Identification of a cytokinin-responsive core promoter and interacting transcription factors of MtCKX1 in Medicago truncatula

Abstract

Background

Cytokinin oxidases/dehydrogenases (CKXs) are key enzymes regulating cytokinin homeostasis and plant development, yet their transcriptional regulation remains poorly characterized, especially in legumes.

Results

Here, we combined promoter truncation, transgenic analysis, and DNA-protein interaction assays to dissect the regulatory mechanism of MtCKX1 in Medicago truncatula. A 1789 bp promoter region was identified, harboring multiple cis-elements responsive to hormones (CK, IAA, ABA, JA, GA) and stresses. Truncation analysis and GUS assay revealed a core cytokinin-responsive region (− 1789 to − 1125 bp), with one construct (CKXp2) functioning as a strong, specific cytokinin-inducible promoter throughout plant development. Using DNA pull-down mass spectrometry, yeast one-hybrid assays, and dual-luciferase reporter, we further demonstrated that C2H2 and GRF transcription factors directly bind to the MtCKX1 promoter.

Conclusions

These findings provide novel insights into the transcriptional regulation of CKX genes and highlight the potential of engineered promoters for precise spatiotemporal control of gene expression in biotechnology applications.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12870-026-08265-8.

Introduction

The Cytokinin oxidase/dehydrogenase (CKX) gene family encodes key enzymes responsible for degrading endogenous cytokinins (CKs), playing a pivotal role in maintaining cellular CK homeostasis and mediating plant adaptation to environmental stresses [1]. In Gossypium hirsutum, GhCKX14 enhances drought tolerance by modulating antioxidant enzyme activities [2]. Conversely, in rice (Oryza sativa), OsCKXs interact with OsPHP1 to negatively regulate cytokinin levels, thereby increasing sensitivity to salt and drought stresses [3]. Genetic studies of rice CKX mutants revealed that OsCKX4 and OsCKX9 suppress tillering, whereas OsCKX1 and OsCKX2 promote this trait [4]. While numerous studies have characterized the functions of the CKX gene and its protein-protein interactions, the regulatory mechanisms of CKX promoters and their upstream DNA-protein interactions remain poorly understood.

Promoters, defined as genomic sequences upstream of transcription start sites (TSS), harbor multiple cis-acting elements that orchestrate complex transcriptional regulation across cellular, tissue-specific, developmental, and environmental contexts [5]. The functional characterization of promoters provides critical insights into gene regulatory networks and facilitates the annotation of gene function [6]. For instance, in chrysanthemum (Chrysanthemum morifolium), the CmHSFA4 protein can directly bind to the heat shock element of the CmMYB121 promoter in the form of a protein complex, regulating the expression level of CmMYB121 and thereby regulating the salt tolerance [7]. In rice, the interaction between OsSPL10 and OsWOX3B attenuates their binding affinity to the HL6 promoter, modulating trichome length [8]. Similarly, in pear (Pyrus bretschneideri), the PbHB.G7.2 transcription factor activates the ethylene biosynthetic gene PbACS1b by binding to its promoter, increasing ethylene production during fruit ripening [9]. These studies underscore the pivotal role of promoters in plant transcriptional regulation and their utility as molecular entry points for dissecting physiological processes.

However, the presence of numerous cis-elements with undefined spatial distributions within promoter regions poses challenges for precisely mapping core regulatory domains and identifying upstream transcription factors (TFs). Promoter truncation analysis has emerged as a powerful strategy to overcome this limitation. By systematically deleting promoter segments and assessing their impact on gene expression, this approach enables the identification of functional cis-regulatory modules and elucidates molecular mechanisms of transcriptional control [10]. For example, in peach (Prunus persica), truncation assays demonstrated that three PpIDD proteins recognize the TTGTC motif in the PpGA20ox1 promoter and form a complex with PpDELLA1 to activate transcription [11]. Similarly, wheat (Triticum aestivum) TaSnRK2.4 promoter truncation analysis revealed that DRE and ABR elements, along with MYB/MYC TF binding sites, coordinately mediate drought and ABA responses [12]. These cases highlight the utility of promoter truncation in deciphering upstream regulatory mechanisms.

Current methods for identifying promoter-interacting TFs include DNA pull-down coupled with mass spectrometry (DNA Pull down-MS) [13] and yeast one-hybrid (Y1H) assays [14]. Among these, DNA Pull down-MS offers superior specificity, sensitivity, and broad applicability, making it widely adopted in staple crops like rice [15] and wheat [16], though its application in alfalfa (Medicago) remains limited. Herein, we employ promoter truncation to delineate core regulatory regions of the MtCKX promoter in Medicago truncatula, followed by DNA Pull down-MS to identify interacting TFs. And we further confirmed the direct binding of them to the CKX1 promoter using yeast one-hybrid and dual-luciferase assays. This study aims to uncover functionally specialized alfalfa promoter sequences and characterize upstream TFs directly regulating MtCKX. Our findings will contribute to a more comprehensive understanding of plant cytokinin regulatory networks and facilitate the development of stress-resilient crops.

Result

The MtCKX1 gene was sensitive to multiple hormones

According to the data of the M. truncatula genome website, the sequence of the promoter region of the MtCKX1 gene was obtained for subsequent analysis by PCR. The elements prediction of this sequence revealed that several cis-acting elements, including ABRE in response to abscisic acid (ABA), O2-site in response to cytokinin (CK), TGACG-motif and CGTCA-motif in response to methyl jasmonate (MeJA), and P-box in response to gibberellin (GA), MYB and MYC elements in response to stress, was discovered in the CKX1 promoter region (Fig. 1a, Table S2), indicating that numerous hormones and stress have the potential to activate the CKX1 gene.

Fig. 1.

Elements analysis and induced expression analysis of MtCKX1. a cis-acting element prediction of MtCKX1 promoter. The arrow indicates the position where CKXp1/2/3/4 is truncated. b-f The expression levels of MtCKX1 at different time points, treated with IAA, 6-BA, ABA, MeJA, and GA, respectively. The lowercase English letters in these pictures represent the significance of the data. Different letters indicate significant differences between two sets of data

Taking into account the multiple hormone-responsive elements in the promoter region of the MtCKX1 gene, we exogenously sprayed IAA, 6-BA, GA, ABA, and MeJA on wild-type Medicago truncatula and collected samples for fluorescence measurement at various intervals. The results indicated that under IAA induction, the expression level of CKX1 exhibited an initial increase, followed by a decrease and a subsequent rise, reaching its first peak at 1 h (Fig. 1b). Under both 6-BA and MeJA treatments, the expression of CKX1 showed an initial upward trend followed by a decline. Notably, the peak expression occurred at 3 h under 6-BA induction, whereas under MeJA induction, it was observed at 9 h (Fig. 1c and e). In contrast, ABA induction led to mild fluctuations in CKX1 expression within the first 6 h, followed by a sharp increase thereafter (Fig. 1d). Under GA treatment, the expression pattern displayed an initial decrease, followed by an increase and a subsequent decline, with troughs observed at 3 h and 9 h, and a peak at 6 h (Fig. 1f). These findings demonstrate that the MtCKX1 gene is indeed responsive to these hormonal signals.

The 1789 bp to 1125 bp region of the MtCKX1 promoter was a critical regulatory region

According to the distribution of cis-acting elements, the promoter of the MtCKX1 gene was divided into four regions with lengths of 1789 base pairs (bp), 1125 base pairs, 467 base pairs, and 136 base pairs, respectively, to further investigate the regulatory mechanisms of the gene (Fig. 2a). Then, CKXp1-4:GUS expression vectors were constructed (Figure S1 a-b) and introduced into the Arabidopsis plant by using the floral dip approach. The results of transgenic line gDNA identification supported the effectiveness of the transformation (Figure S1 c-d).

Fig. 2.

Screening of MtCKX promoter core regulatory region. a elements analysis of CKXp1-4. b the GUS assay of different tissues from the transgenic lines (CKXp1-4) and wildtype (WT) as control. Representative images are shown, as staining was consistent across all independent lines analyzed for each construct

GUS staining assay showed that, in the mature tissues, the highly active GUS protein was observed in stem ends, leaves, petals, fruit pods, and stalks of the CKXp1 lines, while the other lines did not (Fig. 2c). These results above demonstrated that a key region existing in the MtCKX1 promoter from 1789 bp to 1125 bp activated the subsequent GUS expression in the mature tissues. At the seedling stage, GUS protein was detected in the whole plant of CKXp1 and CKXp2, only in the root and hypocotyl of CKXp4, and the other lines were undetectable (Fig. 2c). This situation was different from what happened in mature tissues. These findings implied that distinct portions of the CKX1 gene promoter exhibited diverse regulation patterns depending on the growth stage, whereas CKXp2 could act as a specific promoter at the seedling stage. In conclusion, it can be speculated that the 1789 bp to 1125 bp region of the CKX1 promoter was essential for maintaining the proper function of this promoter, and the 1125 bp to 467 bp region might contain a plant growth stage sensor that suppressed the expression of GUS in various tissues of the CKXp2-4 lines at the maturation stage. Moreover, there might be a root- and hypocotyl-specific responder in the 136 bp to 0 bp region of the MtCKX1 promoter, which promoted the expression of GUS in the hypocotyls and roots of CKXp4 at the seedling stage.

CKXp1–4 could be activated by 6-BA, GA, and IAA

Considering the role of the MtCKX1 gene in cytokinin biosynthesis and the presence of cytokinin response elements in the promoter of this gene, 10 µmol·L^− 1 6-BA was added to each transgenic line, and the expression of GUS protein was analyzed by GUS staining to further explore the regulatory effect of cytokinin on the promoter. The results revealed that CKXp1 and CKXp2 lines exhibited high activity of GUS protein expression in stems, leaves, flowers, and fruit pods, while GUS protein remained undetected in the mature tissues of two other lines. Compared with those under the no-treatment conditions above (Fig. 3a), these results suggested the presence of the segment regulated by cytokinin in the 1125 bp to 467 bp interval upstream of the MtCKX1 gene promoter and CKXp2 was capable of acting as a specific promoter for cytokinin response in mature tissues. In seedlings, all four lines displayed GUS expression, but qualitatively, the lowest expression level was found in CKXp3 (Fig. 3a). Compared with the expression of GUS in partial lines without treatment (Fig. 2c), it could be inferred that cytokinin exerted a certain activation effect on the MtCKX1 promoter, which was consistent with the results of the hormone-induced expression described above.

Fig. 3.

The GUS assay of the transgenic lines (CKXp1-4) and wildtype (WT) as control under the treatment of different hormones. a the GUS analysis of the different tissues from CKXp1-4 and WT with 10 µmol·L^− 1 6-BA treatment. b the GUS analysis of the different tissues from CKXp1-4 and WT with 10 µmol·L^− 1 GA and 10 µmol·L^− 1 IAA treatment, respectively. Representative images are shown, as staining was consistent across all independent lines analyzed for each construct

In addition, the GUS assay was also completed with 10 µmol·L^− 1 GA and 10 µmol·L^− 1 IAA as inducers. The results showed that CKXp1 and CKXp2 lines expressed GUS protein at a higher level in the whole plant under the induction of GA and IAA at the seedling stage (Fig. 3b). In CKXp3, GUS was only observed in the stem and leaf treated by these two hormones, and the expression level was higher under GA treatment. Additionally, with GA treatment, the hypocotyl and stem of CKXp4 had notable GUS expression, while did the roots, stems, and leaves with IAA treatment (Fig. 3b). These results proved that CKXp1–4 could be activated by both GA and IAA, yet they exhibited certain tissue specificity.

C2H2 and GRF transcription factors could bind to the promoter of the MtCKX1 gene directly

To screen for upstream transcription factors that could bind to the MtCKX1 promoter and were associated with cytokinin response, we designed a 330 bp nucleic acid probe sequence from the 1789 bp to 467 bp interval upstream of the MtCKX1 promoter based on the results of GUS expression and cis-acting element analysis. This probe included MYC elements for stress response, ABRE elements for ABA response, O2-site elements for cytokinin response, TGACG-motif and CGTCA-motif elements for MeJA response, and P-box for GA response (Figure S2a), which was used for subsequent DNA-Pull down-MS.

SDS-PAGE gel electrophoresis analysis of elution products during Pull down-MS showed successful pulldown of multiple proteins in the target group and control (Figure S2b). 222 proteins were found in the experimental group and 469 in the control group, according to the results of mass spectrometry. Among them, 25 proteins detected only in the target group were considered potential interacting proteins (Figure S2c). Through the annotation of the NCBI database, four transcription factor proteins were mined, including C2H2, GeBP, GRAS, and GRF (Table 1).

Table 1.

The information of the candidate transcription factors binding to the MtCKX1 promoter

Accession	Description	Score	Coverage	Proteins	UP	Peptides	PSMs	MW (kDa)	calc. pI
Medtr2g084815.1	C2H2 family protein	2.43	4.3	2	1	1	1	30	5.01
Medtr2g010790.1	GeBP family protein	1.86	2.86	1	1	1	1	45.7	5.3
Medtr7g074650.1	GRAS family protein	0	6.34	1	1	1	1	89	6.58
Medtr8g020560.1	GRF family protein	0	25	1	1	1	1	39.9	7.99

UP Unique Peptides, PSMs Peptide Spectrum Matching, MW Molecular Weight, calc. pI Calculated isoelectric point

To corroborate the in-vitro interaction, Y1H assays were employed to verify the binding of C2H2 and GRF to the CKX promoter in yeast. Restriction digestion and PCR confirmed that the bait vector pAbAi-CKXP was successfully linearized and integrated into the Y1H genome (Figure S3a and b). AbA sensitivity tests revealed that Y1H cells carrying the bait could not grow on SD medium containing > 100 ng mL⁻¹ AbA (Figure S3c). This concentration was therefore selected as the selective threshold for subsequent experiments.

Co-transformation of C2H2-CKXP or GRF-CKXP into Y1H cells resulted in robust growth on SD-Leu medium supplemented with 100 ng mL⁻¹ AbA, whereas cells co-transformed with the empty pGADT7 vector failed to grow under the same conditions (Figure S3d). Serial dilution assays showed that the positive control (pGADT7-53 + pAbAi-p53) and the test combinations (C2H2-CKXP and GRF-CKXP) all exhibited dose-dependent growth, while the negative controls (pGADT7 + pAbAi-p53 or pGADT7 + pAbAi-CKXP) remained completely inhibited (Fig. 4a). Collectively, these results demonstrate that C2H2 and GRF bind directly to the CKX promoter in yeast. Dual-luciferase imaging revealed that luciferase activity was restricted to the leaf sectors infiltrated with pGreenII 62-SK-MtC2H2 or MtGRF together with pGreenII 0800-LUC-CKX1p; no signal was detected in any of the control combinations (Fig. 4b). These data demonstrate that MtC2H2 and MtGRF physically interact with the MtCKX promoter in planta.

Fig. 4.

Serial dilution assays during Y1H. a The numerical labels above indicate the dilution factors of the yeast culture. pAbAi-p53 + pGADT7-53 represents the Y1H strain after co-transformation of plasmids (pAbAi- p53 and pGADT7-53) as the positive control. pAbAi-p53/CKXP+pGADT7 represents the Y1H strain after co-transformation of plasmids (pAbAi-CKXP or pAbAi-p53 and pGADT7) as negative control. pAbAi-CKXP+pGADT7-C2H2/GRF represents the Y1H strain after co-transformation of plasmids (pAbAi-CKXP and pGADT7-C2H2 or pGADT7-GRF). SD/-Leu/AbA(100 ng/mL) represents SD medium containing 100 ng/mL AbA but not Leu. b Numbers 1 and 5 denote the agro-infiltration of pGreenII 62-SK empty + pGreenII 0800-LUC-CKX1p. 2 indicates pGreenII 62-SK-MtC2H2 + pGreenII 0800-LUC-CKX1p. 3 corresponds to pGreenII 62-SK-MtC2H2 + pGreenII 0800-LUC empty. 4 and 8 represent pGreenII 62-SK empty + pGreenII 0800-LUC empty. 6 corresponds to pGreenII 62-SK-MtGRF + pGreenII 0800-LUC-CKX1p. 7 denotes pGreenII 62-SK-MtGRF + pGreenII 0800-LUC empty

Discussion

To date, the functions of cytokinin oxidase/dehydrogenase (CKX) gene have been characterized in numerous plant species. For instance, CKX activity has been implicated in regulating vegetative growth under abiotic stress in pea (Pisum sativum) [17], tuber formation in potato (Solanum tuberosum) [18], development of shoot and root meristems in tobacco (Nicotiana tabacum) [19], and floral organ development in rice (Oryza sativa) [20]. These studies collectively demonstrate that CKX genes participate in diverse developmental processes, with functional divergence observed across species. Medicago truncatula, a key legume model plant [21], offers valuable insights applicable to other leguminous species due to its well-characterized genetics and genomic resources. However, functional studies on the CKX gene family in M. truncatula remain limited. Although several MtCKX mutants associated with shoot and root development have been generated [22], further mechanistic investigations and in-depth functional analyses are still lacking. This highlights a significant knowledge gap regarding the roles and molecular regulatory mechanisms of CKX genes in M. truncatula, underscoring the need for comprehensive research in this area.

In this study, we focused on the promoter of the MtCKX1 gene in Medicago truncatula to identify upstream transcription factors regulating its expression, employing a combination of promoter truncation, genetic transformation, GUS staining, and other physiological and genetic approaches. A 1789 bp length genomic sequence upstream of the initiation codon was cloned as the putative promoter region. In silico analysis revealed the presence of multiple cis-acting elements associated with hormone responses (CK, MeJA, ABA, GA), stress signals, and light responsiveness, suggesting that MtCKX1 expression may be induced by diverse hormonal and environmental cues. This hypothesis was further supported by exogenous hormone treatments. Previous studies have shown that auxin (IAA) promotes cell elongation and division, and elevated IAA levels often lead to reduced cytokinin content [23]. Given that CKX catalyzes the irreversible degradation of cytokinins, MtCKX1 may play an active role in IAA-mediated cytokinin homeostasis. Expression of MtCKX1 was also induced by 6-BA, consistent with observations in Poncirus trifoliata [24]. In contrast, exogenous MeJA application in wheat was reported to suppress CKX expression and activity, thereby promoting cytokinin accumulation [25]. This discrepancy may arise from differences in growth stages or functional divergence among non-orthologous CKX family genes. The responsiveness of CKX genes to ABA was further corroborated in strawberry, where only FvCKX1, FvCKX2, FvCKX4, and FvCKX8 exhibited ABA-induced upregulation [26], aligning with our findings. Additionally, studies in pea demonstrated that GA₃ treatment alters endogenous cytokinin levels primarily through modulation of CKX activity, suggesting that CKX serves as a key node in GA–cytokinin crosstalk [27], which further supports our results. Collectively, these data indicate that MtCKX1 expression is responsive to multiple hormones, including IAA, 6-BA, ABA, MeJA, and GA, positioning it as a reliable marker gene for studying hormone-responsive regulatory mechanisms.

To elucidate the core regulatory regions of the promoter and the specific mechanisms of its cis-elements, we performed systematic promoter truncation and subsequent stable transformation in Arabidopsis thaliana. Promoter truncation is widely employed in plant research for two primary purposes: first, to engineer and optimize promoters for efficient and specific regulation of downstream genes, particularly in CRISPR/Cas9 systems where U3/U6 promoters have been extensively modified in species such as cotton and tomato [28, 29]; second, to identify transcription factor binding sites and characterize hormone- or stress-responsive modules within promoters. For instance, in Brassica juncea, truncation and modification of the SOC1 promoter enabled the identification of binding sites for FLC and SVP proteins [30], while in tobacco, deletion analysis of the DHN promoter revealed distinct regions containing response elements for SA, MeJA, and ABA [31]. In this study, we adopted the second strategy. Based on the distribution of predicted cis-acting elements, the MtCKX1 promoter was sequentially truncated into four segments and stably introduced into Arabidopsis. GUS expression assays indicated that the region upstream of the CKX1 start codon contains a repressive segment within the − 1125 bp length region that inhibits GUS expression. Conversely, histochemical GUS staining demonstrated that the region between − 1789 bp and − 1125 bp harbors a cytokinin-responsive element driving strong reporter expression. Notably, the truncated version designated CKXp2 functioned as a particular, cytokinin-inducible promoter throughout the plant.

Promoters can be categorized into three types based on their expression patterns: constitutive, tissue-specific, and inducible promoters [32]. Constitutive promoters, such as the CKXp1 construct in this study, drive high-level transgene expression across all plant organs and are widely used for functional studies and trait improvement [33]. However, they often impose metabolic burdens and may impair normal growth [34]. Tissue-specific promoters, exemplified by CKXp4, restrict expression to particular tissues or developmental stages—in this case, the hypocotyl and roots at the seedling stage—thus minimizing pleiotropic effects [35]. Inducible promoters, such as CKXp3, activate gene expression only upon external stimuli like hormones or stresses [36]; here, CKXp3 exhibited cytokinin-dependent expression specifically in seedlings.

The CKXp2 promoter represents a hybrid type, combining developmental and inducible features: it drives strong constitutive expression during seedling development and remains responsive to cytokinin induction across all organs in mature plants. This dual functionality offers significant biotechnological potential. For example, coupling CKXp2 with a disease-resistance gene could create a modular transgenic system enhancing pathogen resistance during the vulnerable seedling stage, while avoiding unnecessary expression in adult plants to reduce metabolic costs and biological risks. In the event of pathogen attack, targeted cytokinin application could reactivate resistance gene expression, providing an environmentally friendly alternative to chemical pesticides. This will also represent one of the key focuses of our subsequent research.

Based on the results of GUS expression and histochemical staining analyses, the region from − 1789 bp to -1125 bp upstream of the CKX1 start codon was identified as the core regulatory segment of the promoter. A nucleic acid probe was designed within this region for DNA pull-down mass spectrometry (MS) analysis. DNA pull-down MS is a powerful in vitro technique for studying DNA-protein interactions [37]. Using this approach, four TFs, including C2H2, GeBP, GRAS, and GRF, were identified as potential upstream regulators of CKX1. Previous studies have shown that the expression of ZFP6, a C2H2-type zinc finger protein regulating inflorescence stem trichome formation, is influenced by cytokinin levels [38]. Given that CKX1 encodes a rate-limiting enzyme responsible for the irreversible degradation of cytokinins, its potential interaction with a C2H2-type TF suggests that CKX1 may serve as a molecular link between cytokinin signaling and trichome development, possibly mediating the regulation of ZFP6 by cytokinins. Similarly, the identification of GeBP (glabrous-enhancer binding protein), a known key regulator of trichome patterning [39], further supports a role for CKX1 in trichome development. The GRAS family TF detected in this screen has been implicated in diverse biological processes, including plant growth, hormone signaling, and stress responses. For instance, DELLA proteins, members of the GRAS family, act as negative regulators of GA signaling and participate in the crosstalk between GA and jasmonic acid (JA) pathways [40, 41]. Our results suggest that GRAS TFs may also be involved in cytokinin-mediated regulatory networks, extending their functional repertoire beyond GA and JA signaling. Furthermore, GRF TFs play crucial roles in early plant development and organ formation [42]. The interaction between CKX1 and a GRF TF may reflect a novel mechanism through which cytokinins modulate early growth and development. Subsequent yeast one-hybrid and dual-luciferase reporter assays independently confirmed the interaction between C2H2/GRF and the CKX promoter. In contrast, full-length clones of GeBP and GRAS were not obtained, precluding their inclusion in the validation pipeline; their functional relevance therefore remains to be elucidated. These challenges may be attributed to sequence complexity and low expression levels. While their biological relevance remains an open question, the direct binding of C2H2 and GRF to the MtCKX1 promoter, as robustly confirmed by Y1H and DLR assays, provides substantial mechanistic insight into the cytokinin-mediated transcriptional regulation of MtCKX1. The potential roles of GeBP and GRAS constitute a valuable direction for future research. Additionally, the parallel verification of the capture efficiency of biotin-labeled probes using Western blot analysis could be further integrated into the experimental pipeline in future studies. Together, these findings provide new insights into the molecular mechanisms by which CKX1 is transcriptionally regulated and how it may integrate multiple hormonal and developmental signals.

The transcriptional regulatory mechanism of MtCKX1 elucidated in M. truncatula in this study extends beyond a single model species. Given the conservation of cytokinin signaling pathways in plants and the pivotal role of the CKX gene family in regulating key agronomic traits in legume crops, we speculate that the cytokinin-responsive module identified here may function broadly across legumes. Furthermore, the strong, inducible promoter CKXp2 and the tissue-specific promoter CKXp4 characterized in this work provide programmable transcriptional regulatory tools for synthetic biology approaches in legume crop breeding. For instance, using CKXp2 to drive the expression of stress-resistance genes under drought-inducible conditions, or employing CKXp4 to specifically enhance the expression of nutrient-uptake-related genes in roots, holds promise for conferring desirable traits “on-demand” while minimizing the growth penalties often associated with constitutive expression. Future validation of these promoters or editing of homologous regulatory regions in various legume crops will be crucial for translating these fundamental discoveries into agricultural applications.

Materials and methods

Plant materials and growth conditions

The seeds of M. truncatula (R108) and Arabidopsis (Columbia) used in this study were preserved at the Lawn Research Institute of Beijing Forestry University. After vernalization for three days at 4℃, the seeds of M. truncatula were transferred to sterile filter paper in a sealed petri dish at 26/24℃ (light/dark, 16 h/8 h). One week after the seeds germinate, the seedlings had about three to four leaves and were transferred to sterile soil to continue growing.

Isolation and cis-acting elements analysis of MtCKX1 promoter

Total DNA was extracted from the mature leaves of M. truncatula through the CTAB method [43]. Based on the R108 genome sequence database (https://medicagohapmap2.org/), the primers were designed to amplify the about 2000 bp region of the MtCKX1 promoter via PCR (Table S1). The purified PCR products were ligated into the pMD-19T cloning vector from TAKARA and then the generated vector containing the region of MtCKX1 promoter was stored in DH5α E. coli competent cells for further analysis.

Cis-acting elements were predicted by PlantCARE (http://bioinformatics.psb.ugent.be/webtools/plantcare/html/) and the results were visualized with TBtools [44]. According to the distribution of these elements, the region of the MtCKX1 promoter was truncated as four sequences with different lengths using the similar method above.

Plasmid construction and generation of transgenic plants

Using the four cloning vectors above as a template, primers were designed to insert the four sequences into plant expression vector 1391Z to construct the MtCKXp::GUS, named CKXp1 CKXp2, CKXp3, and CKXp4, respectively. These vectors were transformed into Arabidopsis via the floral dip method [45]. T0 seeds obtained via the floral dip infiltration method were sown on half-strength MS solid medium containing 30 mg·L⁻¹ hygromycin. The plates were incubated under controlled conditions: 23 °C/22°C (day/night) with a 16 h light / 8 h dark photoperiod for 7 days. Seedlings exhibiting leaf bleaching or yellowing were considered hygromycin-sensitive (non-transgenic or negative), while the remaining green seedlings were identified as putative positive transformants. These green seedlings were then transferred to sterilized potting mix (peat moss: vermiculite = 3:1) and covered with a transparent film to maintain humidity. The film was removed after 3–4 days. Plants were grown under the same temperature and light conditions until flowering, and T1 seeds were harvested individually. T2 and T3 seeds were obtained using the identical selection method described above. T3 generation transgenic lines were identified by PCR and RT-PCR and cultivated for further treatment and analysis. The thermal cycling program used for both the standard PCR and the RT-PCR analyses was identical: 94℃ 3 min, 94℃ 15s, 55℃ 15s, 72℃ 30s, 35 cycles(To step 2), 75℃ 5 min, 12℃ ∞.

Quantitative Real-Time PCR (qRT-PCR)

Leaves were collected from two-month-old wild-type Medicago truncatula plants at 0, 1, 3, 6, 9, and 12 h after exogenous spraying with 10 µmol/L solutions of IAA, 6-BA, GA, ABA, or MeJA, for hormone-induced expression analysis of MtCKX1. All the collected leaf tissues above were immediately snap-frozen in liquid nitrogen and stored at -80 °C until RNA extraction. The total RNA of these samples was collected with Plant RNA Extraction Kit (TAKARA, Dalian, China). Using MtActin as the internal reference, respectively, the real-time fluorescence quantitative analysis of the CKX1 and GUS gene was completed by the CFX Connect RT-PCR system. The relative expression of them was calculated by the 2^−ΔΔCT method [46]. The qRT-PCR results are presented in the Results section (see Fig. 1). All samples were processed with four biological replicates. The primers used were shown in the supplementary Table s1.

Histochemical GUS detection assay

The seedlings of T3 generation transgenic lines were cultivated in 1/2 MS culture medium for 2 weeks and then randomly selected seedlings were immersed in GUS staining solution and left to stand for 12 h at 37 ℃, protected from light. The remaining plants were moved into the nutrient soil and continued to be cultivated for 4–6 weeks. Roots, stems, leaves, flowers, and fruit pods were collected, and the same treatment as before was applied to these tissues. And they were observed with a microscope, after removing chlorophyll with 75% alcohol.

Protein pull-down, coupled Mass Spectrometry (pull-down-MS)

Wild-type callus of M. truncatula was obtained via plant tissue culture and its nuclear protein was extracted by Nuclear Protein Extraction Kit (Solarbio, Beijing, China) as subsequent experimental material.

Based on the core cis-acting elements screened, primers were designed and then added with biotin tags, purified by High-Performance Liquid Chromatography (HPLC). Nucleic acid products used as decoys were synthesized by PCR (95℃, 30s; 72℃, 2 min; 37℃, 2 min; 25℃, 2 min; 4℃, ∞).

The nuclear protein and the nucleic acid products above were mixed in a 1.5mL centrifuge tube and were incubated on a rotary shaker at room temperature for 30 min. This mixture was purified using BeyoMag™ Streptavidin Magnetic Beads (Beyotime, China). Additionally, the above steps were completed using primers without added biotin tags, but with identical sequences, and the resulting product was used as a negative control. All the end products were subjected to electrophoretic analysis using 10% SDS-PAGE gel. When bands were observed on the silver-stained gel treated with the Fast Silver Stain Kit (Beyotime, China), the gel block was sent to Bio for mass spectrometry analysis (Biomarker, China).

Yeast One Hybrid (Y1H)

The promoter of MtCKX gene was connected to the pAbAi vector by homologous recombination, named as pAbai-CKXP. The constructed vector was linearized by restriction enzyme digestion and subsequently introduced into Y1H yeast competent cells for downstream AbA concentration-based screening. The transformed yeast strains were plated onto SD medium lacking leucine and supplemented with AbA at 0, 100, 200, 300, 500, 700, or 900 ng·mL⁻¹. the minimal inhibitory concentration was determined by monitoring yeast growth after 3–5 d at 30 °C.

Based on the annotation of candidate transcription factors obtained from pull-down-MS analysis, gene-specific primers were designed to amplify four transcription factors, and the corresponding open reading frames of the C2H2- and GRF-type factors were successfully cloned by high-fidelity PCR. The two transcription factors were inserted into the pGADT7 expression vector via homologous recombination and subsequently introduced into Y1H competent cells harboring the pAbAi-CKXP plasmid for one-to-one protein-DNA interaction assays. Co-transformed yeast strains were spotted in parallel onto SD medium lacking leucine and onto the same medium supplemented with 100 ng mL⁻¹ AbA; growth was assessed after 3–5 days of incubation at 30 °C. Yeast clones that grew normally were expanded and serially diluted 10⁻¹-, 10⁻²-, 10⁻³- and 10⁻⁴-fold; 5 µL aliquots of each dilution were spotted onto SD-Leu medium and SD-Leu + 100 ng mL⁻¹ AbA medium to evaluate the interaction between the bait DNA and the candidate proteins. Yeast co-transformed with pGADT7-53 and pAbAi-p53 (TAKARA Bio Inc.) served as the positive control, while strains harboring pGADT7 together with pAbAi-p53 or pGADT7 together with pAbAi-CKXP were used as negative controls.

Dual-Luciferase Reporter assay (DLR)

MtC2H2 and MtGRF were cloned into pGreenII 62-SK, and the CKX promoter into pGreenII 0800-LUC; all constructs were electroporated into Agrobacterium tumefaciens GV3101. Following overnight expansion, bacterial suspensions (OD₆₀₀ = 0.8, 1:1 v/v) were mixed to yield four infiltration combinations: (i) pGreenII 62-SK-MtC2H2/MtGRF + pGreenII 0800-LUC-CKX1p (treatment); (ii) pGreenII 62-SK-MtC2H2/MtGRF + pGreenII 0800-LUC empty; (iii) pGreenII 62-SK empty + pGreenII 0800-LUC-CKX1p; and (iv) pGreenII 62-SK empty + pGreenII 0800-LUC empty (controls). Each mixture was infiltrated into discrete sectors of the same young Nicotiana benthamiana leaf. Plants were maintained at 24 °C in darkness for 48 h. D-Luciferin potassium salt (1 mM, 0.1% Tween-20) was evenly applied to the abaxial surface; after 5 min of dark adaptation, luciferase activity was captured using an in vivo imaging system.

Conclusion

This study delineates a cytokinin-responsive module within the MtCKX1 promoter of Medicago truncatula and identifies C2H2 and GRF transcription factors as direct upstream regulators. Promoter truncation coupled with in vivo GUS assays resolved a core regulatory region (− 1789 to − 1125 bp) that confers strong, cytokinin-inducible and development-stage-specific activity, offering a biotechnological tool for spatiotemporal transgene control in legumes. DNA-pull-down MS, yeast one-hybrid and dual-luciferase experiments collectively demonstrate that MtC2H2 and MtGRF physically interact with this region, thereby modulating MtCKX1 expression and cytokinin homeostasis. These findings expand our understanding of CKX transcriptional networks in legumes and provide molecular targets for engineering cytokinin-mediated stress resilience and yield traits in forage and pasture species.

Abbreviations

CKXs

Cytokinin oxidases/dehydrogenases

CKs

Cytokinins

TSS

Transcription start sites

TFs

Transcription factors

DNA Pull down-MS

DNA pull-down coupled with mass spectrometry

Y1H

Yeast one-hybrid

ABA

Abscisic acid

MeJA

Methyl jasmonate

GA

Gibberellin

IAA

Auxin

GeBP

Glabrous-enhancer binding protein

JA

Jasmonic acid

qRT-PCR

Quantitative real-time PCR

DLR

Dual-luciferase reporter assay

Authors’ contributions

YL, MW and CZ conceptualized and designed the experiments. YL, MW and XS performed the experiments. YL, XS and KT collected and analysed data. YL and MW wrote the original draft. LH reviewed and edited the final manuscript.

Funding

National Key Research and Development Program of China(2023YFD1200302). National Natural Science Foundation of China (32573613). Yunnan Provincial Key Research and Development Program (202403AP140045). National Natural Science Foundation of China (32401481).

Data availability

All data in this study are available in the article and its appendix supplementary files.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.