Integrated Transcriptomics and Metabolomics Analyses Provide Insights into Heat Resistance in Passion Fruit (P. edulis f. flavicarpa)

Abstract

Passion fruit (Passiflora edulis) is an economically important fruit worldwide. However, heat stress severely threatens its production, particularly in tropical and subtropical regions. To elucidate the molecular and metabolic mechanisms underlying heat tolerance, comparative physiological, transcriptomic, and metabolomic analyses were conducted between two yellow passion fruit cultivars: heat-tolerant ‘Summer Queen’ (F2) and heat-sensitive ‘Qinmi 9’ (QM9). Physiological evaluations demonstrated that QM9 exhibited significantly lower heat tolerance than F2, manifesting as severe leaf wilting, impaired photosynthetic efficiency, and elevated reactive oxygen species (ROS) accumulation. F2 exhibited distinct metabolic and transcriptional adaptations under heat stress, particularly in purine metabolism and flavonoid biosynthesis. Metabolites such as glutamine, xanthine, luteoloside, and trifolin were enriched in F2, alongside the upregulation of genes like adenosine kinase (AK), xanthine dehydrogenase (XDH), guanine deaminase (GDA), and flavonoid 3′-hydroxylase (F3′H). Weighted gene co-expression network analysis (WGCNA) highlighted strong associations between these pathways and transcription factors (e.g., MYB, HSF, WRKY), suggesting their pivotal roles in heat adaptation. Exogenous application of xanthine and trifolin markedly enhanced heat tolerance in passion fruit. Furthermore, knockdown of PeGDA and PeXDH markedly altered the heat tolerance of F2. These findings reveal that elevated metabolites in purine metabolism and flavonoid biosynthesis enhance heat tolerance in passion fruit, offering new insights into the molecular mechanisms of heat tolerance and potential targets for breeding climate-resilient passion fruit varieties.

1. Introduction

Global warming and the intensification of the greenhouse effect pose significant challenges to modern agricultural production. Elevated temperatures directly disrupt critical physiological processes in plants, including photosynthesis, respiration, and nutrient assimilation, ultimately leading to reduced crop yields [1]. Among various abiotic stressors, heat stress has become a predominant threat to global food security. As the frequency and intensity of extreme heat events continue to rise, deciphering the molecular and physiological mechanisms of plant heat tolerance is imperative for developing climate-resilient crops. This concern is particularly acute for tropical and subtropical species, which routinely endure intense heat stress that severely constrains their productivity and economic value [2]. Thus, research on plant heat resistance is not only scientifically significant but also urgently needed to address the increasing impacts of climate change on agriculture [3].

Yellow passion fruit (P. edulis f. flavicarpa), a climacteric fruit widely cultivated in tropical and subtropical regions, is highly valuable for its nutritional, medicinal, and economic value [4]. Renowned for its rich content of vitamins, carotenoids, and polyphenols, it has become a staple in various industries [5]. However, its commercial potential, particularly in major producing regions such as Brazil and China, is increasingly threatened by heat stress, which induces flower abscission and fruit set failure under elevated temperatures [1,6]. While research on passion fruit has predominantly focused on purple varieties, yellow passion fruit has recently gained attention because of its distinct flavor profile, high sweetness ratio, and unique growth characteristics, positioning it as a promising candidate for breeding heat-resistant cultivars [5,7]. At the molecular level, transcription factors (TFs) are key regulators of plant responses to abiotic stress, including heat stress. TFs such as heat stress transcription factors (HSFs), MYB, WRKY, and NAC are well-documented regulators that orchestrate the defense mechanisms of plants against environmental stressors [8]. HSFs, in particular, play a central role in the heat response by modulating the expression of heat shock proteins (HSPs), which safeguard cellular integrity under high-temperature conditions [9,10]. Other TF families, including DREB and bZIP, are involved in mediating stress signaling pathways that contribute to heat tolerance by activating stress-responsive genes [11]. Given the pivotal roles of TFs in stress adaptation, elucidating their regulatory networks in yellow passion fruit is essential for advancing the understanding of heat stress tolerance and facilitating the development of climate-resilient varieties.

Although extensive research has been conducted on heat resistance mechanisms in various plant species, studies on passion fruit remain limited. The integration of transcriptomics and metabolomics provides a comprehensive approach to unravel molecular and metabolic responses to heat stress, enabling the identification of differentially expressed genes and metabolic shifts for a deeper understanding of heat resistance mechanisms [12]. Transcriptomics and metabolomics analyses of two yellow passion fruit varieties with contrasting heat tolerances, Qinmi9 (QM9, conventional varieties) and Summer Queen (F2, heat-tolerant), revealed significant differences in gene expression and metabolic accumulation. This study identified key genes and metabolites linked to heat resistance, highlighting the critical roles of purine metabolism and flavone and flavonol biosynthesis pathways. These findings advance the understanding of heat tolerance in passion fruit and provide valuable insights for breeding climate-resilient varieties.

2. Materials and Methods

2.1. Seedling Responses to Controlled Heat Stress

Seedling heat stress assay under controlled conditions: To determine differences in heat tolerance among passion fruit varieties, seedlings of two yellow passion fruit cultivars (P. edulis f. flavicarpa: QM9 and F2) were collected from experimental fields at the Guangxi Academy of Agricultural Sciences and transplanted into soil-filled pots. These plants were acclimated for 3 days in a growth chamber under controlled conditions (25 °C, 16 h light/8 h dark photoperiod) before heat treatment. To detect differences in heat stress between two varieties, seedlings were transferred to a growth chamber at 40 °C for 48 h. Each treatment group included three biological replicates.

2.2. Field-Grown Plant Responses to Natural High-Temperature Conditions

Field sample collection under natural high-temperature conditions for transcriptomics and metabolomics analyses: To investigate differences in gene expression and metabolite accumulation among passion fruit varieties under naturally high-temperature stress, samples from mature QM9 and F2 plants were collected under naturally high temperature conditions (32–36 °C) in the same open-air experimental field at the Guangxi Academy of Agricultural Sciences. Leaves, stems, and fruits were collected during the ripening stage. All fruits were immediately dissected into pericarp and pulp components for subsequent analyses. This analysis focused on comparing transcriptomic and metabolomic differences between the conventional variety QM9 and the heat-tolerant variety F2 under natural high-temperature conditions, rather than examining responses of the same variety to both normal and heat-stress environments. Each sample type was collected with three biological replicates.

2.3. Physiological Measurements

Electrolyte leakage (EL) was measured as described by Dahro et al. [13], with minor modifications. Leaf discs (0.5 cm diameter) were incubated in 20 mL deionized water at 25 °C for 24 h, and initial conductivity (EC1) was recorded. Samples were then heated at 95 °C for 20 min to obtain the final conductivity (EC2). EL (%) was calculated as EC1/(EC1 + EC2) × 100%. The Fv/Fm ratio was assessed using Imaging WinGege software V2.56zn.

Malondialdehyde (MDA) content was measured using the thiobarbituric acid (TBA) colorimetric method. Fresh passion fruit leaves (0.1 g) were placed in grinding tubes containing stainless steel beads, flash-frozen in liquid nitrogen, and homogenized at 60 Hz for 120 s. The samples were mixed with 1 mL of 5% (w/v) trichloroacetic acid (TCA) and incubated on ice for 1 h. After centrifugation at 8000× g for 10 min at 4 °C, 200 μL of the supernatant was combined with 400 μL of 0.6% TBA solution. The mixture was incubated at 95 °C for 60 min, cooled on ice, and centrifuged at 10,000× g for 10 min at room temperature. Absorbance was measured at 532 nm and 600 nm using a microplate reader, with distilled water as a blank. Absorbance differences were calculated as ΔA = (A₅₃₂ − A₅₃₂ blank) − (A₆₀₀ − A₆₀₀ blank), and MDA content was calculated using the instructions provided with the MDA assay kit (BC0025, Beijing Solarbio Science & Technology Co., Ltd., Beijing, China).

To detect reactive oxygen species (ROS), histochemical staining was performed using 3,3′-diaminobenzidine (DAB, CAS#: 91-95-2) and nitroblue tetrazolium (NBT, CAS#: 298-83-9). Heat-stressed leaves were incubated in DAB or NBT staining solutions at 37 °C in the dark on a shaker at 100 rpm for 24 h. When visible staining appeared, the staining solutions were discarded, and the leaves were decolorized in 65 °C anhydrous ethanol until chlorophyll was completely removed. The decolorized leaves were then flattened and photographed.

2.4. Metabolite Extraction and Profiling

To assess metabolite accumulation differences between QM9 and F2, stem, leaf, pericarp, and pulp samples were obtained from field-grown plants under natural high-temperature stress. Three biological replicates were analyzed for widely targeted metabolites and volatile organic compounds. Metabolite extraction, detection, and quantification were performed by Novogene Co., Ltd., using UPLC-MS/MS and GC-MS platforms. Data were processed with Analyst 1.6.3 software. An Orthogonal Partial Least Squares Discriminant Analysis (OPLS-DA) model was used to distinguish metabolic profiles. Metabolites with VIP ≥ 1 and |log₂ fold change| ≥ 1 were defined as differentially abundant metabolites (DAMs). The identified metabolites were annotated using the KEGG compound database and mapped to metabolic pathways via the KEGG pathway database.

2.5. Transcriptome Sequencing and Profiling

To assess gene expression differences between QM9 and F2, stem, leaf, pericarp, and pulp samples were obtained from field-grown plants under natural high-temperature stress. Total RNA was extracted using the RNAprep Pure Plant Kit, and genomic DNA was removed with RNase-Free DNase I (Takara Bio Inc., Kusatsu, Japan) following the manufacturer’s protocol. RNA integrity was confirmed via a 1.0% EB-stained agarose gel, while quality and concentration were measured by Novogene Co., Ltd. (Beijing, China) using the NanoDrop 2000C and Agilent 2100. For cDNA library construction, 3 µg of RNA per sample was processed with the NEBNext Ultra RNA Library Prep Kit (New England Biolabs, Inc., Ipswich, MA, USA). PCR products were purified using AMPure XP beads (Beckman Coulter, Inc., Brea, CA, USA), and library quality was assessed on the Agilent Bioanalyzer 2100 (Agilent Technologies, Inc., Santa Clara, CA, USA). Sequencing was carried out on Illumina HiSeq 2500 or HiSeq X Ten platforms (Illumina, Inc., San Diego, CA, USA), depending on the experimental setup.

Raw RNA-seq reads were quality-checked using FastQC [14]. Adapters and low-quality reads were removed with Trimmomatic [15]. Clean reads were then mapped to the genome using HISAT2 [16] with default parameters. Transcripts were assembled with StringTie [17], and gene expression was quantified in fragments per kilobase of transcript per million fragments mapped (FPKM). DEGs were identified using DESeq2 [18], with thresholds of |log₂ fold change| ≥ 1 and adjusted p < 0.01. DEGs were subsequently annotated with Gene Ontology (GO) and KEGG pathway terms.

2.6. Mapping Correlation Networks Between DEGs and DAMs

Principal component analysis (PCA) and hierarchical clustering were employed to provide a comprehensive comparison of protein-coding gene and metabolite profiles in the sampled tissues (stem, leaf, pericarp, and pulp) of QM9 and F2. The pheatmap package in R was utilized for this analysis. To ensure comparability across samples, gene and metabolite accumulation data underwent transformation and Z-score normalization, facilitating clearer visualization of dataset variance. Enrichment analysis was conducted using the hypergeometric test to identify statistically significant over-representations of specific genes or metabolites. This was further enhanced by pathway enrichment analysis based on the KEGG database, which elucidated the biological pathways influenced over the time course. Additionally, co-expression networks were constructed to investigate co-expression relationships among protein-coding genes, employing the WGCNA package [19]. Visualization of the gene interaction network was accomplished using Cytoscape software V3.10.1 [20].

2.7. Exogenous Spraying Experiment

In the exogenous spraying experiment, the optimal concentrations of xanthine (10 mM) and trifolin (2 mM) were determined through preliminary experiments. 10 mM xanthine solution was prepared by dissolving 15.2 mg of xanthine in 10 mL of 0.025 M NaOH. 2 mM trifolin solution was prepared by dissolving 10 mg of trifolin in a total volume of 11.15 mL, using a solvent mixture containing 5% (v/v) anhydrous ethanol and 0.1% (v/v) Tween-80.

At room temperature, QM9 and gene-silenced F2 plants were exogenously sprayed with water, a 10 mM xanthine solution, and a 2 mM trifolin solution, respectively, using a spray bottle. Spraying was performed twice: a second spray was conducted 24 h after the first spray, and the subsequent heat stress experiment (42 °C for 24 h) was conducted 24 h later.

2.8. Construction of the Gene Silencing System

The VIGS (Virus-Induced Gene Silencing) system was established to target two key genes involved in purine metabolism: guanine deaminase (GDA) and xanthine dehydrogenase (XDH). Target gene fragments were amplified using the following primers to construct the pTRV2 vectors: GDA-F: 5′-aaggttaccgaattctctagaATCGTGACCATGGACCCCG-3′, GDA-R: 5′-cgtgagctcggtaccggatccTTCGTAAGGCCAGATGCGG-3′; XDH-F: 5′-aaggttaccgaattctctagaGAGTATCTTAGAGACTTGGGTCTGACTG-3′, XDH-R: 5′-cgtgagctcggtaccggatccTCCAGGAGTACAAAAACCACACTG-3′.

Recombinants were screened on LB medium containing kanamycin (50 mg/L), and positive clones were verified by colony PCR and sequencing. Correct recombinant plasmids were introduced into Agrobacterium tumefaciens strain GV3101 (carrying the pSoup-p19 helper plasmid). Transformants were selected on LB medium supplemented with kanamycin (50 mg/L) and rifampicin (25 mg/L).

Agrobacterium cultures carrying pTRV1, pTRV2, pTRV2-GDA, and pTRV2-XDH plasmids were grown in YEP medium. Cells were harvested, resuspended in infiltration buffer (10 mM MES, 10 mM MgCl₂, 200 μM acetosyringone), and adjusted to an OD₆₀₀ of ~1.0, then incubated in the dark for 4 h. Equal volumes of Agrobacterium with pTRV1 were mixed with those containing pTRV2-GDA or pTRV2-XDH and infiltrated into one-month-old F2 passion fruit seedlings as experimental groups. A mixture of pTRV1 and empty pTRV2 was used as a negative control, and uninjected plants were used as blanks, with three biological replicates each. After infiltration, plants were kept in the dark at 25 °C for 1 day, then transferred to 25 °C, 50% humidity, under a 16 h light/8 h dark cycle for 2 days before sampling to evaluate gene silencing.

2.9. RT-qPCR Analysis

To validate the successful silencing of the target genes via VIGS, total RNA was extracted from leaves of F2 representing three experimental groups: (1) VIGS-treated plants (pTRV2-GDA and pTRV2-XDH), (2) vector control plants (empty pTRV2), and (3) untreated wild-type plants (mock control). Total RNA was isolated from these samples using TRIzol (Labled, Beijing, China) reagent according to the manufacturer’s instructions. First-strand cDNA was synthesized using the Lablead First-Strand cDNA Synthesis Mix kit following the provided protocol. The reference gene primers [21] were as follows: Forward 5′-AGCTCTTCTACATCTGCGCT-3′ and Reverse 5′-TTCTTGTGCATCTTCCCCCG-3′. Gene-specific primers were as follows: qGDA-F: 5′-AATAACACGGCAGACGGGAGG-3′, qGDA-R: 5′-TTGACCCAAACTGTATGAGCTGCC-3′; qXDH-F: 5′-GGAGGTATTTGCTTCAGAGTTTGTG-3′, qXDH-R: 5′-CCCACCTGAACTTCCCCTTCT-3′. Quantitative PCR reactions were performed using the Lablead 2× Realab Green PCR Fast Mixture according to the manufacturer’s instructions. Relative gene expression levels were calculated using the 2^−ΔΔCt method.

2.10. Statistical Analysis

All experiments were conducted with a minimum of two independent repetitions, each containing three biological replicates. Data are presented as mean ± standard deviation (SD). Statistical significance was determined using Student’s t-test, with significance thresholds set at * p < 0.05, ** p < 0.01, and *** p < 0.001.

3. Results

3.1. F2 Passion Fruit Exhibits Greater Heat Tolerance Compared with QM9

Two yellow-fruited passion fruit (P. edulis f. flavicarpa) cultivars were utilized: ‘Qinmi 9’ (QM9), identified as the normal (heat-sensitive) control, and ‘Summer Queen’ (F2), characterized as heat-tolerant. Both cultivars were authenticated and provided by the Guangxi Academy of Agricultural Sciences. To evaluate heat tolerance, one-month F2 and QM9 passion fruit seedlings were subjected to heat stress (40 °C for 48 h). While both varieties showed comparable performance under normal conditions, QM9 exhibited severe water-soaking and leaf wilting after stress, whereas F2 maintained better morphological integrity (Figure 1a). Contrary to the phenotypic observations, electrolyte leakage (EL) measurements revealed higher values in F2 than in QM9 after heat stress (Figure 1b). However, malondialdehyde (MDA) content—a more specific indicator of membrane lipid peroxidation—showed significantly greater accumulation in QM9 (Figure 1c). This apparent discrepancy suggests that while F2 experienced enhanced ion efflux under heat stress, QM9 suffered more severe oxidative damage to cellular membranes. Chlorophyll fluorescence measurements further corroborated these findings. QM9 exhibited severely impaired fluorescence parameters and significantly reduced Fv/Fm ratios following heat stress, whereas F2 plants maintained relatively stable photosynthetic efficiency with only minor perturbations (Figure 1d). Histochemical staining for reactive oxygen species (ROS) provided additional evidence of differential stress responses. Nitro blue tetrazolium (NBT) and 3′3-diaminobenzidine (DAB) staining revealed substantially lower accumulation of O₂^•− and H₂O₂ in F2 compared to QM9 after heat treatment (Figure 1e). Collectively, these results demonstrate that F2 possesses superior heat tolerance relative to QM9, as evidenced by (1) better maintenance of cellular membrane integrity, (2) preserved photosynthetic function, and (3) more effective ROS scavenging under heat stress conditions.

3.2. Metabolic Profiling of Heat-Tolerant and Conventional Passion Fruit

Metabolite detection mainly investigates the variation in heat tolerance between the two cultivars under these natural high-temperature conditions. To investigate heat-associated metabolic differences between QM9 and F2, samples were collected under natural high-temperature conditions (32–36 °C) with three biological replicates per group (Figure 2a). For comprehensive metabolic profiling, widely targeted metabolites and volatile organic compounds (VOCs) were studied in pericarp (P), pulp (J), stem (S) and leaf (L) tissues using complementary analytical platforms: liquid chromatography–mass spectrometry (LC-MS) for metabolite characterization and gas chromatography–mass spectrometry (GC-MS) for VOC detection [22,23].

A total of 1236 metabolites were identified and categorized into 16 classes, including amino acids and derivatives (16.42%), organic derivatives (21.44%), flavonoids (12.79%), carbohydrates (9.47%), lipids (9.39%), phenolic acids and derivatives (6.31%), nucleotides and derivatives (6.31%), terpenoids (3.96%), phenylpropanoids and polyketides (3.48%), alkaloids (2.67%), amines (2.27%), vitamins (1.70%), phytohormones (1.29%), acids, alcohols, and polyols (1.21%), benzene and derivatives (0.97%), and polyamines (0.32%) (Figure 2b, Table S1). Principal component analysis (PCA) revealed distinct clustering of biological replicates within each group, confirming the separation between groups and validating the reliability and robustness of the metabolomic data for downstream analysis (Figure 2c). Cluster analysis revealed that there were significant differences in metabolite accumulation between the two varieties and their respective tissues (Figure 2d). To identify heat-resistant metabolic changes, differentially accumulated metabolites (DAMs) were compared in each tissue of QM9 and F2 (Table S2). In the pericarp, a total of 89 differentially accumulated metabolites (DAMs) were identified in F2 compared to QM9, including 48 upregulated and 41 downregulated metabolites (Figure 2e). Similarly, in the pulp, 112 DAMs were detected, of which 62 were upregulated and 50 were downregulated in F2 (Figure 2e). Additionally, stem and leaf tissues demonstrated the most DAMs between QM9 and F2 in the stem and leaf (Figure S1a). Cluster heatmap showed that there were significant differences in metabolite accumulation between the two varieties and their respective tissues (Figure 2f and Figure S1b). These findings highlight the distinct metabolic profiles between the two cultivars, providing preliminary insights into the metabolic mechanisms underlying their differential heat tolerances.

3.3. Specific Differences in Metabolites Between Two Passion Fruit Varieties

In the comparison between QM9_P vs. F2_P and QM9_J vs. F2_J, A total of 22 shared DAMs were identified, with 179 DAMs detected across both comparison groups (Figure 3a). To systematically characterize these metabolic changes in response to heat stress, K-means clustering was applied to the 179 DAMs, which were categorized into nine distinct clusters (Figure 3a–c, Table S3). Notably, metabolites in clusters 2, 3, 4, and 8 exhibited significantly higher accumulation in the pericarp and/or pulp of the heat-tolerant F2 compared to QM9. Similarly, analysis of 371 DAMs in stem and leaf tissues revealed nine distinct clusters, with clusters 1, 3, 6, 7, and 8 showing increased accumulation in F2 (Figure S2, Table S3). These heat-induced metabolites were predominantly composed of organic acids and derivatives, phenolic acids and derivatives, flavonoids—compounds known to play critical roles in plant responses to heat stress. Their upregulation in F2 underscores their role as conserved stress-responsive metabolites, potentially contributing to enhanced thermotolerance.

Further metabolomic analysis revealed significant differences in the accumulation of key metabolite classes between QM9 and the heat-tolerant F2 variety. Strikingly, F2 displayed significantly elevated levels of terpenoids, organic acids, alkaloids, flavonoids, heterocyclic compounds, and nitrogen-containing compounds in both the pericarp and pulp compared to QM9 (Figure 3d,e). This enhanced accumulation suggests an active metabolic reprogramming in F2, potentially contributing to its stress-responsive phenotype. A similar trend was observed in stem and leaf tissues, where F2 accumulated higher amounts of these key metabolite classes (Figures S3 and S4). The consistent upregulation of these compounds across different tissues implies a systemic adaptation mechanism in F2, possibly linked to improved stress resilience. Conversely, the levels of alcohol and polyols, organic oxygen compounds, carbohydrates, and lipids in F2 were lower than those in QM9 (Figure 3d,e, Figures S3 and S4). This differential metabolic profile may reflect a trade-off in resource allocation, where F2 prioritizes the biosynthesis of stress-protective secondary metabolites.

3.4. Transcriptomic Analysis of Heat Stress Responses of Heat-Tolerant and Conventional Passion Fruit

To explore the molecular mechanisms underlying metabolic adaptation to heat stress in QM9 and F2 passion fruit, RNA-Seq analysis was conducted alongside metabolome profiling. High-throughput RNA sequencing via the Illumina HiSeq platform generated an average of 7.07 Gb of high-quality reads for QM9 and 6.56 Gb for F2, with average mapping rates of 93.42% and 89.31%, respectively, to the previously assembled yellow passion fruit genome [24] (Table S4). Consequently, a comprehensive dataset was generated for gene expression analysis (Table S5). The heatmap (Figure 4a) depicts the gene expression profiles of QM9 and F2, demonstrating consistent expression patterns within biological replicates and significant differential expression between the two cultivars. PCA revealed clear separation between the groups, with distinct clustering of biological replicates within each group, further validating the reliability of the transcriptomic data (Figure 4b). Additionally, correlation analysis indicated high reproducibility among biological replicates for each treatment, underscoring the robustness of the dataset (Figure 4c). Collectively, these results indicate that gene expression changes under heat stress are both species- and tissue-specific, reflecting corresponding patterns of metabolite accumulation.

To further investigate the transcriptional responses to heat stress in QM9 and F2 passion fruit varieties, differentially expressed genes (DEGs) were identified (Figure 4d). Compared with those in QM9, 3894 genes in F2 were upregulated and 3465 were downregulated in the pericarp. Similarly, in the pulp, 1809 genes were upregulated, whereas 1113 were downregulated. In the stem, 5535 genes presented increased expression, and 4991 genes presented decreased expression. In the leaf tissue, 3901 genes were upregulated, and 3165 were downregulated. Both upregulated and downregulated DEGs were observed in heatmaps across all tissues—pericarp, pulp, stem, and leaf—indicating substantial and tissue-specific shifts in gene expression profiles in response to heat stress (Figure 4e). These findings collectively demonstrate that heat stress induces widespread and tissue-specific transcriptional reprogramming in passion fruit, with the heat-tolerant variety F2 exhibiting distinct gene expression patterns compared with those of QM9.

Gene Ontology (GO) analysis of DEGs showed significant enrichment in lipid metabolism (GO:0006629), cell wall biogenesis (GO:0042546), and transmembrane transporter activity (GO:0022857) across all tissues. Tissue-specific GO terms were also identified: in the pulp, purine nucleoside transmembrane transporter activity (GO:0015211) and the superoxide metabolic process (GO:0006801) were significantly enriched, whereas ribosome biogenesis (GO:0042254) and fatty acid metabolic processes (GO:0006631) were enriched in the pericarp (Figure S5a, Table S6). Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis highlighted amino acid biosynthesis, carbon metabolism, and glycolysis/gluconeogenesis as common heat-responsive pathways. Tissue-specific pathways included calcium and MAPK signaling in the pulp; fatty acid degradation and unsaturated fatty acid biosynthesis in the pericarp; photosynthesis and TCA cycle in the leaf; and carotenoid biosynthesis and flavonoid degradation in the stem (Figure S5b, Table S7). These functional enrichment analyses demonstrated that heat stress triggers extensive reprogramming of metabolic and signaling pathways in passion fruit, with tissue-specific adaptations in the pulp, pericarp, leaf, and stem. The enrichment of pathways related to lipid metabolism, amino acid biosynthesis, and stress signaling underscores the complex regulatory networks involved in heat tolerance, providing valuable insights into the molecular mechanisms underlying the adaptation of passion fruit to high-temperature conditions.

3.5. Investigating the Heat Tolerance Mechanisms in Passion Fruit

To further elucidate the regulatory mechanisms of heat stress-induced metabolic changes and explore the co-expression network of DEGs, a gene co-expression network was constructed using Weighted Gene Co-expression Network Analysis (WGCNA). Genes with similar expression patterns were grouped into modules, and 15 co-expression modules were identified based on gene expression similarity (Figure 5a, Table S8). Compared to QM9, the highly expressed genes in F2 were predominantly enriched in the cyan, turquoise, and green modules (Figure 5a). The module–trait correlation heatmap revealed that genes in the cyan and turquoise modules were significantly correlated with key metabolites in the purine metabolic pathway (including glutamine and xanthine) and the flavonoid and flavonol biosynthesis pathway (including luteoloside and trifoline) (Figure 5b). This strong correlation highlights the potential roles of these pathways in the heat stress response, prompting further investigation into their regulatory mechanisms.

Integrated analysis of DEGs, DAMs, and WGCNA highlighted purine metabolism and flavonoid biosynthesis as key heat tolerance mechanisms in passion fruit. While these pathways are known in other species, their roles in passion fruit are newly identified, suggesting conserved stress responses. Through WGCNA, genes in the cyan and turquoise modules were linked to purine pathways, supporting their role in heat stress adaptation. Notably, three adenosine kinase (AK) genes (02G04075, 01G01990, and 09G01909), xanthine dehydrogenase (XDH) gene (08G01335) guanine deaminase (GDA) gene (05G01808) were upregulated in the F2 (Figure 5c). Additionally, higher levels of metabolites such as glutamine and xanthine were detected in the pericarp and/or pulp of F2 compared to QM9 (Figure 5c). These findings suggest that F2 exhibits adaptive modifications in purine metabolism in response to heat stress. MYB transcription factors are known to interact with HSFs to regulate the expression of HSPs, which function as molecular chaperones involved in protein assembly, stabilization, and maturation [25,26]. HSPs play critical roles not only in plant development but also in responses to abiotic and biotic stresses [27]. In our recently sequenced passion fruit genome, over 700 transcription factors were identified, including 170 belonging to the MYB family [24]. Regulatory network analysis revealed strong associations between key genes in the purine metabolism pathway and several transcription factor families, such as MYB, HSF, WRKY, bZIP, bHLH, and MADS (Figure 5d). This suggests that purine metabolism-related genes may play a pivotal role in the response to heat stress of passion fruit.

In addition to upregulated genes in the purine metabolism pathway, significant upregulation was observed for two flavonoid 3′-hydroxylase (F3′H) genes (08G01441 and 02G00366) in heat-tolerant F2 varieties, highlighting their potential role in flavonoid and flavonol biosynthesis. Downstream metabolites, including luteoloside and trifolin, were highly accumulated in F2, particularly in the pericarp (Figure 5e). Regulatory network analysis revealed strong associations between the F3′H gene and multiple transcription factor families, such as HSF, MYB, and WRKY (Figure 5f). These findings suggest that F3′H and its associated metabolites play a pivotal role in enhancing heat tolerance in passion fruit.

3.6. Functional Validation via Exogenous Application and Gene Silencing

To investigate the role of downstream metabolites from the purine and flavonoid pathways in heat tolerance of passion fruit, optimal concentrations of xanthine (10 mM) and trifoliate (2 mM) were established through preliminary experiments. Exogenous spraying experiments were subsequently performed on QM9 seedlings using these concentrations (10 mM xanthine and 2 mM trifolin, respectively). Under heat stress conditions (42 °C for 24 h), water-pretreated control seedlings exhibited more severe leaf wilting and lower chlorophyll Fv/Fm ratios compared to those pretreated with xanthine or trifolin (Figure 6a,b). Histochemical staining for ROS further corroborated the differential stress responses between treatments. Notably, NBT and DAB staining demonstrated significantly attenuated accumulation of O₂^•− and H₂O₂ in xanthine- and trifolin-pretreated QM9 seedlings compared to water-treated controls following heat stress (Figure 6c). These findings reveal that xanthine and trifolin, as key downstream metabolites in the purine and flavonoid pathways, may play regulatory roles in enhancing passion fruit thermotolerance.

To further investigate the key genes involved in xanthine biosynthesis, peGDA and peXDH were silenced in the F2 cultivar using virus-induced gene silencing (VIGS) to assess their functional roles in heat tolerance. Compared with the mock and TRV control plants, the TRV-peGDA and TRV-peXDH F2 plants exhibited pronounced water-soaking and leaf wilting following heat stress (Figure 6d). The transcript levels of peGDA and peXDH were significantly reduced in the respective silenced lines (Figure 6e). Additionally, exogenous application experiments were performed on F2 seedlings after VIGS treatment using 10 mM xanthine and 2 mM trifolin, respectively. Under heat stress conditions (42 °C for 24 h), water-treated control seedlings showed more severe leaf wilting compared to those pretreated with xanthine or trifolin (Figure 6f). These results suggest that knockdown of peGDA and peXDH may affect the accumulation of xanthine, leading to decreased heat tolerance, while the heat tolerance of gene-silenced F2 can be reversed by the application of clofoline, further emphasizing the role of xanthines and clofoline in the response of passion fruit to heat stress.

4. Discussion

High-temperature stress, a prevalent abiotic factor, significantly disrupts plant metabolic processes, primarily through protein degradation, consequently inhibiting growth and development [28]. With the persistent escalation of global temperatures, heat stress is anticipated to increasingly constrain plant growth [29]. Passion fruit, a tropical and subtropical crop renowned for its distinctive aroma, occupies a pivotal position in the tropical fruit market. However, throughout its growth cycle, the plant is subjected to various environmental stressors, including heat, cold, drought, and pathogen infections [11]. Although passion fruit is native to tropical regions, it exhibits marked sensitivity to high-temperature stress. With the acceleration of global warming, elevated temperatures have become a critical factor threatening passion fruit production. Studies demonstrate that heat stress significantly inhibits photosynthetic efficiency, reduces antioxidant enzyme activity, and modulates the expression of heat-resistant genes such as heat shock proteins (HSPs) [3]. Considerable interspecific variation exists in thermal responses among cultivars: for instance, the optimal photosynthetic temperature for ‘Summer Queen’ and ‘Ruby Star’ is 30 °C [30], whereas the hybrid cultivar ‘Minami-jujisei’ shows pollen germination optimum at 36 °C [31]. Such germplasm diversity provides valuable genetic resources for breeding thermotolerant varieties. However, current research on passion fruit thermotolerance remains predominantly focused on physiological responses, leaving the molecular mechanisms and key regulatory pathways largely unexplored. Further in-depth investigations are urgently needed to elucidate these fundamental aspects.

This study elucidates the molecular and metabolic mechanisms underlying heat tolerance in passion fruit by integrating comparison of physiological differences, transcriptomic, and metabolomic approaches. The results demonstrate that the heat-tolerant variety F2 exhibited enhanced heat tolerance compared with the heat-sensitive QM9, and F2 displayed unique metabolic and gene expression profiles, particularly in purine metabolism and flavonoid biosynthesis. This metabolic profile suggests that heat-tolerant passion fruit varieties may leverage these compounds as part of their adaptive response to heat stress. The increased accumulation of flavonoids and phenolic acids likely enhances antioxidant capacity, protecting cellular structures from heat-induced oxidative damage. F2 exhibits increased organic acids and nitrogen compounds alongside decreased alcohols, polyols, carbohydrates, and lipids, indicating a metabolic adaptation that enhances osmotic regulation and energy metabolism under thermal stress. This metabolic shift may reflect a strategic reallocation of resources in F2, where energy and carbon are preferentially diverted toward the synthesis of heat-protective metabolites (e.g., flavonoids and terpenoids) rather than storage compounds like carbohydrates and lipids. The reduced accumulation of osmoprotective alcohols and polyols in F2 suggests a shift in osmotic adjustment strategy, potentially indicating greater reliance on alternative osmoprotectants such as proline and other stress-responsive amino acids during heat stress. These metabolic adjustments demonstrate that thermotolerant varieties such as F2 selectively enhance biosynthesis of protective metabolites while suppressing metabolic pathways involved in maintaining non-stress-related functions. These insights not only advance the comprehension of passion fruit’s response to elevated temperatures but also provide potential targets for enhancing heat tolerance in other crops.

Purine metabolism regulates essential cellular processes, including signaling, redox balance, and energy metabolism [32,33]. Purine metabolites play vital roles in cellular functions, including energy production, signal transduction, and redox regulation [34]. Previous studies have demonstrated the importance of purine metabolism in plant responses to heat stress. For instance, heat stress in quinoa enriched the purine metabolic pathway and upregulated related metabolites in both heat-sensitive and heat-tolerant varieties [28]. Similarly, tobacco shoots exhibited significant increases in purine metabolites such as adenine, xanthine, hypoxanthine, and guanine under heat stress [35]. Notably, 3′,5′-Cyclic AMP, a key second messenger, has been identified as a regulator of heat stress responses, influencing protein homeostasis, ion regulation, and the ubiquitin–proteasome system [36]. This study revealed purine metabolism as a critical factor in the heat tolerance of the F2 variety. Purine compounds like glutamine and xanthine were more abundant in F2, and exogenous application of xanthine could improve the heat tolerance of QM9. These metabolites play crucial roles in regulating energy metabolism and maintaining redox balance—key processes for mitigating heat-induced stress. The upregulation of purine metabolism-related genes, such as GDA and XDH, further underscores their role in heat adaptation. Gene silencing experiments further demonstrated that downstream genes in this metabolic pathway can influence heat tolerance in passion fruit. Flavonoid biosynthesis was another pathway significantly enriched in F2. Flavonoids are well known for their significant role in enhancing plant stress resistance, contributing to both individual growth and increased tolerance to abiotic stresses. These compounds provide a theoretical foundation for understanding plant resilience mechanisms [37]. Flavonoids, including luteolin and trifolin, accumulated at higher levels in F2, and exogenous application of trifolin could improve the heat resistance of QM9, although the effect was not as obvious as that of xanthine. As potent antioxidants, flavonoids scavenge reactive oxygen species and alleviate oxidative stress, a major consequence of heat exposure [38]. The elevated levels of these compounds in F2 suggest that flavonoid biosynthesis contributes to heat tolerance by reducing oxidative damage, consistent with findings in other plant species under abiotic stress. Integrated transcriptomic and metabolomic analyses established correlations between key flavonoid pathway genes (e.g., F3′H) and differential flavonoid accumulation in F2, further underscoring this pathway’s role in thermal adaptation.

Numerous transcription factors (TFs) play critical roles in plant adaptation to elevated temperatures [39]. Among these, the WRKY gene family is well known for its involvement in stress responses [40,41], while HSFs regulate stress tolerance by controlling the expression of HSPs [42]. Under thermal stress, plants activate the heat shock response, a conserved mechanism that induces heat-stimulated proteins to enhance thermal tolerance [43]. HSFs, as plant-specific TFs, are central to this process, mediating the activation of HSPs and other protective genes during heat stress [44]. In this study, several TFs associated with plant adaptability were identified, many of which are linked to purine metabolism and flavonoid biosynthesis pathways, including MYB, WRKY, HSF, bZIP, bHLH, and MADS. These findings underscore the importance of TFs connected to key metabolic pathways in regulating heat stress responses, providing new insights into the molecular mechanisms of heat tolerance in passion fruit.

However, a limitation of this study is that the conclusions are drawn from comparisons between only two passion fruit varieties (F2 and QM9). While this approach identified key metabolic and transcriptional differences associated with heat tolerance, the generalizability of these findings may be constrained by the limited genetic diversity represented. Future studies incorporating a broader range of heat-tolerant and heat-sensitive varieties would strengthen the validity of the proposed mechanisms and help distinguish universal adaptive traits from variety-specific responses.

5. Conclusions

This study reveals that purine metabolism and flavonoid biosynthesis are key pathways contributing to heat tolerance in yellow passion fruit, with the heat-tolerant variety F2 exhibiting elevated levels of metabolites like xanthine and flavonoids (e.g., trifolin) and upregulation of associated genes. Exogenous application of these metabolites enhanced heat tolerance, while silencing downstream pathway genes compromised thermotolerance, further validating this mechanism. Transcription factors such as WRKY, HSF, and MYB may further regulate these adaptive responses. These findings provide critical molecular targets for breeding heat-resistant passion fruit varieties, offering a strategic foundation to mitigate the impacts of rising global temperatures on tropical crop production. Future research should validate these genes and metabolites to optimize heat stress resilience.