A Loss of DPA4 Mutant Reveals That the Boundary Gene CUC2 Promotes Adventitious Root Formation in Arabidopsis Leaf Explants via Repression of ULT1 and Increased Auxin Biosynthesis

1. Introduction

Plant cells exhibit remarkable developmental plasticity enabling regeneration of new organs and even an entire plant body after suffering wounds [1,2]. The potential capacity of plants for de novo organogenesis finds wide ranging applications in agriculture, biotechnology, and biological studies including tissue culture and vegetative propagation via cuttings and explants [3,4]. Among various forms of plant regeneration, de novo root regeneration (DNRR) from wounded and/or detached plant tissues and organs are commonly utilized in biotechnological breeding and cultivation research [5,6]. In contrast to de novo shoot regeneration that requires a proper ratio of exogenous auxin and cytokinin [7,8], DNRR from Arabidopsis leaf explants can take place on B5 medium without exogenous phytohormones [9,10].

Upon injury, Arabidopsis leaf explants sense a multitude of signals, including wound-derived signals, other stress-related and environmental stimuli, factors triggering endogenous developmental programs [11,12,13,14]. These signals direct the production and transport of auxin towards the future regeneration-competent cells near the wound site, which facilitates adventitious root (AR) formation [15,16]. Once the leaf is detached, the wounding signal and phytohormone jasmonate (JA) promptly triggers auxin biosynthesis [12,17]. Thereby, JA signaling activates the expression of ETHYLENE RESPONSE FACTOR 109 (ERF109) and ABSCISIC ACID REPRESSOR 1 (ABR1/ERF111) [11,12]. The transcriptional activators ERF109 and ABR1 serve as molecular connection between wounding, JA signaling, and auxin biosynthesis by direct activation of ANTHRANILATE SYNTHASE ALPHA SUBUNIT 1 (ASA1) [12,18]. ASA1 catalyzes the rate-limiting step in tryptophan (Trp) biosynthesis, which provides the substrate for auxin biosynthesis. Via the conversion of Trp to indole-3-pyruvate (IPA) by the TAA amino transferases, YUCCA (YUC) flavin-containing monooxygenases convert IPA to auxin indole-3-acetic acid (IAA), one of the major natural auxins [19,20,21]. The newly synthesized auxin is polar-transported to the future rooting site adjacent to the wounding site [22]. The emerging auxin maximum induces the expression of WUSCHEL-RELATED HOMEOBOX 11 (WOX11)/WOX12, which facilitates the cell fate transition from regeneration-competent vasculature-associated pluripotent cells (VPCs) to root founder cells [23,24,25]. Both WOX11/12 and auxin are required to promote the expression of LATERAL ORGAN BOUNDARIES DOMAIN 16 (LBD16) and WOX5, which facilitates the transformation of the root founder cells into root primordia [24,25,26,27]. Disrupting either auxin biosynthesis or transport prevents the establishment of the auxin maxima in the VPCs and, in turn, AR formation [28].

In Arabidopsis, CUP-SHAPED COTYLEDON2 (CUC2) encodes a NAC domain transcription factor that is a key regulator that establishes organ boundaries and participates in multiple developmental pathways [29]. In leaves, CUC2 is specifically targeted by MIR164A, which triggers the cleavage of the CUC2 mRNA, while cuc2-1D that carries a point mutation in the MIR164A-targeting site and mir164a-4 mutants display increased CUC2 expression [30,31]. During leaf margin patterning, CUC2 establishes auxin activity maxima by directing PIN1 localization at the leaf margin, thereby shaping serrated leaf morphology [32,33]. Furthermore, CUC2 restricts leaf expansion primarily through modulating cell proliferation rather than influencing cell expansion [34].

Notably, CUC2 expression is indispensable for de novo shoot organogenesis from callus tissue [35,36,37] and serves as a molecular indicator of regenerative potential in root-derived explants [38]. Collectively, these findings establish CUC2 as a key regulator of auxin distribution patterns. Since auxin is essential for of AR formation [28], it is a reasonable question whether CUC2 has a function in DNRR from leaf explants.

The B3 transcription factor DEVELOPMENT-RELATED POLYCOMB TARGET IN THE APEX4 (DPA4/NGAL3), belonging to the RAV (Related ABI3/VP1) family, has been implicated in multiple developmental processes including leaf margin development [39,40], de novo stem cell formation in axillary meristems [32], and seed size regulation [41,42]. DPA4 is expressed in the shoot apex during primordia formation, while also being expressed in the leaf sinuses coinciding with CUC2 expression domain [30,39]. Genetic study demonstrate that DPA4 suppresses CUC2 expression to regulate leaf morphogenesis [39,40]. Notably, DPA4 expression is early induced by wounding in leaf explants during DNRR (Ye et al. 2020). However, the regulatory role of DPA4 in DNRR remains to be elucidated.

In this study, we reveal that DPA4, encoding a well-known repressor of CUC2 expression, inhibits DNRR from Arabidopsis leaf explants. Beside the loss-of-function mutant dpa4-2, we employed two other CUC2 overexpressing lines, cuc2-1D and mir164a-4, to test whether increased CUC2 function promotes DNRR in general. All three mutants displayed enhanced AR formation suggesting that CUC2 promotes DNRR downstream of DPA4. Furthermore, we found similar expression changes in dpa4-2 and cuc2-1D mutant leaf explants, which include increased mRNA levels of CUC2, ERF109, ASA1, and YUC4/9, while ULT1 was lower expressed compared to the wild-type. The enhanced root regeneration was associated with increased auxin concentrations in both cuc2-1D and dpa4-2 mutant leaf explants, as indicated by intensified DR5::GUS reporter activity especially in and around the rooting side. Genetic analysis of dpa4-2 cuc2-1D double mutants and pharmacological inhibition of YUC-mediated auxin biogenesis supported the hypothesis that increased auxin production and enhanced DNRR are causally linked. These results indicating a mechanistic link between transcriptional regulation of CUC2 by DPA4 and auxin-dependent developmental reprogramming during DNRR. Our work provides new insights into the molecular control and gene regulatory network of AR formation and offers potential biotechnological applications for improving vegetative propagation of crop plants.

2. Results

2.1. DPA4 Suppresses De Novo Root Regeneration (DNRR) from Leaf Explants

A previous study showed that DPA4 belongs to a group of transcription factors genes that are early induced by wounding during DNRR [11]. However, it remained unclear whether DPA4 has positive or negative effects to wound-induced DNRR. To investigate the potential role of DPA4 in AR formation, we conducted DNRR assays following established protocols [9], in which detached leaf explants were cultured on phytohormone-free B5 medium under dark conditions. We quantified the rooting rate and root regeneration capacity in leaf explants of the loss-of-function mutant dpa4-2 and wild-type (Col-0) plants (Figure 1A-D). The time-course analysis revealed that dpa4-2 leaf explants exhibited significantly accelerated root formation rates compared to the wild-type control (Figure 1A-D). Furthermore, the rooting capacity were significantly increased in dpa4-2 mutant leaf explants in comparison with the wild-type (Figure 1D). These findings indicate that DPA4 plays a role in suppressing DNRR in Arabidopsis.

2.2. DPA4 Regulates Genes Known to be Involved in AR Formation and Auxin Biosynthesis

Since DPA4 encode a transcription factor, we investigated the expression changes of putative downstream targets of DPA4 after wounding to better understand its function in suppressing DNRR. For the expression analysis of the candidate genes, we harvested leaf explants at 4 hours after culturing (HAC) on B5 medium. Since DPA4 is a known repressor of the boundary gene CUC2 in leaf margin development [39], we examined CUC2 expression levels in leaf explants. The reverse transcription quantitative PCR (RT-qPCR) analysis revealed significantly increased CUC2 expression in the dpa4-2 mutant compared to the wild-type (Figure 2A), confirming that DPA4 functions upstream of CUC2 in leaf explants (4 HAC). In a previous study, we found that ULT1 is a negative regulator of DNRR likely by repressing ERF109 and, in turn, the auxin biosynthesis gene ASA1 [43]. Since loss of ULT1 phenocopies the acceleration of AR formation in dpa4-2 mutants, we analyzed the expression of ULT1, ERF109, and ASA1 (Figure 2). Our quantitative analysis revealed significantly reduced levels of ULT1 mRNA, while the expression of ERF109 and ASA1 was increased in dpa4-2 leaf explants compared to the wild-type. Like in ult1 mutants, the expression of ABR1, another wounding-induced activator of ASA1, was not significantly changed in dpa4-2 (Figure 2). These results suggest that DPA4 constrains AR formation at least partially by activation of ULT1 that, in turn, represses ERF109, and thereby limits auxin biosynthesis through the direct repression of ASA1 by ERF109. The expression levels of YUC genes have been reported as the rate-limiting step in auxin biosynthesis and DNRR [28,44,45]. Therefore, we examined the expression levels YUC genes in dpa4-2 leaf explants, 4 HAC (Figure 2). Notably, YUC4, YUC6, and YUC9 showed significant upregulation in dpa4-2 compared to the wild-type, while YUC1 expression was not significantly changed. Collectively, these results indicate that the enhanced rooting rate and rooting capacity observed in dpa4-2 mutant leaf explants is likely caused by an increased auxin biosynthesis.

2.3. cuc2-1D Phenocopies the Increased AR Formation Phenotype of dpa4-2 Mutant Leaf Explants, While Both Mutants Display Similar Expression Changes in DNRR-Related Genes

Since CUC2 is a known target of DPA4 in leaf margin development [39] and plays an essential role in de novo shoot regeneration [35,36,37], we hypothesized that the increased CUC2 expression might promote AR formation in dpa4-2 mutants. To test whether increased CUC2 expression levels can promote DNRR, we employed cuc2-1D mutants that carry a mutation in the CUC2 miRNA target site resulting in CUC2 overexpression [31]. We performed DNRR assays to check the rooting rate and capacity in cuc2-1D mutant leaf explants (Figure 3A, E, G). Furthermore, we tested the rooting rate and capacity in mir164a-4 mutant leaf explants (Supplementary Figure S1). Consistent with our hypothesis, cuc2-1D and mir164a-4 mutant leaf explants exhibited similar increased rooting rates and rooting capacity as dpa4-2 mutants, which were significantly higher than the wild-type, suggesting that increased CUC2 expression is sufficient to promote AR formation.

Since cuc2-1D phenocopies the increased AR formation phenotype of dpa4-2 mutants, we analyzed the expression of genes that was misregulated in dpa4-2 in cuc2-1D mutant leaf explants during DNRR (Figure 2). Like in dpa4-2 mutants, ULT1 was downregulated, while CUC2, ERF109, ASA1, YUC4, YUC6, and YUC9 were upregulated in cuc2-1D mutants in comparison to the wild-type indicating that increased auxin levels could cause the increased AR formation phenotype in both mutants. Notably, also in miR164a-4 mutant leaf explants, CUC2 was upregulated and ULT1 were downregulated (Supplementary Figure S2). Although ASA1 were upregulated in all three CUC2 overexpressing mutants, partially different YUC genes were upregulated in dpa4-2, cuc2-1D, and mir164a-4 (Figure 2 and Supplementary Figure S2) indicating that increased CUC2 is not the only factor that influence YUC expression in these three lines.

To support the hypothesis that increased auxin levels cause the enhanced DNRR in cuc2-1D and dpa4-2 mutants, we conducted DNRR assays that either increased the auxin levels by exogenous application of IAA or decreased endogenous IAA levels in the leaf explant by applying the auxin biosynthesis inhibitor yucasin (Figure 3). Treatment with 0.1 μM IAA accelerated ARs formation in all three genotypes without equalizing the rooting rates [43], and cuc2-1D and dpa4-2 leaf explants displayed earlier rooting initiation compared to the wild-type, although the differences in the time-course was reduced relative to mock (Figure 3E-F). Furthermore, the rooting capacity was equalized in wild-type, cuc2-1D, and dpa4-2 leaf explants (Figure 3G) corroborating the idea that accelerated and enhanced AR formation is mainly caused by increased auxin levels in both mutants.

Next we tested the influence of increased and decreased gibberellin (GA) levels on DNRR in dpa4-2 and cuc2-1D mutant leaf explants. The phytohormone GA is well-known for its negative effects on AR formation likely through its impact on auxin transport [27,46]. Although the changes in rooting capacity and the overall effects of the GA synthesis inhibitor PBZ were less distinct, exogenous GA significantly decreased the rooting rates in Col-0, cuc2-1D, and dpa4-2 leaf explants, while the differences in the rooting rate between Col-0 and the both mutants with increased CUC2 levels remained (Supplementary Figure S2) indicating that GA signaling is unaffected during DNRR in cuc2-1D, and dpa4-2 mutants.

Since the expression of several YUC genes is upregulated in dpa4-2 and cuc2-1D mutant leaf explants, we investigated whether YUC-mediated auxin biosynthesis is required for the enhanced AR formation in both mutants. To test this, we used yucasin that is a specific inhibitor of the YUC enzymes reducing endogenous auxin levels during DNRR on B5 medium [28,47]. At a low concentration of 30 μM yucasin, DNRR in wild-type, cuc2-1D, and dpa4-2 leaf explants was partially reduced resulting in decelerated rooting rates over the time-course and lower rooting capacity (Figure 3C, H, J). As expected, combined treatment with 30 μM yucasin and 0.1 μM IAA partially rescued the rooting rate and rooting capacity in all three lines (Figure 3D, I-J). Notably, the combined treatment with yucasin and IAA equalized rooting rate and rooting capacity in wild-type and dpa4-2 leaf explants strongly suggest that the enhanced rooting activity in dpa4-2 leaf explants is primarily caused by increased YUC-dependent auxin biosynthesis.

In contrast, the rooting rate and rooting capacity of cuc2-1D mutant leaf explants remained significantly higher than the wild-type during low (30 μM) yucasin treatments with and without IAA (Figure 3C-D, I-J). This could be either caused by higher auxin levels in cuc2-1D than dpa4-2 leaf explants or an auxin-independent factor that promotes DNRR is increased cuc2-1D but not in dpa4-2. To rule out the latter one, we repeated the DNRR experiment with a higher concentration of 100 μM yucasin with and without 0.1 μM IAA (Figure 3K-M). The single treatment with 100 μM yucasin annulled the rooting activity in all three lines confirming that YUC-dependent auxin biosynthesis is in general essential for AR formation (Figure 3K,M). The addition of IAA partially rescued the disrupted rooting activity caused by 100 µM yucasin, the low rooting rates and rooting capacities were not significant different among all three genotypes (Figure 3L,M). In summary, these findings indicate that the enhanced DNRR ability in cuc2-1D and dpa4-2 mutants results from elevated auxin production. This is in line with the hypothesis that DPA4 regulates DNRR through a CUC2-dependent pathway that controls YUC-mediated auxin biosynthesis.

2.4. DPA4 and CUC2 Promote Endogenous Auxin Levels

To confirm that the enhanced DNRR activity in dpa4-2 and cuc2-1D results from high auxin production in leaf explants during regeneration, we performed a GUS reporter gene assay using DR5::GUS reporter to show the spatiotemporal pattern of auxin in the leaf explants. We analyzed DR5::GUS staining in wild-type, dpa4-2, and cuc2-1D leaf explants at two time points (0 DAC and 1 DAC) (Figure 4). The DR5::GUS reporter activity reflects free auxin levels [48]. As expected, the DR5::GUS expression was low and absence in the potential rooting side in all three genotypes at 0 DAC (Figure 4A-C) [24,28]. After leaf explants were placed on B5 medium (mock) for one day (1 DAC), DR5::GUS staining intensively increased above the wounding sites and was significantly stronger in cuc2-1D and dpa4-2 leaf explants compared to the wild-type, while the staining was stronger in cuc2-1D than dpa4-2 (Figure 4D-F). At 1 DAC, the differences in DR5::GUS expression between dpa4-2 and cuc2-1D remained evident during all treatments with 0.1 μM IAA (Figure 4G-I), 100 μM yucasin (Figure 4J-L), and 100 μM yucasin and 0.1 μM IAA (Figure 4M-N), though the staining was always much weaker in the wild-type. As expected, the IAA treatments increased, while yucasin decreased the DR5::GUS staining intensity, and yucasin + IAA double treatment rescued the DR5::GUS staining near mock levels. Taken together, the differences in the intensity of the DR5::GUS staining (Figure 4) correlate largely with the previous detected differences in rooting rate and rooting capacity caused by the different treatments and genotypes (Figure 3). This strongly support the hypothesis that increased auxin levels at the rooting side, likely caused by increased auxin biosynthesis, result in the enhanced DNRR activity in cuc2-1D and dpa4-2 leaf explants.

2.5. DPA4 and CUC2 Regulate AR Formation Likely Through a Common Genetic Pathway that Controls Auxin Biosynthesis

Given that CUC2 is overexpressed in both cuc2-1D and dpa4-2 (Figure 2), we investigated whether they function in the same genetic pathway by generating cuc2-1D dpa4-2 double mutants. We subsequently examined the rooting rate and capacity in leaf explants from wild-type, cuc2-1D, dpa4-2, and cuc2-1D dpa4-2 double mutants (Figure 5A-C). We found that the dpa4-2 and cuc2-1D single mutants and the cuc2-1D dpa4-2 double mutants exhibited a similar increase in rooting rate and capacity compared to the wild-type; no additive or synergistic effect could be detected. Furthermore, we examined the expression of genes related to DNRR in the cuc2-1D dpa4-2 double mutant leaf explants using RT-qPCR (Figure 5D). We found that ULT1 was reduced to the same level in cuc2-1D, dpa4-2 and cuc2-1D dpa4-2 leaf explants. In contrast, the expression of ERF109, YUC4, and YUC9 was significantly upregulated in cuc2-1D, dpa4-2, and cuc2-1D dpa4-2 compared to the wild-type. Moreover, the expression levels of ULT1, ERF109, YUC4, and YUC9 in the cuc2-1D dpa4-2 double mutants remained similar to those in cuc2-1D and dpa4-2 single mutants, consistent with the observation that the cuc2-1D dpa4-2 double mutants did not show enhanced DNRR compared to the cuc2-1D and dpa4-2 single mutants. Interestingly, ASA1 expression, which was increased in the cuc2-1D and dpa4-2 mutants single in a previous experiment (Figure 2), was only significantly increased in the cuc2-1D dpa4-2 leaf explants (Figure 5D). This might reflect that ASA1 expression levels, which are only slightly increased in ult1-3, cuc2-1D, dpa4-2, and cuc2-1D dpa4-2 double mutants (Figure 2 and Figure 5D) [43], is not solitarily dependent of the expression levels of ERF109, but on variable environmental factors. In conclusion, the absence of additive or synergistic effects in cuc2-1D dpa4-2 double mutants on the rooting phenotype and expression patterns strongly support that DPA4 and CUC2 function in the same genetic pathway that controls the expression of ULT1 and auxin biosynthesis genes regulating DNRR in leaf explants.

3. Discussion

DNRR is a developmental process that enables plants to regenerate new roots post-injury, after detachment, and under other stress conditions, representing a crucial survival strategy of higher plants [9,15]. Mechanical wounding primarily drives this process of AR formation by creating high auxin concentrations near the wounding site [28]. Despite its importance, the molecular pathways coordinating DNRR, which include hormone responses and the control of intrinsic developmental programs by transcription and epigenetic factors [15,16], are poorly characterized.

A previous study showed that DPA4 belongs to a group of transcription factors genes that are early induced by wounding during DNRR, but it remained unclear whether DPA4, which belongs to the B3 transcription factor gene subfamily, has positive or negative effects to wound-induced DNRR [11]. In this study, we investigated the role of DPA4 in AR formation in a series of DNRR assays and found that DPA4 is a suppressor of root regeneration from leaf explants via reduced auxin content. Thereby, DPA4 controls likely auxin biosynthesis through the indirect activation of ULT1, a previously known suppressor of DNRR [43], and the repression of CUC2, a here newly identified promoter of DNRR (Figure 6).

During screening of candidate genes in DNRR assays, we observed significantly enhanced AR formation in dpa4-2 mutant leaf explants compared to wild-type (Figure 1). This findings suggests that DPA4 functions as a key regulator of root regeneration, supporting the emerging view that plant B3 transcription factors play important roles in regulating DNRR [11,17,49]. Our RT-qPCR analysis confirmed increased CUC2 expression in dpa4-2 leaf explants, consistent with previous reports showing that CUC2 is a downstream target of DPA4 in various developmental processes [32,39,40]. Furthermore, we detected enhanced expression of JA-induced ERF109, a member of the AP2/ERF transcription factor subfamily that mediates crosstalk between JA signaling and auxin biosynthesis [12,18]. These observations suggest that DPA4 may regulate DNRR through modulation of auxin biosynthesis.

Furthermore, we found that several YUC genes were significantly upregulated in dpa4-2, cuc2-1D, and miR164a-4 mutants that all displayed increased CUC2 expression (Figure 2 and Supplementary Figure S1). This result strongly suggests that increased CUC2 levels are sufficient to promote DNRR by activating the expression of various YUC genes, corroborating previous findings that YUC1 and YUC4 expression depends on CUC2 during embryogenesis [50]. We found ULT1, which encodes a known negative regulator of DNRR [43], equally downregulated in all three CUC2 overexpression mutants, dpa4-2, cuc2-1D, and miR164a-4 (Figure 2 and Supplementary Figure S1) suggesting that CUC2 is a direct or indirect repressor of ULT1 expression (Figure 6). Since ULT1 is a general repressor of ERF109 [51] that encodes an activator of ASA1 during DNRR [12,43], the low ULT1 expression levels may contribute to the high auxin levels via ASA1-depenent auxin biosynthesis (Figure 6). However, since the increase of ASA1 expression is rather weak and variable, ULT1 may suppresses DNRR also through ASA1-independent pathways. Notably, the enhanced root regeneration phenotype of cuc2-1D dpa4-2 double mutants was very similar to the single mutants and no additive or synergistic effect could be detected (Figure 5) indicating that DPA4 and CUC2 function mainly in the same DNRR pathway (Figure 6).

Auxin levels play a pivotal role in DNRR. Previous studies have demonstrated that both exogenous IAA application and inhibition of auxin production by yucasin significantly affect the root regeneration process [28]. Our findings confirmed that increased auxin biosynthesis essential for the enhanced DNRR in dpa4-2 and cuc2-1D mutants. This conclusion is further supported by our analysis of the DR5::GUS expression patterns in leaf explants in DNRR assays with and without IAA and/or yucasin treatment (Figure 4). The intensity of the DR5::GUS staining correlates largely with the differences in rooting rate and rooting capacity caused by the different treatments and genotypes (Figure 3). Nevertheless, we cannot exclude the possibility that CUC2 and DPA4 regulates root regeneration also through other pathways [12,52,53].

In summary, our findings suggest that DPA4 plays a crucial role in DNRR in Arabidopsis leaf explants by serving as a key player in inhibiting ASA1- and YUC-mediated auxin biosynthesis. Furthermore, DPA4 and CUC2 regulate DNRR through a common auxin-dependent pathway. In our Conceptional model (Figure 6), DPA4 suppresses DNRR by downregulating CUC2, which, in turn, upregulates ULT1 that represses ERF109 and, ultimately, ASA1. Likely independently of ULT1, CUC2 represses YUC-mediated auxin biosynthesis. Overproliferation of ARs would decrease the survival rate of detached aerial organs [43]. Therefore, the DPA4-CUC2-ULT1 module likely evolved to limit DNRR increasing the persistence of detached Arabidopsis leaves.

4. Materials and Methods

4.1. DPA4 and CUC2 Regulate AR Formation Likely Through a Common Genetic Pathway that Controls Auxin Biosynthesis

All plant materials were in the Col-0 background. Seeds of the T-DNA insertion mutant dpa4-2 [39] (N650707) and the miRNA-resistant cuc2-1D allele [31] (N16485) were obtained from the Nottingham Arabidopsis Stock Center. DR5::GUS were kindly provided by Rüdiger Simon. The seeds were surfaced-sterilized by three sequential treatments with 70% ethanol (3 min each), then the seeds were sown on half-strength Murashige and Skoog (1/2 MS) plates containing 1% sucrose, and 1% of agar with pH adjusted to 5.8. After sowing, the plates were first kept at 4 °C for 3 days in darkness, then transferred to growth chambers maintained at 22 °C under long-day photoperiod conditions (16 h light/8 h dark cycle).

4.2. De Novo Root Regeneration (DNRR) Assays with and Without Hormone Treatment

The first pair of primary leaves was cut from 12-day-old seedlings of Arabidopsis plants for the DNRR assay. Leaf explants were cultured following previously described methods [9] on B5 medium (Gamborg B5 medium containing 3% sucrose, 1% agar, and 0.5 g/L MES, with pH adjusted to 5.7) under dark conditions. The rooting rate was calculated as the percentage of leaf explants developing roots at a given time point [9], while regeneration capacity was assessed based on the percentage of leaf explants showing varying numbers of regenerated roots [11]. For IAA and yucasin treatments, we used 0.1 µM IAA (Sigma-Aldrich) and 30 µM or 100 µM yucasin (MACKLIN) in B5 medium (applied continuously after detachment). Each experiment included more than five plates per treatment, with at least eight explants per genotype per plate. A one-tailed Student’s t test was employed to assess the statistical significance between value pairs of different genotypes, resulting in p values. In this study, all error bars represent the mean ± standard error.

4.3. Quantitative RT-PCR and GUS Staining

The leaf explants after 4 HAC (hours after culture) on B5 medium were collected and snap frozen in liquid nitrogen. Total RNA was isolated using the TRIZOL regent (Invitrogen), and cDNA was synthesized with a synthesis kit (Thermo Scientific). RT-qPCR was performed using the SYBR green supermix (Vazyme) on a Biorad CFX96 system (Biorad). The RT-qPCR procedure and the primers for CUC2, ABR1, ERF109, ASA1, ULT1, YUC1, YUC4, YUC6, YUC9, and eIF4A (Supplementary Table S1) have been previously described [28,34,43,54]. All expression data represent the average of four biological replicates.

The detection of β-glucuronidase (GUS) activity in leaf explants was performed according to a published protocol with minor modifications [34,55]. In brief, leaf explants were harvested at specific time points after detachment (time 0 DAC and 1 DAC (day after culturing on B5 medium) and immersed in the GUS assay solution (0.5 mM ferrocyanide, 0.5 mM ferricyanide, 50 mM NaHPO4, 50mM Na2HPO4, and 1% Triton X-100) containing 1 mM X-Gluc (Solarbio). Then, the leaf explants in the GUS staining solution were vacuum-infiltrated for 30 minutes and subsequently incubated overnight at 37°C. To remove the chlorophyll, the stained leaves were passed through an ethanol series and then photographed with a stereomicroscope (Olympus). Finally, the digital photographs were collated with Adobe Photoshop.