1. Bibliographic Information

1.1. Title

Comprehensive characterization and expression analysis of enzymatic antioxidant gene families in passion fruit (Passiflora edulis)

1.2. Authors

Jianxiang Liang, Lin Lu, Wenbin Zhang, Ming Chi, Mengqian Shen, Chang An, Shengzhen Chen, Xiaomei Wang, Ruoyu Liu, Yuan Qin, and Ping Zheng.

The corresponding authors are Yuan Qin (yuanqin@fafu.edu.cn) and Ping Zheng (zhengping13@mails.ucas.ac.cn). Their affiliations are primarily with the College of Horticulture, Fujian Agriculture and Forestry University (FAFU), and other research institutes, indicating expertise in plant genetics, horticulture, and molecular biology.

1.3. Journal/Conference

The paper was published in iScience, a peer-reviewed, open-access scientific journal published by Cell Press. iScience is a reputable, broad-scope journal that covers fundamental and applied research in the life, physical, earth, and health sciences. It is known for publishing methodologically sound and transparent research.

1.4. Publication Year

Published online on October 26, 2023.

1.5. Abstract

The study conducts a comprehensive identification and characterization of enzymatic antioxidant gene families in passion fruit (Passiflora edulis). The authors identified 90 genes belonging to seven families: 11 SODs, 45 APXs, 8 CATs, 7 GPXs, 6 MDHARs, 8 DHARs, and 5 GRs. The phylogenetic analysis revealed that genes with similar subcellular localizations tend to have closer evolutionary relationships. Expression analysis showed that certain antioxidant genes are highly expressed in growing flowers and fruits, suggesting a crucial role in protecting these actively proliferating tissues from oxidative damage. The study also identifies specific candidate genes that may be instrumental in enhancing the resistance of passion fruit to temperature stress (both heat and cold), providing a valuable genetic resource for future breeding programs.

1.6. Original Source Link

The official source link is provided via a DOI (Digital Object Identifier): https://doi.org/10.1016/j.isci.2023.108329. The paper is officially published and available open-access.

2. Executive Summary

2.1. Background & Motivation

Plants, being sessile organisms, are constantly exposed to various environmental stresses, such as extreme temperatures, drought, and high salinity. These stresses lead to the overproduction of Reactive Oxygen Species (ROS), which are highly reactive molecules that can cause significant damage to cells by oxidizing proteins, lipids, and DNA. This phenomenon is known as oxidative stress. To survive, plants have evolved a sophisticated antioxidant defense system. This system includes enzymatic components that detoxify ROS.

Passion fruit (Passiflora edulis) is a commercially valuable tropical fruit crop. However, its growth and productivity are often challenged by environmental stresses, particularly temperature fluctuations. While the antioxidant properties of passion fruit are known, a systematic, genome-wide understanding of the genes that encode its enzymatic antioxidant defense system was lacking. This represents a significant research gap, as identifying and characterizing these genes is a critical first step toward developing more stress-resilient passion fruit cultivars through molecular breeding.

The paper's innovative entry point is to perform the first-ever comprehensive, genome-wide characterization of seven key enzymatic antioxidant gene families in passion fruit. By leveraging the recently available passion fruit genome sequence, the authors aimed to create a foundational genetic resource that details the identity, structure, evolution, and expression patterns of these crucial genes.

2.2. Main Contributions / Findings

The primary contributions and findings of this paper are:

1. Comprehensive Gene Family Identification: The study successfully identified and cataloged 90 members of seven major enzymatic antioxidant gene families in the passion fruit genome:

   • 11 Superoxide Dismutases (PeSODs)
   • 45 Ascorbate Peroxidases (PeAPXs)
   • 8 Catalases (PeCATs)
   • 7 Glutathione Peroxidases (PeGPXs)
   • 6 Monodehydroascorbate Reductases (PeMDHARs)
   • 8 Dehydroascorbate Reductases (PeDHARs)
   • 5 Glutathione Reductases (PeGRs)

2. Systematic Multi-Level Characterization: The authors performed a detailed analysis of these genes, covering their phylogenetic relationships, gene structures (exons/introns), conserved protein motifs, chromosomal locations, gene duplication events, and predicted regulatory elements (transcription factors and microRNAs). A key finding from the phylogenetic analysis is that genes with the same predicted subcellular localization (e.g., chloroplast, mitochondria) tend to cluster together, suggesting functional conservation.

3. Expression Profiling and Functional Implication: Using RNA-sequencing data, the study revealed that many antioxidant genes exhibit tissue-specific expression patterns. Notably, high expression levels were observed in developing flowers and fruits, suggesting their role in protecting these high-energy, rapidly growing tissues from oxidative damage.

4. Identification of Stress-Responsive Candidate Genes: The research identified several antioxidant genes whose expression is significantly altered under heat and cold stress, as well as in response to key plant hormones (ABA, SA, GA, MeJA). These genes are strong candidates for future functional studies and could be targeted in breeding programs to enhance the temperature tolerance of passion fruit.

3. Prerequisite Knowledge & Related Work

3.1. Foundational Concepts

3.1.1. Reactive Oxygen Species (ROS) and Oxidative Stress

Reactive Oxygen Species (ROS) are chemically reactive molecules containing oxygen. Examples include superoxide anion ($O_2^{•-}$), hydrogen peroxide ($H_2O_2$), and hydroxyl radical ($OH^{•}$). In plants, ROS are natural byproducts of metabolic processes like photosynthesis and respiration. At low levels, they act as important signaling molecules in various developmental processes. However, under environmental stress (e.g., heat, cold, drought), ROS production can dramatically increase, overwhelming the plant's detoxification capacity. This leads to oxidative stress, a state where ROS cause widespread damage to cellular components like DNA, proteins, and lipids, ultimately impairing plant growth and survival.

3.1.2. The Enzymatic Antioxidant Defense System

To counteract oxidative stress, plants possess a complex network of enzymatic and non-enzymatic antioxidants. This paper focuses on the key enzymatic components:

• Superoxide Dismutase (SOD): This is the first line of defense. SOD enzymes catalyze the dismutation (a reaction where a substance is both oxidized and reduced) of the highly toxic superoxide anion ($O_2^{•-}$) into oxygen ($O_2$) and the less toxic hydrogen peroxide ($H_2O_2$). $ 2O_2^{•-} + 2H^+ \rightarrow H_2O_2 + O_2 $
• Catalase (CAT) and Ascorbate Peroxidase (APX): These enzymes are responsible for detoxifying the hydrogen peroxide ($H_2O_2$) produced by SOD.
  • CAT directly converts $H_2O_2$ into water ($H_2O$) and oxygen ($O_2$). It is highly efficient but has a lower affinity for $H_2O_2$ compared to APX. $ 2H_2O_2 \rightarrow 2H_2O + O_2 $
  • APX reduces $H_2O_2$ to water using ascorbate (AsA, also known as Vitamin C) as the electron donor. This reaction is a central part of the Ascorbate-Glutathione cycle.

3.1.3. The Ascorbate-Glutathione (AsA-GSH) Cycle

The AsA-GSH cycle is a crucial ROS-scavenging pathway in plants, primarily for detoxifying $H_2O_2$. It involves a series of reactions that recycle the antioxidants ascorbate (AsA) and glutathione (GSH). The enzymes studied in this paper are central to this cycle:

1. Ascorbate Peroxidase (APX): Reduces $H_2O_2$ to water, oxidizing ascorbate (AsA) to monodehydroascorbate (MDHA).
2. Monodehydroascorbate Reductase (MDHAR): Regenerates AsA from MDHA using NAD(P)H as an electron donor.
3. Dehydroascorbate Reductase (DHAR): If MDHA is not quickly reduced, it can disproportionate into AsA and dehydroascorbate (DHA). DHAR reduces DHA back to AsA, using reduced glutathione (GSH) as the electron donor. In this process, GSH is oxidized to oxidized glutathione (GSSG).
4. Glutathione Reductase (GR): Regenerates the reduced form of glutathione (GSH) from its oxidized form (GSSG), using NADPH as an electron donor. This step is critical for maintaining the pool of GSH needed by DHAR and GPX.

• Glutathione Peroxidase (GPX): This enzyme also detoxifies peroxides, including $H_2O_2$, using glutathione (GSH) as a reducing agent.

3.1.4. Regulatory Elements: TFs, miRNAs, and CREs

• Cis-regulatory elements (CREs): These are short sequences of DNA in the promoter region (upstream of a gene's coding sequence) that act as binding sites for proteins to control gene transcription. Examples include light-responsive elements, hormone-responsive elements (e.g., ABRE for abscisic acid), and stress-responsive elements.
• Transcription Factors (TFs): These are proteins that bind to CREs to either activate or repress the transcription of a gene. Families like MYB and bHLH are known to regulate plant development and stress responses.
• microRNAs (miRNAs): These are small, non-coding RNA molecules (around 21-24 nucleotides) that regulate gene expression post-transcriptionally, usually by binding to messenger RNA (mRNA) and causing its degradation or preventing its translation into a protein.

3.2. Previous Works

The authors build upon a large body of research that has characterized these antioxidant gene families in other plant species. Genome-wide studies in model plants like Arabidopsis thaliana and major crops like rice (Oryza sativa) have established the classification, structure, and general functions of SOD, APX, CAT, and other antioxidant genes. These studies serve as a reference for naming conventions and phylogenetic comparisons.

For passion fruit specifically, prior research was more limited. Some studies had investigated the physiological antioxidant responses or used transcriptomics to look at gene expression changes during fruit ripening or under specific light conditions. However, none had performed a systematic, genome-wide identification and multi-faceted characterization of the enzymatic antioxidant gene families. This paper fills that void by providing a complete inventory and foundational analysis.

3.3. Technological Evolution

This study is a product of advancements in genomics and bioinformatics. The availability of a high-quality, chromosome-level reference genome for passion fruit (Ma et al., 2021) was the essential prerequisite. In the past, researchers could only study one or a few genes at a time using laborious methods. Today, with a reference genome, scientists can use bioinformatics tools to perform genome-wide searches for entire gene families using techniques like Hidden Markov Models (HMM) and Basic Local Alignment Search Tool (BLAST).

Furthermore, high-throughput RNA sequencing (RNA-seq) technology allows for the quantification of the expression levels of all genes in a given tissue at a specific time. This enables large-scale expression profiling across different developmental stages and stress conditions, which was not feasible with older methods like Northern blotting.

3.4. Differentiation Analysis

The core innovation of this paper is not the invention of a new method, but rather the first comprehensive application of established genomic and bioinformatic methods to characterize the enzymatic antioxidant system in passion fruit.

Compared to previous work, this study is differentiated by:

• Scope: It is the first to systematically investigate seven major antioxidant gene families in passion fruit at a genome-wide level.
• Depth: It goes beyond simple identification, providing a multi-layered analysis that includes phylogeny, gene structure, conserved motifs, chromosomal organization, gene duplication, regulatory networks (TFs and miRNAs), and protein structure prediction.
• Integration: It integrates genomic data with transcriptomic data (RNA-seq) to link gene identity with expression patterns during development and under stress, leading to the identification of high-priority candidate genes for functional validation.

4. Methodology

The authors employed a systematic bioinformatics pipeline to identify, characterize, and analyze the expression of enzymatic antioxidant gene families in passion fruit.

4.1. Identification of Enzymatic Antioxidant Genes

The first step was to find all the potential members of the seven gene families in the passion fruit genome.

1. Acquiring HMM Profiles: The process started by obtaining Hidden Markov Model (HMM) profiles for the conserved protein domains characteristic of each gene family (e.g., Cu-Zn_superoxide_dismutase domain for CSD genes, APX domain for APX genes). These profiles were downloaded from the Pfam protein family database. An HMM profile is a statistical model that represents the consensus sequence of a protein domain, making it a powerful tool for searching for new family members.
2. HMM Search: The HMM profiles were used to search the entire passion fruit proteome (all proteins encoded by the genome) using the HMMER software. This identified all protein sequences that contained the characteristic domain for each family.
3. BLAST Search (for verification): To ensure no genes were missed, known antioxidant protein sequences from the model plant Arabidopsis thaliana were used as queries to perform a BLASTp (protein-protein BLAST) search against the passion fruit proteome.
4. Validation and Redundancy Removal: The results from both HMM and BLAST searches were combined. Any redundant sequences were removed. The remaining candidate protein sequences were then submitted to the NCBI Conserved Domain Database (CDD) and SMART database to confirm the presence and integrity of the expected antioxidant domains. Only sequences with the correct, complete domains were retained for further analysis.

4.2. Phylogenetic and Physicochemical Analysis

Once the gene families were identified, their evolutionary relationships and basic properties were analyzed.

• Phylogenetic Analysis: To understand the evolutionary history, multiple sequence alignments of the protein sequences from passion fruit and Arabidopsis were performed using MUSCLE. A phylogenetic tree was then constructed using the IQ-TREE software, which employs a Maximum Likelihood method. The $WAG+G$ (Whelan and Goldman + Gamma) model was chosen as the best-fit model for amino acid substitution. The resulting tree was visualized using online tools like Evolview and iTOL. This analysis helps to classify the passion fruit genes into subfamilies based on their relationship to the well-characterized Arabidopsis genes.
• Physicochemical Properties: Basic properties of the predicted proteins, such as molecular weight (MW), isoelectric point (pI), and stability, were calculated using the ProtParam tool on the ExPASy server. Subcellular localization (where the protein functions within the cell, e.g., cytoplasm, chloroplast) was predicted using Cell-PLoc 2.0.

4.3. Gene Structure, Motifs, and Regulatory Elements Analysis

• Gene Structure: The exon-intron organization of each gene was determined by comparing the coding sequences (CDS) with their corresponding genomic DNA sequences. The Gene Structure Display Server (GSDS) was used to visualize these structures. Similar exon-intron patterns among genes often reflect close evolutionary relationships.
• Conserved Motifs: The MEME (Multiple Em for Motif Elicitation) suite was used to identify short, conserved amino acid sequences (motifs) within the protein families. The presence and arrangement of these motifs can provide insights into protein function and evolution.
• Cis-regulatory Elements (CREs): The 2000 bp region upstream of the transcription start site (the promoter region) for each gene was extracted. The PlantCARE database was then used to scan these promoter sequences to identify known CREs related to hormone responses, stress responses, and developmental processes.

4.4. Chromosomal Location and Gene Duplication Analysis

• Chromosomal Mapping: The identified genes were mapped to their physical locations on the nine passion fruit chromosomes using the genome annotation file. The TBtools software was used to create a visual map.
• Gene Duplication Events: The MCScanX toolkit was used to identify gene duplication events, which are a primary driver of gene family expansion. Two main types were considered:
  • Tandem duplication: Two or more homologous genes located close to each other on the same chromosome.
  • Segmental/Whole-Genome Duplication (WGD): Homologous genes found on different chromosomes in syntenic blocks (regions with conserved gene order), often resulting from large-scale duplication events.
• Evolutionary Pressure Analysis (Ka/Ks): For each duplicated gene pair, the rates of non-synonymous substitutions (Ka, changes that alter the amino acid sequence) and synonymous substitutions (Ks, changes that do not alter the amino acid) were calculated using KaKs_Calculator 3.0. The ratio Ka/Ks indicates the type of selective pressure acting on the genes:

  • $Ka/Ks < 1$: Purifying (negative) selection, suggesting the gene's function is conserved.

  • $Ka/Ks = 1$: Neutral selection.

  • $Ka/Ks > 1$: Positive (Darwinian) selection, suggesting the gene is evolving new functions.

    The time of divergence (T) for duplicated pairs was estimated using the formula: $ T = \frac{K_s}{2 \times \lambda} $ where $\lambda$ is the neutral substitution rate, typically assumed to be around $1.5 \times 10^{-8}$ substitutions per site per year for dicot plants.

4.5. Protein Structure and Regulatory Network Analysis

• Protein Structure: Secondary structures (α-helices, β-sheets) were predicted using the SOPMA tool. Three-dimensional (tertiary) protein structures were modeled using SWISS-MODEL and the AlphaFold2 database. The final structures were visualized using PyMOL.
• Transcription Factor (TF) Network: The PlantRegMap database was used to predict which TFs might regulate the antioxidant genes by identifying TF binding sites in their promoter regions. The resulting interaction network was visualized using Cytoscape.
• miRNA Target Prediction: The psRNATarget server was used to predict which miRNAs might target the antioxidant gene transcripts for post-transcriptional regulation.

4.6. Expression Profiling and Validation

• RNA-seq Analysis: Publicly available RNA-seq datasets for passion fruit were used. These datasets included samples from different tissues (leaf, petiole, bud), floral and fruit developmental stages, and plants under heat ($30^\circ\mathrm{C}$) and cold ($20^\circ\mathrm{C}$) stress. The expression levels were quantified as Transcripts Per Million (TPM). A heatmap was generated based on $log2(TPM + 1)$ values to visualize expression patterns.

• qRT-PCR Validation: To experimentally validate the RNA-seq findings and investigate responses to more extreme conditions, quantitative Real-Time PCR (qRT-PCR) was performed. Passion fruit seedlings were subjected to heat ($45^\circ\mathrm{C}$), cold ($8^\circ\mathrm{C}$), and various hormone treatments (GA, SA, ABA, MeJA). RNA was extracted from leaves, and the expression of selected candidate genes was measured. The relative expression level was calculated using the $2^{-\Delta\Delta C_T}$ method.

  The $2^{-\Delta\Delta C_T}$ method calculates the fold change in gene expression relative to a control sample and normalized to a reference (housekeeping) gene. $ \Delta C_T = C_{T(\text{target gene})} - C_{T(\text{reference gene})} $ $ \Delta\Delta C_T = \Delta C_{T(\text{treated sample})} - \Delta C_{T(\text{control sample})} $ $ \text{Fold Change} = 2^{-\Delta\Delta C_T} $ The passion fruit EF1a gene was used as the reference gene.

5. Experimental Setup

5.1. Datasets

• Genomic Data: The chromosome-level reference genome of passion fruit (Passiflora edulis) was obtained from the National Genome Data Center (NGDC) under accession number GWHAZTM00000000. This served as the basis for all gene identification and analysis.
• Transcriptomic Data (RNA-seq): Publicly available RNA-seq data (NGDC accession CNP0002747) were used for expression profiling. This dataset included:
  • Developmental Series: Samples from various floral and fruit development stages (e.g., flower bud, full-bloom flower, young fruit, ripening fruit).
  • Tissue Series: Samples from different tissues including leaves, petioles, bracts, and buds under mild heat ($30^\circ\mathrm{C}$) and cold ($20^\circ\mathrm{C}$) stress conditions over a 24-hour period.
• Plant Materials for qRT-PCR: Passion fruit seedlings were grown in an intelligent greenhouse. For stress treatments:
  • Cold stress: Plants were kept at $8^\circ\mathrm{C}$.
  • Heat stress: Plants were kept at $45^\circ\mathrm{C}$.
  • Hormone treatments: Plants were sprayed with solutions of gibberellin (GA, 100 µM), salicylic acid (SA, 100 µM), abscisic acid (ABA, 100 µM), and methyl jasmonate (MeJA, 100 µM). Leaf samples were collected at 24h and 48h post-treatment.

5.2. Evaluation Metrics

• Gene Expression Level (TPM): Transcripts Per Million is a measure of gene expression that normalizes for both gene length and sequencing depth, allowing for more accurate comparison of expression levels between genes and across samples.
• Relative Gene Expression (Fold Change): Calculated using the $2^{-\Delta\Delta C_T}$ method from qRT-PCR data. This metric quantifies how much the expression of a target gene changes in a treated sample compared to an untreated control.
• Ka/Ks Ratio: A metric used to assess the selective pressure on duplicated genes. The calculation and interpretation are described in the Methodology section (4.4).
• Statistical Significance (p-value): Student's t-test or one-way ANOVA were used to determine if the observed differences in gene expression (from qRT-PCR) were statistically significant. A p-value less than 0.05 ($p < 0.05$) was typically considered significant. The paper uses asterisks to denote levels of significance: * for $p < 0.05$, ** for $p < 0.01$, and *** for $p < 0.001$.

5.3. Baselines

The "baselines" or control groups in this study were the untreated passion fruit plants.

• For the qRT-PCR experiments, the baseline was plants grown under normal conditions (0 hours of treatment, 0 µM hormone concentration). All expression changes under stress or hormone treatments were calculated relative to this control group.
• For the developmental RNA-seq analysis, expression levels at different stages were compared to each other to identify developmental patterns, rather than against a single baseline.

6. Results & Analysis

6.1. Core Results Analysis

6.1.1. Identification and Phylogenetic Analysis of Antioxidant Genes

The study successfully identified 90 enzymatic antioxidant genes belonging to seven families in the passion fruit genome. The APX family was the largest with 45 members, while the GR family was the smallest with 5 members.

The phylogenetic analysis, which included 150 antioxidant proteins from both passion fruit and Arabidopsis, provided key insights. The phylogenetic tree for the SOD family is shown below (Figure 1 from the original paper).

该图像是用于展示超氧化物歧化酶（SOD）及相关酶类的系统发育树，包含多个酶的分类和关系。图中标注了不同基因的名字以及其在不同物种（包括西番莲）的分布情况，帮助理解这些酶在抗氧化反应中的作用。

This tree shows that the SOD genes are classified into three major subfamilies based on their metal cofactor: Cu/Zn-SOD (CSD), Fe-SOD (FSD), and Mn-SOD (MSD). A critical finding across all seven families was that genes with the same predicted subcellular localization tended to cluster together. For example, chloroplast-localized APX genes formed a distinct clade from cytosolic APX genes. This strong correlation suggests that the evolutionary divergence of these genes was closely linked to their functional specialization in different cellular compartments.

6.1.2. Gene Structure, Conserved Motifs, and CREs Analysis

The analysis of gene structure (exon-intron organization) and conserved motifs supported the phylogenetic groupings. As shown in the comprehensive chart (Figure 2 from the paper), genes within the same phylogenetic subclade generally shared similar exon-intron numbers and motif arrangements, while genes in different clades showed significant structural diversity. This structural conservation within subclades further supports their shared ancestry and likely functional similarity.

该图像是一个图表，展示了百香果中多种酶类抗氧化基因家族的系统发生关系。图中包括11条SOD、45条APX、8条CAT、7条GPX、6条MDHAR、8条DHAR和5条GR的进化树及其表达分析，强调了在生长花果过程中的高表达基因对于防御氧化损伤的重要性。

Analysis of the promoter regions revealed a large number of cis-regulatory elements (CREs). The distribution of these CREs is summarized in Figure 3 from the paper.

该图像是图表，展示了酶促抗氧化基因的调控元件（CREs）分布及其与各种生理响应的关系。图中包括了激素响应性、成长与发育及压力响应性分别对应的不同基因数据，并通过维恩图展示了这些基因之间的重叠情况。

The most abundant CREs were light-responsive elements (34.56%), highlighting the close link between the antioxidant system and photosynthesis. Also highly prevalent were CREs related to hormone responsiveness, particularly for MeJA (17.08%) and ABA (12.5%), and stress responsiveness, such as drought-inducibility (MBS element, 10.67%). This indicates that the expression of passion fruit antioxidant genes is likely regulated by a complex interplay of light, hormones, and environmental stress signals.

6.1.3. Chromosomal Location and Collinearity Analysis

The 90 identified genes were distributed across all 9 passion fruit chromosomes. The study found that whole-genome duplication (WGD) or segmental duplication was the primary mechanism for the expansion of these gene families, with 52 duplicated gene pairs identified. Tandem duplication played a smaller role. The Ka/Ks ratios for most of these duplicated pairs were less than 1, indicating that they have been evolving under strong purifying selection, which acts to conserve their original function.

The intraspecies and interspecies synteny analysis (Figure 4) further illustrates these evolutionary relationships.

该图像是图表，展示了在对比基因组学中进行的种内和种间同源性分析。图中展示了不同物种之间的基因位置关系，特别是与百香果（Passiflora edulis）相关的重要基因家族，如SODs、APXs等。

6.1.4. Protein Structure and Regulatory Network Analysis

The predicted 3D structures (Figure 5) of representative proteins from each family revealed conserved structural features essential for their catalytic activity, such as binding sites for cofactors (e.g., $Cu+2$, $Zn+2$) and substrates.

该图像是蛋白质结构示意图，展示了多种酶的三维结构，包括PeMDHAR3、PeDHAR1、PeCSD1、PeAPX1、PeFSD1、PeMSD2、PeGPX2、PeGR4和PeCAT8。图中有不同颜色标识的置信度级别，表明模型结构的稳定性和可靠性。

The regulatory network analysis predicted complex interactions. As shown in Figure 6, a large number of transcription factors, particularly from the ERF, MYB, bHLH, and NAC families, were predicted to regulate the antioxidant genes. These TF families are well-known for their roles in plant development and stress responses.

该图像是图6，展示了酶抗氧化基因的转录因子（TFs）调控网络分析。图中绿色节点表示转录因子，蓝色圆形节点表示酶抗氧化基因，同时包含TFs的词云，字体大小与相应TFs的数量正相关，以及TFs数量的统计结果。

Similarly, the analysis predicted that several miRNAs, such as ped-miR160, ped-miR171, and ped-miR398, target antioxidant genes (Figure 7). For instance, ped-miR398 was predicted to target PeCSD1 and PeCSD2, a regulatory relationship that has been experimentally confirmed in other plants to be crucial for copper homeostasis and oxidative stress tolerance.

该图像是示意图，展示了百香果中各种酶抗氧化基因家族及其与小RNA的相互关系。图A阐述了不同基因之间的网络关系，图B展示了特定基因与其对应的小RNA的靶向互动细节，强调了miRNA和目标基因之间的关系。

6.1.5. Expression Profiles and Functional Enrichment

The RNA-seq data analysis provided significant functional insights. The heatmap in Figure 8 shows the expression of antioxidant genes in different tissues under mild temperature stress.

该图像是图表，展示了在热 (T30, $30^{ ext{°C}}$) 和冷 (T20, $20^{ ext{°C}}$) 应激条件下，各种酶抗氧化基因在叶、茎、苞片和芽中的转录表达水平的热图。数据基于 $ext{log}_2( ext{TPM} + 0.01)$ 处理而成。

Several genes, such as PeCSD3, $PeFSD2/3$, $PeCAT3/4$, and multiple PeAPX genes, were highly expressed across all tissues and developmental stages, suggesting they have fundamental housekeeping roles in managing ROS. Importantly, a subset of genes showed high expression specifically in developing flowers and fruits. This suggests they play a specialized role in protecting these metabolically active and rapidly proliferating tissues from oxidative damage, which is critical for reproductive success and fruit quality.

GO and KEGG enrichment analyses (Figure 9) of the antioxidant genes and their co-expressed partners revealed enrichment in pathways related to photosynthesis, carbon metabolism, and biosynthesis of secondary metabolites. This strongly supports the idea that the antioxidant system is tightly integrated with primary energy metabolism and plant defense mechanisms.

该图像是KEGG通路富集分析的图表，展示了不同酶抗氧化基因家族和共表达基因的富集通路及其相关性。左侧为家族基因的通路分析，右侧为共表达基因的通路分析，气泡大小表示基因数量，颜色代表p值。

6.1.6. Identification and Validation of Stress-Responsive Genes

The study identified several candidate genes for enhancing stress tolerance. The qRT-PCR experiments under more extreme stress conditions ($45^\circ\mathrm{C}$ heat, $8^\circ\mathrm{C}$ cold) validated and extended the RNA-seq findings.

As shown in Figure 10, the expression of genes like PeCSD3, PeCAT1, PeAPX16, PeAPX19, and PeDHAR7 was significantly upregulated by heat stress. In contrast, PeAPX24 was strongly induced by cold stress.

该图像是条形图，展示了不同条件下多个抗氧化酶基因（如PeCS3、PeCAT1、PeDHAR7等）的相对表达水平。结果显示，热和冷处理对这些基因的表达有显著影响，表示不同温度条件下的抗氧化反应变化。

Furthermore, hormone treatments (Figure 11) showed that these genes are part of hormone-mediated stress response pathways. For example, PeAPX16, PeAPX19, and PeCAT1 were induced by ABA and MeJA, two key stress-related hormones. This provides strong evidence that these genes are key players in the response to abiotic stress in passion fruit.

该图像是表达分析图，展示了不同处理下的热带水果百香果（Passiflora edulis）中多种抗氧化酶基因（如PeAPX16、PeAPX19、PeCAT1等）的相对表达水平。这些数据表明不同处理（GA、SA、ABA、MeJA）对基因表达的影响，具有重要的生物学意义。

The findings are summarized in a schematic model (Figure 12), which illustrates how environmental stresses trigger ROS production, and how the enzymatic antioxidant system, regulated by hormones and other factors, works to mitigate the damage and confer stress resistance.

该图像是示意图，展示了外源植物激素在耐逆境应激机制中的作用。图中说明了冷、热等不良环境如何导致活性氧（ROS）的生成，从而引发氧化损伤。外源植物激素如脱落酸（ABA）、生长素（GA）、水杨酸（SA）和美克杰酸（MeJA）被表示为保护因子，通过调节抗氧化防御基因的表达来提高植物的应激耐受性。

7. Conclusion & Reflections

7.1. Conclusion Summary

This study provides the first comprehensive, genome-wide characterization of seven key enzymatic antioxidant gene families in passion fruit. It successfully identified 90 genes and performed a multi-faceted analysis of their phylogeny, gene structure, regulation, and expression. The key conclusions are:

1. The expansion of these gene families in passion fruit was driven primarily by whole-genome or segmental duplications, followed by functional conservation under purifying selection.

2. Phylogenetic relationships are strongly correlated with subcellular localization, suggesting functional specialization within different cellular compartments.

3. A subset of antioxidant genes is highly and specifically expressed in flowers and fruits, indicating a critical role in protecting reproductive and developmental processes from oxidative damage.

4. The study identified and experimentally validated several specific genes (PeCSD3, PeCAT1, $PeAPX16/19/24$, PeDHAR7) that are highly responsive to temperature stress and key stress hormones, making them prime candidates for future genetic improvement of passion fruit.

   This work serves as a fundamental resource for understanding the molecular basis of stress tolerance in passion fruit and provides a valuable toolkit of candidate genes for breeding more resilient cultivars.

7.2. Limitations & Future Work

The authors acknowledge the primary limitation of their study: most of the analyses are based on bioinformatic predictions and transcriptomic data. While the qRT-PCR experiments provide initial experimental validation, the precise biological functions of the identified candidate genes remain to be confirmed.

Future work should focus on:

• Functional Validation: Using techniques like genetic transformation (overexpression or knockout/knockdown via CRISPR-Cas9) to definitively determine the role of the candidate genes (e.g., PeAPX16, PeAPX24) in conferring temperature stress tolerance.
• Protein-level Studies: Investigating the biochemical properties and enzymatic activity of the encoded proteins to confirm their predicted functions.
• Regulatory Network Validation: Experimentally verifying the predicted interactions between TFs, miRNAs, and their target antioxidant genes.

7.3. Personal Insights & Critique

This paper is an excellent example of a foundational genomics study in a non-model but economically important crop. Its strength lies in its comprehensive and systematic approach, integrating multiple layers of bioinformatic analysis to build a holistic picture of the antioxidant gene families.

• Inspiration: The study provides a clear and effective blueprint for how to leverage a newly sequenced genome to gain rapid insights into important gene families. The workflow—from identification and phylogeny to expression analysis and candidate gene validation—can be applied to other gene families and other crops.
• Transferability: The findings have direct implications for applied agricultural science. The identified candidate genes could be used as molecular markers in marker-assisted selection (MAS) or as direct targets for genetic engineering to develop passion fruit varieties with enhanced tolerance to heat and cold, which is increasingly important in the face of climate change.
• Potential Issues/Critique:

  • The study relies on mild stress conditions ($30^\circ\mathrm{C}$/$20^\circ\mathrm{C}$) for the RNA-seq data, which might not fully capture the response to more severe, agriculturally relevant stress. The qRT-PCR at more extreme temperatures ($45^\circ\mathrm{C}$/$8^\circ\mathrm{C}$) helps mitigate this, but a full transcriptomic analysis under these conditions would be more powerful.

  • While the correlation between subcellular localization and phylogeny is a strong finding, the localization itself is based on in silico prediction. Experimental confirmation using techniques like GFP-fusion protein localization would be necessary to solidify this conclusion.

    Overall, this is a rigorous and valuable contribution to passion fruit research. It lays the essential groundwork for future functional genomics studies and provides tangible targets for improving the resilience of this important fruit crop.