantitative RT‐PCR analysis showed that OsZHD1 transcripts were undetectable in both oszhd1‐1 and oszhd1‐2 mutants, indicating that they are knockout mutants (Figure 2b, Figure S2b).

Figure 2.

Figure 2

Open in a new tab

Knockout mutation of OsZHD1 reduces photoperiod sensitivity. (a) Schematic representation of the T‐DNA insertion oszhd1‐1 mutant. Red triangle indicates the T‐DNA insertion site in the genomic DNA of OsZHD1 that is composed of a single exon. Black and red alphabets represent OsZHD1 and inserted T‐DNA fragment nucleotides, respectively. Black and white boxes represent one exon and untranslated regions of OsZHD1, respectively. (b) Semiquantitative RT‐PCR analysis of oszhd1‐1. The parental japonica rice (DJ) and oszhd1‐1 were grown for 60 DAS under NLD. Total RNA was isolated from the second leaves from the top at ZT1. OsUBQ5 (LOC_Os01g22490) was used as a loading control. (c–e) Flowering phenotypes of DJ (left) and oszhd1‐1 (right) grown under SD and LD in the growth chambers (c, d), and under NLD in the rice paddy field (e). White arrowheads indicate emerging panicles. Enlarged panicles were highlighted by dotted circles. White bars = 7 cm. (f) Days to heading of DJ and oszhd1‐1 under SD, LD, and NLD. Means and standard deviations were taken from 30 plants. Asterisks indicate significant difference between DJ and oszhd1‐1 (Student's t‐test, ***p < 0.001). (g) Comparison of leaf emergence rates in DJ (open circles) and oszhd1‐1 (black circles) under SD. Means and standard deviations were taken from 30 plants. Black arrow indicates the mean flowering time of oszhd1‐1.

To investigate the effect of oszhd1 mutation on flowering time, oszhd1 mutants (oszhd1‐1 and oszhd1‐2) and their parental lines (DJ and HY) were grown in the paddy field under NLD as well as in the growth chambers under SD and LD. The oszhd1 mutants flowered approximately 11 days earlier than the parental lines under SD, whereas they flowered approximately 11 days later than the parental lines under both NLD and LD (Figure 2c–f, Figure S2c–f). In addition, to evaluate the possible effect of oszhd1 mutation on the vegetative growth rate, we measured the leaf emergence rate of oszhd1 mutants grown in the growth chamber under SD. The results showed no significant difference in the leaf emergence rate between the parental lines and oszhd1 mutants, indicating that oszhd1 mutation does not affect the vegetative growth rate under SD (Figure 2g, Figure S2g). To further confirm the regulatory role of OsZHD1 in photoperiodic flowering, we generated two transgenic rice lines overexpressing OsZHD1, OsZHD1‐OE1 and OsZHD1‐OE2 (collectively referred to as OsZHD1‐OEs). Total RNA was isolated from the second leaves of DJ and OsZHD1‐OEs at 15 DAS grown under SD. Semi‐quantitative RT‐PCR analysis showed that OsZHD1 transcripts were more abundant in OsZHD1‐OEs than in DJ (Figure S3a). OsZHD1‐OEs flowered later under SD and earlier under LD compared to DJ (Figure S3b–d). These observations suggest that oszhd1 mutation enhances PS, whereas OsZHD1 overexpression reduces PS in rice flowering.

3.3. OsZHD1 Affects the Expression of Ehd1, Hd3a and RFT1

To investigate the underlying mechanisms of OsZHD1, we examined the diurnal expression of flowering time genes. DJ and oszhd1‐1 plants were grown in the growth chamber and their leaf tissues were harvested every 3 h for 24 h at 21 DAS and 56 DAS under SD and LD, respectively. RT‐qPCR analysis showed that in oszhd1‐1 mutants, the transcript levels of the florigen genes RFT1 and Hd3a were higher at ZT1 under SD and lower at ZT4 under LD compared to DJ (Figure 3a–d). We then analyzed the expression levels of Ehd1, an upstream regulator of florigen genes, and Hd1, a key PS regulator. Interestingly, the oszhd1‐1 mutation increased the expression of Ehd1 under SD, which promoted the expression of Hd3a and RFT1, leading to earlier flowering than DJ. In contrast, the oszhd1‐1 mutation decreased Ehd1 expression under LD, resulting in reduced expression of Hd3a and RFT1, and delayed flowering compared to DJ (Figure 3e,f). However, Hd1 expression remained similar between DJ and oszhd1‐1 under both SD and LD (Figure 3g,h). Additional diurnal expression analysis revealed that OsZHD1 overexpression (OsZHD1‐OE1) downregulated the expression of Ehd1, Hd3a, and RFT1 under SD, whereas it upregulated their expression under LD (Figure S4). To further determine how OsZHD1 regulates Ehd1 expression, we examined the expression levels of several upstream regulators of Ehd1, such as OsGI, Ehd2, Ehd4, OsMADS50, OsMADS51, OsCOL4, OsCOL10, DTH8, Ghd7, and OsPRR37 under SD and LD. RT‐qPCR results showed no significant differences in the diurnal expression patterns of the upstream regulatory genes between DJ and oszhd1‐1 under SD and LD (Figure S5). Taken together, these results strongly suggest that OsZHD1 functions directly in the Ehd1‐Hd3a/RFT1 pathway for photoperiodic flowering in rice.

Figure 3.

Figure 3

Open in a new tab

Diurnal expression patterns of flowering time genes in oszhd1‐1. Expression patterns of RFT1 (a, b), Hd3a (c, d), Ehd1 (e, f), and Hd1 (g, h) in DJ (open circles) and oszhd1‐1 (black circles) grown in the growth chambers for 21 DAS under SD (a, c, e, g) or 56 DAS under LD (b, d, f, h). The fully developed second leaves from the top were sampled every 3 h. Black and grey boxes indicate the dark period. Means and standard deviations were taken from three biological replicates.

3.4. OsZHD1 Regulates Ehd1 Expression in a Photoperiod‐Dependent Manner

To determine whether OsZHD1 binds to the Ehd1 promoter, we performed Y1H and ChIP assays (Figure 4a–c). We searched for the potential OsZHD1 binding motifs using The Plant Promoter Analysis Navigator (PlantPAN 4.0, https://plantpan.itps.ncku.edu.tw/plantpan4/index.html). This analysis predicted several TAATC motifs that are evenly distributed throughout the promoter sequences of Ehd1 (Figure 4a). Y1H assays showed that OsZHD1 binds to the pEhd1‐D promoter region of Ehd1 (Figure 4b). ChIP assays confirmed that OsZHD1 binds to the amplicon‐P3 and ‐P4 regions of the Ehd1 promoter, which overlaps with the pEhd1‐D region identified in the Y1H assays (Figure 4c). To determine whether photoperiods affect the expression of flowering time genes in rice protoplasts, we measured the transcript levels of Ehd1, Hd3a, and RFT1 in rice protoplasts isolated from DJ seedlings grown for 10 days under SD and LD. RT‐qPCR analysis revealed that the expression levels of Ehd1, Hd3a, and RFT1 were significantly higher under SD compared to under LD (Figure S6). Next, to investigate whether OsZHD1 regulates Ehd1 expression in a photoperiod‐dependent manner, we performed LUC assays using rice protoplasts isolated from DJ seedlings grown for 10 days under SD and LD. As a result, LUC activity was significantly decreased in the SD‐grown rice protoplasts cotransfected with the pEhd1:LUC and pUbi:OsZHD1‐MYC plasmids compared to those cotransfected with the pEhd1:LUC and pUbi:MYC control plasmids (Figure 4d,e). In contrast, LUC activity was significantly increased when these plasmids were cotransfected into the LD‐grown rice protoplasts (Figure 4d,f). In addition, we performed further LUC assays to investigate whether deletion of the pEhd1‐D region in the Ehd1 promoter (pEhd1‐ΔD) affects the binding affinity of OsZHD1 to the Ehd1 promoter. The results showed that OsZHD1 failed to bind to the Ehd1 promoter region when the pEhd1‐D region was truncated (Figure 4d,g,h). These results suggest that OsZHD1 plays a role in the modulation of PS in the regulation of flowering time in rice.

Figure 4.

Figure 4

Open in a new tab

OsZHD1 regulates Ehd1 transcription. (a) Schematic representation of the Ehd1 promoter fragments used in Y1H (A–D) and ChIP (P1–P4) assays. Orange boxes indicate putative ZF‐HD binding sites (ZF‐HD BS, TAATC) within the Ehd1 promoter. (b, c) Y1H (b) and ChIP (c) assays showed the binding of OsZHD1 to the Ehd1 promoter. (b) Empty bait (AD) and prey (LacZ) were used as negative controls. (c) OsZHD1‐MYC was expressed in rice protoplasts of 10‐day‐old DJ seedlings. Immunoprecipitation was performed using an anti‐MYC antibody. Actin was used as a negative control. (d) Schematic representation of the constructs used in the dual‐LUC reporter assays in rice protoplasts. pEhd1, 2‐kb upstream region of the Ehd1 start codon; pEhd1‐ΔD, pEhd1 promoter truncated by deletion of the D region; OsZHD1, coding region of OsZHD1; rLUC, renilla luciferase gene; pUbi, Ubiquitin promoter; LUC, firefly luciferase gene; NOS, transcription termination sequence of the nopaline synthase gene. (e–h) LUC assays were performed in rice protoplasts isolated from DJ leaves grown under SD (e, g) and LD (f, h) for 10 days. (c, e–h) Means and standard deviations were taken from three biological replicates. (c) Statistical significance was determined by one‐way ANOVA followed by Tukey's HSD test. Different letters indicate statistically significant differences among groups at p < 0.05. (e–h) Asterisks indicate significant difference between rice protoplasts transfected with the pUbi:MYC and pUbi:OsZHD1‐MYC plasmids (Student's t‐test, ***p < 0.001). ns, not significant. [Color figure can be viewed at wileyonlinelibrary.com]

3.5. OsZHD1 Interacts With Hd1 and OsPRR37 Among the Core Photoperiod Regulators

To determine whether OsZHD1 interacts with the core photoperiod regulators, we performed Y2H, BiFC, and LCI assays. In Y2H assays, yeast colony growth on selective media indicated that OsZHD1 interacts with Hd1 and OsPRR37, but not with DHT8 or Ghd7 (Figure 5a,b). To investigate whether conserved domains of Hd1, OsPRR37, and OsZHD1 mediate the physical interaction between these proteins, we performed additional Y2H assays using the partial regions containing their respective conserved domains (Figure 5a). The results showed that OsZHD1‐1 (1–180 aa), containing the ZF‐HD domain, interacts with Hd1‐2 (209–407 aa), containing the CO, CO‐like, and TOC1 (CCT) domain. In addition, OsZHD1‐2 (181–279 aa), containing the homeodomain, interacts with OsPRR37‐1 (1–208 aa), containing the pseudo‐receiver (PR) domain (Figure 5b). In BiFC assays using rice protoplasts, YFP signals were exclusively observed in the nuclei of protoplasts cotransfected with OsZHD1‐YC and Hd1‐YN, or OsZHD1‐YC and OsPRR37‐YN, but not with DTH8‐YC and OsZHD1‐YN, or OsZHD1‐YC and Ghd7‐YN (Figure S7a). In LCI assays, the LUC signals were detected when OsZHD1N (N‐terminal LUC fusion) and Hd1C (C‐terminal LUC fusion) were co‐expressed (Figure S7b), and when OsPRR37N and OsZHD1C were co‐expressed in the N. benthamiana leaves (Figure S7c). To investigate whether OsZHD1 affects the formation of Hd1‐Ghd7 or Hd1‐DTH8 complexes, we performed Y3H assays. The results suggested that OsZHD1 interferes with the formation of both Hd1‐Ghd7 and Hd1‐DTH8 complexes (Figure S8). Taken together, these results demonstrate that OsZHD1 interacts with Hd1 and/or OsPRR37.

Figure 5.

Figure 5

Open in a new tab

OsZHD1 interacts with Hd1 and OsPRR37. (a) Schematic diagrams of the full‐length and truncated fragments of OsZHD1, Hd1, PRR37, DTH8 and Ghd7 used in Y2H assays. ZF, zinc finger domain; HD, homeodomain; B‐box, B‐box domain; PR, pseudo‐receiver domain; CCT, CO, CO‐like, and TOC1 domain; NF‐YB/HAP3, Nuclear Factor Y subunit B/Heme Activator Protein 3. (b) Y2H assays demonstrated interaction between full‐length or truncated fragments of OsZHD1, Hd1, PRR37, DTH8 and Ghd7. The prey and bait constructs were transformed into the yeast strain AH109. Transformants were grown on synthetic defined (sd) medium lacking leucine and tryptophan (sd/‐LT) for selection, and subsequently on sd medium lacking leucine, tryptophan, and histidine (sd/‐LTH) supplemented with 5 mM 3‐amino‐1,2,4‐trizol (3‐AT) to assess protein interaction. Empty prey (pGADT7) and empty bait (pGBKT7) were used as negative controls. [Color figure can be viewed at wileyonlinelibrary.com]

3.6. Natural Variation in OsZHD1 Contributes to the Regional Adaptation by Reducing PS

Natural variation in key genes controlling flowering time has allowed rice varieties to adapt to a wide range of environmental conditions (Koo et al. 2013; Kwon et al. 2014). Therefore, we analyzed the correlation between single nucleotide polymorphisms (SNPs) identified within the OsZHD1 CDS and flowering time in the rice cultivars using the IRGCIS database. Our analysis identified a nonsynonymous SNP (CGT to TGT), resulting in an amino acid change (Arg to Cys), in 2076 rice varieties, which were categorised into two haplotypes, OsZHD1a and OsZHD1b (Figure 6a). A further study was conducted to determine the association between OsZHD1 haplotypes and flowering time in the rice varieties. The results showed that 569 rice varieties with the OsZHD1a haplotype tended to flower earlier than 1473 varieties with the OsZHD1b haplotype (Figure 6b). Interestingly, the OsZHD1a and OsZHD1b haplotypes were predominantly identified in japonica and indica rice varieties, respectively (Figure 6c). When the geographical distribution of each OsZHD1 haplotype was plotted on a world map, the rice varieties with the OsZHD1a haplotype were predominantly grown at higher latitudes compared to those with the OsZHD1b haplotype (Figure 6d). To determine the effect of natural variation in OsZHD1 on flowering time in relation to latitude, the heading dates of rice varieties grown in the regions with the latitudes of ≥ 30° N and < 30° N were examined. At higher latitudes (≥ 30° N), 123 varieties with the OsZHD1a haplotype have a significantly earlier flowering time than 167 varieties with the OsZHD1b haplotype. The mean flowering time for the OsZHD1a‐containing cultivars was 76.60 days, with a maximum of 111 days and a minimum of 50 days. In contrast, the mean flowering time for the OsZHD1b‐containing cultivars was 81.45 days, with a maximum of 126 days and a minimum of 59 days (Figure 6e). At lower latitudes (0° N ~ 30° N), although the 357 cultivars with the OsZHD1a haplotype (mean, 95.84 days; maximum, 160 days; minimum, 62 days) flowered earlier than the 1153 cultivars with the OsZHD1b haplotype (mean, 99.61 days; maximum, 160 days; minimum, 61 days), the difference in flowering time between OsZHD1a and OsZHD1b was reduced compared to the higher latitudes (≥ 30° N), and the minimum flowering time for the OsZHD1b haplotype was lower than that of OsZHD1a (Figure 6f). In addition, to determine whether OsZHD1a and OsZHD1b haplotypes differ in their ability to bind to the Ehd1 promoter, we performed Y1H assays using the pEhd1‐D region of the Ehd1 promoter (Figure S9a). The results showed that OsZHD1a binds to the Ehd1 promoter region, but OsZHD1b does not (Figure S9b).

Figure 6.

Figure 6

Open in a new tab

Natural variation of OsZHD1 alleles in cultivated rice. (a) A nonsynonymous SNP at OsZHD1 was identified in 3000 rice varieties from the IRGCIS database. This SNP classified the cultivated rice into two haplotypes, OsZHD1a and OsZHD1b. The location of each SNP is indicated in relation to the translation initiation site of the OsZHD1 CDS, starting from the start codon and extending to + 1. The OsZHD1a is the japonica “Nipponbare” allele. Details of the nonsynonymous SNP are provided in Table S2. (b) Violin plot showing DTH in 3000 rice varieties as a function of each OsZHD1 haplotype. The number of varieties for each haplotype is given below each violin plot. Rice varieties carrying OsZHD1a and OsZHD1b are shown as light pink and green light dots, respectively. (c) Distribution of cultivated rice subpopulations for each OsZHD1 haplotype. The subpopulations are divided into 12 types: aro, aromatic; aus, aus; ind1, ind2, ind3, and indx are indica varieties; temp, temperate japonica; subtrop, subtropical japonica; trop, tropical japonica; admix, other unassigned varieties. (d) Geographical distribution of cultivated rice varieties harbouring different OsZHD1 alleles worldwide. Each haplotype is shown in different colours in the Venn diagrams, OsZHD1a (light pink) and OsZHD1b (light green). The size of the circles represents the number of varieties in each country. The dashed line represents the equator. (e, f) Violin plot depicting the DTH for each OsZHD1 haplotype in the regions with latitudes of ≥ 30° N (e) and and < 30° N (f). Rice varieties carrying OsZHD1a and OsZHD1b are shown as light pink and green light dots, respectively. (e) Regions with latitudes of ≥ 30° N include South Korea, China, Japan, and the United States. (f) Regions with latitudes of < 30° N include Bangladesh, Bhutan, Brazil, Cambodia, Colombia, Côte d'Ivoire, India, Indonesia, Laos, Madagascar, Myanmar, Nepal, Nigeria, Pakistan, Philippines, Sri Lanka, Taiwan, China, Thailand, and Vietnam. Details of these regional data are provided in Tables S3,S4. (b, e, f) Asterisks denote a significant difference between the OsZHD1a and OsZHD1b haplotypes (Student's t‐test, ***p < 0.001). DTH, days to heading. [Color figure can be viewed at wileyonlinelibrary.com]

3.7. OsZHD1 Alleles in Domesticated Rice Derived From OrZHD1 in O. rufipogon

Rice cultivars were domesticated from the wild rice (O. rufipogon), and several phylogenetic analyses strongly suggested that the indica and japonica subspecies originated from Or‐I and Or‐III, respectively (Huang et al. 2012). We screened SNPs in the CDS of OrZHD1 from the OryzaGenome database (https://shigen.nig.ac.jp/rice/oryzabase/). A nonsynonymous SNP of OrZHD1, causing a change in the 46th amino acid from arginine to cysteine (R46C), was identified in 446 O. rufipogon accessions. This SNP is identical to the SNP found in O. sativa and was categorised into two haplotypes, OrZHD1a and OrZHD1b (Figure 7a). We investigated the distribution of the OrZHD1 haplotypes in the O. rufipogon subspecies. The results showed that, while 51% of the subspecies with the OrZHD1a haplotype belonged to the Or‐III type, whereas 68% of the Or‐I type harboured the OrZHD1b haplotype (Figure 7b). The geographical distribution of the OrZHD1 haplotypes showed that the wild rice carrying the OrZHD1a haplotype was mainly found in southern China, whereas the OrZHD1b haplotype was distributed in South Asia and India (Figure 7c). Taken together, a single SNP in OsZHD1, leading to R46C, is expected to be a key genetic variation that proactively responds to changes in photoperiod and regulates flowering time in rice.

Figure 7.

Figure 7

Open in a new tab

Polymorphic nucleotides at OrZHD1 are present in wild rice O. rufipogon. (a) A nonsynonymous SNP at OrZHD1 was identified in 446 O. rufipogon accessions from the National Institute of Genetics under the National Bioresource Project. This SNP differentiates wild rice into two haplotypes, OrZHD1a and OrZHD1b. The position of each SNP is relative to the translation start site of the OrZHD1 coding region (designated as +1 bp). Details of the nonsynonymous SNP are provided in Table S5. (b) Distribution of wild rice subpopulations according to OrZHD1 haplotypes. The subpopulations are divided into three groups: Or‐I (red), Or‐II (yellow), and Or‐III (green). (c) Geographical distribution of wild rice varieties with different OrZHD1 alleles in Asia. Each haplotype is represented by circles of different colour; OrZHD1a (green), OrZHD1b (red). The dashed line represents the equator. [Color figure can be viewed at wileyonlinelibrary.com]

3.8. Overexpression of OsZHD1 Increases Grain Yield

To investigate whether OsZHD1 affects grain yield, the DJ, oszhd1‐1, and OsZHD1‐OE1 plants were grown in the paddy field under NLD. Several agronomic traits such as panicle length, number of grains per plant, 500‐grain weight, seed setting rate, and grain yield per plant were measured. oszhd1‐1 exhibited longer panicle length compared to DJ and OsZHD1‐OE1 (Figure S10a,f,g). There was no significant difference in 500‐grain weight between the genotypes (Figure S10b). The seed setting rate of OsZHD1‐OE1 was higher than that of oszhd1‐1 but lower than that of DJ (Figure S10c). Interestingly, overexpression of OsZHD1 increased the number of grains per panicle (Figure S10d), which probably contributed to an increase of yield per plant in OsZHD1‐OE1 (Figure S10e).

4. Discussion

4.1. Possible Roles of OsZHD1 in Flowering Time

The flowering process in plants can be divided into two main stages. First, florigen is synthesised in the phloem companion cells of the leaf in response to photoperiodic signals and translocates to the SAM (Komiya et al. 2008; Komiya et al. 2009; Tamaki et al. 2007). Second, upon reaching the SAM, florigen induces the formation of the inflorescence meristem through a complex regulatory network (Taoka et al. 2011).

Some regulators are involved in flowering processes both in the leaf and in the SAM. For instance, suppression of Ehd1 expression by RNA interference (RNAi) downregulates the expression of OsMADS14 and Hd3a (Kim et al. 2007). In addition to the possible function of OsMADS14 in florigen synthesis in the leaf, OsMADS14 expression is induced in the SAM of RFT1‐overexpressed rice plants (Pasriga et al. 2019). Furthermore, downregulation of both RFT1 and Hd3a in RFT1‐Hd3a double RNAi plants reduces OsMADS14 expression (Komiya et al. 2008). These results suggest that OsMADS14 acts as a downstream gene of RFT1 and Hd3a in the SAM.

Recently, OsZHD1 has been characterised as a key regulator of inflorescence development (Yoon et al. 2023). OsZHD1 is highly expressed in the SAM (Yoon et al. 2023). RFT1, a rice florigen, directly binds to the promoter region of OsZHD1 and OsZHD2, leading to an upregulation of their transcript levels in RFT1‐overexpressing shoot callus during the regeneration stage (Yoon et al. 2023). The mutation in both OsZHD1 and OsZHD2 downregulates the expression of genes involved in auxin biosynthesis, which is crucial for rice inflorescence morphogenesis (Yang et al. 2017; Yoon et al. 2023). As a result, the oszhd1 oszhd2 double mutants exhibit defects in inflorescence branching development (Yoon et al. 2023). In this study, we discovered a distinct function of OsZHD1, which acts as a regulator of florigen synthesis in rice. OsZHD1 was highly expressed in the leaf tissues at both vegetative and reproductive stages (Figure 1a). OsZHD1 directly binds to the promoter region of Ehd1 (Figure 4) and differentially regulates RFT1 expression in response to day length (Figure 3a,b, Figure S4a,b). Collectively, these findings demonstrate that OsZHD1 is not only involved in the formation of the inflorescence meristem in the SAM, but also in the production of florigen in the leaves.

4.2. OsZHD1 is a Component of the Core Photoperiod Regulators

Rice has a unique Ehd1‐Hd3a/RFT1 pathway that plays a crucial role in flowering time regulation. This Ehd1‐centred pathway is regulated by Hd1‐mediated regulatory modules. Hd1 differentially regulates rice flowering in response to photoperiod by promoting flowering under SD and delaying it under LD. In contrast to the action of Hd1, OsZHD1 reduced PS, as the oszhd1 mutation resulted in early flowering under SD and delayed flowering under LD (Figure 2). OsZHD1 physically interacted with Hd1 (Figure 5, Figure S7), and both Hd1 and OsZHD1 regulated Ehd1 expression by directly binding to its promoter region (Figure 4; Nemoto et al. 2016). These observations suggest that OsZHD1 finetunes flowering time by antagonistically regulating Ehd1 expression in response to photoperiod.

Hd1 is one of the core photoperiod genes, along with Ghd7, DTH8, and OsPRR37, that collectively increase PS (Qiu et al. 2021; Wang et al. 2021). Functional alleles of these core photoperiod genes strongly inhibit flowering under LD. The Hd1‐Ghd7 complex binds to the promoter region of Ehd1 (Nemoto et al. 2016). The interaction between the transcription‐activating domain of Hd1 and the CCT domain of Ghd7 may inhibit the transcriptional activation activity of Hd1 on Ehd1 expression, thereby delaying flowering under LD (Nemoto et al. 2016). The genetic and molecular evidence shows that DTH8 physically interacts with Ghd7, and the DTH8‐Ghd7‐Hd1 complex increases PS by strongly repressing Ehd1 expression under LD (Li et al. 2015; Cai et al. 2019; Zong et al. 2021). In addition, OsPRR37 functions as a negative regulator of Ehd1 and Hd3a, especially under LD (Koo et al. 2013; Gao et al. 2014). OsPRR37 forms a trimeric complex with DTH8 and NF‐YC subunits and it can also replace Hd1 to form the Hd1‐DTH8‐NF‐YCs complex. This heterotrimer binds to the Hd3a promoter, increases the H3K27me3 levels, and subsequently represses Hd3a expression (Du et al. 2017; Wei et al. 2010). Among the core photoperiod regulators, we found that OsZHD1 interacts with Hd1 and OsPRR37, but not with DTH8 or Ghd7 (Figure 5, Figure S7). OsZHD1 may affect the floral regulatory network in which Hd1, Ghd7, DTH8, and OsPRR37 repress Ehd1 expression. For instance, it is possible that the OsZHD1‐Hd1 complex competes with Hd1‐Ghd7 or DTH8‐Ghd7‐Hd1 complexes. Indeed, Y3H assays showed that OsZHD1 downregulates the formation of both Hd1‐Ghd7 and Hd1‐DTH8 complexes (Figure S8). Interference of OsZHD1 with Hd1‐mediated regulatory modules could potentially restore Hd1 activation of Ehd1 expression (Cai et al. 2019; Zong et al. 2021). In addition, it is possible that the OsZHD1‐OsPRR37 complex prevents the formation of the OsPRR37‐DTH8‐NF‐YC complex. Taken together, OsZHD1 could disrupt the interactions between these flowering regulators and thereby promote rice flowering under LD.

4.3. Natural Variation of OsZHD1 Contributes to Regional Adaptation of Rice Cultivars by Reducing PS

In the tropics, a short photoperiod promotes rice flowering. However, premature flowering can lead to insufficient vegetative growth, resulting in reduced photosynthetic efficiency and impaired nutrient translocation from source to sink (Johansson and Staiger 2015; Shim and Jang 2020). This ultimately reduces rice grain production. In the temperate regions, the long photoperiod in summer delays flowering, which can reduce grain yield as cold weather approaches. Thus, rice varieties grown at higher latitudes carry nonfunctional alleles of Hd1, Ghd7, DTH8, and OsPRR37, resulting in promoting rice flowering under LD (Koo et al. 2013; Li et al. 2015).

In our study, we found that OsZHD1 reduced PS by modulating Ehd1 expression in response to photoperiodic conditions. Under SD, OsZHD1 prevents premature flowering, allowing for a sufficient vegetative growth period, which increases grain yield. Furthermore, OsZHD1 promotes flowering under LD, ensuring that grain filling is completed before onset of cold weather. Thus, rice has evolved OsZHD1‐regulated flowering mechanisms to enhance regional adaptation (Figure S11).

Artificial selection and domestication of the natural alleles have contributed to the expansion of rice cultivation to a wider range of latitudes. First, nonfunctional alleles of LD‐dependent flowering repressors, including Ghd7, DTH8, and OsPRR37, are intensively selected during the domestication process (Koo et al. 2013; Li et al. 2015). Second, natural variation of the OsCCA1 allele, which acts to downregulate LD‐dependent flowering repressors, was derived from O. rufipogon and domesticated into cultivated rice (Lee et al. 2022). The functional allele OsCCA1a upregulates the expression of Ehd1 by repressing the expression of OsPRR37 and DTH8, thereby promoting flowering in rice varieties under LD (Lee et al. 2022). In the present study, the OsZHD1 protein binds directly to the Ehd1 promoter and upregulates its expression under LD (Figure 4). Furthermore, rice varieties carrying the japonica‐type OsZHD1a allele flowered earlier than those carrying the indica‐type OsZHD1b allele and were cultivated at higher latitudes (Figure 6). Therefore, we speculate that the domestication of OsZHD1a from the wild rice (O. rufipogon) OrZHD1a has contributed to the expansion of rice cultivation (Figure 7).

5. Accession Numbers

The gene sequence data from this article can be found in the Rice Annotation Project Database or Rice Genome Annotation Project under the following accession numbers: OsZHD1, LOC_Os09g29130; Hd3a, LOC_Os06g06320; RFT1, LOC_Os06g06300; Hd1, LOC_Os06g16370; Ehd1, LOC_Os10g32600; OsPRR37, LOC_Os07g49460; DTH8, LOC_Os08g07740; Ghd7, LOC_Os07g15770; OsGI, LOC_Os01g08700; OsCOL4, LOC_Os02g39710; OsCOL10, LOC_Os03g50310; OsMADS50, LOC_Os03g03100; OsMADS51, LOC_Os01g69850; Ehd2, LOC_Os10g28330; Ehd4, LOC_Os03g02160.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting Information Table S1.

PCE-48-7392-s002.docx (18.1KB, docx)

Supporting information Table S2‐S4.

PCE-48-7392-s003.xlsx (235KB, xlsx)

Supporting information Figures S1‐11.

PCE-48-7392-s001.docx (3.9MB, docx)

Acknowledgements

We appreciate Dr. Gynheung An for providing the seeds of T‐DNA mutants. This study was carried out with the support of the Cooperative Research Program for Agriculture Science and Technology Development (RS‐2022‐RD010386 to N.‐C.P.), Rural Development Administration, Republic of Korea, and Basic Science Research Program through the National Research Foundation (NRF) of Korea grant funded by the Korean Government (RS‐2023‐00247376 to S.‐H.C), and the Incheon National University Research Grant (20240281 to K.K.).

Yehyun Yim and Hyeryung Yoon contributed equally to this study.

Contributor Information

Kiyoon Kang, Email: kykang@inu.ac.kr.

Nam‐Chon Paek, Email: ncpaek@snu.ac.kr.

Data Availability Statement

The data that supports the findings of this study are available in the supporting material of this article.

References

  1. An, G. , Ebert P. R., Mitra A., and Ha S. B.. 1989. “Binary Vectors.” In Plant Molecular Biology Manual, edited by Gelvin S. B. and Schilperoort R. A., 29–47. Springer. [Google Scholar]
  2. Andrés, F. , and Coupland G.. 2012. “The Genetic Basis of Flowering Responses to Seasonal Cues.” Nature Reviews Genetics 13, no. 9: 627–639. [DOI] [PubMed] [Google Scholar]
  3. Ariel, F. D. , Manavella P. A., Dezar C. A., and Chan R. L.. 2007. “The True Story of the HD‐Zip Family.” Trends in Plant Science 12, no. 9: 419–426. [DOI] [PubMed] [Google Scholar]
  4. Cai, M. , Chen S., Wu M., et al. 2019. “Early Heading 7 Interacts With DTH8, and Regulates Flowering Time in Rice.” Plant Cell Reports 38, no. 5: 521–532. [DOI] [PubMed] [Google Scholar]
  5. Cho, L. H. , Yoon J., and An G.. 2017. “The Control of Flowering Time by Environmental Factors.” The Plant Journal 90, no. 4: 708–719. [DOI] [PubMed] [Google Scholar]
  6. Cho, S.‐H. , Tóth K., Kim D., et al. 2022. “Activation of the Plant Mevalonate Pathway by Extracellular ATP.” Nature Communications 13, no. 1: 450. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Ciftci‐Yilmaz, S. , and Mittler R.. 2008. “The Zinc Finger Network of Plants.” Cellular and Molecular Life Sciences 65: 1150–1160. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Curtis, M. D. , and Grossniklaus U.. 2003. “A Gateway Cloning Vector Set for High‐Throughput Functional Analysis of Genes in Planta.” Plant Physiology 133, no. 2: 462–469. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Doi, K. , Izawa T., Fuse T., et al. 2004. “ Ehd1, a B‐Type Response Regulator in Rice, Confers Short‐Day Promotion of Flowering and Controls FT‐like Gene Expression Independently of Hd1 .” Genes & Development 18, no. 8: 926–936. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Du, A. , Tian W., Wei M., et al. 2017. “The DTH8‐Hd1 Module Mediates Day‐Length‐Dependent Regulation of Rice Flowering.” Molecular Plant 10, no. 7: 948–961. [DOI] [PubMed] [Google Scholar]
  11. Figueiredo, D. D. , Barros P. M., Cordeiro A. M., et al. 2012. “Seven Zinc‐Finger Transcription Factors Are Novel Regulators of the Stress Responsive Gene OsDREB1B .” Journal of Experimental Botany 63, no. 10: 3643–3656. [DOI] [PubMed] [Google Scholar]
  12. Gao, H. , Jin M., Zheng X.‐M., et al. 2014. “ Days to heading 7, a Major Quantitative Locus Determining Photoperiod Sensitivity and Regional Adaptation in Rice.” Proceedings of the National Academy of Sciences 111, no. 46: 16337–16342. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Gao, H. , Zheng X.‐M., Fei G., et al. 2013. “ Ehd4 Encodes a Novel and Oryza‐Genus‐Specific Regulator of Photoperiodic Flowering in Rice.” PLoS Genetics 9, no. 2: e1003281. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Garris, A. J. , Tai T. H., Coburn J., Kresovich S., and McCouch S.. 2005. “Genetic Structure and Diversity in Oryza sativa L.” Genetics 169, no. 3: 1631–1638. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Hayama, R. , and Coupland G.. 2004. “The Molecular Basis of Diversity in the Photoperiodic Flowering Responses of Arabidopsis and Rice.” Plant Physiology 135, no. 2: 677–684. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Hayama, R. , Yokoi S., Tamaki S., Yano M., and Shimamoto K.. 2003. “Adaptation of Photoperiodic Control Pathways Produces Short‐Day Flowering in Rice.” Nature 422, no. 6933: 719–722. [DOI] [PubMed] [Google Scholar]
  17. Huang, X. , Kurata N., Wei X., et al. 2012. “A Map of Rice Genome Variation Reveals the Origin of Cultivated Rice.” Nature 490, no. 7421: 497–501. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Jain, M. , Nijhawan A., Tyagi A. K., and Khurana J. P.. 2006. “Validation of Housekeeping Genes as Internal Control for Studying Gene Expression in Rice by Quantitative Real‐Time PCR.” Biochemical and Biophysical Research Communications 345, no. 2: 646–651. [DOI] [PubMed] [Google Scholar]
  19. Jeon, J. S. , Lee S., Jung K. H., et al. 2000. “T‐DNA Insertional Mutagenesis for Functional Genomics in Rice.” Plant Journal 22, no. 6: 561–570. [DOI] [PubMed] [Google Scholar]
  20. Johansson, M. , and Staiger D.. 2015. “Time to Flower: Interplay Between Photoperiod and the Circadian Clock.” Journal of Experimental Botany 66, no. 3: 719–730. [DOI] [PubMed] [Google Scholar]
  21. Khush, G. S. 1997. “Origin, Dispersal, Cultivation and Variation of Rice.” Plant Molecular Biology 35: 25–34. [PubMed] [Google Scholar]
  22. Kim, S. L. , Lee S., Kim H. J., Nam H. G., and An G.. 2007. “ OsMADS51 Is a Short‐Day Flowering Promoter That Functions Upstream of Ehd1, OsMADS14, and Hd3a .” Plant Physiology 145, no. 4: 1484–1494. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Komiya, R. , Ikegami A., Tamaki S., Yokoi S., and Shimamoto K.. 2008. “ Hd3a and RFT1 Are Essential for Flowering in Rice.” Development 135, no. 4: 767–774. [DOI] [PubMed] [Google Scholar]
  24. Komiya, R. , Yokoi S., and Shimamoto K.. 2009. “A Gene Network for Long‐Day Flowering Activates RFT1 Encoding a Mobile Flowering Signal in Rice.” Development 136, no. 20: 3443–3450. [DOI] [PubMed] [Google Scholar]
  25. Koo, B.‐H. , Yoo S.‐C., Park J.‐W., et al. 2013. “Natural Variation in OsPRR37 Regulates Heading Date and Contributes to Rice Cultivation at a Wide Range of Latitudes.” Molecular Plant 6, no. 6: 1877–1888. [DOI] [PubMed] [Google Scholar]
  26. Kwon, C. T. , Yoo S. C., Koo B. H., et al. 2014. “Natural Variation in Early Flowering1 Contributes to Early Flowering in Japonica Rice Under Long Days.” Plant, Cell & Environment 37, no. 1: 101–112. [DOI] [PubMed] [Google Scholar]
  27. Lee, S. , Kim J., Han J. J., Han M. J., and An G.. 2004. “Functional Analyses of the Flowering Time Gene OsMADS50, the Putative Suppressor of Overexpression OF Co 1/AGAMOUS‐LIKE 20 (Soc1/Agl20) Ortholog in Rice.” Plant Journal 38, no. 5: 754–764. [DOI] [PubMed] [Google Scholar]
  28. Lee, S.‐J. , Kang K., Lim J.‐H., and Paek N.‐C.. 2022. “Natural Alleles of Circadian Clock Associated1 Contribute to Rice Cultivation by Fine‐Tuning Flowering Time.” Plant Physiology 190, no. 1: 640–656. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Lee, Y. S. , Jeong D. H., Lee D. Y., et al. 2010. “ OsCOL4 Is a Constitutive Flowering Repressor Upstream of Ehd1 and Downstream of OsphyB .” Plant Journal 63, no. 1: 18–30. [DOI] [PubMed] [Google Scholar]
  30. Li, X. , Liu H., Wang M., et al. 2015. “Combinations of Hd2 and Hd4 Genes Determine Rice Adaptability to Heilongjiang Province, Northern Limit of China.” Journal of Integrative Plant Biology 57, no. 8: 698–707. [DOI] [PubMed] [Google Scholar]
  31. Mansueto, L. , Fuentes R. R., Borja F. N., et al. 2017. “Rice SNP‐Seek Database Update: New SNPs, Indels, and Queries.” Nucleic Acids Research 45, no. D1: D1075–D1081. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Matsubara, K. , Yamanouchi U., Wang Z.‐X., Minobe Y., Izawa T., and Yano M.. 2008. “ Ehd2, a Rice Ortholog of the Maize INDETERMINATE1 Gene, Promotes Flowering by Up‐Regulating Ehd1 .” Plant Physiology 148, no. 3: 1425–1435. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Nemoto, Y. , Nonoue Y., Yano M., and Izawa T.. 2016. “ Hd1, a CONSTANS Ortholog in Rice, Functions as An Ehd1 Repressor Through Interaction With Monocot‐Specific CCT‐Domain Protein Ghd7.” The Plant Journal 86, no. 3: 221–233. [DOI] [PubMed] [Google Scholar]
  34. Osnato, M. 2023. “Evolution of Flowering Time Genes in Rice: From the Paleolithic to the Anthropocene.” Plant, Cell & Environment 46, no. 4: 1046–1059. [DOI] [PubMed] [Google Scholar]
  35. Park, H. C. , Kim M. L., Lee S. M., et al. 2007. “Pathogen‐Induced Binding of the Soybean Zinc Finger Homeodomain Proteins GmZF‐HD1 and GmZF‐HD2 to Two Repeats of ATTA Homeodomain Binding Site in the Calmodulin Isoform 4 (GmCaM4) Promoter.” Nucleic Acids Research 35, no. 11: 3612–3623. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Pasriga, R. , Yoon J., Cho L.‐H., and An G.. 2019. “Overexpression of Rice Flowering Locus T 1 (RFT1) Induces Extremely Early Flowering in Rice.” Molecules and Cells 42, no. 5: 406–417. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Qiu, L. , Wu Q., Wang X., et al. 2021. “Forecasting Rice Latitude Adaptation Through a Daylength‐Sensing‐Based Environment Adaptation Simulator.” Nature Food 2: 348–362. [DOI] [PubMed] [Google Scholar]
  38. Saleh, A. , Alvarez‐Venegas R., and Avramova Z.. 2008. “An Efficient Chromatin Immunoprecipitation (Chip) Protocol for Studying Histone Modifications In Arabidopsis Plants.” Nature Protocols 3, no. 6: 1018–1025. [DOI] [PubMed] [Google Scholar]
  39. Seo, P. J. , Hong S.‐Y., Kim S.‐G., and Park C.‐M. (2011). Competitive Inhibition of Transcription. [DOI] [PubMed]
  40. Shi, J. , An G., Weber A., and Zhang D.. 2023. “Prospects for Rice in 2050.” Plant, Cell & Environment 46, no. 4: 1037–1045. [DOI] [PubMed] [Google Scholar]
  41. Shim, J. S. , and Jang G.. 2020. “Environmental Signal‐Dependent Regulation of Flowering Time in Rice.” International Journal of Molecular Sciences 21, no. 17: 6155. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Shrestha, R. , Gómez‐Ariza J., Brambilla V., and Fornara F.. 2014. “Molecular Control of Seasonal Flowering in Rice, Arabidopsis and Temperate Cereals.” Annals of Botany 114, no. 7: 1445–1458. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Song, Y. H. , Shim J. S., Kinmonth‐Schultz H. A., and Imaizumi T.. 2015. “Photoperiodic Flowering: Time Measurement Mechanisms in Leaves.” Annual Review of Plant Biology 66, no. 1: 441–464. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Srikanth, A. , and Schmid M.. 2011. “Regulation of Flowering Time: All Roads Lead to Rome.” Cellular and Molecular Life Sciences 68: 2013–2037. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Street, N. R. , Sjödin A., Bylesjö M., Gustafsson P., Trygg J., and Jansson S.. 2008. “A Cross‐Species Transcriptomics Approach to Identify Genes Involved in Leaf Development.” BMC Genomics 9: 589. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Sun, K. , Huang M., Zong W., et al. 2022. “ Hd1, Ghd7, and DTH8 Synergistically Determine the Rice Heading Date and Yield‐Related Agronomic Traits.” Journal of Genetics and Genomics 49, no. 5: 437–447. [DOI] [PubMed] [Google Scholar]
  47. Tamaki, S. , Matsuo S., Wong H. L., Yokoi S., and Shimamoto K.. 2007. “Hd3a Protein Is a Mobile Flowering Signal in Rice.” Science 316, no. 5827: 1033–1036. [DOI] [PubMed] [Google Scholar]
  48. Tan, J. , Jin M., Wang J., et al. 2016. “ OsCOL10, a CONSTANS‐Like Gene, Functions as a Flowering Time Repressor Downstream of Ghd7 in Rice.” Plant and Cell Physiology 57, no. 4: 798–812. [DOI] [PubMed] [Google Scholar]
  49. Taoka, K. , Ohki I., Tsuji H., et al. 2011. “14‐3‐3 Proteins Act as Intracellular Receptors for Rice Hd3a Florigen.” Nature 476, no. 7360: 332–335. [DOI] [PubMed] [Google Scholar]
  50. Tran, L. S. P. , Nakashima K., Sakuma Y., et al. 2007. “Co‐Expression of the Stress‐Inducible Zinc Finger Homeodomain ZFHD1 and NAC Transcription Factors Enhances Expression of the ERD1 Gene in Arabidopsis.” The Plant Journal 49, no. 1: 46–63. [DOI] [PubMed] [Google Scholar]
  51. Wang, X. , He Y., Wei H., and Wang L.. 2021. “A Clock Regulatory Module Is Required for Salt Tolerance and Control of Heading Date in Rice.” Plant, Cell & Environment 44: 3283–3301. [DOI] [PubMed] [Google Scholar]
  52. Wei, X. , Xu J., Guo H., et al. 2010. “ DTH8 Suppresses Flowering in Rice, Influencing Plant Height and Yield Potential Simultaneously.” Plant Physiology 153, no. 4: 1747–1758. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Windhövel, A. , Hein I., Dabrowa R., and Stockhaus J.. 2001. “Characterization of a Novel Class of Plant Homeodomain Proteins That Bind to the C4 PhosphoenolPyruvate Carboxylase Gene of Flaveria trinervia .” Plant Molecular Biology 45: 201–214. [DOI] [PubMed] [Google Scholar]
  54. Xu, Y. , Wang Y., Long Q., et al. 2014. “Overexpression of OsZHD1, a Zinc Finger Homeodomain Class Homeobox Transcription Factor, Induces Abaxially Curled and Drooping Leaf in Rice.” Planta 239: 803–816. [DOI] [PubMed] [Google Scholar]
  55. Xue, W. , Xing Y., Weng X., et al. 2008. “Natural Variation in Ghd7 Is an Important Regulator of Heading Date and Yield Potential In Rice.” Nature Genetics 40, no. 6: 761–767. [DOI] [PubMed] [Google Scholar]
  56. Yang, J. , Yuan Z., Meng Q., et al. 2017. “Dynamic Regulation of Auxin Response During Rice Development Revealed by Newly Established Hormone Biosensor Markers.” Frontiers in Plant Science 8: 256. [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Yano, M. , Katayose Y., Ashikari M., et al. 2000. “ Hd1, a Major Photoperiod Sensitivity Quantitative Trait Locus in Rice, Is Closely Related to the Arabidopsis Flowering Time Gene CONSTANS .” The Plant Cell 12, no. 12: 2473–2483. [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Yano, M. , Kojima S., Takahashi Y., Lin H., and Sasaki T.. 2001. “Genetic Control of Flowering Time in Rice, a Short‐Day Plant.” Plant Physiology 127, no. 4: 1425–1429. [PMC free article] [PubMed] [Google Scholar]
  59. Yoon, J. , Baek G., Pasriga R., et al. 2023. “Homeobox Transcription Factors OsZHD1 and OsZHD2 Induce Inflorescence Meristem Activity at Floral Transition in Rice.” Plant, Cell & Environment 46, no. 4: 1327–1339. [DOI] [PubMed] [Google Scholar]
  60. Yoon, J. , Cho L.‐H., Yang W., et al. 2020. “Homeobox Transcription Factor OsZHD2 Promotes Root Meristem Activity In Rice by Inducing Ethylene Biosynthesis.” Journal of Experimental Botany 71, no. 18: 5348–5364. [DOI] [PMC free article] [PubMed] [Google Scholar]
  61. Zhang, Y. , Su J., Duan S., et al. 2011. “A Highly Efficient Rice Green Tissue Protoplast System for Transient Gene Expression and Studying Light/Chloroplast‐Related Processes.” Plant Methods 7: 30. [DOI] [PMC free article] [PubMed] [Google Scholar]
  62. Zhao, J. , Chen H., Ren D., et al. 2015. “Genetic Interactions Between Diverged Alleles of Early Heading Date 1 (Ehd1) and Heading Date 3A (Hd3a)/Rice Flowering Locus T1 (RFT1) Control Differential Heading and Contribute to Regional Adaptation In Rice (Oryza sativa).” New Phytologist 208, no. 3: 936–948. [DOI] [PubMed] [Google Scholar]
  63. Zhao, T. , Hu J., Gao Y., et al. 2018. “Silencing of the SL‐ZH13 Transcription Factor Gene Decreases the Salt Stress Tolerance of Tomato.” Journal of the American Society for Horticultural Science 143, no. 5: 391–396. [Google Scholar]
  64. Zong, W. , Ren D., Huang M., et al. 2021. “Strong Photoperiod Sensitivity Is Controlled by Cooperation and Competition Among Hd1, Ghd7 and DTH8 in Rice Heading.” New Phytologist 229, no. 3: 1635–1649. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supporting Information Table S1.

PCE-48-7392-s002.docx (18.1KB, docx)

Supporting information Table S2‐S4.

PCE-48-7392-s003.xlsx (235KB, xlsx)

Supporting information Figures S1‐11.

PCE-48-7392-s001.docx (3.9MB, docx)

Data Availability Statement

The data that supports the findings of this study are available in the supporting material of this article.

Articles from Plant, Cell & Environment are provided here courtesy of Wiley

RESOURCES