In annual crop, spikelets are typically begun during the initial reproductive phases, while primordia form inside the spikelets and determine spike fertility later in the repro-ductive phase (Ochagavía et al. 2018). In contrast to the development of leaves and spikelets, when all initiated organs are visible in fully grown plants, floret development involves a period of floret initiation prompted by a phase of floret death (Kim and Khan, et al. 2021). Generally, 6-12 floret primordia are formed in each spikelet; however, most degenerate before reaching the anthesis stage (Sibony and Pinthus 1988). The proper rate of this mortality and the onset of floret mortality are most likely related to the amount of assimilates available to the developing juvenile spike (González et al. 2003; Guo et al. 2016). As a result of the floret's production and degeneration, the potential grain number is determined during the pre-anthesis stage (Slafer and Savin 1994). Moreover, the most of these processes occur while the spikes are connected alongside the culm during the pre-anthesis stage; the quantity of grains is related to the number of fertile florets that emerge during stem elongation (Ferrante et al. 2010, 2013). The number of fertile florets is usually determined by the number of remaining florets rather than the total number of florets produced (Ferrante et al. 2013). Thus, the total number of spikelet on each spike and the number of fertile florets per spikelet typically determine the wheat grain yield potential (Li et al. 2019).

In Arabidopsis, the time to bolting or the rosette leaf number when the first floral bud emerges are two often utilized macroscopic markers of phase shift in reproductive phase or floral transitions (Pouteau and Albertini 2009). However, in temperate grasses such as wheat (Tritium aestivum) and barley (Hordeum vulgare), most phases of reproductive development, such as blooming, occur within the leaf sheath and can be scored most efficiently when the shoot is dissected. Waddington et al. (1983) produced a quantified index for wheat and barley development based on carpel and shoot morphogenesis. This scale is associated with the development of advanced floret primordium and/or floret. In wheat and barley, flowering is generally scored as heading, however the several stages of shoot apex development differ in their sensitivity to environmental stimuli and are influenced by unique genetic variables; hence, floral transition and flowering time may not be linked or separated (Digel et al. 2015).

Flowering time and/or maturity are determined by an extensive network of genetic regulators in plants that res-pond to external stimuli and endogenous genes (Kong et al. 2018). The cross-talk network between VERNALIZATOIN genes such as VRN1, VRN2, VRN3, and VRN-D4 in wheat and barley consolidates vernalization and photo-period pathway-related flowering time (Li et al. 2020). However, in Arabidopsis, the flowering regulatory network modulates the MADS-domain transcription factor Flowering LOCUS C (FLC), which influences large numbers of target genes and is also known as a fundamental regulator of flowering.

In all eukaryotic cells, the ubiquitin (Ub) 26S pro-teasome system is a critical major mechanism for protein degradation (Cho et al. 2006). This system degrades a large number of proteins in the cytoplasm and nucleus and is important in the regulation of cell metabolism such as cell differentiation, endocytosis, embryogenesis, senescence, hormone responses, and pathogen defense (Smalle and Vierstra 2004; Frugis and Chua 2002). In particular, the ubiquitination system required E1 (activating enzyme), E2 (conjugating enzyme), and E3 (ubiquitin ligase). An E3 ligase is an essential enzyme that recognizes and controls substrate proteins (Morreale and Walden 2016). The RING (Really Interesting New gene)-type is one of the E3 ligases, and it consists of eight zinc-coordinating C (Cys) and H (His) residues (in the order C C-X2-C-X(9–39)-C-X(1–3)-H-X(2–3)-Cys-X2-C-X(4–8)-C-X2-C (X can be any amino acid) that forms a complex two zinc atoms (Saurin et al. 1996). Many RING E3 ligases have been found to be involved in the regulation of plant growth and develop-ment, hormone signaling, and responses to biotic and abiotic stresses (Hwang et al. 2003; Cho et al. 2006; Park et al. 2018; Kim and Khan et al. 2021; Kim and Kim et al. 2021). Despite extensive research into the function of flowering related pathways in other plants, little known about the physiological process underlying E3 ligase, particularly in wheat.

In this study, we demonstrate the function of Wheat Membrane Associated Ring-CH-Type Finger 9, TaMARCH9, whose expression level was highly increased from differen-tiation to the end of floret initiation stages (WS3-6). The plasma membrane localized TaMARCH9 protein possessed E3 ligase activities. Moreover, transgenic TaMARCH9 in Arabidopsis plants had earlier bolting phenotype than wild type plant.

Plant material and growth conditions

The wheat cultivar (T. aestivum L., cv. Yeongkwang) seeds were vernalized for 50 days in a dark room at 4°C. At the end of vernalization, the plants were transferred to soil pots in a greenhouse at 25°C under 16 hours/8 hours light/dark photoperiod conditions. Wheat primordia tissues at various developmental stages (WS1.5 to 6). were collected for analysis. As previously mentioned, each stage of wheat primordia and floret measurement and sampling informa-tion was clarified by Kim and Khan et al. 2021.

Gene expression studies

To investigate the relative expression level, total RNA extraction, cDNA synthesis, and quantitative Real-time PCR (qRT-PCR) were used. TRIzol^Ⓡ (Invitrogen, Carlsbad, CA, USA) was used to extract total RNA from 40 to 50 wheat primordia tissues at various developmental stages (WS1.5 to 6). First-strand cDNA was created from the total RNA (2 mg) using the Power cDNA Synthesis kit (iNtRON Biotechnology, Seongnam, Gyeonggi, Korea) following the manufacturer’s instructions. The qRT-PCR findings were examined using a CFX96 real-time PCR (Bio-Rad, Hercules, CA, USA). As an internal control, TaActin was used. Supplementary Table S1 contains a list of all primers used in this work.

Subcellular localization

To determine the subcellular localization of TaMARCH9, TaMARCH9 was cloned into green fluorescent protein (GFP)-vector (pMDC43) that containing CaMV 35S pro-moter using LR reaction (Invitrogen, Carlsbad, CA, USA). The constructs generated with 35S:TaMARCH9-GFP and 35S:GFP (empty) were applied to polyethylene glycol- transfected wheat protoplasts for transient experiments. Subcellular localization was performed as previously described by Kim (Kim et al. 2021). Fluorescent signals were conducted by fluorescence microscopy (LSM 700; Carl Zeiss, Jena, Germany).

In vitro ubiquitination assay

To examine the ubiquitination activities of TaMARCH9, maltose binding protein (MBP) tagged vector (pMAL-C-5X) was constructed. TaMARCH9-MBP was expressed by E. coli strain (BL21) and purified by amylose resin (NEB, Ipswich, MA, USA). TaMARCH9-MBP protein were co-incubated 3 hours in absence or presence of E1, E2, ubiquitin, and ubiquitination reaction butter [1 M Tris-HCl, pH 7.5; 40 mM ATP; 100 mM MgCl₂; 40 mM dithiothreitol (DTT)) at 30°C degree. Anti-MBP (NEB, Ipswich, MA, USA) and anti-Ubi (Sigma-Aldrich, St Louis, MO, USA) were used to immunoblot the reaction products.

Overexpression of TaMARCH9 in Arabidopsis

Arabidopsis (ecotype type Columbia-0) was used to generate TaMARCH9 overexpressing plants. Arabidopsis flowers were soaked with Agrobacterium tumefaciens cells (GV3101) with recombinant vectors (TaMARCH9-GFP) suspended in 5% sucrose and 0.05 percent Silwet-L77 (Sigma-Aldrich, St Louis, MO, USA). All of the steps involved in Arabidopsis transformation were carried out exactly as described previously (Ho et al. 2022). For transgenic plant selection, T₁ generation seeds were allowed to grow on MS medium with 50 mg/L hygromycin. Independent T₃ overexpression lines (OE1, OE2) were used in this study.

Expression analysis of TaMARCH9 at floret development stages in wheat

In previous study, the expression level of 34 wheat RING type E3 ligase protein that were selected based on domain analysis were investigated under floret primordia development stages (Kim and Khan et al. 2021). Of these, we verified the transcription pattern of 34 wheat RING type E3 ligase under primordia developmental stages. Among them, a transcript level of TraesCS7A02G283700 (TaMARCH9) was highly induced at reproductive development phases. After the initiation of reproductive phase, TaMARCH9 was up-regulated from WS 3 to WS 6 (about 4 to 12-fold) (Fig. 1).

Figure 1. Expression pattern of TaMARCH9 was regulated different stage of primordia development. (A) TaMARCH9 transcripts quantified by qRT-PCR in wheat primordia samples (WS: Waddington scale). Each bar represents the mean ± SD of three inde-pendent experiments.

Characterization and ubiquitination activity of TaMARCH9

To characterize the TaMARCH9, we designed gene specific primers based on Triticum aestivum L. in the NCBI sequence database to isolate the TaMARCH9 (Supplementary Table 1). Se-quence analysis revealed that the full-length of TaMARCH9 protein encodes 311 amino acid residues, which contains RING-CH-type- domain (Fig. 2A). Multiple sequence align-ment indicated that TaMARCH9 had a high identity with Brachypodium distachyon MARCH9 (BRADI_ 3g34880v3) (92%) over the protein size as well as a similar level of identity with Zea mays RING type E3 ligase protein (Zm00001eb055230) (87%) (Supplementary Fig. 1A).

Figure 2. Domain analysis in TaMARCH9 and ubiquitination assay. (A) Schematic diagram of full-length TaMARCH9 protein. The box depicts the coding region and the solid lines represent the 5’- and 3’-untranslated regions. (B) In vitro ubiquitination assay with TaMARCH9. The ubiquitination reaction mixture contained E1 (Human), E2 (Arabidopsis), maltose-labeled protein TaMARCH9 (MBP-TaMARCH9), ubiquitin, and ATP. Polyubiquitin chains were visualized via immunoblotting using an anti-ubiquitin and anti-MBP antibody.

To determine the ubiquitination activities of TaMARCH9, we constructed MBP-tagged TaMARCH9 recombinant pro-tein. Ubiquitination chains were observed in the presence of all of these components with MBP-tagged TaMARCH9, according to immunoblot analysis using both anti-Ub and anti-MBP, respectively (Fig. 2B).

Subcellular localization of TaMARCH9

In order to explore the transient expression of TaMARCH9 protein, the full length TaMARCH9 gene was fused into it with GFP to generate 35S:TaMARCH9-GFP. In wheat protoplast, the GFP tagged TaMARCH9 protein was tran-siently expressed using transfection methods. TaMARCH9 was mainly observed on plasma-membrane, whereas the empty GFP protein was appeared in the nucleus as well as plasma-membrane (Fig. 3).

Figure 3. Subcellular localization of TaMARCH9. Subcellular localization of 35S:GFP-TaMARCH9 in wheat pro-toplasts. Bars = 10 mm. The fluorescence signal was detected using a confocal laser-scanning micro-scope.

TaMARCH9-overexpressing in Arabidopsis showed accelerated flowering phenotype

Since TaMARCH9 transcription level was shown to be considerably increased during the differentiation stage (WS 3 to 6), we predicted that TaMARCH9 would be connected to flowering time. Moreover, the cauliflower mosaic virus 35S constitutive promoter (CaMV35S) is a foundational promoter that could be found in a variety of cell types, tissues, and organs in plants (Li et al. 2011). To examine the phenotype of TaMARCH9, CaMV35S was used to create transgenic Arabidopsis plants that overex-pressed TaMARCH9. Two independent T₃ transgenic lines with considerably enhanced level of TaMARCH9 tran-script level were chosen for further investigation (Fig. 4A). During our investigation for phenotypic difference, we found that under the normal growth condition (16 hours light/8 hours dark), the TaMARCH9 overexpression lines (OE1, 2) displayed early flowering phenotype (Fig. 4B). In this regard, we also recorded rosette leaf number and flowering time to determine the impact of TaMARCH9 overexpression on phase transitions in Arabidopsis. Com-pare to wild type (WT), TaMARCH9 OE1, 2 showed decreased rosette leaf number and reduced time to flowering (Fig. 4B, C). TaMARCH9 OE was therefore able to accelerate Arabidopsis flowering.

Figure 4. Early flowering phenotype of TaMARCH9 in Arabidopsis. (A) Semi-quantitative RT-PCR of TaMARCH9 overexpressing Arabidopsis (B) Flower-ing phenotype of the transgenic plants overex-pressed TaMARCH9. The WT and 35S:TaMARCH9- GFP (OE) plants in a chamber (LD, 16 hours light/8 hours dark photoperiod at 22°C/18°C). Bars = 3.5 cm. Rosette leaf number at bolting (C) Days of opening of first flower (D) in the WT and OE line in plants. Data represent the mean ± SD of ten biological replicates, and asterisks re-present significant differences (*P value < 0.05, and **P value < 0.01 Student’s t-test).

Expression analysis of TaMARCH9-overexpressing in Arabidopsis

Fig. 4 shows the phenotypic differences in flowering time between the WT and TaMARCH9 OE in Arabidopsis. As rosette leaf number and bolting time was significantly reduced in OE lines, we hypothesized that TaMARCH9 might be explained by autonomous pathways. To check the expression of flowering-time induced genes, we confirmed some of floral accelerate genes (FT and SOC) and repressor genes (FLC and MAF3). The transcript levels of floral accelerator genes (FT and SOC), was considerably increased (about 3 to 10-fold) in the OE lines (Fig. 5). On the other hands, the expression of floral repressors genes (FLC and MAF3) was reduced in OE lines (Fig. 5). These findings showed that TaMARCH9 was implicated in the transcrip-tion levels of flowering time-related genes in response to flowering.

Figure 5. Expression analysis of TaMARCH9 in Arabidopsis. Expression pattern of predicted downstream genes in the WT and OE1, 2 plants.

Flowering time and shorter periods of vegetative stages might become particularly significant for wheat produc-tivity in condition of abiotic stress since they can reduce harsh conditions during the sensitive flowering and post- anthesis grain filling periods (Shavrukov et al. 2017). As a result, early flowering is a useful method for producing improved wheat cultivars.

Previous studies verified that different types of E3 ligases were highly induced at floral transition period (Kim and Khan et al. 2021; Ho et al. 2022). Similarly, our finding reveals that the TaMARCH9 was highly up-regulated at reproductive stages, causing it to be involved in flowering time in wheat. During its life cycle, Arabidopsis passes through at least two phase transitions: vegetative and reproductive (Bäurle and Dean 2006). The vegetative phase transition is characterized by a shift from the juvenile to the adult vegetative stage, which is generally recognized by the production of leaves with abaxial trichomes; in contrast, the substantial vegetative to reproductive floral transition, is recognized by bolting, flowering, and seed set for the next generation (Bäurle and Dean 2006; Li et al. 2011).

Endogenous regulator that are independent of the photoperiod, temperature, and gibberellin pathways are referred to as the autonomous pathway for regulating flowering. All of the genes in the autonomous pathway suppress FLC expression. FLC is a core floral repressor that prevents the induction of flowering during vegetative growth by blocking the expression of floral activators such as FT and SUPPRESSOR OF OVEREXPRESSION OF CO1 (SOC1) (Srikanth and Schmid 2011). The inhibition of FLC promotes activation of FT and SOC expression level (Michaels et al. 2005). In addition, the additional members of the FLC gene family, denoted MAFs (MAF1-5), can operate as floral repressors (Ratcliffe et al. 2001). Similarly, early flowering phenotype TaMARCH9 showed that the mRNA level of FLC and the transcription level of FT and SOC were reversed (Fig. 5).

Various type of E3 ligases, which contains RING, U-box, SCF complex domain, regulate plant external and internal stimuli (Hwang et al. 2003; Li et al. 2012; Ho et al. 2022). Moreover, E3 ligase encoded proteins such as RING, U-box, F-box type has been linked to the regulation of reproductive phase development, including flowering time, maturity, reproductive organ development (Cao et al. 2008; Li et al. 2012; AN et al. 2017; Hong et al. 2020). This study demonstrates that TaMARCH9 protein encoded RING domain called RING-CH-type at its N-terminal side (Fig. 2A). The TaMARCH9 exhibited ubiquitination chains when combined with E1 and E2 enzymes and also was localized in plasma-membrane (Fig. 2B, 3). Plasma-mem-brane in plant cells serves key functions in response to transporting solution and sensory transductions (Sussman and Harper 1989).

This work was supported by a grant from the Korea University and Basic Science Research Program through the National Research Foundation of Korea, which is funded by the Ministry of Science, ICT & Future Planning (grant number 2017R1D1A1B06030349).