Rice OsFKF1 Promotes Degradation of OsPRR37 for Short-Day Early Flowering Authors: Hyeryung Yoon, Sang-Ji Lee, Jinku Kang, Sung-Hwan Cho, Nam-Chon Paek Published: 2025 ABSTRACT Many plant species use changes in day length to fine-tune flowering time and achieve successful reproduction under optimal conditions. Extensive molecular genetic studies have identified key regulatory genes that repress flowering under non-inductive long-day (LD) conditions in rice (Oryza sativa), a facultative short-day (SD) plant. However, understanding how these regulators are modulated at the post-translational level in response to photoperiodic conditions remained elusive. In this study, we showed that the rice F-box protein FLAVIN-BINDING, KELCH REPEAT, F-BOX 1 (OsFKF1) interacts with and promotes the degradation of the LD-dependent flowering repressor PSEUDO-RESPONSE REGULATOR 37 (OsPRR37)/Heading date 2 (Hd2). Genetic analysis using osfkf1 osprr37 double mutants suggested that OsFKF1 destabilizes OsPRR37 to upregulate Early heading date 1 expression for early flowering under SD. In addition, we found that Hd1, which is a flowering inducer in SD but acts as a flowering repressor in LD, interacts with OsPRR37, suggesting that the Hd1-OsPRR37 complex inhibits flowering under LD. Our results provide new insights into the post-translational regulation of OsFKF1-OsPRR37-Hd1 in photoperiodic flowering in rice.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12284-025-00880-7. Background Photoperiod is a critical environmental signal that regulates flowering time (Gardner and Allard 1922). Plants monitor seasonal changes in day length to induce flowering under optimal conditions (Song et al. 2015). The three plant groups were classified according to their responses to photoperiod (Thomas and Vince-Prue 1997). Short-day (SD) plants flower when the photoperiod is shorter than a certain threshold, known as the critical day length. In contrast, long-day (LD) plants flower when the day is longer than the critical day length. Day-neutral plants flower regardless of photoperiod. Variations in photoperiod sensitivity play a crucial role in the adaptation of crop species to different environments, ultimately affecting yield and quality (Gao et al. 2014; Song et al. 2015). The molecular mechanisms underlying photoperiodic flowering have been extensively studied in the facultative LD plant, Arabidopsis (Arabidopsis thaliana), and the facultative SD plant, rice (Oryza sativa) (Song et al. 2015; Zhu et al. 2021; Wang et al. 2021). In Arabidopsis, FLOWERING LOCUS T (FT) encodes a protein known as the florigen, which acts as a mobile flowering signal to induce flowering. Florigen is produced in the leaves and transported to the shoot apical meristem, where it induces flowering (Corbesier et al. 2007). FT expression is primarily controlled by the zinc finger transcriptional activator CONSTANS (CO), which contains two B-boxes and a CCT (CO, CO-LIKE, and TOC1) domain (Putterill et al. 1995). CO binds to the FT promoter to activate its expression (Adrian et al. 2010; Tiwari et al. 2010). CO transcription and protein stability are regulated by the circadian clock and light signaling for LD-dependent activation of FT expression (Yanovsky and Kay 2002; Valverde et al. 2004). The blue light receptor FLAVIN-BINDING, KELCH REPEAT, F-BOX 1 (FKF1) plays a crucial role in promoting flowering by positively modulating CO at both transcriptional and post-translational levels (Sawa et al. 2007; Song et al. 2012; Lee et al. 2017; Yan et al. 2020). Under LD, blue light-activated FKF1 forms a ubiquitin ligase complex with GIGANTEA (GI), facilitating the degradation of CYCLING DOF FACTORs (CDFs), which act as transcriptional repressors of CO, thereby activating CO expression in the late afternoon (Imaizumi et al. 2005; Sawa et al. 2007; Fornara et al. 2009). In contrast, under SD, FKF1 is mainly expressed at night, reducing its binding affinity to GI and thus stabilizing CDFs. FKF1 interacts with CO to increase its stability, thereby promoting FT expression under LD (Song et al. 2012). Furthermore, FKF1 stabilizes the CO protein in the afternoon under LD by inhibiting the homodimerization of CONSTITUTIVE PHOTOMORPHOGENIC 1 (COP1), which is responsible for CO degradation in the dark (Lee et al. 2017). Therefore, CO mRNA expression and protein accumulation during the day induce FT expression under LD. The Arabidopsis CO-FT module is conserved in rice. Two florigen genes, Heading date 3a (Hd3a) and RICE FLOWERING LOCUS T 1 (RFT1), have been identified in rice (Komiya et al. 2008). Heading date 1 (Hd1), a putative orthologous gene of Arabidopsis CO, activates Hd3a and RFT1 to induce flowering under inductive SD (Yano et al. 2000; Izawa et al. 2002; Hayama et al. 2003). In addition to this conserved pathway, rice also has a unique pathway that controls photoperiodic flowering. The B-type response regulator, Early heading date 1 (Ehd1), promotes flowering by activating the expression of Hd3a and RFT1 (Doi et al. 2004; Cho et al. 2016). In contrast to Arabidopsis CO, Hd1 acts as a flowering repressor that represses the expression of Ehd1, Hd3a, and RFT1 under LD (Izawa et al. 2002; Ishikawa et al. 2005; Komiya et al. 2008). In addition, the CCT domain-containing proteins, PSEUDO-RESPONSE REGULATOR 37 (OsPRR37, also known as Hd2, DTH7, and Ghd7.1) and Ghd7 (for grain number, plant height and heading date; also known as Hd4), together with DAYS TO HEADING 8 (DTH8, also known as Hd5 and Ghd8) encoding a CCAAT box-binding NF-YB transcription factor, repress the expression of Ehd1 under LD (Xue et al. 2008; Wei et al. 2010; Yan et al. 2011; Koo et al. 2013; Gao et al. 2014; Liu et al. 2018). Recent studies have shown that Hd1 primarily functions as a flowering activator and switches to a repressor by interacting with OsPRR37, Ghd7, and DTH8 under LD (Nemoto et al. 2016; Goretti et al. 2017; Zhang et al. 2017, 2019b; Zhu et al. 2017; Du et al. 2017; Cai et al. 2019; Zong et al. 2021). Interestingly, natural variation in these genes contributed to the expansion of rice cultivation areas during domestication by reducing photoperiod sensitivity (Xue et al. 2008; Koo et al. 2013; Gao et al. 2014; Zhang et al. 2015, 2019a; Zheng et al. 2016; Wu et al. 2020). Recent studies have shown that post-translational modification plays a critical role in regulating key photoperiodic flowering regulators in rice. For example, Hd1 is degraded at night by the C3HC4 RING domain-containing E3 ubiquitin ligase Heading date Associated Factor 1 (HAF1) via the 26 S proteasome pathway to precisely regulate Hd1 accumulation (Yang et al. 2015). It has also been reported that Hd1 interacts with the autophagy protein OsATG8 and is degraded by the nuclear autophagy pathway in the dark (Hu et al. 2022). HAF1 also modulates the circadian accumulation of an LD-dependent flowering activator, OsELF3, through ubiquitination to control flowering under LD (Zhu et al. 2018). OsELF3 directly represses the expression of the LD-dependent flowering repressors Ghd7 and OsPRR37 within the evening complex, whereas the photoreceptor OsPhyB inactivates OsELF3 through post-translational modifications in response to light under LD (Andrade et al. 2022). In addition, Ghd7 is degraded by OsGI through a 26 proteasome-dependent pathway while stabilized by the antagonistic actions of OsPhyA and OsPhyB (Zheng et al. 2019). A recent study also revealed that the E3 ligase IPA INTERACTING PROTEIN1 (IPI1) mediates the sucrose-induced polyubiquitination of Ghd7, suggesting that sucrose signaling induces flowering by facilitating Ghd7 degradation (Cho et al. 2024). We have previously shown that the rice FKF1 (OsFKF1) acts as a flowering activator upstream of Ehd1 and Ghd7 (Han et al. 2015). Moreover, we have shown that OsPRR37 is a key LD-dependent flowering repressor that has contributed to rice adaptation at high latitudes by reducing photoperiod sensitivity (Koo et al. 2013). Although the role of FKF1 in the post-translational modification of flowering regulators has been reported in Arabidopsis, the detailed molecular mechanism by which OsFKF1 controls photoperiodic flowering in rice remains elusive. In this study, we showed that OsFKF1 interacts with and degrades OsPRR37 to promote flowering under SD. In addition, OsPRR37 interacts with Hd1, suggesting that OsPRR37 functions with Hd1 to suppress flowering under LD. Therefore, our results suggest that the degradation of OsPRR37 by OsFKF1 and Hd1-OsPRR37 interaction are critical molecular mechanisms underlying the photoperiodic control of flowering in rice. Results OsFKF1 Interacts with OsPRR37 In our previous study, we demonstrated that OsFKF1 acts as a flowering activator in rice (Han et al. 2015). To further investigate the post-translational regulation of OsFKF1 in photoperiodic flowering, we examined its interactions with several known flowering regulators using yeast two-hybrid (Y2H) assays. Our analysis revealed that OsFKF1 interacts with OsPRR37 (Fig. 1). The Y2H results showed that the LOV domain of OsFKF1 (aa 54–123) interacts with the N-terminal region (aa 1–200) of OsPRR37, which contains the pseudo-receiver (PR) domain (Fig. 1A, B). To further confirm the interactions in planta, we performed BiFC assays in rice protoplasts. OsPRR37 and OsFKF1 were fused to the C-terminal (YC) and N-terminal (YN) portions of YFP, respectively. Co-expression of OsPRR37-YC and OsFKF1-YN resulted in the YFP fluorescence in the nucleus (Fig. 1C), indicating that OsFKF1 interacts with OsPRR37 in the nucleus. In addition, we performed split LCI assays on the N. benthamiana leaves to further verify the OsFKF1-OsPRR37 interaction. OsPRR37 and OsFKF1 were fused to the N- and C-terminal fragments of luciferase (LUC), OsPRR37N and OsFKF1C, respectively. The LCI assays showed that co-expression of OsPRR37N and OsFKF1C produced a strong LUC signal in the leaf sectors, confirming the in vivo interaction OsFKF1 and OsPRR37 (Fig. 1D). In addition, Co-IP assays verify that OsFKF1 and OsPRR37 interact in vivo (Fig. 1E). Taken together, these results demonstrated that OsFKF1 interacts with OsPRR37. Fig. 1OsFKF1 interacts with OsPRR37. A Schematic diagram of the truncated fragments of OsFKF1 and OsPRR37 in Y2H assays. Numbers indicate amino acid (aa) positions. LOV, LOV domain; F-box, F-box domain; Kelch, Kelch repeats; OsPRR37-N, N-terminal region containing Pseudo-Receiver (PR) domain; OsPRR37-M, middle region; OsPRR37-C, C-terminal region containing CCT domain. B Interaction of OsFKF1-LOV and OsPRR37-N in Y2H assays. OsPRR37 interacts with the LOV domain of OsFKF1 via its N-terminal region containing the PR domain. β-Galactosidase (β-gal) activity was measured for each transformant. -LT, synthetic defined (sd) medium lacking Leu/Trp; -LTHA, sd medium lacking Leu/Trp/His/Ade. Asterisks indicate significant differences compared to the negative control (Student’s t-test, ***P < 0.001). C BiFC assays confirms that the OsFKF1-OsPRR37 interaction occurs in the nucleus. OsPRR37 and OsFKF1 fused to C-terminal YFP (YC) and N-terminal YFP (YN), respectively, were expressed in rice protoplasts. The protoplasts co-expressing OsPRR37 and OsFKF1 showed YFP signals in the nuclei. The nuclear localization signal (NLS)–red fluorescence protein (RFP) was used as a nuclear marker. White bars = 5 mm. D Split-luciferase complementation assays of the N. benthamiana leaves co-infiltrated with Agrobacterium strains harboring OsPRR37 fused to the N-terminal fragment of luciferase (OsPRR37N)/OsFKF1 fused to the C-terminal domain of luciferase (OsFKF1C), OsPRR37N/vecC, vecN/OsFKF1C, and vecN/vecC. Dotted circles represent the leaf panels infiltrated with Agrobacterium carrying the respective constructs. The relative luminescence unit (RLU) intensity of OsPRR37N/OsFKF1C was quantified using the ImageJ software. Different letters indicate significantly different values according to one-way ANOVA and Duncan’s least significant range test (P < 0.05). E Co-IP assays shows the OsFKF1 and OsPRR37 interaction in vivo. OsFKF1-HA and OsPRR37-Myc fusion proteins were co-express in rice Oc cell protoplasts. Co-IP was performed with anti-Myc antibody OsFKF1 interacts with OsPRR37. A Schematic diagram of the truncated fragments of OsFKF1 and OsPRR37 in Y2H assays. Numbers indicate amino acid (aa) positions. LOV, LOV domain; F-box, F-box domain; Kelch, Kelch repeats; OsPRR37-N, N-terminal region containing Pseudo-Receiver (PR) domain; OsPRR37-M, middle region; OsPRR37-C, C-terminal region containing CCT domain. B Interaction of OsFKF1-LOV and OsPRR37-N in Y2H assays. OsPRR37 interacts with the LOV domain of OsFKF1 via its N-terminal region containing the PR domain. β-Galactosidase (β-gal) activity was measured for each transformant. -LT, synthetic defined (sd) medium lacking Leu/Trp; -LTHA, sd medium lacking Leu/Trp/His/Ade. Asterisks indicate significant differences compared to the negative control (Student’s t-test, ***P < 0.001). C BiFC assays confirms that the OsFKF1-OsPRR37 interaction occurs in the nucleus. OsPRR37 and OsFKF1 fused to C-terminal YFP (YC) and N-terminal YFP (YN), respectively, were expressed in rice protoplasts. The protoplasts co-expressing OsPRR37 and OsFKF1 showed YFP signals in the nuclei. The nuclear localization signal (NLS)–red fluorescence protein (RFP) was used as a nuclear marker. White bars = 5 mm. D Split-luciferase complementation assays of the N. benthamiana leaves co-infiltrated with Agrobacterium strains harboring OsPRR37 fused to the N-terminal fragment of luciferase (OsPRR37N)/OsFKF1 fused to the C-terminal domain of luciferase (OsFKF1C), OsPRR37N/vecC, vecN/OsFKF1C, and vecN/vecC. Dotted circles represent the leaf panels infiltrated with Agrobacterium carrying the respective constructs. The relative luminescence unit (RLU) intensity of OsPRR37N/OsFKF1C was quantified using the ImageJ software. Different letters indicate significantly different values according to one-way ANOVA and Duncan’s least significant range test (P < 0.05). E Co-IP assays shows the OsFKF1 and OsPRR37 interaction in vivo. OsFKF1-HA and OsPRR37-Myc fusion proteins were co-express in rice Oc cell protoplasts. Co-IP was performed with anti-Myc antibody OsFKF11 Promotes Degradation of OsPRR37 OsFKF1 contains an F-box domain that forms the SKP-cullin-Rbx-F-box (SCF) E3 ubiquitin ligase complex, indicating its potential role in protein degradation (Nelson et al. 2000; Imaizumi et al. 2005; Han et al. 2015). Based on this, we hypothesized that OsFKF1 promotes the degradation of OsPRR37 to promote flowering in rice. To investigate the effect of OsFKF1 on the stability of OsPRR37, we transiently expressed the OsPRR37-Myc protein in rice protoplasts prepared from WT and osfkf1 plants. Notably, the accumulation of OsPRR37-Myc protein was significantly higher in the osfkf1 protoplasts than in the WT protoplasts (Fig. 2A). In addition, treatment with 20 mM MG132 further increased the level of the OsPRR37-Myc, indicating that OsPRR37 is degraded via the 26 S proteasome pathway (Fig. 2B). In contrast, co-expression of the OsPRR37-Myc and OsFKF1-HA in the rice protoplasts promoted the degradation of OsPRR37-Myc protein (Fig. 2C). These results suggested that OsFKF1 possibly mediates the degradation of OsPRR37. Fig. 2OsFKF1 mediates the degradation of OsPRR37. A Western blot analysis showing the accumulation of OsPRR37-Myc in the WT and osfkf1 protoplasts. OsPRR37-Myc was transiently expressed in rice protoplasts isolated from WT and osfkf1. B OsPRR37 is degraded via the 26 S proteasome pathway. Protoplasts transiently expressing OsPRR37-Myc were incubated overnight with or without 20 mM MG132. MG132 treatment increased the OsPRR37-Myc levels, indicating proteasome-dependent degradation. C Co-expression of OsFKF1 promotes the degradation of OsPRR37. Co-expression of OsFKF1-HA reduced OsPRR37-Myc protein levels, while treatment with 20 mM MG132 enhanced its accumulation. OsPRR37-Myc co-expressed with OsHAL3-HA (Su et al. 2016) was used as a negative control to exclude non-specific dosage effects. All Myc- or HA-tagged constructs were driven by the maize Ubiquitin promoter. OsPRR37 Histone H3 was used as a loading control. Numbers at the bottom of the gel image indicate the relative intensities of protein bands (anti-Myc/anti-H3) calculated using the ImageJ OsFKF1 mediates the degradation of OsPRR37. A Western blot analysis showing the accumulation of OsPRR37-Myc in the WT and osfkf1 protoplasts. OsPRR37-Myc was transiently expressed in rice protoplasts isolated from WT and osfkf1. B OsPRR37 is degraded via the 26 S proteasome pathway. Protoplasts transiently expressing OsPRR37-Myc were incubated overnight with or without 20 mM MG132. MG132 treatment increased the OsPRR37-Myc levels, indicating proteasome-dependent degradation. C Co-expression of OsFKF1 promotes the degradation of OsPRR37. Co-expression of OsFKF1-HA reduced OsPRR37-Myc protein levels, while treatment with 20 mM MG132 enhanced its accumulation. OsPRR37-Myc co-expressed with OsHAL3-HA (Su et al. 2016) was used as a negative control to exclude non-specific dosage effects. All Myc- or HA-tagged constructs were driven by the maize Ubiquitin promoter. OsPRR37 Histone H3 was used as a loading control. Numbers at the bottom of the gel image indicate the relative intensities of protein bands (anti-Myc/anti-H3) calculated using the ImageJ Genetic Interaction Between OsFKF1 and OsPRR37 To investigate the genetic relationship and molecular mechanisms of the OsFKF1-OsPRR37 interaction, we generated an osfkf1 osprr37 double mutants by crossing the T-DNA insertion mutants of osfkf1 and osprr37 (Fig. S1A) (Koo et al. 2013; Han et al. 2015). The homozygous osfkf1 osprr37 mutants were identified by genomic PCR (Fig. S1B). Previous studies have shown that osfkf1 mutants flowered later than WT under both SD and LD, whereas osprr37 mutants flowered earlier than WT mainly under LD (Koo et al. 2013; Han et al. 2015). WT, osfkf1, osprr37, and osfkf1 osprr37 were grown either in the paddy field under natural long-day (NLD, > 14 h light/day, 37°N latitude, Suwon, Korea) or in the growth chambers under LD (14 h light/10 h dark) and SD (10 h light/14 h dark) to evaluate the flowering time phenotypes (Fig. 3; Fig. S2). Fig. 3Flowering time of osfkf1 osprr37 double mutant. Flowering time phenotypes and days to heading in WT, osfkf1, osprr37, and osfkf1 osprr37 under LD (A, B) and SD (C, D) conditions (n = 10). The photographs were taken one week after the heading of the osprr37 mutant under SD conditions (A) and ten days after the heading of the osprr37 mutant under LD conditions (C). Red arrows indicate panicles in heading. Box-and-whisker plots show the maximum and minimum, 25–75th percentiles (box), and median (middle line). Different letters indicate significantly different values according to one-way ANOVA and Duncan’s least significant range test (P < 0.05). DTH, days to heading. White Scale bars = 10 cm Flowering time of osfkf1 osprr37 double mutant. Flowering time phenotypes and days to heading in WT, osfkf1, osprr37, and osfkf1 osprr37 under LD (A, B) and SD (C, D) conditions (n = 10). The photographs were taken one week after the heading of the osprr37 mutant under SD conditions (A) and ten days after the heading of the osprr37 mutant under LD conditions (C). Red arrows indicate panicles in heading. Box-and-whisker plots show the maximum and minimum, 25–75th percentiles (box), and median (middle line). Different letters indicate significantly different values according to one-way ANOVA and Duncan’s least significant range test (P < 0.05). DTH, days to heading. White Scale bars = 10 cm Under NLD, osfkf1 (124 days to heading [DTH]) flowered later than WT (113 DTH), whereas osprr37 (99 DTH) flowered earlier than WT, which is consistent with the previous reports (Fig. S2A-D) (Koo et al. 2013; Han et al. 2015; Liu et al. 2018). osfkf1 osprr37 double mutants (106 DTH) flowered much earlier than WT and osfkf1 under NLD, implicating that OsPRR37 is epistatic to OsFKF1. However, osfkf1 osprr37 flowered 7 days later than osprr37, suggesting that OsFKF1 promotes flowering through other pathways in addition to negative post-translational regulation of OsPRR37. Similarly, osprr37 and osfkf1 osprr37 flowered earlier than WT in the growth chambers under LD, whereas osfkf1 flowered later than WT (Fig. 3A, B), further supporting that OsFKF1 is an upstream repressor of OsPRR37. Moreover, osfkf1 (90 DTH) flowered much later than WT (63 DTH) under SD, while osprr37 (99 DTH) showed no significant difference in flowering time compared to WT, as previously reported (Koo et al. 2013; Han et al. 2015). osfkf1 osprr37 (75 DTH) flowered later than WT and osprr37, but flowered earlier than osfkf1 under SD (Fig. 3C, D; Fig. S2E), also suggesting that OsFKF1 promotes flowering through other regulatory pathways in addition to the OsPRR37 degradation process. Taken together, these results indicate that OsFKF1 promotes flowering not only by destabilizing OsPRR37, but also through other flowering induction pathways. We examined the diurnal expression of flowering time genes in WT, osfkf1, osprr37, and osfkf1 osprr37 using RT-qPCR (Fig. 4; Fig. S3-5). OsFKF1 exhibited a diurnal expression pattern with a peak just before dusk at ZT13 under LD and at ZT10 under SD (Fig. 4A, B). Notably, OsFKF1 exhibited a similar diurnal rhythm in osprr37. In addition, OsPRR37 displayed rhythmic expression with a peak at ZT10 under both LD and SD in both WT and osfkf1 (Fig. 4C, D). These results suggest that OsFKF1 does not affect the expression levels of OsPRR37 and vice versa. Fig. 4Diurnal expression of flowering time genes in osfkf1, osprr37, and osfkf1 osprr37 under LD and SD. Relative transcript levels of flowering time genes including OsFKF1 (A, B), OsPRR37 (C, D), Hd3a (E, F), RFT1 (G, H), Ehd1 (I, J), and Hd1 (K, L) in the WT, osfkf1, osprr37, and osfkf1 osprr37 mutant plants under LD and SD conditions. Leaf samples were collected every 3 h at 70 days after sowing (DAS) under LD and 30 DAS under SD. Expression levels of flowering time genes were determined by RT-qPCR analysis and normalized to OsUBQ5 as an internal control. The gray boxes represent the dark period, and the gray, orange, green, and blue lines represent WT, osfkf1, osprr37, and osfkf1 osprr37 double mutants, respectively. Data are shown as mean ± standard deviation using three biological replicates. Asterisks indicate significant differences compared to DJ, as analyzed by Student’s t-test (*P < 0.05, **P < 0.01, ***P < 0.001) Diurnal expression of flowering time genes in osfkf1, osprr37, and osfkf1 osprr37 under LD and SD. Relative transcript levels of flowering time genes including OsFKF1 (A, B), OsPRR37 (C, D), Hd3a (E, F), RFT1 (G, H), Ehd1 (I, J), and Hd1 (K, L) in the WT, osfkf1, osprr37, and osfkf1 osprr37 mutant plants under LD and SD conditions. Leaf samples were collected every 3 h at 70 days after sowing (DAS) under LD and 30 DAS under SD. Expression levels of flowering time genes were determined by RT-qPCR analysis and normalized to OsUBQ5 as an internal control. The gray boxes represent the dark period, and the gray, orange, green, and blue lines represent WT, osfkf1, osprr37, and osfkf1 osprr37 double mutants, respectively. Data are shown as mean ± standard deviation using three biological replicates. Asterisks indicate significant differences compared to DJ, as analyzed by Student’s t-test (*P < 0.05, **P < 0.01, ***P < 0.001) Fig. 5OsPRR37 interacts with Hd1. Flowering time phenotypes and days to heading in WT, osprr37, hd, and osprr37 hd1 under natural long-day (NLD) (A, B) and short-day (SD) (C, D) conditions. The photograph was taken before the WT plant flowered under NLD conditions (A) and 7 days after the WT plant flowered under SD conditions (C). To analyze heading date, WT, osprr37, hd1, and osprr37 hd1 plants were grown in the paddy field under natural long-day (NLD) conditions (B) and in the growth chamber under SD conditions (D). Days to heading were recorded when the first panicle emerged (n = 20 for NLD; n = 12 for SD). Box-and-whisker plots show maximum and minimum, 25–75th percentiles (box), and median (middle line). Different letters indicate significantly different values according to one-way ANOVA and Duncan’s least significant range test (P < 0.05). A Y2H assays showing the interaction between the N-terminus of the OsPRR37 fragment and full-length Hd1. -LT, synthetic defined (sd) medium lacking Leu/Trp; -LTH, sd medium lacking Leu/Trp/His. B BiFC assays demonstrating the interaction between OsPRR37 and Hd1 in the nucleus. OsPRR37 and Hd1 fused to C-terminal YFP (YC) and N-terminal YFP (YN), respectively, were expressed in rice protoplasts. The protoplasts co-expressing OsPRR37 and Hd1 showed YFP signals in the nuclei. The nuclear localization signal (NLS)–red fluorescence protein (RFP) was used as a nuclear marker. White bars = 5 μm OsPRR37 interacts with Hd1. Flowering time phenotypes and days to heading in WT, osprr37, hd, and osprr37 hd1 under natural long-day (NLD) (A, B) and short-day (SD) (C, D) conditions. The photograph was taken before the WT plant flowered under NLD conditions (A) and 7 days after the WT plant flowered under SD conditions (C). To analyze heading date, WT, osprr37, hd1, and osprr37 hd1 plants were grown in the paddy field under natural long-day (NLD) conditions (B) and in the growth chamber under SD conditions (D). Days to heading were recorded when the first panicle emerged (n = 20 for NLD; n = 12 for SD). Box-and-whisker plots show maximum and minimum, 25–75th percentiles (box), and median (middle line). Different letters indicate significantly different values according to one-way ANOVA and Duncan’s least significant range test (P < 0.05). A Y2H assays showing the interaction between the N-terminus of the OsPRR37 fragment and full-length Hd1. -LT, synthetic defined (sd) medium lacking Leu/Trp; -LTH, sd medium lacking Leu/Trp/His. B BiFC assays demonstrating the interaction between OsPRR37 and Hd1 in the nucleus. OsPRR37 and Hd1 fused to C-terminal YFP (YC) and N-terminal YFP (YN), respectively, were expressed in rice protoplasts. The protoplasts co-expressing OsPRR37 and Hd1 showed YFP signals in the nuclei. The nuclear localization signal (NLS)–red fluorescence protein (RFP) was used as a nuclear marker. White bars = 5 μm Next, we measured the transcript levels of Hd3a, RFT1, Ehd1, and Hd1 under LD and SD (Fig. 4E-L; Fig. S3-5). Previous studies have shown that OsPRR37 downregulates Ehd1, whereas OsFKF1 upregulates it under LD (Koo et al. 2013; Han et al. 2015; Liu et al. 2018). Consistently, the expression levels of Hd3a, RFT1, and Ehd1 were upregulated in osprr37 but downregulated in osfkf1 compared to those in WT under LD (Fig. 4E, G, I). The expression levels of these genes were higher in osfkf1 osprr37 than in WT, indicating that OsFKF1 upregulates Ehd1 expression by repressing OsPRR37 under LD. However, the upregulation of Hd3a, RFT1, and Ehd1 in osfkf1 osrpr37 was less significant than that in osprr37, suggesting that OsFKF1 promotes flowering through additional pathways in addition to the repression of OsPRR37. Under SD, the expression levels of Hd3a, RFT1, and Ehd1 remained unchanged in osprr37, but were reduced in osfkf1 (Fig. 4F, H, J), as previously reported (Han et al. 2015). Similarly, the expression levels of Hd3a, RFT1, and Ehd1 were reduced in osfkf1 osprr37, indicating that OsFKF1 also plays a role in promoting flowering under SD. However, the expression levels of Hd1 were not significantly different among osprr37, osfkf1, and osfkf1 osprr37 under both LD and SD (Fig. 4K, L). These genetic and molecular analyses strongly suggested that OsFKF1 upregulates Ehd1 expression through both OsPRR37 degradation and other flowering induction pathway(s). OsPRR37 Interacts with Hd1 Previous studies have suggested that OsPRR37, together with Ghd7 and DTH8, is involved in the functional switch of Hd1, converting it from a flowering inducer under SD to a flowering repressor under LD (Fujino et al. 2019; Zhang et al. 2019b). To investigate the genetic interaction between OsPRR37 and Hd1, we generated osprr37 hd1 double mutants of japonica cv. Dongjin using the CRISPR/Cas9 system. Sequencing analysis confirmed a 1-bp insertion and a 2-bp deletion in the target regions of OsPRR37 in osprr37 single and osprr37 hd1 double mutants, respectively. A 2-bp deletion and a 1-bp insertion were identified in the Hd1 target regions of hd1 single and osprr37 hd1 double mutants, respectively (Fig. S6). Under NLD, osprr37 flowered earlier than WT, but later than hd1 (osprr37, 99 DTH; WT, 116 DTH; hd1, 83 DTH) (Fig. 5A, B), indicating that the flowering repression of Hd1 is weakened in the absence of OsPRR37, and that additional factors may contribute additively to the repressive activity of Hd1. In contrast, osprr37 hd1 flowered 3 days later than hd1, and the difference was small (osprr37 hd1, 86 DTH) (Fig. 5B), suggesting that OsPRR37 does not actively repress flowering in a genetic background lacking Hd1. These results are consistent with previous studies suggesting that OsPRR37, along with Ghd7 and DTH8, contributes to the conversion of Hd1 into a flowering repressor under LD (Fujino et al. 2019; Zhang et al. 2019b). Under SD, hd1 flowered later than WT, whereas osprr37 flowered similarly to WT (hd1, 71 DTH; WT, 57 DTH; osprr37, 59 DTH) (Fig. 5C, D). The osprr37 hd1 double mutants flowered only 3 days later than the hd1 single mutants, showing no obvious differences (osprr37 hd1, 74 DTH) (Fig. 5D). To investigate whether OsPRR37 directly contributes to the repressive function of Hd1 under LD, we examined the interaction between OsPRR37 and Hd1 using Y2H assays. The results showed that the N-terminal PR domain of OsPRR37 interacts with Hd1 (Fig. 5E). Next, we performed BiFC assays to verify the in vivo interaction of OsPRR37 and Hd1. Co-expression of Hd1-YN and OsPRR37-YC in the rice protoplasts resulted in YFP signals in the nucleus (Fig. 5F), indicating that OsPRR37 interacts with Hd1 in the nucleus. Natural Variations of OsFKF11 Previous studies have shown that natural variation of OsPRR37 alleles contributes to the adaptation of cultivated rice in northern-limit areas (Koo et al. 2013; Yan et al. 2013; Gao et al. 2014; Zheng et al. 2016). We thus analyzed the natural variation of OsFKF1 in 2,034 cultivated rice accessions using the International Rice Genebank Collection Information System (IRGCIS) database to investigate the potential role of natural alleles of OsFKF1 on flowering time in rice (Mansueto et al. 2017). Two nonsynonymous single nucleotide polymorphisms (SNPs) at 564 and 1554 bp from the translation start site of OsFKF1 and three haplotypes (Types 1–3) were identified in the coding region of OsFKF1 (Fig. S7A). We also examined the distribution of rice subpopulations for each OsFKF1 haplotype. The majority of japonica varieties carried OsFKF1 Type 1, whereas Type 2 and Type 3 were found mainly in indica varieties (Fig. S7B). We then analyzed the mean days to heading (DTH) of rice accessions carrying each haplotype to investigate the association between the flowering time of rice varieties and OsFKF1 haplotypes. However, no significant difference in flowering time was observed among the OsFKF1 haplotypes. These results suggest that natural variation in OsFKF1 may not be associated with flowering time in cultivated rice (Fig. S7C). Discussion Flowering plants have intricate mechanisms to precisely regulate flowering time in response to seasonal changes in day length. The facultative SD plant, rice, has specific flowering repression pathways that have not been identified in the facultative LD plant, Arabidopsis. Several LD-specific flowering repressors, including OsPRR37, Ghd7, and DTH8, have been reported to confer photoperiod sensitivity in rice (Xue et al. 2008; Wei et al. 2010; Yan et al. 2011; Koo et al. 2013; Gao et al. 2014; Liu et al. 2018). In this study, we show a post-translational regulatory mechanism underlying photoperiod-sensitive flowering in rice (Fig. 6). We found that OsFKF1 interacts with OsPRR37 and promotes its degradation. In addition, OsPRR37 interacts with Hd1, suggesting that OsPRR37 is associated with the LD repressor function of Hd1. Therefore, these results suggest that both OsFKF1-mediated degradation of OsPRR37 and Hd1-OsPRR37 interaction are important for the photoperiodic control of flowering time in rice. Fig. 6A proposed model for the post-translational regulation of OsPRR37 in photoperiodic flowering in rice. As OsPRR37 functions as an LD-specific flowering repressor but has no function under SD conditions, we propose that OsFKF1 may mediate the degradation of OsPRR37 under SD conditions. Under SD conditions, the transcript levels of both OsPRR37 and OsFKF1 peak just before dusk, possibly allowing their interaction at night, subsequently leading to the degradation of OsPRR37 and early flowering (A). Under LD conditions, OsPRR37 is expressed earlier than OsFKF1 in the afternoon, allowing OsPRR37 protein to accumulate. Since Hd1 accumulates during the daytime and is degraded in the dark (Hu et al. 2022), OsPRR37 and Hd1 are likely to interact during the daytime to form a flowering repressor complex, which contributes to delayed flowering under LD conditions (B). Under SD and LD conditions, OsFKF1 acts as a positive regulator of flowering independent of OsPRR37-mediated regulation (dotted black arrow). Dotted circle of OsPRR37 indicates OsFKF1-mediated degradation A proposed model for the post-translational regulation of OsPRR37 in photoperiodic flowering in rice. As OsPRR37 functions as an LD-specific flowering repressor but has no function under SD conditions, we propose that OsFKF1 may mediate the degradation of OsPRR37 under SD conditions. Under SD conditions, the transcript levels of both OsPRR37 and OsFKF1 peak just before dusk, possibly allowing their interaction at night, subsequently leading to the degradation of OsPRR37 and early flowering (A). Under LD conditions, OsPRR37 is expressed earlier than OsFKF1 in the afternoon, allowing OsPRR37 protein to accumulate. Since Hd1 accumulates during the daytime and is degraded in the dark (Hu et al. 2022), OsPRR37 and Hd1 are likely to interact during the daytime to form a flowering repressor complex, which contributes to delayed flowering under LD conditions (B). Under SD and LD conditions, OsFKF1 acts as a positive regulator of flowering independent of OsPRR37-mediated regulation (dotted black arrow). Dotted circle of OsPRR37 indicates OsFKF1-mediated degradation OsPRR37 is Controlled by Post-Translational Modification The circadian clock is an endogenous time-keeping system that enables plants to synchronize physiological events with the environmental day/night cycle (Schultz et al. 2001). OsPRR37, a member of the Pseudo-Response Regulator (PRR) family, is a component of the circadian clock in rice (Koo et al. 2013). In Arabidopsis, the PRR family genes play essential roles in photoperiodic flowering and are regulated by transcriptional and post-translational feedback mechanisms (Farré and Liu 2013). Although OsPRR37 is known to be a key LD-dependent flowering repressor, its post-translational regulation remains elusive. In this study, we propose a post-translational regulatory mechanism by which OsPRR37 is degraded by OsFKF1. OsPRR37 accumulated at higher levels in the osfkf1 protoplasts than in the WT protoplasts, whereas co-expression of OsPRR37 and OsFKF1 reduced OsPRR37 protein abundance (Fig. 2), suggesting that OsFKF1 may facilitate OsPRR37 degradation. Although our study does not directly demonstrate OsFKF1-mediated degradation of OsPRR37, similar post-translational regulatory mechanisms have been reported in Arabidopsis. In Arabidopsis, the PRR family proteins TIMING OF CAB EXPRESSION1 (TOC1) and PRR5 are degraded in a dark-dependent manner by the F-box proteins ZTL, FKF1, and LKP2 (Más et al. 2003; Kiba et al. 2007; Fujiwara et al. 2008; Baudry et al. 2010). These studies showed that the circadian oscillations of TOC1 and PRR5 were disrupted in the ztl, ztl lkp2, ztl fkf1, and ztl fkf1 lkp2 mutants, suggesting that their circadian oscillation is controlled by post-translational regulation. Therefore, OsFKF1-mediated degradation of OsPRR37 may similarly play a critical role in regulating the circadian oscillation of OsPRR37, possibly promoting flowering under SD conditions in rice. Further studies are needed to provide direct biochemical evidence for OsFKF1-mediate degradation of OsPRR37, such as cell free degradation assays and in vitro ubiquitination assays. In addition, it is essential to determine OsPRR37 levels in osfkf1 or OsFKF1-overexpressing plants under different photoperiodic conditions to substantiate our proposed model. In addition to ubiquitination, phosphorylation may modulate the activity or stability of OsPRR37. Hd2/OsPRR37 is phosphorylated by two kinases Hd6/CKIIa and Hd16/CKI (Kwon et al. 2015) and both Hd6 and Hd16 repress flowering under LD and act upstream of OsPRR37 (Yamamoto et al. 2000; Ogiso et al. 2010; Hori et al. 2013; Kwon et al. 2014; Shibaya et al. 2011). This suggests that phosphorylation of OsPRR37 may positively influence its activity and/or stability. Phosphorylation of TOC1 facilitates both its degradation and stabilization in Arabidopsis (Fujiwara et al. 2008). Phosphorylation of TOC1 enhances its interaction with ZTL or PRR3, depending on the relative abundance of ZTL and phosphorylated PRR3, resulting in either promotion or inhibition of TOC1 degradation by ZTL. Therefore, while this study focused on OsFKF1-mediated OsPRR37 degradation, it is necessary to determine whether Hd6- and Hd16-mediated phosphorylation of OsPRR37 under LD conditions antagonistically inhibits OsFKF1-mediated OsPRR37 degradation or disrupts the OsFKF1-OsPRR37 interaction, resulting in late flowering. Generating phospho-mimetic or phospho-defective OsPRR37 mutants and studying their effects on degradation and protein interactions would provide valuable insights for future research aimed at a deeper understanding of the molecular mechanisms underlying photoperiodic flowering in rice. OsFKF1 Promotes Flowering Through Transcriptional and Post-Translational Regulation The external coincidence model proposes that the photoperiodic response is activated when light signals coincide with the endogenous circadian rhythm (Pittendrigh and Minis 1964). This model has been strongly supported by molecular genetic studies in Arabidopsis. In Arabidopsis, FKF1 interacts with GI to degrade CDFs under LD, when FKF1 expression peaks during the light period in the late afternoon (Sawa et al. 2007). The degradation of CDFs leads to the activation of CO, which subsequently activates FT expression and promotes flowering. By contrast, our study demonstrates that OsFKF1 promotes flowering in rice by degrading OsPRR37, a repressor of the rice-specific flowering activator Ehd1, under LD conditions. While FKF1-mediated photoperiodic flowering in Arabidopsis is primarily central to activate FT expression, in rice, it is crucial for promoting Ehd1 expression, which is unique to monocots (Song et al. 2015; Cho et al. 2016). These findings suggest that OsFKF1 regulates the monocot-specific photoperiodic flowering mechanism. These differences therefore highlight an important evolutionary divergence between monocots and dicots in photoperiodic regulation. In this study, we speculate that OsFKF1 may degrade OsPRR37 under SD to induce flowering, as OsPRR37 functions as a flowering repressor under LD, but has no function under SD. Notably, another LD-dependent flowering repressor Ghd7 has been shown to be degraded by OsGI, probably in conjunction with OsFKF1, via the 26 S proteasome-dependent pathway (Zheng et al. 2019). Ghd7 protein levels were significantly increased under LD, but decreased under SD, indicating a day-length-dependent accumulation of Ghd7. Similarly, we speculate that OsPRR37 accumulates under LD while it is degraded under SD. In Arabidopsis, FKF1 and all five PRR proteins have been shown to follow their rhythmic mRNA expression patterns (Kiba et al. 2007; Fujiwara et al. 2008; Song et al. 2012). Both OsFKF1 and OsPRR37 exhibited diurnal expression patterns with peaks at ZT10, just before dusk, under SD (Fig. 4B-D). In contrast, under LD, OsPRR37 expression peaks at ZT10, which is earlier than the peak of OsFKF1 expression at ZT 13 in the afternoon (Figs. 4A-C and 6). Therefore, we propose that the coincidence of OsFKF1 and OsPRR37 expression at night is likely critical for their interaction and for the degradation of OsPRR37 under SD, which may promote flowering. In contrast, OsPRR37 may accumulate in the late afternoon under LD conditions, leading to the suppression of flowering. To validate this hypothesis, future studies examining the protein levels of OsFKF1 and OsPRR37 under SD and LD conditions are required. Taken together, we suggest that OsFKF1 may mediate the degradation of key flowering repressors to promote flowering. Our genetic analysis revealed that osfkf1 osprr37 exhibited delayed flowering compared to osprr37, which showed no significant difference in flowering time compared to WT under SD, but exhibited an early flowering phenotype under LD (Fig. 3). These results suggest that in addition to the degradation of OsPRR37, OsFKF1 may promote flowering through additional regulatory pathways, as FKF1 activates FT expression through multiple feed-forward mechanisms in Arabidopsis (Song et al. 2012). Previously, we showed that OsFKF1 is involved in the blue light gating of Ehd1 and interacts with OsGI and OsCDF1, suggesting that OsFKF1 may degrade OsCDF1 in a blue light-dependent manner to activate Ehd1 expression (Han et al. 2015). Moreover, it has been suggested that OsFKF1 may be involved in the degradation of Ghd7 with OsGI via the 26 S proteasome-dependent pathway (Zheng et al. 2019). These findings imply that OsFKF1 modulates flowering via diverse pathways beyond OsPRR37 degradation. Further in-depth studies are necessary to identify additional OsFKF1 targets and to elucidate the multiple regulatory pathways through which OsFKF1 controls photoperiodic flowering in rice. Natural variation in key photoperiodic flowering regulators, OsPRR37, Hd1, Ghd7, and DTH8, has significantly influenced the flowering time diversity in rice and contributed to the expansion of the rice cultivation area (Xue et al. 2008; Koo et al. 2013; Gao et al. 2014; Zheng et al. 2016; Wu et al. 2020). In this study, we analyzed the natural variation of OsFKF1 alleles to investigate its potential contribution to the flowering time diversity among 3,000 rice cultivars. It revealed that OsFKF1 has three haplotypes with two nonsynonymous SNPs in the coding regions (Fig. S7). However, there is no significant association between the allelic variation of OsFKF1 and the flowering time of rice varieties. In addition to controlling flowering time, FKF1 is known to regulate various biological processes in Arabidopsis, such as the circadian clock, cellulose synthesis, and chloroplast translocation (Baudry et al. 2010; Yuan et al. 2019, 2023). Although the function of OsFKF1 has been characterized mainly in the context of photoperiodic flowering in rice, future research is needed to determine whether it participates in other biological pathways. Moreover, investigating whether natural OsFKF1 haplotypes influence variation in additional agronomic or physiological traits could shed light on the broader evolutionary significance of OsFKF1 allelic diversity. Post-Translational Modification of OsPRR37 may Affect the Functional Conversion of Hd1 Previous genetic studies have suggested that OsPRR37, together with Ghd7 and DTH8, plays a role in switching of Hd1 function from a flowering activator to a flowering repressor (Du et al. 2017; Fujino et al. 2019; Zhang et al. 2019b; Zong et al. 2021). Hd1 functions as a flowering activator under both SD and LD in the genetic background of ghd7 and osprr37 mutations (Fujino et al. 2019). Furthermore, Hd6/CKIIα enhances the repressive activity of Hd1 under LD, although Hd1 is not a target of Hd6 phosphorylation. This suggests that OsPRR37/Hd2, a target of Hd6, is required for the functional conversion of Hd1 under LD (Ogiso et al. 2010; Kwon et al. 2014). In this study, we found that OsPRR37 interacts with Hd1 (Fig. 5A, B), suggesting that the accumulation of OsPRR37 under LD may enhance the flowering repressor activity of Hd1. A previous study reported that Hd1 accumulates during the daytime and is degraded by autophagy in the dark (Hu et al. 2022). OsPRR37 may accumulate in the afternoon and interact with Hd1 during the daytime to repress flowering under LD conditions. Similarly, Ghd7 accumulates under LD conditions and interacts with Hd1 to repress flowering, while it is degraded under SD conditions (Zheng et al. 2019; Liu et al. 2022). In Arabidopsis, homodimerization and heterodimerization of TOC1 with PRRs modify its transcriptional activity by altering its DNA-binding affinity or recruiting cofactors that can convert transcriptional repressors into activators and vice versa (Somers 2012). Hd1 is a transcriptional activator, while OsPRR37 is a transcriptional repressor (Yano et al. 2000; Liu et al. 2018). Further investigation is required to determine whether the interaction of Hd1 with OsPRR37 changes the Hd1 function from a transcriptional activator to a transcriptional repressor. Therefore, we speculate that the degradation of OsPRR37 by OsFKF1 under SD may cause Hd1 to function as a flowering activator, whereas the accumulation of OsPRR37 and Hd1-OsPRR37 interaction under LD may convert Hd1 to a flowering repressor. Conclusions In this study, we reveal that OsFKF1 interacts with and degrades a flowering repressor OsPRR37. Our results suggest that OsFKF1-mediated degradation of OsPRR37 induces flowering under SD conditions. In addition, OsPRR37 interacts with Hd1 to repress flowering under LD conditions. Recently, Vicentini et al. (2023) proposed that CO plays a central role in photoperiodic flowering in Arabidopsis, whereas two CCT domain proteins, Ghd7 and OsPRR37, are central in rice in addition to Hd1. Our study further demonstrated the importance of OsFKF1-mediated degradation of OsPRR37 in the photoperiodic control of flowering in rice, in addition to the post-translational regulation of Ghd7 (Zheng et al. 2019). These findings will provide more insight into elucidating the molecular basis of photoperiodic flowering in the facultative SD plants. Moreover, these results imply that OsFKF1 is a promising target for breeding early-flowering rice varieties using CRISPR/Cas9-mediated gene editing in elite rice cultivars, enabling the optimization of flowering time under different photoperiodic conditions. Materials and Methods Plant Materials and Growth Conditions T-DNA insertion mutant lines of OsFKF1 (LOC_Os11g34460; 1B-03633; osfkf1) and OsPRR37 (LOC_Os07g49460; 1B-02503; osprr37) in the japonica cultivar ‘Dongjin’ (hereafter referred to as the wild type; WT) were obtained in our previous studies (Jeon et al. 2000; Koo et al. 2013; Han et al. 2015). The osfkf1 osprr37 double mutant was generated by crossing the homozygous osfkf1 and osprr37 plants, both of which had previously been characterized for their flowering time phenotypes through the segregation of WT, heterozygous, and homozygous plants in our earlier studies (Koo et al. 2013; Han et al. 2015). To identify homozygous osfkf1, osprr37, and osfkf1 osprr37 double mutants, genomic PCR was performed using the T-DNA plasmid pGA2715 border primers, pGA2715_RB and pGA2715_LB1, in combination with OsFKF1 and OsPRR37 gene-specific primers, respectively (Table S1). To generate the osprr37, hd1, and osprr37 hd1 mutants using CRISPR/Cas9 technology, the specific 20-nt spacer sequences were designed using the CRISPRdirect software (https://crispr.dbcls.jp/) (Naito et al. 2015) (Table S1) and introduced into the pRGEB32 vector (Xie et al. 2015). Flowering-time phenotypes were analyzed using T2 generation plants. WT and mutant plants were grown in a paddy field under NLD conditions (> 14 h light/day, 37°N, Suwon, Korea). Day lengths during rice growth in the paddy field were recorded as 14.2 h in May, 14.7 h in June, 14.5 h in July, and 13.6 h in August (measured in the middle of each month). Rice plants were also grown in growth chambers under LD (14.5 h light at 30 °C/9.5 h dark at 24 °C) or SD (10 h light at 30 °C/14 h dark at 24 °C) conditions. Flowering time was recorded from sowing to the emergence of the first panicle in the main stem. RNA Extraction and RT-qPCR Analysis Total RNA was extracted by homogenizing the samples to a powder using a TissueLyser II (QIAGEN), followed by the MG RNA Extraction Kit (Macrogen). The concentration of extracted total RNA was quantified using a NanoDrop spectrophotometer (Thermo Fisher Scientific). First-strand cDNA was synthesized from 2 µg of total RNA using M-MLV reverse transcriptase and oligo(dT)15 primers (Promega) in a total volume of 25 µL. The synthesized cDNA was diluted with 75 µL of distilled water. RT-qPCR was performed using a LightCycler 2.0 instrument (Roche Diagnostics) with 2 µL of the diluted cDNA mixture, 10 µL of 2X GoTaq Master Mix (Promega), and 1 µL of 10 pM primer pairs in a total volume of 20 µL. The qPCR cycling conditions were an initial denaturation at 95 °C for 2 min, followed by 42 cycles at 95 °C for 5 s, 59 °C for 15 s, and 72 °C for 10 s. Rice UBIQUITIN5 (OsUBQ5) was used as an internal control. The relative expression level of each gene was determined using the 2−ΔΔCT method, as previously described (Livak and Schmittgen 2001). The primer sequences used for RT-qPCR are listed in Table S1. Bimolecular Fluorescence Complementation (BiFC) Assays For BiFC assays, the full-length coding sequences of OsPRR37, OsFKF1, and Hd1 were amplified from rice cDNAs using gene-specific primers (Table S1). The amplified products were subcloned into the pCR8/GW/TOPO vector (Invitrogen). The cDNAs were then cloned into the pAM-PAT-35 S::YFPn and pAM-PAT-35 S::YFPc vectors, respectively, using the Gateway LR clonase II enzyme mix (Invitrogen). Rice protoplasts were isolated from the Oc cell and transfected with the recombinant YN and YC vectors using a 40% PEG solution. After overnight incubation in the dark, fluorescence signals were observed using a confocal laser scanning microscope (SP8 X; Leica). YFP fluorescence was detected at an excitation wavelength of 488 nm and an emission wavelength of 500–590 nm. RFP signal was detected at an excitation wavelength of 561 nm and an emission wavelength of 580–620 nm. Yeast Two-Hybrid (Y2H) Assays The Y2H assays were conducted according to the Yeast Protocols Handbook (Clontech, Kyoto, Japan). The coding sequences of the full-length OsFKF1 and OsPRR37 fragments were amplified by reverse transcription PCR using gene-specific primers (Table S1) and then inserted into the pGBKT7 or pGADT7 vectors (Clontech) to generate the GAL4 DNA-BD (binding domain) and GAL4-AD (activation domain) constructs, respectively. The respective plasmid combinations were co-transformed into yeast strain AH109. Positive transformants were grown on a synthetic defined (SD) medium lacking the amino acids leucine and tryptophan. Interaction screening was performed on two types of selective media (SD/-Leu/-Trp/-His/-Ade or SD/-Leu/-Trp/-His with 5 mM 3-amino-1,2,4-triazole) for 3 d. β-Galactosidase activity was determined using a liquid assay using chlorophenol red-β-D-galactopyranoside (CPRG; Roche Applied Science) as the substrate, as described in the Yeast Protocols Handbook (Clontech). Split Luciferase Complementation Imaging (LCI) Assays The coding sequences of OsPRR37 and OsFKF1 were cloned into pCAMBIA-NLuc and pCAMBIA-CLuc vectors, respectively. Agrobacterium tumefaciens GV3101, transformed with the indicated constructs, were incubated in infiltration buffer (10 mM MES pH 5.7, 10 mM MgCl2, and 150 µM acetosyringone) for 2 h in the dark. Agrobacterium was then infiltrated into the leaves of 3-week-old Nicotiana benthamiana, and luciferase activity was measured after 2 d. Leaves were sprayed once with 1 mM d-luciferin (Goldbio) and kept in the dark for 5 min to allow chlorophyll luminescence to decay. Luminescence was monitored using a chemiluminescence imaging system (FUSION FX7; Vilber, Germany), and the relative luminescence unit (RLU) intensities were calculated using the Image J software. Protein Degradation Assays in Rice Protoplasts Rice protoplasts were isolated from 10-d-old WT and osfkf1 seedlings grown under LD (14.5 h light/9.5 h dark). The coding sequences of OsFKF1 and OsPRR37 were amplified by reverse transcription PCR using gene-specific primers (Table S1) and cloned into the pGA3817 or pGA3818 vectors (Kim et al. 2009), driven by maize Ubiquitin promoter, to generate Myc- or HA-tagged constructs. 5 × 106 cells mL− 1 of rice protoplasts were transfected with 15 µg of each vector using the polyethylene glycol (PEG)-mediated transfection method (Yoo et al. 2007). Transfected protoplasts were suspended in the protoplast incubation solution (0.5 M mannitol, 20 mM KCl, and 4 mM MES, pH 5.8) and kept in the dark at room temperature for 16 h to induce protein expression. Protoplasts were then harvested and transiently expressed proteins were detected by immunoblotting with anti-Myc and anti-HA antibodies. Immunoblot Analysis Total proteins were isolated from rice protoplasts using 50 µL of lysis buffer (50 mM Tris-HCl, pH 7.5, 150 mM NaCl, 0.5% NP-40, and 1 mM EDTA). After centrifugation twice at 12,000 g for 15 min, the supernatants were then boiled with an SDS sample buffer at 95 °C for 8 min. Total proteins were separated using 10% or 15% SDS-PAGE and transferred to PVDF membranes (Merck Millipore). The membranes were immunoblotted with anti-Myc (1:10000, #2040; Cell Signaling Technology) or anti-HA (1:10000, #2367; Cell Signaling Technology) antibodies. Bound proteins were detected using the Western blot detection reagent WESTSAVE (AbFrontier) according to the manufacturer’s protocol. Proteins were also immunoblotted with an H3 antibody as a loading control. The relative amounts of proteins (anti-Myc/anti-H3) were calculated using ImageJ software (https://imagej.nih.gov/ij/). Co-Immunoprecipitation Assays OsFKF1-HA and OsPRR37-Myc fusion proteins were transiently expressed in rice protoplasts prepared from the Oc cells. After overnight incubation, the protoplasts were harvested and lysed in immunoprecipitation (IP) buffer containing 75 mM NaCl, 50 mM Tris-HCl [pH 7.5], 5 mM EDTA, 1% Triton X-100, 1 mM dithiothreitol (DTT), 1 mM phenylmethanesulfonyl fluoride, 2 mM NaF, 20 µM MG132, and protease inhibitor cocktail (Roche). The cell lysates were clarified by centrifugation at 12,000 rpm for 5 min at 4 °C, and the supernatants were incubated with anti-HA agarose beads (M20013; ABmart) for 2 h at 4 °C. The beads were washed three times with wash buffer containing 75 mM KCl, 50 mM Tris-HCl [pH 7.5], 5 mM EDTA, 0.1% Triton X-100, 1 mM DTT, and 1 mM phenylmethanesulfonyl fluoride. Finally, the beads were boiled with 20 µL of protein loading buffer for 5 min to elute the bound proteins. The eluted proteins were separated by electrophoresis on a 10% SDS-polyacrylamide gel and detected by immunoblot analysis. Analysis of Single Nucleotide Polymorphism (SNP) To investigate the genetic variation of OsFKF1 in cultivated rice (Oryza sativa), we searched for SNPs within the coding sequence of OsFKF1 using the 3 K Rice Genome Project database from the International Rice Genebank Collection Information System (IRGCIS; https://snp-seek.irri.org/) (Mansueto et al. 2017). Japonica cultivar ‘Nipponbare’ was used as the reference genome for SNP analysis. Flowering time data for the different rice cultivars were obtained from the IRGCIS phenotype database. Statistical Analyses The statistical significance of the differences in mean values was determined using two-tailed Student’s t-tests with P values < 0.05. One-way analysis of variance (ANOVA) with the Duncan’s least significant range test (P < 0.05) was used to determine the statistical significance of differences among more than two populations. All central values shown in figures are means of biological or experimental replicates. Box plots were constructed to represent flowering time distributions (Williamson 1989; Spitzer et al. 2014). In the box plots, the central lines indicate the medians and plus signs (+) indicate means; box boundaries indicate the 25th and 75th percentiles. Statistical analyses were performed using RStudio software (https//www.rstudio.com/) (R version 4.2.2). Supplementary Information Supplementary Material 1. Supplementary Material 1. Supplementary Material 2. Supplementary Material 2.