Analysis of the IPT gene family reveals the critical role of the JrERF113-JrIPT1 module in cold resistance of walnut (Juglans regia)

Abstract

Walnut (Juglans regia), a globally significant nut-bearing tree species, holds paramount economic and ecological importance. However, it is susceptible to abiotic stress, especially low-temperature stress. Cytokinin (CTK), a central phytohormone orchestrating plant growth, developmental plasticity, and stress resilience. As the rate-limiting enzyme in CTK biosynthesis, isopentenyltransferase (IPT) mediates adaptive responses to environmental challenges. In this study, eight IPT family members were identified from the walnut genome, with phylogenetic analysis classifying them into two evolutionary clades. Notably, JrIPT1, belonging to the ATP/ADP-IPT subclass, exhibited broad tissue-specific expression and a marked response to cold treatment. Functional characterization revealed that JrIPT1-overexpressing transgenic Arabidopsis displayed an early flowering phenotype accompanied by significantly upregulated expression of CTK signaling pathway-related flowering genes. Further analysis demonstrated that JrIPT1 overexpression promoted CTK accumulation in both transgenic Arabidopsis and walnut plants, leading to substantially enhanced survival rates, photosystem activity, and reactive oxygen species (ROS) scavenging capacity under cold stress. Conversely, virus-induced gene silencing (VIGS)-mediated suppression of endogenous JrIPT1 in walnut reduced CTK levels and increased cold sensitivity. Mechanistic insights revealed that the cold-induced ERF transcription factor JrERF113 directly binds to the GCC-box motif in the JrIPT1 promoter and then activates its transcription, also endows walnuts with cold resistance. This study establishes a ‘JrERF113-JrIPT1’ regulatory module that enhances walnut cold tolerance by boosting CTK biosynthesis. The elucidated CTK metabolic regulatory network not only advances the understanding of molecular mechanisms underlying plant cold adaptation but also provides potential genetic targets and theoretical foundations for breeding cold-resistant walnut varieties.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12870-025-07656-7.

Introduction

Cytokinin (CTK), one of the six major classes of plant hormones, is renowned for its capacity to promote cell division and play indispensable roles in regulating meristem maintenance, organogenesis, floral induction, fruit set, and postharvest storage. The involvement of CTK in floral induction has been well-documented. For instance, exogenous application of the synthetic cytokinin 6-benzyladenine (6-BA) accelerates flowering transitions and induces the expression of flowering-related genes [1]. CTK specifically accumulates in shoot apical meristems, where it upregulates flowering genes to expedite floral initiation [2]. Elevated CTK levels in lateral buds promote flower bud formation in Japanese apricot [3], while multiple flowering pathway genes are differentially expressed in the Jatropha curcas inflorescence buds after treatment with exogenous CTK [4]. Beyond developmental regulation, CTK contributes to stress tolerance against drought [5], salinity [6], waterlogging [7], and cold [8].

Structural genes directly modulate endogenous CTK levels, with isopentenyltransferases (IPTs), cytokinin oxidase/dehydrogenases (CKXs), and LONELY GUY (LOG) proteins being well-characterized regulators. The rate-limiting step in CTK biosynthesis relies on IPT enzymes. The Arabidopsis genome encodes nine IPT members categorized into ATP/ADP-IPTs (AtIPT1, 3, 4–8) and tRNA-IPTs (AtIPT2 and 9), with the former primarily responsible for bioactive CTK production. Functional characterization of Arabidopsis IPTs through knockout mutants and overexpression lines has demonstrated their roles in CTK homeostasis, shoot regeneration, embryogenic callus formation, and hormonal crosstalk [9, 10].

Ethylene response factors (ERFs), a large transcription factor (TF) family, regulate downstream gene expression by binding canonical ethylene response element motifs (GCC-box). Beyond developmental regulation, ERF genes mediate responses to environmental stresses, including drought [11], salinity [12], and waterlogging [13]. For example, tomato ERF.D2 integrates ABA and JA signaling to enhance drought tolerance [14]. Multiple ERFs also participate in cold stress responses, such as BpERF13 from white birch (Betula platyphylla), which activates cold-responsive genes and scavenges reactive oxygen species (ROS) to confer freezing tolerance [15].

Walnut (Juglans regia), a foundational nut species with dual economic value for nut production and high-quality timber, faces mounting challenges from climate change-induced abiotic stresses. Low temperature represents a severe environmental stress in North China, which will disrupt reproductive development and lead to a reduction in timber yield. Research on plant cold adaptability in model organisms has been relatively thorough [16]. The CBF gene (C-repeat Binding Factor) plays a pivotal role in response to low-temperature stress, and there exist both CBF-dependent and CBF-independent signaling pathways [17, 18]. While genomic advancements have enabled functional characterization of walnut gene families like JrAHL and JrLOG [19, 20], the regulatory networks governing CTK-mediated stress adaptation remain poorly understood. As rate-limiting enzymes in CTK biosynthesis, IPT genes represent critical genetic hubs balancing growth-defense tradeoffs in woody perennials. However, their evolutionary conservation and functional divergence in walnut remain unexplored. This study presents the first genome-wide identification of the IPT family in walnut, revealing JrIPT1 as a cold-responsive hub through transcriptional profiling. Mechanistic investigations demonstrate that JrIPT1 transcription is directly activated by the AP2/ERF transcription factor JrERF113, establishing a novel ‘JrERF113-JrIPT1’ regulatory module that coordinates CTK homeostasis with freezing tolerance. This axis not only elucidates the molecular basis of CTK-mediated stress resilience in trees but also provides genetic targets for accelerating climate-resilient walnut breeding. Our findings offer critical genomic resources and mechanistic insights for enhancing abiotic stress tolerance in perennial species, with significant implications for sustainable tree management amid global climate change.

Materials and methods

IPT gene family identification

Arabidopsis thaliana IPT protein sequences were retrieved from the TAIR database and used as queries for BLASTP alignment (E-value threshold: 1E − 5) against protein-coding genes in the walnut genome sequenced by Huang’s laboratory. Domain presence in BLAST hits was validated using Pfam (http://pfam-legacy.xfam.org/) and NCBI’s CDD (http://www.ncbi.nlm.nih.gov/Structure/cdd/wrpsb.cgi). Candidate genes were renamed based on chromosomal localization, and the physicochemical properties of walnut IPT proteins were analyzed using ExPASy tools (https://www.expasy.org/). A phylogenetic tree was constructed via the Neighbor-Joining (N-J) method in MEGA7.0 with 1000 bootstrap replicates and gamma distribution modeling. The coding sequence (CDS) and intron structures of JrIPT genes were visualized using GSDS 2.0 (https://gsds.cgrpoee.top/). Syntenic relationships among walnut chromosomes and between walnut and Arabidopsis were analyzed using MCScanX, with duplication events displayed via Advanced Circos plots.

Plant material and growth conditions

Experimental materials included three-year-old potted Juglans regia ‘Lvling’ trees, one-month-old tissue-cultured J. regia ‘Lvling’ seedlings, Arabidopsis (ecotype Columbia-0), and tobacco (Nicotiana benthamiana) plants.

For cold stress analysis, potted walnut trees (maintained at the Hebei Agricultural University experimental orchard, No. 2596, Lekai South Street, College of Forestry, Hebei Agricultural University, Baoding, Hebei, China) were transferred to a 4 °C growth chamber. Leaf samples were harvested at 0, 2, 4, 8, and 16 h post-treatment, flash-frozen in liquid nitrogen, and stored at −80 °C for further analyses. All cultivars used are domesticated varieties and are not listed under the Convention on International Trade in Endangered Species of Wild Fauna and Flora (CITES).

Tissue-cultured ‘Lvling’ walnut seedlings were propagated under standard conditions (25 °C, 16 h light/8 h dark) in a growth cabinet. Arabidopsis wild-type (Col-0) and JrIPT1-overexpressing lines (JrIPT1-OE) were cultivated in controlled chambers (23 °C, 16 h light/8 h dark). Tobacco plants for subcellular localization, GUS staining, and dual-luciferase assays were grown under previously validated conditions [21].

RNA extraction, cDNA synthesis, and gene expression analysis

Total RNA was extracted using the E.Z.N.A. Plant RNA Kit (Omega Bio-tek, Norcross, GA, USA). First-strand cDNA was synthesized using the TransScript One-Step gDNA Removal and cDNA Synthesis SuperMix (TransGen Biotech, Beijing, China). Quantitative real-time PCR (qRT-PCR) was performed on an Applied Biosystems 7500 system with 18 S rRNA and Actin as internal controls [19].

Expression profiling utilized public RNA-seq datasets: 19-tissue transcriptome atlas [22], endocarp development series (35–147 days post-anthesis) [23], and Colletotrichum gloeosporioides infection time-course [24].

Gene cloning and sequence analysis

Full-length coding sequences (CDS) of JrIPT1 and JrERF113 were amplified from the cDNA of 2 h cold-treated walnut leaves. Three-dimensional structural prediction employed AlphaFold v2.1.1. Multiple sequence alignments for ERF proteins were performed using DNAMAN v9.0 with default parameters.

Subcellular localization

Target gene CDS (without stop codon) was fused to GFP in pCAMBIA2300-derived vectors under CaMV35S promoter control. Recombinant constructs were transiently expressed in N. benthamiana leaves via Agrobacterium tumefaciens (GV3101) infiltration. Subcellular localization was visualized using a Leica TCS-SP8 confocal microscope following protocols described previously [25].

Genetic transformation, phenotypic analysis, and cold tolerance assays

Arabidopsis transformation utilized the floral dip method with proJrIPT1::GUS and 35 S::JrIPT1 cassette in pCAMBIA1381 and pCAMBIA2300 constructs, respectively. Homozygous T₃ lines were selected via hygromycin/kanamycin resistance. Floral induction assays and analysis of flowering-related gene expression were conducted following previously described protocols [26]. Cold resistance assays followed established protocols [27]: Arabidopsis seedlings underwent cold acclimation at 4 °C for 1 d, followed by exposure to −8 °C for 5 h. Subsequently, they were allowed to recover at 4 °C for 12 h and then grown at 23 °C for 5 days, during which the survival rate was calculated. Additionally, Arabidopsis seedlings were exposed to 4 °C for three days, following which the maximum quantum yield of photosystem II (Fv/Fm) was measured, and leaves were stained with nitroblue tetrazolium (NBT) and 3,3’-diaminobenzidine (DAB) as previous description [27].

For walnut transformation, overexpression constructs and virus-induced gene silencing (VIGS) vectors were utilized. The overexpression vector was identical to that employed in subcellular localization experiments. For VIGS-mediated knockdown, target-specific cDNA fragments (408 bp of JrERF113 and 442 bp of JrIPT1) were cloned into the tobacco rattle virus vector pTRV2. Recombinant pTRV2 constructs (containing target sequences) and helper plasmid pTRV1 were transformed into Agrobacterium GV3101. One-month-old walnut seedlings underwent vacuum infiltration with Agrobacterium cultures at 0.08 MPa for 30 min. Post-infiltration, plants were cultured on DKW medium in darkness (12 h) followed by a 16 h photoperiod (25 °C) for 48 h. Transient expression efficiency was validated via RT-qPCR using gene-specific primers (Table S2). Transformed plants were subsequently exposed to 4 °C for 72 h before NBT and DAB staining protocols.

Phytohormone quantification

Endogenous cytokinin levels in one-month-old Arabidopsis and genetically modified walnut leaves were quantified via ESI-HPLC-MS/MS using an Agilent 1260 UPLC system coupled to a 6495 C triple quadrupole mass spectrometer, following previously described protocols [27].

Protein-DNA interaction assays

Recombinant JrERF113-HIS protein was expressed in E. coli BL21(DE3) (0.5 mM IPTG, 37 °C, 6 h) and purified using Ni-NTA agarose (Qiagen). Electrophoretic mobility shift assays (EMSA) were performed using the LightShift Chemiluminescent EMSA Kit (Pierce) with biotin-labeled JrIPT1 promoter probes (Table S3).

GUS histochemical staining and dual-luciferase reporter assay

Transient GUS expression in N. benthamiana leaves was analyzed through histological staining and fluorometric assays [21]. GUS staining in transgenic Arabidopsis was carried out according to previously validated methods [26]. In dual-luciferase assays, the JrIPT1 promoter-driven luciferase reporter and JrERF113 effector were co-expressed in tobacco leaves. Fluorescence signals and relative luciferase activities were quantified using an in vivo imaging system and detection kit [28].

Accession number

AtIPT1 (AT1G68460), AtIPT2 (AT2G27760), AtIPT3 (AT3G63110), AtIPT4 (AT4G24650), AtIPT5 (AT5G19040), AtIPT6 (AT1G25410), AtIPT7 (AT3G23630), AtIPT8 (AT3G19160), AtIPT9 (AT5G20040), OsIPT1 (AB239797), OsIPT2 (AB239798), OsIPT3 (AB239799), OsIPT4 (AB239800), OsIPT5 (AB239801), OsIPT6 (AB239803), OsIPT7 (AB239804), OsIPT8 (AB239805), OsIPT9 (AB239806), OsIPT10 (AB239807), VviERF105 (VIT_16s0013g01060), PtrERF110 (Pt5g006240.2), PtrERF108 (Pt7g003200), OsDREB2B (Q5W6R4.2), OsERF52 (LOC_Os05g49700).

Statistical analysis

Statistical analysis was conducted using one-way ANOVA with significant differences set at P < 0.05. All experiments were performed with at least three biological replicates.

Results

Identification of IPT family gene members in walnuts

Through homology alignment with Arabidopsis AtIPT members and validation via the Pfam database, eight JrIPT genes were identified in the walnut genome (Table S1). Phylogenetic analysis classified the JrIPT family into two subclasses: tRNA isopentenyltransferase (tRNA-IPT) and ATP/ADP isopentenyltransferase (ATP/ADP-IPT), with the tRNA-IPT subclass containing two members and the ATP/ADP-IPT subclass comprising six members (Fig. 1A). Multiple sequence alignment revealed highly conserved N-terminal amino acid sequences in IPT proteins (Fig. 1B), with all members containing the IPPT functional domain. Except for JrIPT7, all other members possessed the conserved IPT domain (Fig. 1C, middle panel). Gene structure analysis showed that only JrIPT3 in the ATP/ADP-IPT subclass contained one intron, while the remaining members were intron-less. In contrast, tRNA-IPT subclass members carried 9 and 10 introns, respectively (Fig. 1C, right panel). Chromosomal localization indicated the non-uniform distribution of JrIPT genes, with chromosomes 1, 2, 7, 11, 13, and 15 each harboring one member, and chromosome 8 containing two members (Fig. 1D). Synteny analysis identified five pairs of JrIPT paralogous genes in the walnut genome (Fig. 1D). Additionally, multiple orthologous gene pairs were detected between walnut and Arabidopsis genomes (Fig. 1E).

Fig. 1.

Identification and evolutionary analysis of the JrIPT1 gene family. A Phylogenetic analysis of IPT members. Phylogenetic tree of IPT genes from walnut, rice, and Arabidopsis constructed using the neighbor-joining method. B Multiple sequence alignment of IPT amino acid sequences from walnut and Arabidopsis.C Evolutionary relationships (left), conserved domain architecture (middle), and exon-intron structures (right) of JrIPT genes. D Genomic collinearity analysis of JrIPT gene family members in walnut. E Cross-species collinearity analysis comparing IPT gene orthologs between walnut and Arabidopsis

Analysis of gene expression patterns in the JrIPT genes

Spatiotemporal expression profiles of the JrIPT family were analyzed using transcriptomic data. ATP/ADP-IPT subclass members exhibited tissue-specific expression patterns: JrIPT1, JrIPT3, and JrIPT8 were highly expressed in mature leaves; JrIPT2 showed predominant expression in hull cortex and hull peel; JrIPT4 was enriched in callus exterior and interior; and JrIPT5 displayed minimal expression in mature packing tissue. In contrast, tRNA-IPT members JrIPT6 and JrIPT7 exhibited broad expression patterns (Fig. 2A). During endocarp development, JrIPT5 was expressed only at early stages, JrIPT8 exclusively at late stages, while JrIPT6 and JrIPT7 showed sustained expression with peak levels at early stages. Notably, JrIPT1 expression initially increased and then decreased, suggesting potential involvement in critical endocarp developmental regulation (Fig. 2B). Under Colletotrichum gloeosporioides infection, JrIPT2 was significantly downregulated in susceptible germplasm F26, and JrIPT4 was downregulated in susceptible germplasm F423, indicating possible roles in disease resistance. Meanwhile, tRNA-IPT members JrIPT6 and JrIPT7 maintained stable expression without significant changes (Fig. 2C).

Fig. 2.

Expression characteristics of JrIPT gene in different tissues (A), during the development stage of endopleura (B), and in F26 and the F423 fruits in response to anthracnose infection (C). The heat map was generated based on the transcriptomic data. The numbers within the boxes represent FPKM values, and the color scale in the annotation on the right is based on the logarithm (base 2) of the values, with the first column serving as the reference ratio for calculation

RT-qPCR revealed that cold stress (4 °C) induced rapid transcriptional upregulation of JrIPT1 and JrIPT8 within 2 h in walnut seedlings, with JrIPT8 remaining induced at 4 h. JrIPT3, JrIPT4, and JrIPT6 showed weak upregulation at 4 h, while JrIPT5 was significantly downregulated. Notably, JrIPT1 exhibited the highest fold change at 2 h, suggesting its involvement in cold stress tolerance (Fig. 3).

Fig. 3.

Expression dynamics of JrIPT genes under cold stress. Temporal expression profiles of JrIPT genes in walnut tissues under 4 °C cold stress, sampled at 0, 2, 4, 8, and 16 h post-treatment. Data represent mean ± SD (n = 3). Significant differences (p < 0.05) between time points are denoted by distinct letters (one-way ANOVA, Tukey’s test)

JrIPT1 gene cloning and expression characteristics analysis

The coding sequence of JrIPT1 was cloned via RT-PCR (Fig. 4A), revealing a 990 bp open reading frame encoding 329 amino acids (Supplementary File) with a predicted molecular weight of 37.25 kDa and theoretical pI of 8.79, classifying it as an ATP/ADP-IPT member. Protein structure prediction identified α-helices, β-sheets, and random coils, with the N-terminal GXXGXGK[S/T] motif located in the β-sheet region forming substrate-binding sites for ATP/ADP/AMP and DMAPP (Fig. 4B). Subcellular localization showed a uniform distribution of JrIPT1-GFP in the cytoplasm and nucleus (Fig. 4C). Spatiotemporal expression analysis using JrIPT1 promoter-driven GUS reporter in transgenic Arabidopsis revealed GUS signals in four-leaf stage seedlings (Fig. 4D), roots (Fig. 4E), stem internodes (Fig. 4F), cauline and rosette leaves (Fig. 4G-H), and reproductive organs including axillary and apical inflorescences and siliques (Fig. 4I-K), indicating multi-organ developmental regulation. Additionally, cold stress significantly activated JrIPT1 promoter activity (Fig. 4L-M).

Fig. 4.

JrIPT1 gene cloning, subcellular localization, and expression pattern analysis. A Agarose gel electrophoresis showing PCR products of JrIPT1 gene cloning (B) Predicted 3D structure of JrIPT1 protein, with the N-terminal GXXGXGK[S/T] catalytic motif highlighted. C Subcellular localization of JrIPT1-GFP fusion protein in transiently transformed tobacco epidermal cells. (D)-(K) Tissue-specific expression patterns revealed by GUS staining in transgenic Arabidopsis lines driven by the JrIPT1 promoter: (D) four-leaf stage seedling, (E) root, (F) mature stem, (G) cauline leaf, (H) rosette leaf, (I) axillary inflorescence, (J) apical inflorescence, (K) silique. (L)-(M) GUS staining intensity (L) and quantitative activity (M) under normal and cold conditions

Overexpression of JrIPT1 promotes CTK accumulation and early flowering in Arabidopsis

JrIPT1-overexpressing (JrIPT1-OE) Arabidopsis exhibited an early-flowering phenotype characterized by reduced rosette leaf number at bolting (Fig. 5A-B). Flowering-related genes AtFD (FLOWERING LOCUS D), AtTSF (TWIN SISTER OF FT), and AtSOC1 (SUPPRESSOR OF OVEREXPRESSION OF CONSTANS1) were significantly upregulated in transgenic plants. Endogenous hormone analysis confirmed elevated levels of cytokinins (N⁶-(Δ²-isopentenyl)-adenine, kinetin, and zeatin) in transgenic lines (Fig. 5D-F), validating JrIPT1’s role in cytokinin biosynthesis.

Fig. 5.

JrIPT1 overexpression alters flowering time and cytokinin homeostasis in Arabidopsis. A Representative images of flowering phenotypes in wild-type (Col-0, WT) and JrIPT1-transgenic plants. B Quantification of rosette leaf number at bolting stage. C qRT-PCR analysis of flowering-related gene expression. D-F Endogenous cytokinin (CTK) content measurements

Overexpression of JrIPT1 enhances low-temperature tolerance of Arabidopsis

JrIPT1-OE transgenic plants were treated with a freezing treatment to identify the function of JrIPT1 in modulating cold tolerance. No phenotypic differences were observed between transgenic and wild-type (WT, Col-0) plants before stress (Fig. 6A, upper panel). Although different plants exhibited comparable levels of leaf wilting following freezing treatment (Fig. 6A, middle panel), the JrIPT1-OE lines demonstrated notably superior recovery compared to the WT one-week post-treatment (Fig. 6A, lower panel). Consistent with these observations, survival rate analysis revealed a significantly higher survival rate in transgenic lines relative to WT following freezing exposure (Fig. 6B).

Fig. 6.

JrIPT1 enhances cold tolerance in Arabidopsis. A Phenotypic comparison of wild type (WT, Col-0) and transgenic lines following freezing stress (−8 °C, 6 h). B Post-recovery survival rates. C Chlorophyll fluorescence imaging (Fv/Fm) assessing photosynthetic efficiency under cold stress. D Relative Fv/Fm ratio (normalized to pre-cold-treated WT = 1). E-H Oxidative stress assessment via NBT (superoxide, E-F) and DAB (H₂O₂, G-H) staining, with quantitative intensity analysis (ImageJ, normalized to WT). Significant differences (p < 0.05) are denoted by distinct letters (one-way ANOVA, Tukey’s test)

Concurrently, low-temperature stress induces a reduction in photosystem II (PSII) maximum photochemical efficiency (Fv/Fm), accompanied by impaired PSII activity and compromised photosynthetic capacity. Notably, transgenic plants exhibit significantly higher Fv/Fm values under cold stress compared to WT (Fig. 6C), indicating superior maintenance of photosynthetic apparatus functionality.

NBT histochemical staining was performed to assess reactive oxygen species (ROS) accumulation in leaves of JrIPT1-OE transgenic and WT plants. Staining of the leaves of WT and transgenic lines was not significantly different under normal conditions (Fig. 6E, upper panel). After the cold treatment, WT leaves exhibited extensive dark blue spots under NBT staining, whereas transgenic plant leaves showed markedly fewer staining foci (Fig. 6E, lower panel; Fig. 6F). Meanwhile, similar results were also obtained by DAB staining (Fig. 6G and H). Collectively, these results demonstrate that JrIPT1 overexpression enhances ROS-scavenging capacity under cold stress conditions.

JrERF113 targets and activates JrIPT1

To elucidate the transcriptional regulatory network underlying JrIPT1-mediated cold resistance, in silico analysis using the PlantPAN tool identified an ethylene-responsive factor (ERF) family protein. Subsequent gene cloning confirmed a 681 bp open reading frame encoding 226 amino acids, which was designated as JrERF113 (Fig. 7A, Supplementary File). Consistent with the expression dynamics of JrIPT1, JrERF113 transcription exhibited rapid induction within 2 h of low-temperature exposure (Fig. 7B). Promoter-driven GUS staining assays and quantitative measurements of relative GUS activity in tobacco revealed that exposure to low temperatures significantly induced promoter activation (Fig. 7C). Multiple sequence alignment demonstrated significant conservation between JrERF113 and cold-responsive ERF orthologs from diverse plant species (Fig. 7D). JrERF113 contains an N-terminal AP2 domain - a conserved DNA-binding module found in plant regulatory proteins including APETALA2 and EREBP, and three-dimensional structural modeling predicted the AP2 domain localizes to the protein’s central core (Fig. 7E), suggesting structural conservation critical for DNA interaction. Subcellular localization via transient expression assays confirmed the nuclear enrichment of JrERF113 (Fig. 7F), consistent with its proposed role as a transcriptional regulator.

Fig. 7.

JrERF113 gene cloning, sequence analysis, subcellular localization. A JrERF113 gene cloning. B Expression pattern analysis of JrERF113 under cold treatment. C GUS staining and activity analysis were used to explore the response of the JrERF113 promoter to cold treatment. D Multiple sequence alignment of the AP2 domain in JrERF113 and ERF transcription factors from other species. E 3D structural model of the JrERF113 transcription factor protein, with an enlarged view of the AP2 domain (right). F Subcellular localization analysis of JrERF113

AlphaFold-mediated DNA-protein docking simulations predicted specific recognition of the GCC-box element within the JrIPT1 promoter by JrERF113. Modeling results indicated high-affinity interaction with the wild-type sequence (Fig. 8A, left), whereas mutation of the core GCC-box disrupted binding (Fig. 8A, right). To experimentally validate these predictions, double-stranded probes corresponding to wild-type and mutant sequences were synthesized (Fig. 8B) for EMSAs. EMSA analysis confirmed a direct interaction between JrERF113 and the wild-type probe (Fig. 8C), with binding specificity demonstrated through competitive inhibition with an unlabeled probe and complete loss of signal with the mutated sequence (Fig. 8C). Collectively, these results establish direct physical interaction between JrERF113 and the JrIPT1 promoter through the GCC-box.

Fig. 8.

Binding and transcriptional activation analysis of JrERF113 on the JrIPT1 promoter. A Molecular docking simulation of JrERF113 protein binding to the GCC-box (ERF binding site) in the JrIPT1 promoter, showing docking modes for wild-type (left) and GCC-box mutant (right) sequences. B Schematic diagram of GCC-box location in the JrIPT1 promoter, with sequences of wild-type and mutant probes used for EMSA. C EMSA demonstrating JrERF113 binding to the JrIPT1 promoter in a GCC-box-dependent manner. D-F JrERF113 enhances JrIPT1 promoter activity. Agrobacterium strains harboring reporter (proJrIPT1::LUC) and effector (JrERF113) plasmids (D) were infiltrated into tobacco leaves. Luciferase signals were visualized 3 days post-infiltration (E), and luciferase activity ratios were quantified (F)

To validate the JrERF113-mediated transcriptional regulation of JrIPT1, a dual-luciferase reporter system was employed to assess promoter responsiveness. The JrIPT1 promoter region was cloned into a reporter construct, while the JrERF113 coding sequence (CDS) was inserted into an effector vector (Fig. 8D). Following transient co-expression in tobacco leaves, fluorescence signal detection (Fig. 8E) and quantitative luciferase assays (Fig. 8F) demonstrated significantly elevated transcriptional activity of the JrIPT1 promoter when co-expressed with JrERF113, confirming JrERF113 functions as a transcriptional activator of JrIPT1.

JrERF113 and JrIPT1 regulate the CTK accumulation and cold tolerance in walnut

To elucidate the functional roles of JrERF113 and JrIPT1 in walnut cold responses, we generated transient overexpression and VIGS transgenic lines using tissue-cultured walnut seedlings. RT-qPCR analysis confirmed significant transcriptional upregulation of JrERF113 and JrIPT1 in their respective overexpression lines, with concurrent activation of JrIPT1 transcription in JrERF113-overexpressing plants. Conversely, VIGS-mediated gene silencing resulted in transcriptional suppression of both targets, accompanied by reduced JrIPT1 expression in JrERF113-silenced plants (Fig. 9A). Endogenous cytokinin profiling revealed marked elevations in N6-(Δ²-isopentenyl) adenine, kinetin, and zeatin levels in overexpression lines, whereas VIGS plants exhibited significant reductions in these hormone accumulations (Fig. 9B).

Fig. 9.

JrERF113 and JrIPT1 modulate cytokinin accumulation and cold tolerance in walnut. A Detection of gene expression levels in walnut leaves overexpressing JrIPT1, JrERF113, or subjected to virus-induced gene silencing (VIGS). B Quantification of CTK contents. C, D DAB staining (C) and relative staining intensity quantification (D) under cold stress. (E)-(F) NBT staining (E) and relative staining intensity quantification (F) under cold stress

To investigate the regulatory impact of these genes on cold tolerance, transgenic lines were subjected to low-temperature stress followed by DAB and NBT histochemical staining to visualize reactive oxygen species (ROS) accumulation. Under control conditions, no differential staining patterns were observed. Following cold exposure, leaves transformed with the empty (pC2300-EV) vector exhibited extensive brown (DAB staining) and dark blue (NBT staining) areas indicative of H₂O₂ and superoxide (O₂⁻) accumulation, respectively. In contrast, overexpression lines exhibited significantly fewer stained regions (left panels, Fig. 9C, E, D and F), suggesting the reduced ROS accumulation. Conversely, VIGS lines showed intensified staining (right panels, Fig. 9C, E, D and F), indicating elevated ROS levels. Collectively, these findings demonstrate that JrERF113 and JrIPT1 coordinate the modulation of endogenous cytokinin levels and maintenance of ROS homeostasis, thereby conferring enhanced cold tolerance in walnut.

Discussion

Cytokinins (CTKs) are involved in multiple aspects of plant growth, development, and environmental responses. Enzymes participate in CTK biosynthesis, activation, and inactivation. Previously, we identified LOG family members responsible for CTK activation in walnut and demonstrated their tissue- and development-specific expression patterns [20]. IPT serves as the rate-limiting enzyme catalyzing CTK synthesis. In this study, eight JrIPT genes were identified in the walnut genome, and phylogenetic analysis classified them into two major subfamilies: tRNA-IPT and ATP/ADP-IPT. Closely clustered members exhibit similar conserved domain compositions and gene structural frameworks (Fig. 1C). While gene duplication preserves sequence conservation in syntenic genes, it also introduces structural and expression pattern diversity during family expansion. Notably, duplication events not only retained critical functional domains (e.g., IPPT and IPT conserved domains) but also drove functional diversification between and within subfamilies through non-coding sequence variations (e.g., intron insertion/deletion) and regulatory element differentiation in promoter regions. For instance, despite sharing 66.85% amino acid identity, JrIPT4 lacks an intron compared to JrIPT3. This structural difference in non-coding regions may alter transcription factor binding site accessibility, shifting expression from leaf-specific (JrIPT3) to callus-specific (JrIPT4) patterns (Fig. 2A). Analogously, intronic regions of the gibberellin oxidase-encoding gene ga2ox1 may serve as upstream transcription factor binding targets, influencing its transcription [29]. Even single-nucleotide variations in promoter regions can affect gene expression [30], further emphasizing the role of non-coding sequences in spatiotemporal gene regulation.

CTK homeostasis is decisive for fruit development. During walnut endocarp development, JrIPT1 and JrIPT2 (both ATP/ADP-IPT subfamily members with 74.33% amino acid identity) exhibit distinct expression patterns: JrIPT1 expression peaks during critical endocarp stages (initiation to mid-stage), while JrIPT2 shows pericarp-specific expression (Fig. 2B). This divergence likely stems from cis-regulatory element differences in their promoters—JrIPT1 contains root-specific (ROOTMOTIFTAPOX1), light-responsive (GATA-box), and stress-responsive (GT1CONSENSUS) elements (Table S4), suggesting integration of environmental cues in endocarp regulation. Notably, transient CTK elevation during early fruit development correlates with cell division activity, and exogenous CTK application or IPT overexpression enhances seed yield [31], providing functional insights into JrIPT1’s role in endocarp development.

Anaerobic induction, MeJA-responsiveness, gibberellin-responsive, zein metabolism regulation, and circadian control elements were also predicted in the JrIPT1 promoter. Widespread promoter activity in roots, stems, leaves, and reproductive organs (Fig. 4D-K), combined with tissue-specific elements (e.g., pollen-specific POLLEN1LELAT52) (Table S4), suggests multi-organ developmental roles. The tissue-specific expression profile based on transcriptomic data reveals that JrIPT1 exhibits distinct tissue preference in its expression (Fig. 2). However, GUS staining results demonstrate strong promoter activity in different tissues and at various developmental stages (Fig. 3). This discrepancy may arise from significant differences in the gene regulatory networks between walnut and Arabidopsis, or from the extremely high sensitivity of the GUS reporter gene assay. In woody plants, exogenous 6-BA (a CTK analog) promotes flowering [1]. Here, JrIPT1-overexpressing Arabidopsis displayed early flowering with elevated endogenous CTK levels (Fig. 5). In Arabidopsis, CTK promotes flowering independently of the florigen gene FT but via its paralogs TSF, transcription factor FD, and floral integrator SOC1 [32]. Similarly, elevated CTK correlates with transient DlFT1 upregulation in longan [2]. The significant upregulation of CTK pathway genes (AtFD, AtTSF, AtSOC1) in our transgenic lines suggests conserved mechanism (Fig. 5D-F). Additionally, microRNAs (miRNAs) modulate CTK-mediated flowering transitions [33], and CTK pathway activation here offers new perspectives on woody plant flowering.

JrIPT1 expression is rapidly and strongly induced by cold stress, unlike other IPT members (Fig. 3), and its promoter activity is cold-responsive (Fig. 4L-M). This prompted an investigation into JrIPT1’s role in cold tolerance, a critical abiotic stress in northern China. Cold damages photosynthetic systems [34]. Transgenic Arabidopsis exhibited higher survival ratio post-freeze treatment, and better photosynthetic performance as well as lower ROS accumulation post-cold treatment (Fig. 6), corroborated by similar results in walnut (Fig. 9). The ectopic expression of MdIPT5b enhances cold resistance in transformed apple calli and tomato plants [35]. Overexpression of PhIPT5 in Populus hopeiensis enhances cold hardiness [36], and CTK signaling mutants in Arabidopsis alter cold responses [37]. Conversely, disrupted zeatin homeostasis in tea (Camellia sinensis) impairs antioxidant defense under cold stress [38]. This study demonstrated that JrIPT1-silenced walnut exhibited significantly reduced CTK content and compromised cold tolerance (Fig. 9), suggesting the existence of a conserved CTK-mediated cold response mechanism across plant species.

The expression of IPT genes responds to various environmental signals and exogenous hormones [26] and is also subject to direct transcriptional regulation by numerous transcription factors. For example, AtIPT7 in Arabidopsis is activated by the STM transcription factor [39], while LsKN1 in lettuce directly binds to the promoter of the cytokinin biosynthesis gene LsIPT3 and upregulates its expression [40]. An ERF-binding site was predicted in the JrIPT1 promoter, which is bound and activated by the cold-induced JrERF113 transcription factor. Molecular docking predictions and EMSA assays further confirmed that this binding is dependent on the ERF-binding site element (Fig. 8A). Consistently, JrERF113 overexpression alters JrIPT1 expression, CTK levels, and cold tolerance in walnut (Fig. 9). ERF homologs in other species (Fig. 7C) (e.g., PtrERF108 and PtrERF110 from Poncirus trifoliata and citrus, respectively) regulate cold tolerance by regulating raffinose, sugar, and sterol biosynthesis, respectively [41, 42]. Furthermore, rice ERF transcription factors OsDREB2B and OsERF52 modulate chilling tolerance [43, 44], indicating distinct transcriptional regulatory networks in different species to cope with cold stress.

In walnut, the JrERF113-JrIPT1 module not only modulates CTK levels but also enhances cold tolerance through ROS scavenging (Fig. 9), linking CTK signaling to antioxidant systems. Collectively, this study elucidates the evolutionary traits of the walnut IPT family and the biological functions of JrIPT1, revealing a novel JrERF113-JrIPT1 module that confers cold tolerance by regulating endogenous CTK levels.

Conclusions

The walnut genome encodes eight JrIPT family genes, among which cold-inducible JrIPT1 exhibits ubiquitous expression patterns and accelerates flowering in transgenic Arabidopsis. Ectopic overexpression of JrIPT1 drives a marked elevation in endogenous CTK levels, concomitant with enhanced cold tolerance. Conversely, genetic suppression of JrIPT1 expression results in diminished CTK accumulation and compromised cold resilience. Mechanistically, the cold-responsive ERF transcription factor JrERF113 directly activates JrIPT1 transcription by binding to the GCC-box element within its promoter region, thereby forming a regulatory circuit that amplifies CTK biosynthesis and cold hardiness in walnut. These findings uncover a novel JrERF113-JrIPT1 regulatory module that confers enhanced cold tolerance through CTK pathway modulation, providing a molecular framework for engineering climate-resilient walnut germplasm.

Authors’ contributions

P. J. and G. Q. planned and designed the research. T. Z., X. Z., R. Y., B. L., and X. D. performed experiments, conducted fieldwork, analyzed data etc. T. Z., P. J. and G. Q. wrote the manuscript.

Funding

This study was financially supported by Modern Seed Industry Technological Innovation (21326304D-2).

Data availability

All data and materials that support the findings of this study are included in the manuscript or in the Supplementary Materials.

Declarations

Ethics approval and consent to participate

This study did not include human or animal subjects.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.