Genome-wide identification of HXK gene families in cucumber and watermelon: CsHXK1 and ClHXK6 play key roles in responding to multiple stresses

Abstract

Background

Hexokinase (HXK) plays a key role in plant sugar metabolism and signal transduction, and is involved in regulating plant growth, development, and stress responses. To date, the HXK gene family in cucurbit crops has only been identified in melon (Cucumis melo), while systematic identification and analysis of HXK genes in cucumber (Cucumis sativus) and watermelon (Citrullus lanatus) remain lacking.

Result

In this study, six HXK genes were identified in both the cucumber (CsHXK1–CsHXK6) and watermelon (ClHXK1–ClHXK6) genomes. Phylogenetic analysis classified these genes into two types (Type A and Type B). Analyses of gene structure and conserved motifs confirmed high conservation among family members. Collinearity analysis further revealed that segmental duplication was the main driver of HXK family expansion in these two crops. Additionally, cis-acting element analysis showed that the promoter regions of CsHXK and ClHXK genes contained abundant stress-responsive elements. Using 7 publicly available cucumber transcriptome datasets and 14 publicly available watermelon transcriptome datasets, we analyzed the tissue-specific expression patterns of CsHXK and ClHXK genes, as well as their expression responses to abiotic and biotic stresses. The results showed that CsHXK1 was highly expressed in cucumber roots and flowers, while ClHXK6 exhibited broad expression across multiple watermelon tissues. Functionally, CsHXK1 was significantly up-regulated under five stress conditions (salt, waterlogging, downy mildew, powdery mildew, and angular leaf spot), and ClHXK6 showed significant differential expression under six stress conditions (salt, drought, osmotic stress, Fusarium wilt-1, Fusarium wilt-3, and squash vein yellowing virus infection). qRT-PCR validation was performed for CsHXK1 under salt stress and for ClHXK6 under salt and drought stresses, and the results consistently supported the transcriptome data, confirming the critical roles of CsHXK1 and ClHXK6 in mediating stress responses in cucumber and watermelon, respectively.

Summary

This study represents the first comprehensive and systematic identification of the HXK gene family in cucumber and watermelon, coupled with an analysis of their expression patterns under diverse abiotic and biotic stresses. It provides a theoretical basis for in-depth exploration of HXK gene functions and offers valuable candidate genes (CsHXK1 and ClHXK6) for improving stress resistance in cucumber and watermelon breeding programs.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12870-025-08038-9.

Introduction

Sugars are pivotal regulators of plant growth, development, and metabolic homeostasis. As primary photosynthates, they not only serve as core substrates for plant energy metabolism but also function as signaling molecules that modulate organogenesis, stress acclimation, and reproductive processes [1]. Moreover, through an intricate sugar metabolic network, plants spatiotemporally allocate photoassimilated carbon to sink tissues (e.g., roots, fruits, and developing seeds) to meet the energetic and biosynthetic demands of growth and development. Sugar signaling pathways crosstalk with hormone signaling networks to orchestrate plant physiological and molecular responses to environmental cues (e.g., biotic and abiotic stresses) [2].

Hexokinase (HXK; EC 2.7.1.1) is a conserved rate-limiting enzyme in plant sugar metabolism that catalyzes the irreversible phosphorylation of hexoses (e.g., glucose and fructose) to produce hexose-6-phosphates (H6P). These products act as a critical metabolic branch point linking glycolysis, the pentose phosphate pathway, and nucleotide biosynthesis [3]. This reaction serves dual roles: it initiates central carbon metabolism and transduces sugar signals to regulate plant growth, development, and stress responses [4, 5]. Based on N-terminal targeting sequences, plant HXK genes are classified into three types: Type A (e.g., AtHXK3 and OsHXK4), which contain chloroplast transit peptides and localize to the plastid stroma; Type B (e.g., AtHXK1, AtHXK2, AtHKL1/2/3, and OsHXK2/3/5/6/9/10), which possess a hydrophobic membrane-anchoring domain and associate with mitochondrial outer membranes [6, 7]; and Type C (e.g., OsHXK1/7/8), which lack both membrane-anchoring domains and chloroplast transit peptides and have been characterized primarily in mosses and rice [8, 9].

Advances in high-throughput sequencing and bioinformatics have facilitated genome-wide identification and characterization of the HXK gene family across diverse plant species, including Arabidopsis thaliana [6], rice (Oryza sativa) [7], maize (Zea mays) [10], and apple (Malus domestica) [11]. In A. thaliana, six HXK genes (AtHXK1-3 and AtHKL1-3) have been functionally validated: these genes integrate nutrient, photosynthetic, and hormonal signals while acting as dual-function molecules with both catalytic and hexose-sensing activities. AtHXK1-3 encode catalytically active proteins capable of hexose phosphorylation, whereas AtHKL1-3 produce hexokinase-like proteins (HKLs) devoid of kinase activity [6, 12]. Notably, AtHXK1 protein predominantly localizes to mitochondria but can also translocate to the nucleus, with both forms exhibiting gene expression regulatory capacity [13, 14]. Beyond metabolic roles, AtHXK1 plays pivotal roles in photomorphogenesis and stomatal conductance regulation. Its weak glucose-binding affinity may establish a feedback mechanism that suppresses plant growth, thereby functioning as a negative growth regulator [15]. Rice harbors 10 OsHXK genes, among which OsHXK1 is a unique single-exon gene while the remaining OsHXK members exhibit highly conserved exon-intron structures [7]. OsHXK1 acts as a positive regulator of leaf senescence by mediating glucose accumulation and inducing reactive oxygen species (ROS) production [16]. OsHXK5 and OsHXK6 function as glucose sensors, with OsHXK5 being essential for pollen maturation, germination, and tube elongation [17, 18]. OsHXK7 enhances anaerobic germination by upregulating glycolysis-mediated fermentation genes [19]. RNAi-mediated suppression of OsHXK10 leads to non-dehiscent anthers and reduced pollen germination [20]. Maize contains nine ZmHXK members. Among them, ZmHXK4 (localized to mitochondria) exhibits strong activity during early germination, while ZmHXK7 (localized to the cytosol) is critical for germination initiation and seedling growth [10, 21]. Similarly, apple contains nine MdHXK genes, among which MdHXK1 promotes anthocyanin biosynthesis by directly phosphorylating and stabilizing the MdbHLH3 transcription factor (TF) [11, 22].

Beyond growth and development, HXK genes also play critical roles in mediating plant stress responses, with accumulating evidence from diverse species. For abiotic stresses: in A. thaliana, cold and salt stresses significantly downregulate AtHXK3 and AtHKL1 expression while remarkably upregulating AtHXK2 [23, 24]; transgenic A. thaliana expressing Glycine max GmHXK2 exhibits enhanced salt tolerance, whereas GmHXK2-silenced plants show reduced expression of salt tolerance-related genes (e.g., GmSOS1, GmSALT3) and increased salt sensitivity [25]; ectopic expression of Prunus HXK3 in A. thaliana significantly improves dual tolerance to salt and drought stresses [26]; AtHXK1-transgenic tobacco tolerates salt and drought stresses attributable to reduced transpiration rates [27]; PdHXK1 (Poplar Hexokinase 1) interacts with the GATA family transcription factor PdGNC and enhances poplar drought resistance by increasing HXK activity [28]; in Jatropha curcas, four HXK genes (JcHXK1, JcHXK2, JcHXK3, and JcHXK4) are significantly upregulated in leaves after cold stress, with JcHXK3 showing distinct cold-induced expression in roots [29]. For biotic stresses: overexpression of OsHXK1 induces ROS production and enhances rice resistance to rice black-streaked dwarf virus (RBSDV) infection [30]; in apple, overexpression of MdHXK1 enhances resistance to the fungal pathogen Botryosphaeria dothidea by regulating the salicylic acid (SA) signaling pathway and ROS metabolism [31]; in Populus trichocarpa, the expression levels of PtHXK2 and PtHXK6 are significantly upregulated following infection by the pathogenic fungus Fusarium solani [32]. Collectively, these findings confirm that HXK genes are pivotal for both abiotic and biotic stress responses.

Cucumber (Cucumis sativus) and watermelon (Citrullus lanatus) are economically important cucurbit crops, cultivated globally as agricultural commodities and integral components of human diets. However, they are highly sensitive to abiotic stresses (e.g., salt, drought) and biotic stresses (e.g., powdery mildew, Fusarium wilt), which severely limit yield and quality in global cultivation. Among cucurbit crops, the HXK gene family has only been identified in melon (Cucumis melo) [33], while systematic characterization and functional analysis of HXK genes in cucumber and watermelon remain limited. The present study aims to: (1) systematically identify HXK gene family members in cucumber and watermelon via genome-wide screening; (2) conduct comprehensive analyses of their gene structure, multiple sequence alignment, conserved motifs, cis-acting elements, and collinearity to infer evolutionary relationships; (3) analyze tissue-specific expression patterns and stress-responsive profiles of HXK genes using large-scale transcriptome sequencing data (7 datasets for cucumber, 14 datasets for watermelon); and (4) validate candidate HXK genes via qRT-PCR to screen for key stress-resistant members. Ultimately, this research will provide a theoretical foundation for elucidating the roles of HXK genes in cucumber and watermelon growth, development, and stress responses, and offer candidate resistance genes to facilitate the genetic improvement of stress tolerance in these two crops.

Materials and methods

Identification of HXK genes in cucumber and watermelon.

Arabidopsis thaliana HXK gene family information was referenced from previous studies [6]. Six A. thaliana HXK protein sequences were downloaded from the TAIR database (https://www.arabidopsis.org/download/list?dir=Proteins%2FTAIR10_protein_lists) [34]. Genome-associated files of cucumber (ChineseLong_v3; http://cucurbitgenomics.org/v2/ftp/genome/cucumber/Chinese_long/v3/) [35] and watermelon (97103_v2.5; http://cucurbitgenomics.org/v2/ftp/genome/watermelon/97103/v2.5/) [36], including coding DNA sequences (CDS), protein sequences, and General Feature Format (GFF) annotation files, were retrieved from the CuGenDBv2 database (http://cucurbitgenomics.org/v2/) [37]. Subsequently, using the six A. thaliana HXK protein sequences as queries, BLASTp searches (E-value ≤ 1e^{− 5}) were performed against the cucumber and watermelon protein databases using the BLASTp plugin in TBtools-Ⅱ (v2.322) software [38]. Initially, 8 cucumber candidate HXK genes (CsaV3_1G039830, CsaV3_1G043010, CsaV3_2G001860, CsaV3_2G001890, CsaV3_2G001900, CsaV3_3G045830, CsaV3_6G039140, CsaV3_6G044550) and 6 watermelon candidate HXK genes (Cla97C01G017270, Cla97C02G049120, Cla97C03G063270, Cla97C03G066410, Cla97C05G099620, Cla97C11G208480) were identified. To validate these HXK candidate genes, Hidden Markov Model (HMM) profiles corresponding to the HXK-specific conserved domains (Hexokinase-1: PF00349; Hexokinase-2: PF03727) were downloaded from the Pfam database (http://pfam.xfam.org/) [39]. HMMER3.0 software was used to scan the cucumber and watermelon protein databases for these domains (E-value ≤ 1e^{− 10}) [40]. Perl scripts were used to extract the protein sequences of all HXK candidates, and these sequences were further validated using three online tools: NCBI-CDD (https://www.ncbi.nlm.nih.gov/Structure/bwrpsb/bwrpsb.cgi) [41], Pfam (https://www.ebi.ac.uk/interpro/search/sequence/) [42], and SMART (https://smart.embl.de/) [43]. Two cucumber candidates (CsaV3_2G001890 and CsaV3_2G001900) lacking the Hexokinase-1 domain were excluded. Finally, 6 HXK gene family members were retained in cucumber (designated as CsHXK) and 6 in watermelon (designated as ClHXK), with each of these genes encoding proteins that contain both the Hexokinase-1 and Hexokinase-2 domains. The CDS and protein sequences of these 6 CsHXK and 6 ClHXK genes are provided in Tables S1 and S2.

Analysis of physicochemical properties and chromosomal localization of HXK proteins in cucumber and watermelon

Physicochemical properties of CsHXK and ClHXK proteins, including the number of amino acids, molecular weight (MW), isoelectric point (pI), instability index, aliphatic index, and grand average of hydropathicity (GRAVY), were analyzed using the Protein Parameter Calc (ProtParam-based) plugin in TBtools-Ⅱ (v2.322). The subcellular localization of CsHXK and ClHXK proteins was predicted using the online tool WoLF PSORT (https://www.genscript.com/wolf-psort.html) [44, 45]. Genomic positions of CsHXK and ClHXK genes were retrieved from the GFF annotation files of cucumber (ChineseLong_v3) and watermelon (97103_v2.5) genomes, respectively, and chromosomal localization maps of these genes were then generated using the Gene Location Visualize (Advanced) plugin in TBtools-Ⅱ (v2.322).

Phylogenetic analysis and amino acid sequence alignment of HXK proteins in cucumber and watermelon

Based on previous studies of the HXK gene families in A. thaliana, rice, maize, and apple [6, 7, 10, 11], 6 HXK protein sequences from A. thaliana, 10 from rice, 9 from maize, and 9 from apple were downloaded, respectively. These sequences were used for phylogenetic tree construction together with CsHXK and ClHXK proteins. Multiple sequence alignment of all HXK protein sequences in the dataset was performed using the ClustalW algorithm in MEGA-X software (v10.2.6) [46]. A neighbor-joining (NJ) phylogenetic tree was then constructed with the following parameters: Poisson model for amino acid substitution, pairwise deletion of gap-containing sites, and 1000 bootstrap replicates to evaluate node confidence. The generated phylogenetic tree was visualized and optimized using the online tool tvBOT (https://www.chiplot.online/tvbot.html) [47]. For detailed characterization of CsHXK and ClHXK proteins, their amino acid sequences were individually aligned using MEGA-X, and the alignment results were visualized with GENEDOC software (v2.7.000) [48]. Homologous regions in CsHXK and ClHXK sequences were identified by comparison with the HXK2 protein from Saccharomyces cerevisiae [49], which is a well-characterized hexokinase used as a reference. Key functional regions, including phosphate 1, phosphate 2, connect 1, connect 2, substrate recognition domain, and adenosine phosphate-binding domain, were labeled using Adobe Illustrator 2020 software (v24.0.1) [50] to clarify the conservation of functional domains in cucumber and watermelon HXK proteins.

Gene structure, and conserved motif analyses of HXK genes in cucumber and watermelon

To analyze the gene structures of CsHXK and ClHXK genes, the positions of exons, introns, and untranslated regions (UTRs) were retrieved from the GFF annotation files of the cucumber (ChineseLong_v3) and watermelon (97103_v2.5) genomes, respectively. Conserved motifs in CsHXK and ClHXK proteins were identified using the MEME Suite website (http://meme-suite.org/) [51], with the following parameter settings: maximum number of motifs = 10, minimum motif width = 6, and maximum motif width = 100. The exon/intron structures of CsHXK and ClHXK genes, as well as the conserved motifs of their encoded CsHXK and ClHXK proteins, were visualized using the Gene Structure View plugin in TBtools -Ⅱ (v2.322) software.

Analysis of cis-acting elements in the promoter regions of HXK genes in cucumber and watermelon

For each CsHXK and ClHXK gene, a 1500 bp promoter region upstream of the translation initiation site was extracted from the respective cucumber (ChineseLong_v3) and watermelon (97103_v2.5) genome files. These promoter sequences were then submitted to the PlantCARE database (http://bioinformatics.psb.ugent.be/webtools/plantcare/html) [52] for the identification of potential cis-acting elements. A total of 34 cis-acting elements were systematically examined and classified into three functional categories: abiotic and biotic stress-responsive cis-acting elements (ARE, DRE core, LTR, MBS, MYB, MYC, STRE, TC-rich repeats, W box, WUN-motif), phytohormone-responsive cis-acting elements (ABRE, as-1, CGTCA-motif, ERE, P-box, TCA-element, TGACG-motif, TGA-element), and plant growth and development-related cis-acting elements (A-box, AE-box, Box 4, CAT-box, circadian, GA-motif, GATA-motif, G-box, GT1-motif, I-box, MRE, ATCT-motif, AuxRR-core, chs-CMA1a, Sp1, TCT-motif). For data visualization and quantification, all identified cis-acting elements in the promoter regions of CsHXK and ClHXK genes were further categorized by their functional classes, quantified, and visualized using TBtools-Ⅱ (v2.322) software.

Collinearity analysis of HXK family genes in cucumber and watermelon

Intraspecific collinearity of HXK family genes in cucumber and watermelon was analyzed separately using the Multiple Collinearity Scan toolkit MCScanX (https://github.com/wyp1125/MCScanX) [53], and Circos software was used to visualize the intraspecific collinearity results. For interspecific collinearity analysis, CsHXK and ClHXK genes were compared with HXK genes from three groups of reference species: the dicot model plant A. thaliana, the monocot model plant rice, and other Cucurbitaceae crops (e.g., melon (Cucumis melo), pumpkin (Cucurbita maxima)). This interspecific collinearity analysis was performed using the One Step MCscanX plugin in TBtools-Ⅱ (v2.322) software. To ensure accurate visualization of collinearity results, the GFF3/GTF Gene Position (Info.) Parse plugin in TBtools-Ⅱ (v2.322) was used to precisely locate the genomic positions of all target HXK genes (including CsHXK, ClHXK, and HXK genes from reference species). Finally, the Advanced Circos plugin in TBtools-Ⅱ (v2.322) was employed to integrate collinearity data and genomic position information, and visualize the comprehensive collinearity results.

Reanalysis of transcriptome data of cucumber and watermelon

Transcriptome sequencing data of cucumber and watermelon were downloaded from the NCBI SRA database (https://www.ncbi.nlm.nih.gov/sra), respectively. First, the downloaded SRA format files were converted to FASTQ format using fasterq-dump (v2.11.0) (https://www.ebi.ac.uk/interpro/). FastQC (v0.11.9) was then used to generate quality assessment reports for the FASTQ files, enabling evaluation of base quality, sequence duplication rate, and adapter contamination [54]. Low-quality sequences (e.g., reads with Phred quality score < 20, adapter sequences, and short reads < 50 bp) were filtered out using Trimmomatic (v0.39) to obtain high-quality clean reads [55]. The filtered FASTQ files were aligned to their respective species’ reference genomes (cucumber: ChineseLong_v3; watermelon: 97103_v2.5) using STAR (v2.7.11b), generating alignment results in SAM format [56]. The SAM files were further processed with SAMtools (v1.18) to convert them into sorted BAM format (sorted by genomic coordinate), which facilitates subsequent expression quantification [57]. StringTie (v2.2.1) was employed to quantify the expression levels of transcripts based on the sorted BAM files, generating a gene count matrix that reflects the read count of each gene across samples [58]. Subsequently, the gene count matrix was imported into DESeq2 (v1.40.2) for normalization and identification of differentially expressed genes (DEGs) [59, 60], The screening criteria were set as |log₂ fold change| ≥ 1 and adjusted P-value (padj) < 0.05 [61].

Analysis of expression patterns of HXK genes in cucumber and watermelon across different tissues and under various abiotic and biotic stresses

To analyze the expression patterns of CsHXK and ClHXK genes across different tissues and under stress conditions, large-scale transcriptome sequencing datasets (7 sets for cucumber, 14 sets for watermelon) were downloaded from the NCBI SRA database. These datasets cover two categories: diverse tissues and both abiotic and biotic stress treatments, with detailed information provided in Table 1. The downloaded transcriptome data were reanalyzed following the aforementioned bioinformatics pipeline to obtain reliable gene expression values of CsHXK and ClHXK genes. The differentially expressed CsHXK and ClHXK genes were specifically analyzed in each stress treatment. Subsequently, the HeatMap plugin in TBtools-Ⅱ (v2.322) software was used to generate expression heatmaps, which visually exhibit the expression levels of CsHXK and ClHXK genes across different tissues and under various abiotic and biotic stresses.

Table 1.

Transcriptome datasets of cucumber and watermelon for analyzing the expression patterns of CsHXK and ClHXK genes in diverse tissues and under abiotic and biotic stresses

Species	Project	No.	Experiment	Sampled tissue	Accession number	Reference
Cucumber (Cucumis sativus)	Tissue-specific expression	1	Different tissues	Root, stem, male flower, female flower, ovary, ovary unfertilized, ovary fertilized, leaf, tendril, tendril base	PRJNA80169	[62]
		2	Root development	Root	PRJNA271595	[63]
	Abiotic stresses	3	Salt	Root	PRJNA437579	[64]
		4	Waterlogging	Root	PRJNA678740	[65]
	Biotic stresses	5	Downy mildew	Leaf	PRJNA285071	[66]
		6	Powdery mildew	Leaf	PRJNA321023	[67]
		7	Angular leaf spot	Leaf	PRJNA704621	[68]
Watermelon (Citrullus lanatus)	Tissue-specific expression	1	Different tissues	Root, stem, male flower, female flower, fruit, leaf, tendril	PRJNA1031825	[69]
		2	Fruit development	Fruit flesh	PRJNA682019	[70]
	Abiotic stresses	3	High temperature	Ovule	PRJNA504354	-
		4	Cold	Leaf	PRJNA328189	[71]
		5	Salt	Leaf	PRJNA609260	[72]
		6	Low light	Fruit	PRJNA602124	[73]
		7	Nitrogen treatment	Leaf, root	PRJNA422970	[74]
		8	Drought	Root	PRJNA326331	[75]
		9	Osmotic stress	Leaf	PRJNA770012	[76]
	Biotic stresses	10	Fusarium wilt-1	Root	PRJNA641525	[77]
		11	Fusarium wilt-2	Root	PRJNA783543	[78]
		12	Fusarium wilt-3	Root	PRJNA973274	-
		13	Cucumber green mottle mosaic virus	Leaf	PRJNA534308	-
		14	Squash vein yellowing virus	Hypocotyl and true leaves	PRJNA1086032	[79]

GO enrichment analysis of HXK genes in cucumber and watermelon

To clarify the functional characteristics of HXK genes in cucumber and watermelon, the entire set of genome-wide protein sequences of these two crops were first submitted to the eggNOG-mapper database (https://eggnog-mapper.embl.de/) for functional annotation [80]. The raw annotation results were then imported into TBtools-Ⅱ, where the eggNOG-mapper Helper plugin was used to convert the full annotation dataset into a format compatible with GO enrichment analysis. Afterwards, when performing GO enrichment analysis using the GO Enrichment plugin of TBtools-Ⅱ, we focused on the CsHXK and ClHXK gene subsets by selecting their corresponding entries from the converted full protein annotation results, and conducted the GO enrichment analysis for each gene family respectively. With an adjusted P-value (padj) < 0.05 as the criterion for screening significant enrichment, the top 30 significantly enriched GO terms were selected for each gene family after sorting by padj values in ascending order. Finally, the online tool tvBOT (https://www.chiplot.online/tvbot.html) was used to visualize the enrichment results of CsHXK and ClHXK genes separately.

Plant materials and stress treatments

To investigate the expression patterns of CsHXK and ClHXK genes under abiotic stress, cucumber cultivar ‘Jinyan 4’ and watermelon inbred line ‘W-22-13’ were used as experimental materials. Seeds were surface-sterilized in 55 °C water for 10 min, imbibed in distilled water for 8 h, and germinated in a constant-temperature incubator at 28 °C. After germination, seedlings were transplanted into 2 × 3-cell seedling trays (individual cell size: 5 cm × 5.5 cm) pre-filled with a standard growth substrate (peat: vermiculite: perlite = 3:1:1). All seedling trays were placed in shallow catchment trays (to maintain humidity and catch leached solution) and then transferred to controlled-environment chambers. The chamber conditions were set as follows: 14 h light/10 h dark photoperiod, 27 °C (day)/25°C (night) temperature, 25,000 lx light intensity, and 60% relative humidity. When seedlings reached the two-true-leaf stage, uniformly developed plants were selected for stress treatments with 24 seedlings per group. Cucumber was subjected to two stress types: salt stress (irrigated with 500 mmol·L⁻¹ NaCl solution) and drought stress (irrigated with 20% PEG6000 solution). Watermelon was treated with four stress types, including high temperature (45 °C), low temperature (4 °C), salt stress (500 mmol·L⁻¹ NaCl solution), and drought stress (20% PEG6000 solution). Specifically, for temperature stress, plants with their seedling containers were transferred to climate chambers set to the target temperatures, maintaining a 14 h light/10 h dark photoperiod with day/night temperatures of 45 °C/45°C for high temperature and 4 °C/4°C for low temperature [71, 81]. For salt and drought stresses, 10 mL of the corresponding stress solution was applied once to the root zone of each seedling to ensure full contact with the root system. Samples were collected at four time points: 0 h (stress initiation, used as control), 6 h, 12 h, and 24 h after stress application. Only root tissues were collected from cucumber, while root and leaf tissues were collected separately from watermelon. At each time point, tissues from 2 seedlings were pooled to form one biological replicate, with three such replicates prepared for each tissue type. All collected samples were immediately flash-frozen in liquid nitrogen and stored at -80 °C for subsequent RNA extraction and gene expression analysis.

Total RNA extraction and qRT-PCR analysis

Total RNA was extracted from the collected cucumber and watermelon tissues using the FastPure Universal Plant Total RNA Isolation Kit (Vazyme Biotech Co., Ltd, Nanjing, China). The concentration of extracted RNA was determined with a NanDrop 2000c spectrophotometer (Thermo Scientific, Waltham, MA, USA), and RNA integrity was verified via 1% agarose gel electrophoresis. Subsequently, approximately 1000 ng of total RNA per sample was reverse-transcribed into complementary DNA (cDNA) using the HiScript^® III RT SuperMix for qPCR (+ gDNA wiper) kit (Vazyme). Quantitative real-time PCR (qRT-PCR) reactions were performed on a ViiA7 real-time fluorescent quantitative PCR system (Applied Biosystems, Waltham, MA, USA) using ChamQ Universal SYBR qPCR Master Mix (Vazyme) to detect the relative expression levels of CsHXK and ClHXK genes under abiotic stresses. Cucumber CsUBI [82] and watermelon β-Actin [83] were used as reference genes for normalization, respectively, and relative gene expression levels were calculated using the 2^⁻ΔΔCT method [84]. Gene-specific primers for qRT-PCR were designed with Primer Premier 6 software, and their specificity was verified via BLASTn in the CuGenDBv2 database (Table S3). Additionally, melting curve analysis was performed for all qRT-PCR reactions to confirm the absence of non-specific amplification. Statistical analysis of gene expression data was conducted using one-way analysis of variance (ANOVA) combined with Tukey’s multiple comparison test. Significance levels were defined as follows: p > 0.05 (ns), p < 0.05 (*), p < 0.01(**), p < 0.001 (***), and p < 0.0001 (****). All statistical analyses were performed using GraphPad Prism 10.1.2 software.

Results

Basic information and physicochemical properties of CsHXK and ClHXK genes and their encoded proteins

Genome-wide identification revealed six HXK genes in both cucumber (designated CsHXK1-CsHXK6) and watermelon (designated ClHXK1-ClHXK6). The CDS lengths of CsHXK genes ranged from 1467 bp to 1524 bp, encoding 488–507 amino acids (aa), while those of ClHXK genes spanned 1467 bp to 2031 bp, encoding 488–676 aa, with ClHXK4 producing the longest peptide (676 aa). The molecular weights of CsHXK proteins fell between 52.75 kD and 55.09 kD, whereas those of ClHXK proteins ranged from 53.22 kD to 73.48 kD, with the ClHXK4-encoded protein exhibiting the highest molecular weight (73.48 kD). The pI values of CsHXK proteins varied from 5.77 to 7.15, mostly clustering around 6.00. For ClHXK proteins, pI values ranged from 5.34 to 6.36, also centering near 6.00, with ClHXK5 showing the lowest pI (5.34) and ClHXK3 the highest (6.36). The instability indices of CsHXK proteins lay within 24.02–45.59, including 4 stable proteins (CsHXK1, CsHXK3, CsHXK4, CsHXK6) and 2 unstable ones (CsHXK2, CsHXK5). For ClHXK proteins, instability indices ranged from 27.22 to 46.55, comprising 2 stable proteins (ClHXK4, ClHXK6) and 4 unstable ones (ClHXK1, ClHXK2, ClHXK3, ClHXK5). The aliphatic indices of CsHXK proteins spanned 89.72–99.76, while those of ClHXK proteins covered 88.74–102.37. The GRAVY values for CsHXK proteins ranged from − 0.142 to 0.024, with 4 hydrophilic proteins (CsHXK1, CsHXK3, CsHXK4, CsHXK6) and 2 hydrophobic ones (CsHXK2, CsHXK5). For ClHXK proteins, GRAVY values fell between − 0.121 and 0.159, also including 4 hydrophilic proteins (ClHXK1, ClHXK4, ClHXK5, ClHXK6) and 2 hydrophobic ones (ClHXK2, ClHXK3). All CsHXK proteins were predicted to localize to chloroplasts. Among ClHXK proteins, all were targeted to chloroplasts except ClHXK3, which was predicted to localize to the plasma membrane (Table 2).

Table 2.

Basic information and physicochemical properties of CsHXK and ClHXK proteins

Species	Gene name	Locus name	CDS size (bp)	Number of amino acids (aa)	Molecular weight (kD)	pI	Instability index	Aliphatic index	Grand average of hydropathicity	Prediction of subcellular location
Cucumber (Cucumis Sativus)	CsHXK1	CsaV3_1G039830	1497	498	54.23	6.00	34.78	92.03	-0.07	Chloroplast
	CsHXK2	CsaV3_1G043010	1524	507	55.09	6.07	45.59	98.44	0.024	Chloroplast
	CsHXK3	CsaV3_2G001860	1497	498	53.95	6.34	24.02	95.96	-0.017	Chloroplast
	CsHXK4	CsaV3_3G045830	1467	488	52.75	5.91	39.63	93.52	-0.007	Chloroplast
	CsHXK5	CsaV3_6G039140	1509	502	54.63	7.15	41.49	99.76	0.022	Chloroplast
	CsHXK6	CsaV3_6G044550	1503	500	54.76	5.77	36.65	89.72	-0.142	Chloroplast
Watermelon (Citrullus lanatus)	ClHXK1	Cla97C01G017270	1668	555	61.21	5.87	41.29	91.69	-0.121	Chloroplast
	ClHXK2	Cla97C02G049120	1566	521	56.57	5.99	42.13	101.38	0.06	Chloroplast
	ClHXK3	Cla97C03G063270	1725	574	62.91	6.36	46.55	102.37	0.159	Plasma membrane
	ClHXK4	Cla97C03G066410	2031	676	73.48	6.09	32.06	88.74	-0.001	Chloroplast
	ClHXK5	Cla97C05G099620	1467	488	53.22	5.34	41.29	90.51	-0.072	Chloroplast
	ClHXK6	Cla97C11G208480	1497	498	54.09	6.30	27.22	94.8	-0.045	Chloroplast

Chromosomal localization of CsHXK and ClHXK genes

Based on the chromosomal localization analysis of HXK family genes in cucumber and watermelon, the genomic distribution maps of CsHXK and ClHXK genes on chromosomes were constructed. CsHXK genes showed an uneven distribution, localized on chromosomes 1 (CsHXK1, CsHXK2), 2 (CsHXK3), 3 (CsHXK4), and 6 (CsHXK5, CsHXK6). Notably, two pairs of segmentally duplicated genes were identified among these genes, namely CsHXK1/CsHXK3 and CsHXK2/CsHXK5 (Fig. 1A). ClHXK genes also exhibited an uneven distribution pattern, mainly localized on chromosomes 1 (ClHXK1), 2 (ClHXK2), 3 (ClHXK3, ClHXK4), 5 (ClHXK5), and 11 (ClHXK6). Similarly, two pairs of segmentally duplicated genes were identified in ClHXK genes, specifically ClHXK2/ClHXK3 and ClHXK4/ClHXK6 (Fig. 1B).

Fig. 1.

Chromosomal localization and distribution of CsHXK and ClHXK genes in cucumber and watermelon. A Chromosomal localization and distribution of CsHXK genes in cucumber. B Chromosomal localization and distribution of ClHXK genes in watermelon. The blue color gradients along chromosomes indicate gene density (higher density corresponds to darker blue). Within both CsHXK and ClHXK gene families, genes marked in green and red each correspond to a pair of segmentally duplicated genes

Phylogenetic and multiple sequence alignment analyses of CsHXK and ClHXK proteins

To investigate the phylogenetic relationships of HXK family genes among cucumber, watermelon, and other species, a phylogenetic tree was constructed using HXK proteins from cucumber, watermelon, along with 6 HXK proteins from A. thaliana, 10 from rice, 9 from maize, and 9 from apple. These HXK proteins were divided into five subfamilies, with members of the cucumber and watermelon HXK families distributed across subfamilies Ⅰ–Ⅳ. CsHXK4, ClHXK5, and Type A HXK proteins containing chloroplast transit peptides (AtHXK3 and OsHXK4) clustered in subfamily Ⅳ. CsHXK proteins (CsHXK1, CsHXK2, CsHXK3, CsHXK5, CsHXK6) and ClHXK proteins (ClHXK1, ClHXK2, ClHXK3, ClHXK4, ClHXK6) were distributed in subfamilies Ⅰ, Ⅱ, and Ⅲ. These three subfamilies all contained Type B HXK proteins, including AtHXK1, AtHXK2, AtHKL1, AtHKL2, AtHKL3 (from A. thaliana), and OsHXK2, OsHXK3, OsHXK5, OsHXK6, OsHXK9, OsHXK10 (from rice). Thus, the CsHXK and ClHXK proteins in subfamilies Ⅰ–Ⅲ were identified as Type B HXK proteins, sharing a common hydrophobic membrane anchor domain. OsHXK1, OsHXK7, and OsHXK8 were classified as Type C HXK proteins, which lack both membrane anchor domains and chloroplast transit peptides and cluster into subfamily Ⅴ, where no cucumber or watermelon HXK proteins were found (Fig. 2A).

Fig. 2.

Phylogenetic tree of HXK proteins across multiple plant species and multiple sequence alignment of CsHXK and ClHXK proteins in cucumber and watermelon. A Phylogenetic tree of HXK proteins from Arabidopsis, rice, maize, apple, cucumber and watermelon. Different colored symbols represent proteins from distinct species (see the top-right legend for symbol-species correspondence). Bootstrap values are indicated at branch nodes to show the reliability of the phylogenetic tree. B Multiple sequence alignment of CsHXK and ClHXK proteins. Black lines demarcate functionally distinct domains (e.g., phosphate 1, substrate recognition, connector 1, etc., as labeled in the figure). Color gradients represent amino acid conservation levels: blue for 100% conservation, orange for 80% conservation, and green for 60% conservation

Sequence alignment of CsHXK and ClHXK proteins revealed conserved structural features based on homology with HXK2 from Saccharomyces cerevisiae. All CsHXK and ClHXK proteins contained the characteristic functional domains: phosphate 1, phosphate 2, connect 1, connect 2, substrate recognition domain, and adenosine phosphate-binding region. A conserved low-complexity region was identified at the N-terminus of the adenosine phosphate-binding region across all CsHXK and ClHXK proteins. Notably, CsHXK2, CsHXK5, ClHXK2, and ClHXK3 proteins exhibited a distinct 9-amino acid indel mutation within the adenosine phosphate-binding region (Fig. 2B). Phylogenetic and structural analyses classified these CsHXK and ClHXK proteins into two functional types: Type A HXK (CsHXK4, ClHXK5) and Type B HXK (CsHXK1–CsHXK3, CsHXK5–CsHXK6, ClHXK1–ClHXK4, ClHXK6).

Analyses of gene structure and conserved motifs of CsHXK and ClHXK genes

Gene structure analysis revealed that the number of exons in CsHXK and ClHXK genes ranged from 9 to 11, with intron numbers ranging from 8 to 10. Specifically, ClHXK3 consisted of 11 exons and 10 introns, ClHXK1 contained 10 exons and 9 introns, while all other CsHXK and ClHXK genes were composed of 9 exons and 8 introns. Conserved motif analysis identified 10 conserved motifs (designated as Motif 1–10; Table S4) in CsHXK and ClHXK proteins, among which Motifs 3–4, 7, and 9 contained the HXK-1 conserved domain, Motifs 1–2 and 5–6 harbored the HXK-2 conserved domain, and Motifs 8 and 10 lacked HXK conserved domains. Notably, CsHXK4 and ClHXK5 each lacked one Motif 2, CsHXK6 lacked one Motif 4 but contained an additional Motif 10, while all other HXK proteins contained Motifs 1–10 (Fig. 3). Collectively, these results indicated that the structures of CsHXK and ClHXK genes were highly conserved.

Fig. 3.

Gene structures of CsHXK and ClHXK genes and conserved motifs in their encoded proteins. The left panel displays the gene structures of CsHXK and ClHXK genes, where blue regions represent untranslated regions (UTRs), orange regions represent exons, and black lines represent introns. The right panel shows the conserved motifs in the proteins encoded by CsHXK and ClHXK genes; each colored box corresponds to a specific motif (Motif 1 to Motif 10, with colors as indicated in the embedded legend)

Analysis of cis-acting elements in the promoter regions of CsHXK and ClHXK genes

A total of 34 distinct cis-acting elements were identified in the 1500 bp upstream regions of CsHXK and ClHXK genes. These cis-acting elements were further classified into three categories: abiotic and biotic stress-responsive elements, phytohormone-responsive elements, and plant growth and development-related elements. The enrichment levels of different cis-acting elements across various genes were visualized using a heatmap, where deeper colors indicated higher enrichment levels (Fig. 4). Among the three classes, plant growth and development-related cis-acting elements were the most abundant. Within this class, Box 4 exhibited the highest enrichment level across all CsHXK and ClHXK genes. Notably, ClHXK1 and ClHXK3 showed high enrichment of cis-acting elements involved in light signal transduction (specifically Box 4 and G-box). This observation suggested that ClHXK1 and ClHXK3 may play crucial roles in regulating plant growth and development processes mediated by light signals. The second most abundant class was abiotic and biotic stress-responsive cis-acting elements. Among these elements, the MYC cis-acting element had the highest enrichment level in CsHXK and ClHXK genes. Most genes, including CsHXK2 to CsHXK5 and ClHXK1 to ClHXK5, exhibited the strongest enrichment of MYC. This cis-acting element is known to be associated with responses to salt, drought, and cold stress. This result suggested that these CsHXK and ClHXK genes may be involved in responding to multiple abiotic stress conditions. Phytohormone-responsive cis-acting elements were the least abundant among the three classes. Within this class, the ethylene-responsive element (ERE) showed the highest enrichment level across CsHXK and ClHXK genes. Several genes, such as CsHXK2 to CsHXK3, CsHXK6, ClHXK1, and ClHXK3, displayed high enrichment of cis-acting elements involved in abscisic acid (ABA) regulation (ABRE and ERE). This finding indicated that these specific CsHXK and ClHXK genes may participate in the ABA signaling pathway to modulate hormone-mediated physiological processes.

Fig. 4.

Abundance and functional classification of cis-acting elements in CsHXK and ClHXK gene promoters. The color gradient in each grid cell represents the number of a specific cis-acting element in the promoter of CsHXK or ClHXK genes (darker blue indicates higher abundance, as indicated by the color scale on the right)

Collinearity analysis of CsHXK and ClHXK genes

Two pairs of segmentally duplicated gene pairs were identified in both the CsHXK and ClHXK gene families. Specifically, the two pairs in CsHXK gene family were CsHXK1/CsHXK3 and CsHXK2/CsHXK5, while those in ClHXK gene family were ClHXK2/ClHXK3 and ClHXK4/ClHXK6 (Fig. 5A, B). Notably, no tandemly duplicated gene pairs were found in either family, which indicated that the segmental duplication mechanism played a crucial role in the evolutionary expansion of CsHXK and ClHXK genes.

Fig. 5.

Collinearity analysis of CsHXK and ClHXK genes in cucumber and watermelon. A Intraspecific collinearity of CsHXK genes in cucumber. Red lines connect collinear CsHXK gene pairs within cucumber chromosomes. B Intraspecific collinearity of ClHXK genes in watermelon. Red lines connect collinear ClHXK gene pairs within watermelon chromosomes. C Interspecific collinearity of HXK genes between cucumber and two model species (Arabidopsis thaliana, Oryza sativa). D Interspecific collinearityof HXK genes between cucumber and two Cucurbitaceae species (Cucumis melo, Cucumis moschata). E Interspecific collinearity of HXK genes between watermelon and two model species (Arabidopsis thaliana, Oryza sativa). F Interspecific collinearity of HXK genes between watermelon and two Cucurbitaceae species (Cucumis sativus, Cucumis melo). Red lines connect collinear gene pairs across species

To further analyze the evolutionary relationships of CsHXK and ClHXK genes across species, systematic collinearity analyses were performed. For CsHXK genes, collinearity was analyzed with HXK genes in A. thaliana, rice, melon, and pumpkin. Results showed 5 orthologous HXK gene pairs with both A. thaliana and rice, 9 with melon, and the highest number (15 pairs) with pumpkin (Fig. 5C, D; Table S5). For ClHXK genes, collinearity was analyzed with HXK genes in A. thaliana, rice, cucumber, and melon, which revealed 5 orthologous HXK gene pairs with A. thaliana, 2 with rice, and 9 pairs each with cucumber and melon (Fig. 5E, F; Table S6). These results collectively highlighted the conservation and differentiation of CsHXK and ClHXK genes across species during evolution.

Expression pattern analysis of CsHXK and ClHXK genes in different tissues and developmental stages

Among CsHXK and ClHXK genes, expression was detected in all tested samples (FPKM > 0). For CsHXK genes, the expression levels of CsHXK1 in roots and flowers were significantly higher than those in other tissues, while CsHXK3 showed a tissue-specific expression pattern in male and female flowers. The expression levels of CsHXK4 in stems and tendrils were significantly higher than those in other tissues. The expression levels of the remaining CsHXK genes in various tissues were relatively low (Fig. 6A). In different regions of the root (meristematic zone, elongation zone, differentiation zone), the expression levels of CsHXK1 and CsHXK3 were relatively high across multiple zones, suggesting that these two genes might be involved in basic physiological processes of root growth. CsHXK5 exhibited a relatively high expression level only in the root differentiation zone (Fig. 6B), which is a region where root hairs form and nutrient absorption begins. This implied that CsHXK5 might play a specialized role in root nutrient uptake or root hair development, which are key functions of the root differentiation zone.

Fig. 6.

Expression heatmap of CsHXK and ClHXK genes in diverse tissues and developmental stages. A Tissue-specific expression heatmap of CsHXK genes across diverse cucumber organs. B Expression heatmap of CsHXK genes in different cucumber root zones. C Tissue-specific expression heatmap of ClHXK genes across diverse watermelon organs. D Expression heatmap of ClHXK genes during watermelon fruit development. DAP: days after pollination. Original FPKM values are shown in the heatmap boxes

For ClHXK genes, the expression levels of ClHXK1 in female flowers were slightly higher than those in other tissues, while ClHXK4 and ClHXK6 showed high expression patterns in all tissues. Between these two genes, the expression of ClHXK4 in roots was particularly prominent, and ClHXK6 showed a relatively high expression level in tendrils. The expression levels of the remaining ClHXK genes in various tissues were relatively low (Fig. 6C). During fruit development, the expressions of ClHXK2 and ClHXK5 were relatively high at the early stage (11 DAP) but decreased significantly at the late development stages (20 DAP, 30 DAP, 40 DAP) (Fig. 6D). This indicated that these two genes might play important roles in the early stage of fruit development, while their roles weakened in the later stage. In contrast, ClHXK4 and ClHXK6 maintained high expression levels in all stages of fruit development, which indicated that they might continue to play important roles throughout the entire process of fruit development. These results indicated that the expressions of CsHXK and ClHXK genes were tissue-specific and developmental stage-specific, providing important clues for understanding the functions of these genes in plant growth and development.

Expression patterns of CsHXK and ClHXK genes under abiotic stresses via transcriptome reanalysis

To investigate the expression patterns of CsHXK and ClHXK genes under abiotic stresses, this study analyzed the expression of CsHXK genes under salt and waterlogging stresses using two sets of cucumber transcriptome data, and examined the expression profiles of ClHXK genes under multiple stress conditions (high temperature, low temperature, salt, low light, high nitrogen, drought, and osmotic stress) with 7 sets of watermelon transcriptome data.

In cucumber, under salt stress, compared with the control, the expression of CsHXK1 and CsHXK2 was significantly up-regulated, while that of CsHXK4 was significantly down-regulated (Fig. 7A). Under waterlogging stress, CsHXK1 and CsHXK2 were both significantly up-regulated relative to the control, including the 7-day waterlogging treatment and the cyclic treatment (7 days of waterlogging, followed by 14 days of recovery, then another 7 days of waterlogging). Meanwhile, CsHXK4 and CsHXK5 were significantly down-regulated in these two waterlogging treatments (Fig. 7B).

Fig. 7.

Expression heatmaps of CsHXK and ClHXK genes under abiotic stresses. A Expression patterns of CsHXK genes under salt stress. CT: control treatment; Salt: salt stress. B Expression patterns of CsHXK genes under waterlogging stress. S: sensitive plants; T: tolerant plants; Ctrl: untreated plants cultivated under optimal conditions; 1xH: non-primed plants waterlogged for 7 days only once; Rec: plants after 7 days of waterlogging and 14 days of recovery; 2xH: primed plants first waterlogged for 7 days, followed by 14 days of recovery, then waterlogged again. C Expression patterns of ClHXK genes under high temperature stress. CT: control treatment; HT-4 h, HT-8 h, HT-12 h and HT-24 h: high temperature treatment for 4, 8, 12 and 24 h. D Expression patterns of ClHXK genes under cold stress. CT: control treatment; Cold: cold stress. E Expression patterns of ClHXK genes under salt stress. CT: control treatment; Salt: salt stress. F Expression patterns of ClHXK genes under low light stress. CT: control treatment; LL: low light treatment; 0DAP, 3DAP, 9DAP and 15DAP: 0, 3, 9 and 15 days after pollination. G Expression patterns of ClHXK genes under nitrogen treatment. LLN: leaves under low nitrogen treatment; LHN: leaves under high nitrogen treatment; RLN: roots under low nitrogen treatment; RHN: roots under high nitrogen treatment. H Expression patterns of ClHXK genes under drought stress. D-0 h, D-1 h, D-6 h, D-12 h and D-24 h: drought treatment for 0, 1, 6, 12 and 24 h. I Expression patterns of ClHXK genes under osmotic stress. CT: control treatment; OS-2 h and OS-4 h: osmotic stress for 2 and 4 h. Original FPKM values are shown in heatmap boxes; differentially expressed genes are high-lighted in red (up-regulation) and green (down-regulation) with log₂(fold-change) values

In watermelon, under high temperature stress, the expression of ClHXK2 was significantly and continuously up-regulated at all treatment stages (4 h, 8 h, 12 h, 24 h) compared with the control (Fig. 7C). Under cold stress, ClHXK5 expression was significantly down-regulated relative to the control (Fig. 7D). Under salt stress, compared with the control, ClHXK2 and ClHXK3 expression was significantly up-regulated, while ClHXK6 expression was significantly down-regulated (Fig. 7E). Under low light stress, ClHXK2 was significantly down-regulated at 3 days of treatment compared with the control, with no significant changes observed at other stages (Fig. 7F). Under high nitrogen stress, ClHXK2 was significantly up-regulated in roots (Fig. 7G). Under drought stress, compared with the control: ClHXK3 was significantly down-regulated only at 6 h; ClHXK4 was significantly up-regulated at 6 h, 12 h, and 24 h; ClHXK5 was significantly down-regulated at 1 h, 6 h, and 12 h; and ClHXK6 was significantly up-regulated at 12 h and 24 h (Fig. 7H). Under osmotic stress, compared with the control, ClHXK2 expression was significantly down-regulated at both 2 h and 4 h of treatment, while ClHXK6 expression was significantly up-regulated at the same two time points (Fig. 7I).

Expression patterns of CsHXK and ClHXK genes under biotic stresses via transcriptome reanalysis

To investigate the expression characteristics of CsHXK and ClHXK genes under biotic stresses, this study analyzed the expression patterns of CsHXK genes under downy mildew, powdery mildew, and angular leaf spot stresses using 3 sets of cucumber transcriptome data. Additionally, the expression patterns of ClHXK genes were examined under stresses including Fusarium wilt, Cucumber green mottle mosaic virus (CGMMV), and squash vein yellowing virus (SqVYV) with 5 sets of watermelon transcriptome data (Fig. 8).

Fig. 8.

Expression heatmaps of CsHXK and ClHXK genes under biotic stresses. A Expression patterns of CsHXK genes under downy mildew stress. S: susceptible plants; R: resistant plants; CT: control treatment; 1 dpi, 2 dpi, 3 dpi, 4 dpi, and 6 dpi: 1, 2, 3, 4, and 6 days post-inoculation. B Expression patterns of CsHXK genes under powdery mildew stress. S: susceptible plants; R: resistant plants; CT: control treatment; 48 hpi: 48 h post-inoculation. C Expression patterns of CsHXK genes under angular leaf spot stress. Gy14: resistant plants; B10: susceptible plants; CT: control treatment; 1 dpi and 3 dpi: 1 and 3 days post-inoculation. D Expression patterns of ClHXK genes under Fusarium wilt-1 stress. R: watermelon-oilseed rape rotation cropping; C: continuous watermelon monocropping. E Expression patterns of ClHXK genes under Fusarium wilt-2 stress. F0, F3, F5 and F8: 0, 3, 5 and 8 days post inoculation with Fusarium oxysporum f. sp. niveum. F Expression patterns of ClHXK genes under Fusarium wilt-3 stress. S-CT: susceptible cultivar under non-inoculated control conditions; S-F: susceptible cultivar inoculated with Fusarium oxysporum f. sp. niveum; R-CT: resistant cultivar under non-inoculated control conditions; R-F: resistant cultivar inoculated with Fusarium oxysporum f. sp. niveum. G Expression patterns of ClHXK genes under Cucumber green mottle mosaic virus stress. CT: control treatment; 48hpi and 25dpi: 48 h and 25 days post-inoculation. H Expression patterns of ClHXK genes under squash vein yellowing virus stress. S: susceptible plants; R: resistant plants; 0dpi, 5dpi, 10dpi and 15dpi: 0, 5, 10 and 15 days post-inoculation. Original FPKM values are shown in heatmap boxes; differentially expressed genes are high-lighted in red (up-regulation) and green (down-regulation) with log₂(fold-change) values

In cucumber, under downy mildew stress, CsHXK1 was significantly up-regulated in susceptible cultivars at 2, 3, 4, and 6 days post-inoculation (dpi), and in resistant cultivars at 3 and 6 dpi. In contrast, CsHXK2 was significantly down-regulated in susceptible cultivars at 2 dpi and in resistant cultivars at 6 dpi (Fig. 8A). Under powdery mildew stress, both CsHXK1 and CsHXK5 were significantly up-regulated in both susceptible and resistant cultivars at 48 h post-inoculation (hpi) (Fig. 8B). Under angular leaf spot stress, CsHXK1 was significantly up-regulated in both resistant and susceptible cultivars at 1 and 3 dpi (Fig. 8C).

In watermelon, under Fusarium wilt stress, ClHXK5 was significantly down-regulated and ClHXK6 was significantly up-regulated in continuous watermelon monocropping compared to watermelon-oilseed rape rotation (Fig. 8D). ClHXK1 was significantly down-regulated at 8 days post-inoculation with Fusarium oxysporum f. sp. niveum (Fig. 8E). Compared with the control, ClHXK6 was significantly up-regulated only in susceptible cultivars post-inoculation with Fusarium oxysporum f. sp. niveum, with no differential expression in resistant cultivars (Fig. 8F). Under CGMMV stress, ClHXK4 was significantly up-regulated at 48 hpi but significantly down-regulated at 25 dpi (Fig. 8G). Under SqVYV stress, ClHXK6 was significantly up-regulated only in susceptible cultivars at 5, 10, and 15 dpi, with no differential expression in resistant cultivars. The other ClHXK genes showed no significant differential expression in either susceptible or resistant cultivars post-inoculation (Fig. 8H).

Comprehensive analysis of the expression characteristics of CsHXK and ClHXK genes under abiotic and biotic stresses

Based on the analysis of CsHXK and ClHXK gene expression patterns under abiotic and biotic stresses, differentially expressed CsHXK and ClHXK genes were identified, and a heatmap was generated to visualize these results (Fig. 9). Among CsHXK genes, all except CsHXK3 and CsHXK6 exhibited significant differential expression under various stress conditions (Fig. 9A), while all ClHXK genes displayed significant differential expression under different stress conditions (Fig. 9B).

Fig. 9.

Comprehensive analysis of the expression patterns of CsHXK and ClHXK genes under abiotic and biotic stresses. A Expression patterns of CsHXK genes under abiotic and biotic stresses. B Expression patterns of ClHXK genes under abiotic and biotic stresses. Gray indicates no significant differential expression, red indicates significant up-regulation, green indicates significant down-regulation, and blue indicates significant up-regulation in one time point and significant down-regulation in another time point within the same stress treatment

Some CsHXK and ClHXK genes exhibited stress-type-specific expression patterns. For instance, CsHXK4, ClHXK2, and ClHXK3 showed differential expression only under abiotic stress (e.g., salt, drought, osmotic stress, high temperature), while ClHXK1 was differentially expressed exclusively under biotic stress (Fusarium wilt). The remaining differentially expressed CsHXK and ClHXK genes responded to both abiotic and biotic stresses, with distinct expression patterns across stress types. Notably, CsHXK1 was significantly up-regulated under all tested stress conditions, and ClHXK6 responded to three types of abiotic stresses (salt, drought, osmotic stress) and three types of biotic stresses (Fusarium wilt-1, Fusarium wilt-3, SqVYV). These two genes are thus members of the HXK family with the broadest range of stress response types in this study.

The observed differences in the stress responsiveness of these genes are presumably tightly correlated with the types of cis-acting elements contained in their promoter regions (Fig. 4). Specifically, the promoter sequences of CsHXK4, ClHXK2, and ClHXK3 are enriched in a large number of cis-acting elements associated with abiotic stress responses, such as ARE, DRE core, MYB, MYC, and STRE; these elements are well-recognized as key regulatory components that mediate plant adaptive responses to various abiotic stresses including salt, drought, osmotic, and heat stress. In contrast, for ClHXK1, which exclusively responds to biotic stress, cis-acting elements related to pathogen defense (e.g., TC-rich repeats) were detected in its promoter region, further validating the correlation between gene expression patterns and the repertoire of cis-acting elements. The remaining CsHXK and ClHXK genes that exhibit dual responsiveness to both abiotic and biotic stresses all harbor a combination of abiotic and biotic stress-responsive cis-acting elements within their promoters. Taking CsHXK1 and ClHXK6 as examples, their promoter regions contain not only abiotic stress-responsive elements such as ARE, MBS, MYB, and MYC but also biotic stress-responsive elements including W box, TC-rich repeats, and WUN-motif. This constitutes the molecular basis underlying their broad-spectrum responsiveness to diverse stress types, while the differential expression patterns observed under distinct stress conditions may be attributed to the variation in the copy number and regulatory efficiency of these individual elements.

GO enrichment analysis of CsHXK and ClHXK genes

Through the GO enrichment analysis of CsHXK and ClHXK genes, we screened out the top 30 significantly enriched GO terms and presented them in ascending order of adjusted P-value (padj). The results demonstrated that these two gene families were significantly enriched in two core categories: biological process and molecular function. In the biological process category, the enriched terms of CsHXK and ClHXK genes were highly concentrated in the key metabolic process of intracellular glucose homeostasis (padj < 1e^− 14) (Fig. 10A, B). This result indicates that these two types of genes maintain the stability of cellular osmotic pressure and the balance of energy supply by regulating the uptake, metabolism, and storage of intracellular glucose, thereby enhancing the stress tolerance of plants. Consequently, they are core functional genes in the glucose metabolic network. In the molecular function category, the enrichment characteristics of the two gene families also showed a clear core orientation, mainly focusing on hexokinase activity, mannokinase activity, and carbohydrate kinase activity (padj < 1e^− 14) (Fig. 10A, B). Under abiotic stress conditions, plants rely on these kinases to regulate the phosphorylation process and utilization efficiency of carbohydrates: this process not only provides energy for stress responses but also modulates cellular osmotic potential. These kinase activities are directly involved in the phosphorylation modification of carbohydrates, further confirming that CsHXK and ClHXK genes play a core role in regulating carbohydrate metabolism by catalyzing specific phosphorylation reactions in the carbohydrate metabolic pathway.

Fig. 10.

GO enrichment analysis of CsHXK and ClHXK genes. A GO enrichment analysis of CsHXK genes. B GO enrichment analysis of ClHXK genes. The circle size indicates the number of genes annotated to each GO term, and the color gradient reflects the range of adjusted P-value (padj)

qRT-PCR analysis of temporal expression dynamics of selected CsHXK and ClHXK genes under abiotic stresses

To detect the expression dynamics of CsHXK genes in cucumber roots under salt and drought treatments, two genes (CsHXK1 and CsHXK3) were selected based on prior transcriptome reanalysis. Both genes exhibited relatively high basal expression in roots (Fig. 6A), with CsHXK1 additionally showing significant differential expression under multiple abiotic stresses (Fig. 9A). Their expression dynamics were analyzed via qRT-PCR at four time points (0 h, 6 h, 12 h, and 24 h) post-treatment. Under salt stress, the expression patterns of CsHXK1 and CsHXK3 varied across time points compared to the 0 h control. Both genes reached a statistically significant expression peak at 6 h, with significantly up-regulated expression compared to 0 h; after 6 h, CsHXK1 expression gradually decreased at 12 h and 24 h, whereas CsHXK3 showed an oscillatory response, with expression decreasing at 12 h and increasing again at 24 h (Fig. 11A). Under drought stress, compared to 0 h, both CsHXK1 and CsHXK3 exhibited characteristics of a long-term stress response. Specifically, CsHXK1 showed no significant expression difference at 6 h and 12 h compared to 0 h, but was significantly up-regulated at 24 h, while CsHXK3 began to show significant up-regulation at 12 h and further increased to a significant peak at 24 h (Fig. 11B).

Fig. 11.

Expression dynamics of selected CsHXK and ClHXK genes in response to diverse abiotic stresses. A Relative expression dynamics of CsHXK1 and CsHXK3 in roots under salt stress (500 mmol·L⁻¹ NaCl) at 0 h, 6 h, 12 h, and 24 h. B Relative expression dynamics of CsHXK1 and CsHXK3 in roots under drought stress (20% PEG6000) at 0 h, 6 h, 12 h, and 24 h. C Relative expression dynamics of ClHXK2 and ClHXK6 in leaves and roots under high temperature stress (45 °C) at 0 h, 6 h, 12 h, and 24 h. D Relative expression dynamics of ClHXK2 and ClHXK6 in leaves and roots under low temperature stress (4 °C) at 0 h, 6 h, 12 h, and 24 h. E Relative expression dynamics of ClHXK2 and ClHXK6 in leaves and roots under salt stress (500 mmol·L⁻¹ NaCl) at 0 h, 6 h, 12 h, and 24 h. F Relative expression dynamics of ClHXK2 and ClHXK6 in leaves and roots under drought stress (20% PEG6000) at 0 h, 6 h, 12 h, and 24 h

To further detect the tissue-specific expression dynamics of ClHXK genes in watermelon leaves and roots under high temperature, low temperature, salt, and drought stresses, two genes (ClHXK2 and ClHXK6) were selected based on transcriptome reanalysis. These two genes were chosen for their relatively high basal expression in leaves and roots (Fig. 6C), as well as significant differential expression under multiple abiotic stresses (Fig. 9B). Their expression dynamics were analyzed via qRT-PCR at 0 h, 6 h, 12 h, and 24 h post-treatment. Under high temperature stress, ClHXK2 expression in leaves showed a significant stepwise up-regulation pattern with prolonged treatment time, whereas ClHXK6 expression remained unchanged (no significant difference compared to 0 h), and their expression patterns in roots were consistent with those in leaves (Fig. 11C). Under low temperature stress, compared to 0 h, ClHXK2 expression in leaves was significantly down-regulated at 6 h, recovered at 12 h, but significantly decreased again at 24 h, while ClHXK6 remained significantly down-regulated throughout the entire stress period; in roots, ClHXK2 showed no overall significant difference across time points compared to 0 h, and ClHXK6 was significantly down-regulated only at 24 h (Fig. 11D). Under salt stress, ClHXK2 in leaves was significantly up-regulated at all time points (peaking at 12 h) compared to 0 h, while ClHXK6 showed a significant stepwise down-regulation; in roots, ClHXK2 expression rapidly increased to a significant level at 6 h and then decreased, and ClHXK6 was continuously down-regulated (most significant reduction at 24 h) (Fig. 11E). Under drought stress, ClHXK2 expression in leaves gradually and significantly down-regulated with increasing treatment time compared to 0 h, whereas ClHXK6 expression gradually and significantly up-regulated; in roots, ClHXK2 showed the most significant down-regulation at 24 h, while ClHXK6 exhibited significant up-regulation at all time points, with the highest significant up-regulation at 12 h (Fig. 11F).

Discussion

Hexokinase (HXK), a key glucose signaling protein and core enzyme in carbohydrate metabolism, regulates metabolic flux in higher plants by controlling the rate of glucose entry into glycolysis [12, 26, 85]. Although the HXK gene family has been systematically identified in various plant species, including Arabidopsis thaliana (6 members) [6], rice (10 members) [7], maize (9 members) [10], apple (9 members) [11], cassava (7 members) [86], sugarcane (20 members) [87], and melon (6 members) [33], no systematic identification or functional characterization of the HXK gene family in cucumber and watermelon had been reported prior to this study. This research fills this critical gap by identifying 6 HXK genes each in cucumber and watermelon (designated as CsHXK1 to CsHXK6 and ClHXK1 to ClHXK6, respectively). This number falls within the typical range (3–10 members) reported for most plant species [86] and is consistent with the number of HXK genes in melon, confirming the evolutionary conservation of HXK genes within the Cucurbitaceae family.

Beyond gene number, analyses of gene structure, phylogeny, and collinearity collectively reveal the evolutionary patterns of the HXK gene family in cucumber and watermelon. Gene structure analysis shows that all identified genes, except ClHXK1 and ClHXK3, exhibit a conserved structure consisting of 9 exons and 8 introns. This structure is highly consistent with that of HXK genes in species such as Arabidopsis thaliana, rice, and cassava [6, 7, 9, 86, 88]. Phylogenetic analysis classifies CsHXK and ClHXK proteins into two functional types, Type A and Type B. Among them, CsHXK4 and ClHXK5 are identified as Type A HXK proteins, which contain chloroplast transit peptides; the remaining members are Type B HXK proteins with membrane anchor domains. Notably, no Type C HXK proteins were detected in cucumber and watermelon, which stands in sharp contrast to rice, a species with multiple Type C HXK members. This finding reveals significant differences in the differentiation of this gene subtype between monocotyledonous and dicotyledonous plants [8]. Further exploration of family expansion revealed segmental duplication as the primary driver of CsHXK and ClHXK family expansion. Interspecific collinearity analysis identified 15 HXK homologous gene pairs between cucumber and pumpkin, which is consistent with a shared evolutionary trajectory in Cucurbitaceae. In contrast, only 2 pairs of homologous HXK genes were found between watermelon and rice. This substantial difference further highlights the lineage-specific expansion pattern of the HXK gene family following the divergence of monocotyledonous and dicotyledonous plants.

Additionally, analysis of the 1500 bp promoter region upstream of the transcription start site of CsHXK and ClHXK genes reveals the presence of multiple stress-related cis-acting elements, such as ARE, MYB, MYC, LTR, STRE, and WUN-motif. These elements likely serve as the molecular basis for the involvement of HXK genes in stress response regulation, a finding consistent with studies in other species. For example, the promoter region of CmHXK1 in melon also contains cis-acting elements such as LTR, STRE and WUN-motif, and this gene was strongly induced under salt, low temperature and drought stresses [33]. The promoter regions of BdHXK23 genes in Buchloe dactyloides contain ARE and TC-rich repeats, endowing these genes with strong salt stress tolerance [89].

A key innovation of this study is the systematic characterization of the expression profiles of HXK genes in cucumber and watermelon and their potential roles in stress responses through large-scale transcriptome analysis. Tissue-specific expression analysis shows that CsHXK1 is highly expressed in roots and flowers, CsHXK3 is specifically expressed in flowers, and ClHXK4 and ClHXK6 exhibit broad tissue expression patterns. These expression profiles share significant similarities with those of HXK genes reported in other species, reflecting the functional conservation and divergence of this gene family. For instance, AtHKL3 in A. thaliana is reported to be specifically expressed only in floral organs [6]; OsHXK10 in rice shows specific high expression in pollen [7]; in tomato, SlHXK4 is highly expressed in petals, while SlHXK1 is predominantly expressed in stamens [24, 90]; and in cassava, most MeHXK genes are expressed in multiple organs, including leaves, stems, flowers, and fruits [86]. These comparative analyses further confirm that the HXK gene family exhibits “conserved yet diverse” expression characteristics in plants.

Of particular note, by integrating 5 sets of cucumber transcriptome data, 12 sets of watermelon transcriptome data, and qRT-PCR validation results, we identified CsHXK1 and ClHXK6 as multi-stress regulators for the first time. This unique expression pattern has not been previously reported in studies of HXK genes in cucumber and watermelon, providing a new breakthrough for deciphering the stress resistance mechanisms of Cucurbitaceae crops. Specifically, CsHXK1 is continuously upregulated under five different stresses (salt, waterlogging, downy mildew, powdery mildew, and angular leaf spot), indicating its broad-spectrum regulatory capacity. Similarly, ClHXK6 shows significant responses to three abiotic stresses (salt, drought, osmotic stress) and three biotic stresses (Fusarium wilt-1, Fusarium wilt-3, squash vein yellowing virus). Such extensive responsiveness is rare among known HXK genes, suggesting that these two genes may play core roles in the stress response networks of cucumber and watermelon.

Compared with previously reported HXK genes in other species, CsHXK1 and ClHXK6 identified in this study exhibit broader and more comprehensive stress response profiles. For example, overexpression of ZmHXK7 in maize enhances salt tolerance [21]; overexpression of GmHXK15 in soybean significantly improves the alkali stress tolerance of transgenic plants [91]; MdHXK1 in apple interacts with Na⁺/H⁺ exchangers and phosphorylates the Ser275 residue to enhance salt tolerance [92]; the expression levels of HXK1 and HXK2 in peach were significantly increased after inoculation with Monilinia fructicola, which might enhance disease resistance by regulating sugar metabolism and oxidative stress responses [93]. In contrast, CsHXK1 and ClHXK6 identified in this study can respond to multiple abiotic and biotic stresses simultaneously a groundbreaking feature that greatly expands our understanding of the functional diversity of HXK genes.

This unique functional characteristic is strongly supported by studies on homologous genes in the model plant A. thaliana. AtHXK1 and AtHXK2 in A. thaliana are homologs of CsHXK1 and ClHXK6, and their well-documented stress resistance functions provide important references for deciphering the potential mechanisms of CsHXK1 and ClHXK6. Previous studies have shown that overexpression of AtHXK1 directly regulates stomatal movement: it induces stomatal closure to reduce plant transpiration, ultimately significantly enhancing drought tolerance in A. thaliana, which provides direct evidence for the involvement of HXK genes in abiotic stress responses [94]. In addition to regulating abiotic stress responses, AtHXK1 also plays a key role in biotic stress defense. Studies have found that AtHXK1 can significantly enhance plant defense against Pseudomonas syringae pv. tomato DC3000 infection by positively regulating PAMP-triggered immunity (PTI) and partially engaging in effector-triggered immunity (ETI) through glucose (Glc)-induced effects mediated by AtHXK1-related pathways [95]. Furthermore, overexpression of AtHXK2 (another homologous gene in A. thaliana) also enhances resistance to this pathogen by promoting the accumulation of hydrogen peroxide (H₂O₂) and upregulating defense-related genes [96].

Homologous genes typically retain core functional domains and regulatory pathways during evolution. Combining the significant stress response characteristics of CsHXK1 and ClHXK6 observed in this study with functional studies of their homologs in A. thaliana, we infer that CsHXK1 and ClHXK6 likely inherit similar stress resistance regulatory functions: they may not only participate in abiotic stress responses (e.g., drought tolerance) by regulating stomatal behavior but also mediate broad-spectrum defense against pathogens by modulating immune response pathways and reactive oxygen species metabolism. This potential for synergistic regulation of abiotic and biotic stresses is highly consistent with the broad-spectrum response characteristics observed in transcriptome data, further supporting the core roles of these two genes in the stress response networks of cucumber and watermelon, and laying a foundation for subsequent functional validation and deciphering the stress resistance regulatory networks of cucumber and watermelon. In summary, this study fills the gap in research on the HXK gene family in Cucurbitaceae. The identification of CsHXK1 and ClHXK6, which are key genes with multi-stress response potential, provides a new entry point for deciphering HXK-mediated plant stress response mechanisms and offers valuable resources for stress resistance breeding in Cucurbitaceae crops. Future studies will validate the functions of these genes and explore their regulatory mechanisms using techniques such as gene editing and overexpression, providing theoretical and technical support for the genetic improvement of crop stress resistance.

Conclusion

This study clarified the evolutionary characteristics of the HXK gene family in cucumber and watermelon and identified its core stress-responsive members: 6 HXK genes were found in each crop (CsHXK1–CsHXK6, ClHXK1–ClHXK6) and classified into Type A/B, with family conservation confirmed via structural and motif analyses, and segmental duplication identified as the main expansion driver. Abundant stress-related cis-elements in HXK promoters, combined with transcriptome (5 cucumber/12 watermelon datasets) and qRT-PCR validation, further pinpointed CsHXK1 (5-stress response) and ClHXK6 (6-stress response) as core regulators. These findings fill the HXK research gap in the two crops, providing multi-stress responsive candidates to address the bottleneck of low efficiency in Cucurbitaceae stress-resistant breeding, while future work will use CRISPR-Cas9, Agrobacterium-mediated transformation, and yeast two-hybrid assays to verify gene functions and dissect regulatory networks.

Supplementary Information

Below is the link to the electronic supplementary material.

Authors’ contributions

J.G. performed bioinformatics analyses, conducted qRT-PCR experiments, analyzed data, and drafted the manuscript. Z.W. and R.C. assisted with the experiments. Y.S., C.Y. and L.J. provided guidance on experimental design and data interpretation. K.Z. conducted a comprehensive review of the manuscript and provided revision suggestions. All authors read and approved the final manuscript.

Funding

This work was funded by the Anhui Provincial Outstanding Youth Research Project (2024AH030012), the National Natural Science Foundation of China (32002061), the Key Discipline Construction Fund for Crop Science of Anhui Science and Technology University (No. XK-XJGF001), the Open Research Fund Program of Key Laboratory of Horticultural Crop Germplasm Innovation and Utilization (Co-construction by Ministry and Province) (AHYY2023008), and the Talent Foundation of Anhui Science and Technology University (NXYJ202103).

Data availability

All data generated or analyzed during this study are included in this published article and its supplementary information files.

Declarations

Ethical approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.