Isolation of CsHB33 from Cucumber and Its Ectopic Expression in Arabidopsis Promotes Leaf Growth but Impairs Drought Tolerance

Abstract

The ZF-HD transcription factors play key roles in plant development and stress responses, yet their functions in cucumber remain poorly understood. Here, we characterized a cucumber ZF-HD gene, CsHB33, and investigated its role in leaf development and drought tolerance. CsHB33 was highly expressed in developing leaves. Its expression was significantly suppressed by abscisic acid (ABA) and down-regulated under drought stress. Heterologous overexpression of CsHB33 in Arabidopsis promoted leaf growth by increasing cell size, but simultaneously enhanced sensitivity to osmotic and drought stress, accompanied by higher stomatal aperture and water loss. Transcriptomic analysis revealed that CsHB33 overexpression up-regulated genes associated with leaf enlargement, while under drought it led to constitutive activation of aromatic amino acid biosynthesis, particularly tryptophan metabolism. This metabolic disturbance likely contributes to the drought-sensitive phenotype. Our findings reveal that CsHB33 exerts opposing effects on leaf growth and drought tolerance, providing new insights into ZF-HD gene function in cucumber and identifying a potential target for improving stress resilience in breeding.

1. Introduction

Leaf morphology, particularly leaf size, is a key determinant of plant growth and agricultural productivity [1]. The leaf serves as the primary site for both photosynthesis and transpiration [2], and also acts as a critical sensor enabling plants to perceive and respond to environmental changes such as drought stress. Leaf size directly influences photosynthetic area and water-use efficiency, thereby regulating crop yield and biomass accumulation [3,4,5]. Thus, identifying key genes that coordinately regulate leaf size and drought response is essential for elucidating plant adaptation mechanisms and developing drought-resistant crop varieties.

Transcription factors (TFs) play a central role in regulating plant growth, development, and stress responses, functioning by binding to specific cis-regulatory elements within target gene promoters to modulate transcriptional activity [6]. The homeodomain (HD), initially discovered in Drosophila melanogaster, represents one of the most widespread DNA-binding motifs in both animals and plants [7]. Structurally, the HD domain is characterized by a conserved 60-amino-acid sequence that folds into three α-helices, with the third helix (recognition helix) and an N-terminal disordered arm (extending beyond helix I) mediating high-affinity, sequence-specific DNA binding [8,9]. Based on the size, location, and structure of HDs and their associations with other domains, HD-containing proteins are classified into six major families: PHD finger (a finger-like domain associated with an HD), WOX (WUSCHEL-related Homeobox), KNOX (Knotted-related Homeobox), ZF-HD (Zinc Finger-Homeodomain), HD-ZIP (Leucine Zipper-related Homeodomain), and Bell-type homeodomain [8]. Among these, the ZF-HD family constitutes a well-studied, plant-specific group of transcription factors, with genes identified in all sequenced land plant genomes [7]. The typical ZF-HD protein features two conserved structural domains: a non-canonical C5H3 Zinc Finger (ZF) and a DNA-binding homeodomain (HD). Additionally, the ZF-HD family includes the Mini Zinc Finger (MIF) subfamily, which retains only the ZF domain and is hypothesized to have evolved from ZF-HD via HD domain loss [7,10].

The ZF-HD family has gained significant attention in recent research due to its pivotal roles in regulating plant development and coordinating responses to environmental stresses [7]. ZF-HD genes have been functionally characterized in a range of plant species, including Arabidopsis, rice, wheat, tomato and cucumber, where it regulates key developmental processes such as leaf morphogenesis, seed longevity, and floral organ development, as well as responses to abiotic stresses like drought, salinity, heat, etc. [7,11,12,13,14,15,16]. For example, OsZHD1 and OsZHD2 in rice regulate leaf development, and their overexpression leads to abaxial leaf curling due to an increased number and disordered arrangement of bulliform cells in the leaf [11]. Similarly, Arabidopsis HB33 functions as a key regulator of leaf size by coordinately controlling cell number and cell size. Reduced expression of HB33 leads to decreases in both cellular parameters and overall leaf area, whereas its moderate elevation promotes leaf expansion [17]. In addition to their role in plant development, ZF-HD genes also participate in abiotic stress responses. Arabidopsis ZFHD1 functions as a transcriptional activator that specifically binds to the promoter of EARLY RESPONSIVE TO DEHYDRATION STRESS 1 (ERD1), and its expression is induced under drought, high salinity, and abscisic acid treatments [18]. The SL-ZH13 gene in tomato is upregulated by drought stress, and silencing SL-ZH13 reduces drought tolerance [19]. In rice, seven ZF-HD transcription factors have been identified that bind to the promoter of DROUGHT RESPONSE ELEMENT BINDING 1 (OsDREB1B) and regulate responses to multiple stresses, including cold, drought, and mechanical stress [20]. In conclusion, although the ZF-HD family functions in many plants, its role in the Cucurbitaceae family, especially in cucumbers, remains largely unexplored.

Cucumber (Cucumis sativus L.), a globally important vegetable crop in the Cucurbitaceae family, is widely cultivated and plays a crucial role in global vegetable production. However, its growth and yield are highly susceptible to abiotic stresses, such as drought, low temperature, and salinity [21,22,23,24]. For instance, drought stress can severely inhibit leaf photosynthesis, impairing plant growth and development and ultimately reducing fruit yield and quality [21,25]. While ZF-HD proteins are known to play critical roles in diverse stress responses across multiple plant species, and leaf morphogenesis represents a key target for breeding stress-tolerant varieties, the specific functions and regulatory mechanisms of ZF-HD genes in cucumber leaf development and stress adaptation remain poorly understood and require systematic investigation. In this study, we characterized the ZF-HD family gene CsHB33 from cucumber, analyzing its sequence features, phylogenetic relationships, and cis-acting elements. Its expression patterns were investigated across different organs, at various leaf developmental stages, and in response to ABA and drought. Heterologous overexpression of CsHB33 gene in Arabidopsis was used to detect its role in leaf development and stress response. The related gene network of CsHB33 involved in leaf development and drought stress response was preliminarily explored by transcriptome analysis. These findings provide a foundation for understanding CsHB33 function and offer a new perspective for its application in breeding stress-resistant cucumber varieties.

2. Materials and Methods

2.1. Characterization of CsHB33 and Its Promoter

The CsHB33 gene (CsaV3_2G028770.1) and its 2-kb upstream region promoter were amplified from cucumber genomic DNA and cDNA using high-fidelity PCR (P505, Vazyme, Nanjing, China). Specific primers were designed based on the cucumber genome database (http://cucurbitgenomics.org/organism/20, accessed on 15 June 2025) and are shown in Supplementary Table S1. The promoter sequence was analyzed for cis-regulatory elements via PlantCARE (https://bioinformatics.psb.ugent.be/webtools/plantcare/html/, accessed on 15 June 2025) [26], and the annotation of identified motifs was categorized by functional annotation (Supplementary Table S2).

2.2. Bioinformatics Analysis

Gene structure, molecular weight, and isoelectric point (pI) of CsHB33 were predicted via NCBI (https://www.ncbi.nlm.nih.gov/) and Expasy Protparam (http://web.expasy.org/translate/, accessed on 18 June 2025). Secondary and tertiary structures were modeled using SOPMA (https://npsa-prabi.ibcp.fr/cgi-bin/npsa_automat.pl?page=npsa_sopma.html, accessed on 18 June 2025) and PHYRE2 (http://www.sbg.bio.ic.ac.uk/phyre2/html/page.cgi?id=index, accessed on 18 June 2025). Homologous sequences were identified via BLASTP (v2.16.0) against NCBI, and a phylogenetic tree was constructed using MEGA7 (v7.0.26).

2.3. Expression Profiling

To analyze the tissue-specific expression pattern of CsHB33, cucumber plants at the fruiting stage were sampled to collect apical buds, leaves, stems, tendrils, female and male flowers, and roots. For spatial expression analysis in leaves, developing leaves were dissected into three distinct regions along the proximodistal axis: the proximal region (characterized by active cell proliferation), the medial region (representing a transition zone of proliferation and expansion), and the distal region (dominated primarily by cell expansion). Three samples were taken from each region in the transverse direction. For analysis of root zones, surface-sterilized cucumber seeds were germinated on 1/2 MS medium. Root tips were micro-dissected to isolate the meristematic zone (small, densely packed cells), elongation zone (evident cell elongation), and mature zone (root hair development). To examine hormonal regulation, cucumber seedlings at the two-leaf stage were uniformly sprayed with 100 μM abscisic acid (ABA). Leaf samples were collected at 0 h (untreated control), 3 h, 6 h, 9 h, 12 h, and 24 h after treatment. Three independent plants were used as biological replicates.

For gene expression analysis, total RNA was extracted using the RNA-Easy Isolation Reagent (R701-02; Vazyme). Reverse transcription was performed to synthesize cDNA, which was then used as the template for quantitative PCR via Taq Pro Universal SYBR qPCR Master Mix (Q712-02; Vazyme). qRT-PCR was carried out on a Roche LightCycler^® 96 system (Roche Diagnostics, Basel, Switzerland) in a 20 μL reaction volume containing 2× ChamQ Universal SYBR qPCR Master Mix. All reactions were performed in three technical replicates for each biological replicate. The expression level of CsHB33 was normalized to the internal reference genes ACTIN, and relative expression was calculated using the 2^−ΔΔCt method. Primer sequences are listed in Supplementary Table S1.

2.4. Plant Materials and Plant Transformation

Cucumber (Cucumis sativus L.) inbred line ‘649’ was used in this study. The cucumber compact plant-type mutant ‘cpa1’ used in this experiment was obtained from previous studies [27]. The cucumber seeds were germinated on moist filter paper at 28 °C in the dark for 2 days. The germinated seeds were then transferred into soil and grown in a controlled chamber 25 °C day/18 °C night, 16 h light/8 h dark photoperiod) until the two-true leaf stage. The plants were subsequently cultivated at the Huasui Base of Hunan Agricultural University (Changsha, China).

For functional validation, the coding sequence of CsHB33 (without stop codon) was amplified using a high-fidelity DNA polymerase (KMM-201, Toyobo, Osaka, Japan), and the purified PCR product was subsequently inserted into the BamHI site of the pCAMBIA1305-FLAG vector to generate the 35S::CsHB33-FLAG overexpression construct. The resulting plasmid was confirmed by Sanger sequencing to ensure the correct insertion and reading frame (Tsingke Biotechnology Co., Ltd., Tianjin, China). The recombinant plasmid was introduced into Agrobacterium tumefaciens GV3101 and transformed into Arabidopsis thaliana ecotype Columbia-0 (Col-0) via the floral dip method. Transgenic T₁ seeds were selected on 1/2 Murashige and Skoog (1/2 MS) medium containing hygromycin. The presence of the transgene in T₁ plants was confirmed by genomic DNA PCR. T₂ seeds harvested from self-pollinated T₁ plants were also subjected to hygromycin selection, and all surviving T₂ individuals were self-pollinated to generate T₃ seeds. Two T₃ lines displaying complete hygromycin resistance, which indicated homozygosity, were selected for use in subsequent experiments. These homozygous T₃ lines, designated OE#1 and OE#2, were identified by PCR and confirmed by qRT-PCR for subsequent assays. Primers used for PCR and qRT-PCR in this study are listed in Supplementary Table S1.

Arabidopsis plants were grown in soil-filled pots (with four plants per pot) under controlled conditions (20 °C, 16 h light/8 h dark photoperiod, 8000 lx light intensity, 60% relative humidity).

2.5. Arabidopsis Phenotypic and Cellular Morphology Analysis

Phenotypic measurements of Arabidopsis transgenic lines and the wild type (Col-0) were taken approximately 45 days after transplanting. For each line (OE#1, OE#2, and Col-0), at least 15 individual plants served as biological replicates. For each plant, rosette traits including petiole length, rosette leaf number, and total rosette leaf area were recorded. The largest fully expanded rosette leaves were harvested at the same developmental stage from both wild-type (Col-0) and transgenic lines for cellular analysis. Epidermal peels from the abaxial leaf surface were prepared and observed under a Mshot digital microscope (Guangzhou Mingmei Optoelectronic Technology Co., Ltd., Guangzhou, China). Cell density and individual cell area were quantified from representative microscopic fields using ImageJ software (v1.47, National Institutes of Health, Bethesda, MD, USA).

2.6. Arabidopsis Stress Treatment

For osmotic stress treatment, surface-sterilized seeds of homozygous T₃ transgenic and wild-type Arabidopsis lines were sown on 1/2 MS medium supplemented with 0 mM (control), 100 mM, or 200 mM mannitol. Plates were placed vertically and grown under standard conditions for 15 days. Root length was measured and samples were collected when clear phenotypic differences were observed between the treatments and the control.

Soil-grown Arabidopsis plants at the pre-bolting stage were subjected to drought stress without watering. Plants were monitored daily; the largest rosette leaves were sampled once visible wilting, leaf rolling, or loss of turgor occurred for subsequent stomatal assays. The experiment was conducted using 26-day-old soil-grown plants subjected to a continuous drought treatment of approximately 12 days. Following a sustained water deficit of about one week, plants were re-watered, and their fresh weight was measured 3 days later to assess recovery capacity.

2.7. Arabidopsis Stomatal Aperture Assay

For the stomatal aperture assay, fully expanded rosette leaves were collected from two independent homozygous T₃ transgenic lines (OE#1 and OE#2) and wild-type (Col-0) plants. Epidermal peels were prepared from the largest rosette leaves of soil-grown plants that had been subjected to natural drought stress. Peels were immediately observed under a light microscope to assess the stomatal aperture under drought conditions. Stomatal aperture was quantified as the width-to-length ratio using ImageJ software (https://imagej.net). Three plants per genotype (OE#1, OE#2, and Col-0) were used as biological replicates for this assay. Three independent microscopic fields were analyzed per replicate, with approximately three stomata measured per field.

2.8. Arabidopsis Leaf Water-Loss Rate

For the leaf water-loss rate assay, fully expanded rosette leaves were collected from the same two independent homozygous T₃ transgenic lines (OE#1 and OE#2) and wild-type (Col-0) plants used in leaf water-loss rate analyses. Fully expanded leaves of uniform size were detached and placed horizontally in a growth chamber (25 °C, 10,000 lx, 50% relative humidity). Fresh weight was recorded at 0 h, 0.5 h, 1 h, 2 h, 3 h, 4 h, 5 h, and 6 h after detachment. Three biological replicates were measured per genotype at each time point.

2.9. Arabidopsis Transcriptome Profiling and Data Analysis

Transcriptome sequencing was performed on leaf samples from both well-watered and drought-stressed plants to identify genes associated with leaf development and drought response. For normal growth conditions, leaf samples were collected from 38-day-old soil-grown wild-type (Col-0) and CsHB33-overexpressing (OE#1) plants. For drought conditions, water withholding was initiated at 26 days after transplanting. Leaf samples were harvested from plants subjected to 12 days of water withholding. Three biological replicates were included per genotype and treatment. Total RNA was extracted, and samples with RNA Integrity Number (RIN) > 7 and concentration ≥ 600 ng·μL⁻¹ were used for library construction. Libraries were sequenced on an Illumina HiSeq platform. Clean reads were aligned to the Arabidopsis reference genome (TAIR10), and differential gene expression analysis was conducted following established methods [27]. Genes with an absolute log₂ fold change (|log₂FC|) ≥ 1 and an adjusted p-value (padj) < 0.05 were considered significantly differentially expressed. These raw data have been uploaded to the NCBI website (PRJNA1433039).

To validate the RNA-seq results, differentially expressed genes were selected for confirmation by qRT-PCR. The primer sequences used for qRT-PCR analysis to validate the RNA-seq results are listed in Supplementary Table S1. The lists of differentially expressed genes identified under normal and drought conditions are provided in Supplementary Tables S3 and S4, respectively.

3. Results

3.1. Sequences and Phylogenetic Analysis of CsHB33

Based on previous transcriptomic studies which identified CsaV3_2G028770 as significantly downregulated in the compact plant-type mutant ‘cpa1’ [27], we first confirmed this expression pattern by qRT-PCR, showing that CsaV3_2G028770 transcript levels were indeed markedly lower in ‘cpa1’ compared to the wild-type inbred line ‘649’ (Figure 1A–C). We then cloned and verified this gene, designated CsHB33 (CsaV3_2G028770), encoding a typical ZF-HD family transcription factor (Figure 1). The coding sequence (CDS) of CsHB33 was amplified from the leaf cDNA of cucumber inbred line ‘649’ and showed 100% identity with the annotated CsaV3_2G028770 in the cucumber genome (Supplementary Figure S1A). The CDS is 963 bp in length and encodes a protein of 320 amino acids (Supplementary Figure S1C).

Phylogenetic analysis indicated that CsHB33 shares the highest homology with Arabidopsis AtHB33 (AT1G75240), followed by AtHB24, AtHB34, and other members of the ZF-HD family (Figure 1E). Multiple sequence alignment with orthologs from Arabidopsis, tomato, and rice revealed that CsHB33 contains the two characteristic conserved domains of this family: the ZF-HD_dimer domain and the homeo_ZF_HD domain, which are highlighted by a blue box and a red box in the alignment figure, respectively (Figure 1D,F). The high conservation of these domains across species supports the functional conservation of this transcription factor family during plant evolution [8,28]. Secondary structure prediction via SOPMA showed that CsHB33 consists of 15.94% α-helices, 4.69% extended strands, and 79.38% random coils (Supplementary Figure S1D). The predicted tertiary structure is consistent with this composition, further supporting its classification as a canonical ZF-HD transcription factor (Supplementary Figure S1E). Additionally, cloning and analysis of the approximately 2 kb promoter region upstream of the CsHB33 translation start site identified multiple cis-regulatory elements, including hormone-responsive motifs (e.g., salicylic acid (SA), gibberellin (GA), and methyl jasmonate (MeJA)) and stress-related sites (such as ARE and MYB/MYC binding sites) (Supplementary Table S2). This suggests that CsHB33 may be involved in hormonal signaling and stress-adaptation pathways.

3.2. Expression Patterns of the CsHB33 in Cucumber

To investigate the spatial and temporal expression pattern of CsHB33, qRT-PCR was performed on various cucumber tissues and developmental stages. The expression of CsHB33 was highest in leaves, followed by roots (Figure 2A). Given its preferential expression in leaves, we further analyzed its transcript levels at different leaf positions, corresponding to distinct developmental stages. The highest expression was detected in the developing leaves at the 10th node, whereas mature leaves (e.g., 2nd node and 4th node) showed significantly lower expression (Figure 2B,C). We subsequently dissected developing leaves along the proximodistal axis into three regions: the proximal zone (characterized by active cell proliferation), the medial zone (a transition between proliferation and expansion), and the distal zone (dominated by cell expansion). Strikingly, CsHB33 expression was highest in the middle region of the leaf and significantly lower in both the proximal and distal zones (Figure 2D,E). Moreover, to assess whether this expression pattern reflects a broader role in growth processes, we examined CsHB33 expression in roots. A comparable spatial pattern was observed in roots. CsHB33 expression was highly expressed in the elongation and maturation zones but low in the meristematic region (Figure 2F), reinforcing its association with cell growth processes.

In parallel, we assessed the hormonal regulation of CsHB33 expression. ABA is one of the most abundant hormones found in plants and serves as a key regulator in their response to abiotic stresses such as drought and salinity [29]. Exogenous ABA treatment led to a significant suppression of CsHB33 transcript levels (Figure 2G). Together with its developmental expression profile, this hormone responsiveness suggests that CsHB33 may serve as an integrator of developmental and environmental signals to modulate leaf morphogenesis.

3.3. Leaf Phenotype of CsHB33 in ARABIDOPSIS

To investigate the role of CsHB33 in leaf morphogenesis, we generated two independent T₃ homozygous CsHB33-overexpressing lines (OE#1 and OE#2) in Arabidopsis, which were confirmed by DNA PCR (Figure 3A). qRT-PCR analysis verified that CsHB33 transcript levels were significantly elevated in both transgenic lines compared to the wild type (Col-0) (Figure 3B). At the mature stage, phenotypic analysis revealed that CsHB33-OE plants exhibited visibly enlarged leaves compared to the Col-0, as shown by whole-plant architecture and individual leaf morphology (Figure 3C,D). Quantitative assessment further showed that the overexpression lines possessed significantly longer petioles and a larger rosette leaf area, whereas the number of rosette leaves remained statistically unchanged (Figure 3E–G). To dissect the cellular basis of this phenotype, the epidermal cells of the largest rosette leaves were examined. Microscopic observation revealed that epidermal cells in OE lines were significantly larger in cell size but exhibited reduced cell number per unit area compared to Col-0 (Figure 3H–J). Collectively, these results demonstrate that CsHB33 overexpression promotes leaf expansion and petiole elongation predominantly through enhancing cell enlargement rather than increasing cell proliferation. This identifies CsHB33 as a critical positive regulator of cell size during leaf development in Arabidopsis.

3.4. Overexpression of CsHB33 Impairs Drought Tolerance in Arabidopsis

The ZF-HD transcription factor family has been implicated in regulating both plant development and abiotic stress responses. Given our previous observation that the expression of CsHB33 was inhibited by ABA treatment, we further examined its transcript levels under drought treatment. qRT-PCR analysis confirmed that CsHB33 transcript levels were significantly downregulated in wild-type cucumber seedlings under drought stress (Figure 4A), consistent with its repression during ABA treatment (Figure 2G).

To elucidate the functional role of CsHB33 in drought responses, we performed osmotic (mannitol) and natural drought stress assays using two independent T₃ homozygous CsHB33-overexpressing Arabidopsis lines and wild-type Col-0 plants. For osmotic stress, 15-day-old seedlings were grown on 1/2 MS medium supplemented with 100 mM, 200 mM and 300 mM mannitol (Figure 4B,E). The CsHB33-OE lines exhibited greater sensitivity than Col-0 under both 100 mM and 200 mM mannitol treatments (Figure 4B,E). Specifically, the whole-plant fresh weight of the two CsHB33-OE lines was reduced to only 83% and 77% of Col-0 under 100 mM mannitol, and to 81% and 70% of Col-0 under 200 mM mannitol (Figure 4C). Most of plants died under 300 mM mannitol, but the survival rate was lower in CsHB33-OE lines compared to wild-type Col-0 (Figure 4B,D). Under 100 mM mannitol treatment, root growth analysis revealed that CsHB33-OE plants displayed severe growth inhibition, with primary root lengths reduced to 77% (OE#1) and 73% (OE#2) of Col-0 values (Figure 4F). Root growth was severely inhibited in all plants under 200 mM mannitol, and thus, no quantitative data were presented.

In parallel, we performed natural drought assays on 26-day-old soil-grown plants. After 12 days of water deprivation, we recorded phenotypes, extended the drought for ~1 week, then rewatered and captured phenotypes again after 3 days (Figure 5A,C). The CsHB33-OE lines lost water more rapidly during the drought period, exhibiting significantly higher water loss rates in leaves than Col-0 (Figure 5B). Microscopic analysis of stomatal apertures under drought conditions further revealed significantly wider openings in CsHB33-OE lines (Figure 5D). The stomatal apertures were 40% in OE#1 and 37% in OE#2 versus 14% in Col-0 (Figure 5E), providing a physiological basis for the heightened water loss phenotype. These results demonstrate that ectopic expression of CsHB33 impairs drought tolerance in Arabidopsis, primarily through defective stomatal regulation and compromised water conservation capacity.

3.5. CsHB33 Regulates the Expression of Genes Involved in Cell Enlargement

To further elucidate the gene regulatory network associated with CsHB33 in the control of leaf morphogenesis, we performed transcriptome profiling of the wild-type (Col-0) and CsHB33-overexpression line (OE#1). A total of 347 differentially expressed genes (DEGs) were identified, including 76 up-regulated genes and 271 down-regulated genes (Figure 6A). KEGG pathway enrichment analysis indicated significant enrichment in processes such as protein processing in the endoplasmic reticulum and ubiquinone and other terpenoid-quinone biosynthesis (Figure 6B). However, as the expression trends of genes within these KEGG-enriched pathways did not directly align with the observed leaf size phenotype, our subsequent functional analysis focused on the most significantly altered DEGs, which revealed a coherent transcriptional program promoting cell expansion.

Among the top up-regulated genes, we identified a core set of regulatory genes that directly orchestrate cell expansion and morphogenesis (Figure 6C). For example, two pectate lyases (AT5G48900, ATPLL21 and AT1G67750, PLL16) and an aspartyl protease (AT5G07030, AP25) were significantly upregulated. The pectate lyases are known to promote wall loosening [30]. Concurrently, Cyclin-D3-1 (AT4G34160, CYCD3-1), a regulator of cell division, and Formin-like protein 3 (AT4G15200, FH3), involved in cytoskeletal reorganization, were also upregulated. The coordinated regulation of these genes involved in cell wall remodeling and cellular structural dynamics strongly supports the conclusion that CsHB33 induces expression changes conducive to cellular enlargement and leaf growth.

To validate the RNA-seq data, we selected four DEGs, including representatives from the cell wall modification, for qRT-PCR analysis. The expression patterns confirmed the transcriptome findings (Figure 6D–G).

3.6. CsHB33 Modulates Genes in Tryptophan Metabolism Under Drought Stress

To investigate the role of CsHB33 in drought-responsive gene regulation, we conducted transcriptome analysis on Col-0 and CsHB33-OE (OE#1) plants under drought stress. A total of 914 DEGs were identified, including 299 up-regulated genes and 615 down-regulated genes in CsHB33-OE plants compared to Col-0 (Figure 7A). KEGG pathway analysis of these DEGs revealed that ‘Biosynthesis of amino acids’, ‘Phenylalanine, tyrosine and tryptophan biosynthesis’, and ‘Tryptophan metabolism’ were among the top 20 enriched pathways (Figure 7B).

Notably, within the core aromatic amino acid biosynthesis pathway ‘Phenylalanine, tyrosine and tryptophan biosynthesis’, we found that 12 core structural genes in the shikimate pathway were differentially expressed in drought-stressed CsHB33-OE plants (Figure 7B). This set included key enzymes such as 2-deoxy-D-arabino-heptulosonate 7-phosphate synthase (AT4G39980, DHS1), anthranilate synthase (AT5G05730, ASA1), and tryptophan synthase (AT5G17990, TRP1; AT3G54640, TSA1 and AT4G27070, TSB2), which catalyze critical steps in the synthesis of tryptophan, phenylalanine, and tyrosine. Together, these transcriptional changes suggest that CsHB33 may perturb aromatic amino acid homeostasis under drought, potentially contributing to altered stress adaptation.

To confirm the transcriptome findings, we performed qRT-PCR on four selected genes involved in these metabolic pathways (Figure 7D–G). The results demonstrated significant expression differences between Col-0 and CsHB33-OE plants under drought stress, which validated the RNA-sequencing data.

4. Discussion

The ZF-HD gene family plays a crucial role in regulating plant growth, development, and environmental adaptation, and its members have been extensively studied in various plant species such as Arabidopsis thaliana and rice [13,15,20]. In this study, a significantly downregulated gene, CsHB33 (CsaV3_2G028770.1), was identified from the transcriptome data of the cucumber compact plant-type mutant ‘cpa1’ [27]. Sequence and phylogenetic analyses confirmed that CsHB33 encodes a typical ZF-HD family transcription factor (Figure 1). The conserved domain of cucumber CsHB33 exhibits high consistency with that of Arabidopsis, tomato, rice and other species (Figure 1F), supporting the evolutionary functional conservation of this gene family [12,14,28].

Tissue-specific expression analysis showed that CsHB33 is preferentially expressed in cucumber leaves, with the highest transcript level detected in the 10th-node functional leaves and stem tip, suggesting its important role in leaf morphogenesis (Figure 2A,B). This expression pattern is similar to Arabidopsis homolog AtHB33, which is also highly expressed in young leaves and apex [17], further supporting a conserved role in early leaf development. Exogenous hormone treatments revealed that CsHB33 expression responds rapidly to ABA (Figure 2G). Given the well-documented involvement of ZF-HD family genes in stress responses [7,12,14,20] and the central role of ABA as a stress hormone, these results imply that CsHB33 may integrate ABA signaling with the regulation of leaf growth. Promoter cis-acting element analysis further supports this hypothesis, as the CsHB33 promoter region contains diverse regulatory sequences associated with plant hormone signaling and abiotic stress responses, including ARE and MYB/MYC binding sites (Supplementary Table S2). The presence of MYB/MYC binding sites suggests that CsHB33 expression is likely regulated by the ABA signaling pathway, which is consistent with previous studies that MYB transcription factors mediating ABA-dependent gene expression [31,32,33]. In Arabidopsis, AtHB33 is repressed by ARF2 in an ABA-dependent manner to modulate seed germination and primary root growth [34]. Elevated ABA levels induce ARF2 expression, which in turn suppresses AtHB33, thereby inhibiting germination and root elongation to promote stress adaptation [34]. Considering the conserved stress-related functions of ZF-HD transcription factors, CsHB33 may similarly function as a molecular mediator linking ABA signaling to stress-responsive growth regulation in cucumber.

In order to verify the biological function of CsHB33, we overexpressed it in Arabidopsis thaliana. Phenotypic analysis showed that the transgenic lines overexpressing CsHB33 showed significant leaf size increases compared to the wild-type control plants in Arabidopsis (Figure 3). These findings confirm the causal relationship between the downregulated expression of CsHB33 in the cucumber ‘cpa1’ mutant and its compact plant type [27], and also demonstrate that CsHB33 acts as a key regulator of leaf expansion, similarly in Arabidopsis [17]. Transcriptomic analysis of these Arabidopsis larger-leaved CsHB33-OE plants provided a direct molecular explanation. The expression of CsHB33 upregulated a coordinated set of genes that execute cell expansion, which was characterized by the induction of key cell wall-loosening enzymes (pectate lyases and aspartyl protease) [35], a cell cycle promoter (CYCD3-1) [36], and cytoskeletal regulators (FH3) [37]. Together, these changes drive the cellular enlargement that underlies the increase in leaf size.

Studies have shown that many ZF-HD transcription factors genes play an important regulatory role in plant stress response. Under drought, high salt and other stresses, the expression of ZF-HD genes is up-regulated in many species, such as several ScZHDs in rye and SlZHDs in tomato [38]. For example, under PEG-simulated drought, the expression of BnZHD3 in rapeseed showed complex temporal and spatial patterns: it continually increased in roots, decreased first and then increased in stems, and increased first and then decreased in leaves [38]. In contrast to these common upregulation patterns, we found that the expression of CsHB33 in cucumber leaves was down-regulated under drought stress (Figure 4A). Further phenotypic analysis showed that CsHB33-OE exhibited increased sensitivity to drought and mannitol stress, showing more severe growth inhibition than wild-type plants (Figure 4 and Figure 5). Further physiological assays revealed that CsHB33-overexpressing Arabidopsis plants maintained a higher stomatal aperture under drought stress, leading to increased water loss (Figure 5B,E). This impaired stomatal closure likely compromises dehydration avoidance, a key component of drought tolerance [39,40]. Although stomatal closure defects may be a key contributor to drought-sensitive phenotypes in CsHB33-overexpressing plants, the precise molecular mechanism linking it to the ABA signaling remains to be elucidated.

Transcriptomic data reveal a molecular mechanism linking this phenotype to drought stress. Arabidopsis CsHB33-OE plants showed a coordinated upregulation of genes in the aromatic amino acid biosynthesis pathway (Figure 7B,C). Although drought stress typically elevates the levels of tryptophan, phenylalanine, and tyrosine in plants [41], which serve as precursors of protective secondary metabolites [42], the constitutive overexpression of this entire biosynthetic module in CsHB33-OE plants suggests a disruption of metabolic homeostasis. For example, key metabolites derived from this pathway, such as auxin, melatonin, and serotonin, are known to be crucial for integrating stress adaptation with processes like antioxidant defense and photosynthesis [43]. However, the broad and sustained induction of the pathway in CsHB33-OE plants implies a loss of metabolic balance rather than a fine-tuned protective response. Notably, drought typically triggers a conserved adaptive response, which is inhibiting leaf elongation by reducing both cell production and final cell size to conserve resources [44]. In contrast, CsHB33-OE plants maintain a growth-promoting transcriptional program while failing to activate essential drought avoidance mechanisms, thereby disrupting the critical balance between growth and stress resilience.

As a member of the ZF-HD family, CsHB33 likely exerts its function by binding to specific cis-elements and modulating the expression of downstream target genes [7,20]. Our transcriptomic profiling reveals that CsHB33 overexpression leads to the concerted upregulation of the core shikimate pathway genes, including AtTSA1 and AtTSB2 (Figure 7), suggesting that these biosynthetic genes may be direct or indirect targets of CsHB33-mediated regulation. TSB1 has been reported to inhibit ABA activity by interacting with β-glucosidase 1 (BG1), thereby coordinating the homeostasis of tryptophan and ABA [45]. Under stress, decreased TSB1 expression releases BG1 inhibition, promoting ABA accumulation and enhancing stress tolerance at the cost of reduced growth [45]. In parallel, in CsHB33-OE plants, the continuous up-regulation of genes such as TSB2 may lead to an imbalance in tryptophan metabolic flux, which may not only affect normal osmotic regulation due to excessive consumption of precursors, but also interfere with such a fine hormone cross-regulation network, thereby destroying the balance between growth and stress resistance.

In summary, this study identifies CsHB33 as a key ZF-HD transcription factor regulating both leaf development and drought adaptation in cucumber. Its functional characterization reveals a new role for this gene family in coordinating morphogenesis and environmental adaptation, providing an important candidate gene and theoretical foundation for molecular design breeding in cucumber—particularly for developing compact, water-efficient, and drought-resistant germplasm. Future work employing CRISPR/Cas9-mediated editing in cucumber, combined with ChIP-seq and RNA-seq analyses, will help systematically unravel the direct target genes and regulatory networks of CsHB33, offering a robust basis for optimizing plant architecture and stress resilience in molecular breeding programs.