A positive feedback loop of cytokinin signaling ensures efficient de novo shoot regeneration in Arabidopsis

Plants possess a remarkable ability to regenerate tissues, which enables the healing of wounds and the induction of de novo organogenesis. In vitro plant tissue culture techniques are based on the regenerative capacity of plants and facilitate the reprogramming of differentiated somatic cells into a new organ or even an entire plant (Sugimoto et al., 2010). Differentiated plant tissues are used as explants to generate a pluripotent cell mass, called callus, on auxin-rich callus-inducing medium (CIM) (Ikeuchi et al., 2013; Zhai & Xu, 2021; Yin et al., 2024). Subsequently, the callus undergoes de novo shoot regeneration on cytokinin-rich shoot-inducing medium (SIM) (Che et al., 2007). A particular emphasis has been placed on de novo shoot organogenesis because the low shoot regeneration rate frequently limits in vitro plant regeneration in many species (Ijaz et al., 2012; Zimik & Arumugam, 2017). Consistent with the fact that de novo shoot regeneration during in vitro tissue culture involves the conversion from callus cells to shoot meristem (Meng et al., 2017; Ogura et al., 2023), key regulators of shoot apical meristem (SAM) establishment are implicated in de novo shoot regeneration (Ikeuchi et al., 2016; Eshed Williams, 2021; Mathew & Prasad, 2021). The PLETHORA 3 (PLT3), PLT5, and PLT7 genes, which are expressed in the whole process of plant regeneration, play a particular role in shoot progenitor formation. Upon transferring to SIM, they are specifically expressed in shoot progenitor cells and promote promeristem formation by activating CUP-SHAPED COTYLEDON 1 (CUC1) and CUC2 (Kareem et al., 2015). The CUC1 and CUC2 proteins are involved in promoting SHOOT MERISTEMLESS (STM) expression and polarizing PIN-FORMED 1 (PIN1) localization to initiate shoot meristem development (Hibara et al., 2003; Bilsborough et al., 2011; Kamiuchi et al., 2014; Kareem et al., 2015). CUC2 also activates the expression of XYLOGLUCAN ENDOTRANSGLUCOSYLASE/HYDROLASE 9 (XTH9) encoding a cell wall-loosening enzyme in nonprogenitor cells and contributes to establishing cell polarity for meristem formation (Varapparambath et al., 2022). Additionally, the main cytokinin regulatory axis is linked to the establishment of shoot stem cells in callus. Type-B ARABIDOPSIS RESPONSE REGULATORs (ARRs), positive regulators of cytokinin signaling, directly promote the expression of WUSCHEL (WUS), which unequivocally regulates the formation of the shoot stem cell niche (Meng et al., 2017; Zhang et al., 2017). Accordingly, mutations in type-B ARRs lead to impaired de novo shoot regeneration (Meng et al., 2017). By contrast, type-A ARRs, which play a negative role in cytokinin signaling, repress de novo shoot regeneration (Buechel et al., 2010). The APETALA2 (AP2)/ETHYLENE RESPONSE FACTOR (ERF)-type transcription factor gene ENHANCER OF SHOOT REGENERATION 1 (ESR1)/DORNROSCHEN (DRN) is involved in diverse aspects of plant regeneration, including wound-induced callus formation and de novo shoot regeneration (Iwase et al., 2017). In particular, during in vitro tissue culture, ESR1 is induced in response to cytokinin and promotes de novo organogenesis from callus (Banno et al., 2001; Iwase et al., 2017). Ectopic expression of ESR1 substantially enhances de novo shoot regeneration, whereas esr1 mutants display reduced de novo shoot formation from calli (Banno et al., 2001; Iwase et al., 2017). Despite the importance of ESR1 in de novo shoot regeneration, its modes of action in the plant cell remain unclear. In this study, we report that ESR1 stimulates cytokinin signaling and ensures efficient de novo shoot regeneration. ESR1 directly activates type-B ARR genes, which ultimately activate WUS. Notably, type-B ARRs also bind to the ESR1 promoter and activate its expression, establishing a positive feedback loop of cytokinin signaling. Collectively, the ESR1–type-B ARR module acts as a crucial player in the process of de novo shoot regeneration by strongly activating cytokinin responses to maximize the plant regeneration efficiency. Arabidopsis thaliana ecotype Columbia-0 was used in all experiments. The arr1 arr12 (CS6981), arr12-1 (CS6978), ARR1-Ypet (CS71599), and ARR12-Ypet (CS71601) seeds were obtained from the Arabidopsis Biological Resource Center. drn-1/esr1-2 and ProESR1:ESR1-GFP were gifts from Dr John Chandler (Chandler et al., 2007) and Dr Akira Iwase (Iwase et al., 2017), respectively. For the generation of 35S:ARR1-MYC and 35S:MYC-ESR1 transgenic plants, 35S:ARR1-MYC (35S:ARR1-10×MYC) and 35S:MYC-ESR1 constructs were cloned into pGWB520 (Invitrogen) and Myc-pBA plasmids, respectively, and transformed into Arabidopsis via Agrobacterium-mediated transformation. Plants were grown under long-day (LD; 16 h : 8 h, light : dark) conditions with white fluorescent light (150 μmol photons m−2 s−1) at 22–23°C. To induce callus formation, leaf explants were obtained from 2-wk-old plants, whereas hypocotyl and root explants were obtained from 7-d-old plants. Tissue explants were placed on CIM (MS medium, including vitamins, 3% w/v sucrose, 0.1 μg ml−1 2,4-dichlorophenoxyacetic acid (2,4-D), and 0.05 μg ml−1 kinetin) and incubated at 22°C in the dark for the indicated time period. To induce shoot regeneration, calli preincubated on CIM for 7 d were transferred to SIM (MS medium, including vitamins, 3% w/v sucrose, 0.9 μmol l−1 indole-3-acetic acid (IAA), and 2.5 μmol l−1 2-isopentenyladenine (iPA)) and incubated at 25°C under continuous light conditions. Total RNA was extracted from plant materials using TRI reagent (TaKaRa Bio, Tokyo, Japan) according to the manufacturer's recommendations. First-strand complementary DNA was synthetized using Moloney Murine Leukemia Virus reverse transcriptase (Enzynomics, Seoul, South Korea) according to the manufacturer's protocol. Reverse transcription quantitative polymerase chain reaction (RT-qPCR) analyses were performed using TOPreal SYBR Green qPCR High-ROX PreMIX (Enzynomics). Primers used for qPCR are listed in Supporting Information Table S1. The abundance of PCR products by each primer pair was normalized relative to the EUKARYOTIC TRANSLATION INITIATION FACTOR 4A1 (eIF4A) gene (At3g13920). The comparative ΔΔCt method was used to evaluate the relative quantity of each amplified product in the sample. The threshold cycle (Ct) was automatically determined for each reaction with the analysis software using default parameters. Specificity of the RT-qPCR reactions was determined by melt curve analysis of the amplified products. The chromatin immunoprecipitation (ChIP) assay was performed as previously reported (Lee et al., 2021). Transgenic plant samples containing epitope-tagged proteins were cross-linked with 1% formaldehyde. The cross-linked tissue samples were ground into a fine powder in liquid nitrogen and then sonicated. The sonicated chromatin complexes were precipitated with Pierce Anti-c-Myc Magnetic Beads (88842; Invitrogen). Precipitated DNA was purified using a DNA Purification Kit (Cosmogenetech, Seoul, South Korea). The abundance of specific DNA fragments in the precipitate was quantified by qPCR, and values were normalized relative to that of eIF4a (Table S2). Reporter and effector plasmids were constructed to perform transient expression assays using Arabidopsis protoplasts. To construct reporter plasmids, the promoter and/or genic sequences of ARR1, ARR12, ESR1, and WUS were separately cloned into the modified pCAMBIA1305 vector containing the minimal 35S promoter sequence and the GUS coding sequence. To construct effector plasmids, ARR1, ARR12, and ESR1 cDNAs were cloned into the pBA002 plasmid containing the CaMV 35S promoter. Protoplast isolation was performed based on a previous report with some modifications (Jeong et al., 2021). Ten-day-old whole seedlings were soaked in 20 ml of enzyme solution containing 2% (w/v) viscozyme, 1% (w/v) cellulase, 1% (w/v) pectinase, 10 mM MES (pH 5.7), 0.47 M d-mannitol, and 10 mM CaCl2 for 6 h at room temperature. The enzyme solution was sieved through a 70 μm nylon mesh (Carolina Biologicals, Burlington, NC, USA) and spun at 100 g for 7 min at room temperature. Recombinant reporter and effector plasmids were co-transformed into Arabidopsis protoplasts by polyethylene glycol (PEG)-mediated transformation. GUS activity in the transformed protoplasts was measured after 16 h of incubation in the dark at 23°C. A 35S:luciferase (LUC) construct was also co-transformed, along with reporter and effector plasmids, into Arabidopsis protoplasts as an internal control. The LUC assay was performed using the Luciferase Assay System Kit (Promega). Protoplasts were isolated and transfected as described above. One milliliter of protoplasts (1 × 107 ml−1) was transfected with 100 μg of plasmid DNA. After incubation, the transfected protoplasts were harvested and lysed in 400 μl of IP buffer (50 mM Tris–HCl pH 7.5, 150 mM NaCl, 0.1% NP-40, 1× protease inhibitor cocktail). The lysate was rotated at 12 rpm for 1 h at 4°C to ensure complete lysis. Cell debris was removed by centrifugation at 12 000 g for 10 min at 4°C. The resulting supernatant (400 μl) was carefully transferred to a new tube and immediately incubated with 30 μl of Anti-MYC magnetic beads (88842; Invitrogen) for 2 h at 4°C with gentle rotation at 10 rpm. The beads were then collected using a magnetic stand and washed three times with IP buffer. Immunoprecipitated proteins were eluted by boiling the beads in 45 μl of 2× sodium dodecyl sulfate (SDS)–polyacrylamide gel electrophoresis (PAGE) loading buffer for 10 min. The eluted proteins were then subjected to SDS–PAGE and subsequently immunodetected with an anti-MYC (05-724; Millipore) or anti-GFP antibody (ab290; Abcam, Cambridge, UK). Etiolated hypocotyl explants of ProESR1:ESR1-GFP, ARR1-Ypet, and ARR12-Ypet were fixed using 4% (w/v) formaldehyde and then cleared with ClearSee solution (031-25151; Wako Chemicals, Osaka, Japan). The cell walls were stained with 1% Calcofluor White solution (4404-43-7; MP Bio, Santa Ana, CA, USA). Fluorescence signals were observed using a confocal microscope (ECLIPSE Ti2; Nikon, Tokyo, Japan). GFP and YFP proteins were excited at a wavelength of 488 nm with a laser power of 6.5, and the emitted light was detected at 502/41 nm. Calcofluor White was excited at 405 nm with a laser power of 0.1, and the emitted light was detected at 429/74 nm. To determine the key molecular components specifically associated with de novo shoot organogenesis in the two-step plant regeneration process, we assembled a list of genes that primarily respond to cytokinin and are also known as regulators of shoot development during normal plant growth. Among others, we were particularly interested in ESR1 (Iwase et al., 2017), because its mechanism of action in the de novo shoot regeneration process of a two-step in vitro tissue culture remains unclear, despite its biological importance. To validate the results of previous studies, we examined whether ESR1 responds to cytokinin. Indeed, the expression of ESR1 was increased specifically by exogenous cytokinin treatment and showed a negligible response to auxin in young seedlings (Fig. 1a). In agreement with this result, while incubation of leaf explants on auxin-rich CIM negligibly influenced ESR1 expression, ESR1 was transiently induced upon incubation on SIM (Fig. 1b). Confocal imaging analysis using ProESR1:ESR1-GFP transgenic calli showed that ESR1 expression was initially observed in inner cell layers but later detected mainly in outer cell layers in callus during incubation on SIM (Fig. 1c). Given that the outer cell layer gives rise to SAM (Ogura et al., 2023), the spatial expression of ESR1 in the outer cell layers suggests its role in de novo shoot regeneration. To estimate the impact of ESR1 on shoot regeneration, we tested the shoot regeneration ability of ESR1-deficient esr1-2 mutant calli (Chandler et al., 2007; Iwase et al., 2017). Leaf explants were preincubated on CIM for 7 d and then transferred to SIM. The number of regenerated shoots was counted to quantify the de novo shoot organogenesis capability of callus tissues. The esr1-2 mutant calli produced almost no shoots, while wild-type (WT) calli regenerated approximately four to five shoots per callus at 21 d after incubation on SIM (DAS) (Fig. 1d,e), consistent with the previous result (Iwase et al., 2017). Similar results were obtained from hypocotyl and root explants (Fig. 1d,e). However, it should be noted that ESR1 did not participate in callus proliferation, as esr1-2 exhibited no clear alterations in callus formation compared with the WT (Fig. S1). To further support these results, we examined the de novo shoot regeneration capability of leaf explant-derived calli using 35S:ESR1 transgenic plants. The ectopic expression of ESR1 increased shoot regeneration, albeit with reduced callus formation (Fig. S2; Banno et al., 2001; Iwase et al., 2017). These observations suggest that ESR1 is an essential regulator of de novo shoot regeneration during in vitro tissue culture. We next asked what molecular components are mainly regulated by ESR1 during de novo shoot organogenesis. We harvested WT and esr1-2 mutant calli at 4 DAS, when ESR1 was highly expressed in WT callus (Fig. 1b). As ESR1 plays a particular role in cytokinin-dependent shoot regeneration, we examined gene expression profiles, which are known as core regulators of cytokinin signaling as well as de novo shoot organogenesis, such as ARR1, ARR12, CUC1, CUC2, KIP-RELATED PROTEIN 1 (KRP1), STM, and TEOSINTE BRANCHED1/CYCLOIDEA/PROLIFERATING CELL FACTOR 3 (TCP3) (Daimon et al., 2003; Liu et al., 2016; Meng et al., 2017; Yang et al., 2020; Wu et al., 2022). RT-qPCR analysis revealed that the expression of a couple of genes was selectively altered in esr1-2 mutant calli compared with WT calli incubated on SIM: ARR1 and ARR12, positive regulators of cytokinin signaling, were downregulated regardless of the origin of tissue explants, while the other genes were not influenced by ESR1 mutation (Figs 2a, S3). It was notable that while CUC1 is known to be upregulated by ectopic expression of ESR1 (Matsuo et al., 2009), it was unchanged in esr1-2 mutant (Fig. 2a), which may explain that ESR1 indirectly affects CUC expression. The WOX5 gene, which is involved in pluripotency acquisition, was also uninfluenced in the esr1-2 mutant (Fig. S4). Time-course expression analysis showed that transcript accumulation of ARR1 and ARR12 was transiently increased in WT calli during 2–4 DAS, but their transient induction was impaired in esr1-2 mutant calli (Fig. 2b). Consistently, expression of WUS, a target of type-B ARRs, was also reduced in esr1-2 mutant calli at 7 DAS (Fig. 2b), indicating that ESR1 promotes de novo shoot regeneration primarily by activating the type-B ARR–WUS module. We next examined whether the ESR1 transcription factor directly binds to the promoters of genes whose expression was altered in the esr1-2 mutant (Fig. 2a). Chromatin immunoprecipitation-quantitative PCR (ChIP-qPCR) analysis using 35S:MYC-ESR1 transgenic calli incubated on SIM (Fig. S5) showed that ESR1 selectively binds to the ARR1 and ARR12 loci (Fig. 2c,d). To further support the direct binding of ESR1 to type-B ARR loci, we conducted transient gene expression assays using Arabidopsis protoplasts. A reporter construct carrying the ESR1-binding ARR1 or ARR12 genomic sequence fused to the GUS reporter was co-transfected with the effector construct expressing the ESR1 gene into Arabidopsis protoplasts (Fig. 2e). As a result, co-expression of these constructs significantly increased GUS activity, compared with the vector control (Fig. 2f), indicating that ESR1 activates type-B ARR expression by directly binding to their loci. Although ESR1 is known to be regulated by the WOUND INDUCED DEDIFFERENTIATION 1 (WIND1) transcription factor, which is responsible for wound-induced cellular reprogramming (Iwase et al., 2017), it seems likely that SIM-induced ESR1 expression would not be fully regulated by WIND1, because expression patterns of WIND1 and ESR1 varied considerably during de novo shoot regeneration (Fig. S6). While WIND1 expression persisted during incubation on SIM, ESR1 was transiently upregulated (Fig. S6). Given that transient activation of ESR1 occurred in response to cytokinin (Fig. 1a,b), we asked which cytokinin signaling factor(s) promotes ESR1 expression upon incubation on SIM. Notably, previous studies showed that type-B ARR genes were co-expressed with ESR1 (Meng et al., 2017; Fig. 1b), which are highly expressed upon incubation on SIM, especially at 4 DAS (Fig. S7). Furthermore, spatial expression of type-B ARRs was observed in outer cell layers of callus, especially at later stages of shoot regeneration, which overlapped with ESR1 (Fig. 3a). In support, shoot emergence from callus was severely impaired in arr1 arr12 double mutants, as observed in esr1-2 mutant (Fig. 3b,c), raising the possibility that they may be interdependent in the plant regeneration process. We thus analyzed ESR1 expression in arr1 arr12 mutant calli during shoot regeneration and found that while ESR1 expression was gradually increased during shoot regeneration in WT calli, its expression was almost completely diminished in arr1 arr12 mutant calli (Fig. 3d). Next, the cis-element analysis of the ESR1 promoter revealed the presence of binding motifs of type-B ARR proteins (Fig. 3e; Zubo et al., 2017). To verify whether type-B ARRs directly bind to the ESR1 locus, we conducted ChIP-qPCR analysis using 35S:ARR1-MYC transgenic plants. Direct binding analysis showed that ARR1 directly binds to the ESR1 promoter (Fig. 3f). Similarly, the ARR12 was also shown to associate with the transcription start site of ESR1 (Fig. S8). Furthermore, transient expression assays using Arabidopsis protoplasts also supported that expression of type-B ARRs increased the promoter activity of ESR1 (Fig. 3g,h). Altogether, these results indicate that ARRs and ESR1 regulate each other, establishing a positive feedback loop of cytokinin signaling to promote de novo shoot regeneration. The ESR1 and type-B ARR genes are likely to form a positive feedback loop of cytokinin responses. To verify genetic relationship between ESR1 and type-B ARRs in de novo shoot regeneration, we crossed 35S:ESR1 and 35S:ARR12 transgenic plants and investigated the shoot regeneration capacity. Consistent with the fact that one can activate the other via positive feedback loop (Fig. S9), 35S:ESR1 × 35S:ARR12 calli showed a shoot regeneration capacity comparable to their respective transgenic calli (Fig. S10), supporting that ESR1 and type-B ARRs function in same genetic pathway to promote de novo shoot regeneration. In addition, we also found that enhanced shoot regeneration of 35S:ARR12 transgenic calli was impaired in esr1-2 × 35S:ARR12 calli (Fig. 4a,b), which was also supported by WUS expression (Fig. 4c). Similarly, increased shoot regeneration of 35S:ESR1 was compromised in the arr12 mutant background (Fig. 4d–f), indicating that both ESR1 and type-B ARRs are required for robust activation of de novo shoot regeneration. Consistent with the close linkage between ESR1 and type-B ARRs in the control of de novo shoot regeneration, we found that the ESR1 and type-B ARR proteins interact with each other (Fig. S11). Furthermore, in addition to their mutual transcriptional activation, they also directly bind to the WUS locus (Figs S12, S13) (see the Discussion section). Taken together, our study demonstrates that ESR1 and type-B ARRs form a positive feedback loop to maximize the cytokinin responses and thereby accelerate the progression of de novo shoot organogenesis during plant regeneration (Fig. 4g). Cytokinin is perceived by three sensor histidine kinases, including ARABIDOPSIS HISTIDINE KINASE2 (AHK2), AHK3, and AHK4, and the binding of cytokinin to AHKs allows autophosphorylation of their receiver domains (Liu et al., 2019). Subsequently, the phosphoryl group is transferred to HISTIDINE PHOSPHOTRANSFER PROTEINs (AHPs) that translocate into the nucleus to activate type-B ARRs in a phosphorylation-dependent manner (To & Kieber, 2008). This main regulatory axis of cytokinin signaling is involved in diverse aspects of plant growth and development, especially shoot development (Leibfried et al., 2005; Riefler et al., 2006; Gordon et al., 2009). For example, the cytokinin signaling circuit determines SAM growth, and genetic mutations of cytokinin receptors and signaling components lead to reduced SAM size (Higuchi et al., 2004; Kurakawa et al., 2007). Consistently, the type-B ARRs directly bind to and activate the WUS gene, which is a pivotal regulator for SAM maintenance (Meng et al., 2017; Zhang et al., 2017). The cytokinin signaling axis is also critical for de novo shoot regeneration during in vitro plant regeneration. Upon the incubation on SIM, cytokinin signaling is provoked and stimulates shoot stem cell establishment and shoot emergence. In particular, the type-B ARR transcription factors directly activate WUS expression in callus to establish the shoot progenitor (Meng et al., 2017). In support, the arr1 arr12 mutant exhibits defects in shoot regeneration from callus, which can be rescued by ectopic expression of WUS (Meng et al., 2017). The cytokinin-responsive ESR1 protein is also known as a core regulator of de novo shoot regeneration, because the loss-of-function mutants exhibit completely impaired shoot regeneration (Iwase et al., 2017; Fig. 1d,e), although its position on the cytokinin signaling network has been elusive. Here, we demonstrate that ESR1 forms a positive feedback loop with type-B ARRs. ESR1 directly activates type-B ARRs, which also directly bind to the ESR1 locus to promote expression. The ESR1–type-B ARR module reinforces cytokinin responses, and thus, the progression of de novo shoot regeneration with enhanced WUS expression. In support, both esr1-2 and arr1 arr12 mutants exhibited complete defects in shoot regeneration (Meng et al., 2017; Ogura et al., 2023). Consistent with the previous studies showing that a positive feedback loop strengthens stimuli-responsive signal transduction (Pandey et al., 2018; Ohashi-Ito et al., 2019), the positive feedback loop constituted by ESR1 and type-B ARRs likely maximizes cytokinin responses, as evidenced by their interdependence in de novo shoot regeneration. Furthermore, the mutual activation of ESR1 and type-B ARRs is also crucial for full activation of WUS expression. They form a protein complex and directly bind to the WUS locus, synergistically promoting expression. ESR1 is particularly important for specifying the WUS promoter region, where both ESR1 and type-B ARRs bind. Binding of type-B ARRs to WUS was diminished in the esr1-2 mutant (Fig. S13), whereas type-B arr mutants had a relatively lower impact in ESR1 binding to WUS (Fig. S13). This may account for the complete loss of regeneration capacity in esr1-2 mutant overexpressing type-B ARR gene (Fig. 4a,b). In addition, we suspected that the role of ESR1 in shoot stem cell specification is likely dependent on cell types. The ESR1–type-B ARR module is expressed mainly in the outer cell layer in callus, where shoot meristem niches are formed, at a later stage of SIM incubation. Considering that the constitutive expression of ESR1 throughout callus sometimes represses de novo shoot regeneration (Temman et al., 2023), the cell type-specific function of ESR1 may explain the controversial effect of ESR1 on shoot regeneration (Banno et al., 2001; Iwase et al., 2017; Temman et al., 2023). The role of ESR1 in cytokinin signaling also appears to be related to wound-induced callus formation. Wound responses depend on cytokinin signaling to trigger cell cycle reentry (Dewitte et al., 2007; Ikeuchi et al., 2017). Consistently, wound-induced callus formation is impaired in cytokinin signaling mutants, such as the arr1 arr12 double mutant (Ikeuchi et al., 2017). Indeed, the wound-responsive WIND1 transcription factor, a key player of wound-induced cellular reprogramming, acts through cytokinin signaling, as WIND1-induced callus formation is compromised in arr1 arr12 mutants (Iwase et al., 2011). Given that WIND1 directly activates the ESR1 gene (Iwase et al., 2017), whose product binds to type-B ARR loci to promote expression, WIND1 regulation of cytokinin signaling may depend on ESR1 during wound-induced cellular reprograming. In agreement with the integrative role of ESR1 in wounding and cytokinin signaling, esr1 mutants have severe defects in wound-induced callus formation (Iwase et al., 2017). Furthermore, considering that ESR1 is also involved in protoplast regeneration (Xu et al., 2021), the ESR1–ARR module is anticipated to be important for a wide range of cytokinin-dependent plant regeneration. Based on the importance of ESR1 in plant regeneration as well as its conservation across a wide range of plant species (Xu et al., 2021; Larriba et al., 2022), our knowledge can be widely applied to improve the efficiency of various in vitro tissue culture applications, because a low efficiency of de novo shoot organogenesis is a key hurdle in protoplast regeneration, in vitro organogenesis, and crop transformation. We thank Dr John Chandler (Cologne Biocenter, Germany) for providing drn-1/esr1-2 seeds and Dr Akira Iwase (Center for Sustainable Resource Science, Japan) for providing ProESR1:ESR1-GFP seeds. This work was supported by the Basic Science Research (NRF-2022R1I1A1A01071792 to KL and NRF-2022R1A2B5B02001266 to PJS) and Basic Research Laboratory (NRF-2022R1A4A3024451) programs of the National Research Foundation of Korea, and by the New Breeding Technologies Development Program (RS-2024-00322275) of the Rural Development Administration. None declared. PJS conceived and designed the study. KL, HY and O-SP conducted the experiments. PJS and KL wrote the manuscript. All authors read and approved the manuscript. KL and HY contributed equally to this work. Research data are contained within the article or Supporting Information (Figs S1–S13; Tables S1, S2). Fig. S1 Callus formation in hypocotyl and root explants of esr1-2. Fig. S2 Callus formation and de novo shoot regeneration in hypocotyl explants of 35S:MYC-ESR1. Fig. S3 Type-B ARR expression in root explant-derived calli of esr1-2. Fig. S4 WOX5 expression in esr1-2 calli. Fig. S5 Protein accumulation of ESR1 in 35S:MYC-ESR1 transgenic plants. Fig. S6 Expression of ESR1 and WIND1 during in vitro tissue culture. Fig. S7 Expression of ARR1 and ARR12 during in vitro tissue culture. Fig. S8 Binding of ARR12 to the ESR1 locus. Fig. S9 Positive feedback regulation between ESR1 and ARR12. Fig. S10 De novo shoot regeneration in hypocotyl explants of 35S:ESR1 × 35S:ARR12. Fig. S11 Interactions between ESR1 and ARR12. Fig. S12 Binding of ESR1 to the WUS locus. Fig. S13 Requirement of ESR1 to ARR12 binding to WUS locus. Table S1 Primers used in RT-qPCR analysis. Table S2 Primers used in chromatin immunoprecipitation assays. Please note: Wiley is not responsible for the content or functionality of any Supporting Information supplied by the authors. Any queries (other than missing material) should be directed to the New Phytologist Central Office. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article. The New Phytologist Foundation remains neutral with regard to jurisdictional claims in maps and in any institutional affiliations.