ABSTRACT

   Soluble sugars, including sucrose, are vital to the overall quality of
   apricot fruits, influencing attributes such as sweetness, flavor, and
   texture. As the primary soluble sugar, sucrose plays a central role in
   determining fruit quality. The synthesis and accumulation of sucrose,
   as well as its interaction with ascorbic acid metabolism, are regulated
   by various enzymes, including sucrose synthase and invertase. While
   several genes related to sucrose metabolism have been identified in
   apricots, the mechanisms by which sucrose and ascorbic acid work
   together to regulate fruit quality remain “is not fully elucidated”. In
   the present investigation, the changes in soluble sugar and ascorbic
   acid levels throughout fruit development were examined, and the
   mechanisms underlying the changes in soluble sugar and ascorbic acid
   levels were examined in ‘Wuyuexian’ apricots (WYX) and ‘Lanzhoudajie’
   apricots (LZ). The data revealed that the sucrose content of ‘WYX’ was
   significantly higher than that of ‘LZ’, whereas the sorbitol levels of
   ‘LZ’ were higher than ‘WYX’. The fructose contents of both varieties
   increased later in growth, and the ascorbic acid content of ‘LZ’ was
   three times higher than that of ‘WYX’ during the early stage of fruit
   growth. These trends became intensified in both varieties after the
   rapid growth stage, with ‘WYX’ having the highest sucrose content.
   Additionally, transcriptomic analysis, K‐means analysis, and weighted
   correlation network analysis (WGCNA) howed the expression levels of
   genes related to sucrose and ascorbic acid metabolism, the synergistic
   effects of gene clusters, and the complex associations among genes.
   Moreover, the key hub genes HSFB1 and FTIP4 were identified, which play
   crucial roles in regulating sucrose and ascorbic acid metabolis.
   However, transcriptome analysis of gene expression only partially
   reflects metabolism regulation, and further in‐depth investigations are
   required to elucidate other regulatory mechanisms. Our research
   revealed a relationship between the synthesis of ascorbic acid and
   soluble sugars in apricots and offers new perspectives on molecular
   breeding and germplasm development in apricots and establishment of
   germplasm resources.

   Keywords: apricot, ascorbic acid content, fruit developmental stage,
   soluble sugar content, transcriptome sequencing, WGCNA
     __________________________________________________________________

   This study investigates the dynamic changes in soluble sugar and
   ascorbic acid contents during apricot fruit development, focusing on
   two varieties, ‘Wuyuexian’ (WYX) and ‘Lanzhoudajie’ (LZ). By combining
   transcriptomic analysis and metabolic profiling, the research uncovers
   the regulatory mechanisms behind sugar and ascorbic acid synthesis,
   identifying key genes such as HSFB1 and FTIP4 that play pivotal roles
   in metabolic processes. The findings reveal significant differences in
   sugar accumulation and ascorbic acid content between the varieties,
   with WYX showing higher sucrose levels and LZ exhibiting higher
   ascorbic acid during early growth stages. This work provides valuable
   insights into apricot fruit quality enhancement and lays the foundation
   for future breeding programs focused on improving nutritional content
   and fruit quality.

   graphic file with name FSN3-13-e70899-g006.jpg
     __________________________________________________________________

Abbreviations

   LZ
          ‘Lanzhoudajie’ apricots

   WYX
          ‘Wuyuexian’ apricots

1. Background

   The apricot (Prunus armeniaca L.) is an important species belonging to
   the Rosaceae family and the “genus Prunus”. It is cultivated throughout
   the Mediterranean region, South Africa, South America, and North
   America and is a popular seasonal fruit that people enjoy eating
   worldwide. It is the third most extensively grown drupe fruit tree in
   the world (Alajil et al. [34]2021; Fan et al. [35]^2017). Apricots were
   originally domesticated in China, and their cultivation can be traced
   back 2600 years (Karatas [36]2022). In addition to their wonderful
   flavor, appealing aroma, and vibrant color, apricot fruits are also
   packed with minerals, including ascorbic acid, β‐carotene, and
   phenolics, which are valued for their nutritional and economic benefits
   for their nutritional and economic value (Zhou et al. [37]2020; Jiang
   et al. [38]^2019; Crouzet et al. [39]^1990).

   As one of the essential components of plant metabolism, sucrose serves
   as a key component for n plant growth and development (Ackermann
   et al. [40]1992; Kroger et al. [41]^2006; Pangborn [42]^2006; Ren
   et al. [43]^2018; Lingle and Dunlap [44]^1987). Sucrose synthase (SUS),
   which regulates the synthesis of UDP‐glucose, is primarily responsible
   for controlling the synthesis of sucrose in plants (Koch [45]2004). The
   level of the rate‐limiting enzyme in sucrose synthesis and the enzyme
   that breaks down sucrose, sucrose phosphate synthase (SPS), is
   positively correlated with the accumulation of sucrose (Wei
   et al. [46]2020; Su et al. [47]^2021). Invertases convert sucrose to
   fructose and glucose. Furthermore, fructokinase and hexokinase can
   phosphorylate glucose and fructose to produce glucose‐6‐phosphate and
   fructose‐6‐phosphate (Rolland et al. [48]2006).

   Recent studies, high‐throughput sequencing and biotechnological
   techniques have revealed numerous enzyme‐encoding genes linked to
   sucrose synthesis and metabolism in apricot fruits (Aslam
   et al. [49]2019; Zhang et al. [50]^2019). During the total sugar and
   sucrose accumulation stages of apricot fruit development, the
   expression of sucrose synthase genes PaSUS1, PaSUS3, ParSuSy5,
   ParSuSy6, and ParSuSy7 was markedly upregulated in response to high
   sucrose concentrations. It was suggested that these genes are involved
   in the synthesis of proteins by sucrose synthase (Iqbal
   et al. [51]2020). Lombardo et al. (Lombardo et al. [52]2011) reported
   that after fruit discoloration, the transcript levels of sucrose
   synthase and sucrose phosphate synthase dramatically increased, but the
   transcript levels of sorbitol dehydrogenase did not significantly
   change over time. Wang et al. (Wang et al. [53]2019) reported a
   correlation between the sugar content of ripe apple fruits and the
   expression of the apple hexose transporter protein MdHT2.2, whose
   primary function is to uptake and move glucose and fructose from the
   cell wall space into cells, which helps maintain the ability to cells
   to release hexose.

   Ascorbic acid (AsA) is abundant in apricot fruit and is one of the most
   widely distributed water‐soluble antioxidants found in both plants and
   animals. As an enzyme cofactor, scavenger of free radicals, and
   electron donor and acceptor in electron transport across membranes or
   chloroplasts, AsA is essential for life. It plays important roles in
   various plant processes, including growth and development, hormone
   signaling, cell division and proliferation, and reactive oxygen species
   scavenging. As a result, it protects DNA, proteins, and lipids from
   oxidative damage during photosynthesis and under abiotic stress (Davey
   et al. [54]2000; Mellidou and Kanellis [55]^2017; Fenech
   et al. [56]^2018). The pathways involved in the synthesis of
   D‐mannose/L‐galactose (D‐Man/L‐Gal), D‐galacturonic acid (D‐GalA),
   L‐Gulose, and inositol (Myo‐inositol) are also involved in the
   production of ascorbic acid in plant fruits (Paciolla et al. [57]2019).
   Additionally, although ascorbic acid is an essential nutrient, the body
   is unable to synthesize ascorbic acid and must instead obtain it from
   external sources. As the apricot fruit develops, its ascorbic acid
   level changes from high to low (Li [58]2013). Research on genetic and
   biochemical techniques for plant regulation, particularly for ascorbic
   acid biosynthesis, is relatively limited, and the enzymes involved in
   ascorbic acid biosynthesis remain unidentified (Yin et al. [59]2024).

   In this study, the ascorbic acid and soluble sugar contents of two
   varieties of apricot fruits at five developmental stages were examined.
   Although significant progress has been made in understanding the
   individual roles of sugars and ascorbic acid in fruit development,
   there remains a critical knowledge gap regarding their synergistic
   regulation. This gap is particularly evident in apricots, where the
   combined influence of sugars and ascorbic acid on fruit quality has not
   been fully explored. This study aims to address this gap by
   investigating the interactions between sugars and ascorbic acid
   throughout fruit development. Furthermore, we employed transcriptome
   analysis, weighted correlation network analysis (WGCNA), and RT–qPCR to
   identify the major genes involved in the synthesis of soluble sugars
   and ascorbic acid in apricots. This study will advance of artificial
   breeding and germplasm resources and enhance the nutritional value and
   overall quality of apricots.

2. Methods

2.1. Plant Materials

   The apricot grove in the Wanjiagou Autonomous Region of Inner Mongolia,
   Tuzuo Banner, Hohhot, was the source of the Wuyuexian apricot ‘WYX’ and
   Lanzhoudajie apricots ‘LZ’. Fruiting started on the sixteenth day after
   flowering, and sampling was performed every seven days after fruiting.
   For each variety, three apricot trees with consistent growth conditions
   were chosen. Undamaged fruits of similar size and appearance were
   chosen for sampling in each stage. The samples were sliced quickly into
   uniformly small pieces, snap‐frozen in liquid nitrogen, covered with
   tin foil, and stored at −80°C. Every sample had three biological
   replicates.

2.2. Detection of Glucose, Sucrose, Fructose, and Sorbitol Contents

   Under the action of specific enzymes, sucrose is converted into glucose
   and fructose, and glucose, through the enzyme complex including
   hexokinase, reduces NADP+ to NADPH. By measuring the increase in NADPH
   at 340 nm, the contents of sucrose, glucose, and fructose can be
   determined separately (GERUISI, China, G0560F).

   Tissue Sample: Weigh approximately 0.1 g of tissue, add 1 mL of
   distilled water, and homogenize on ice. Centrifuge at 12,000 rpm at
   room temperature for 10 min. Add reagents as instructed in the manual
   for detection. All steps are performed strictly according to the
   manual.

2.3. Determination of Ascorbic Acid Content

   Reduced ascorbic acid (AsA) can reduce ferric ions (Fe^3+) to ferrous
   ions (Fe^2+), and the ferrous ions react with red phenantroline to form
   a red complex. This complex has a characteristic absorption peak at
   534 nm, and its intensity is directly proportional to the content of
   reduced ascorbic acid (GERUISI, China, G0201F).

   Tissue Sample Treatment: Weigh approximately 0.1 g of tissue, add 1 mL
   of pre‐cooled extraction solution, and homogenize on ice. After
   extracting at room temperature for 10 min, centrifuge at 12,000 rpm,
   4°C for 10 min. Add reagents as instructed in the manual for detection.
   All steps are performed strictly according to the manual.

2.4. RNA Extraction, RNA‐Seq and Sequence Assembly

   Twenty‐four libraries were created via RNA extraction using a Tiangen
   polysaccharide polyphenol kit (TIANGEN, Beijing). Three biological
   replicates of each sample were included in the analysis. Gel
   electrophoresis (voltage 180 V for 16 min) using 1% agarose gel and an
   Agilent 2100 Bioanalyzer was performed to test the purity and integrity
   of the extracted RNA. RNA purity was measured using a NanoPhotometer
   (OD260/280 and OD260/230). NA concentration was measured using Qubit
   4.0 fluorescence. RNA integrity was assessed using a Qsep400
   Bioanalyzer.

   Following total RNA extraction, mRNA was enriched using Oligo (dT)
   magnetic beads and fragmented with a fragmentation buffer. Short
   segments of mRNA were used as templates to synthesize first‐strand
   cDNA. The Cdna was then purified by adding buffer, dNTPs, and DNA
   polymerase I to create double‐stranded cDNA. Finally, the purified cDNA
   was with a poly‐A tail and end repair for sequencing. These processes
   in addition to sequence splicing, fragment size selection using AMPure
   XP beads, and PCR were the methods used to construct the cDNA library.
   After the insert size satisfied expectations, the synthesis of the cDNA
   library was completed by quantifying the library concentration. The
   Illumina HiSeq platform was then used to sequence the mixed samples.
   RNA‐seq analysis was conducted with three biological replicates, and
   differential gene expression was identified using the DESeq2 package.
   False Discovery Rate (FDR) correction was applied, with a significance
   threshold set at p < 0.05.

   StringTie 2.1.6 assembly software was used to splice the raw data after
   they had been filtered, the sequencing error rate and GC content
   distribution were examined, and clean reads were obtained. After
   hierarchical clustering by Corset (official website:
   [60]https://code.google.com/p/corset‐project/), the spliced transcript
   sequences were inputted into the UniGene database to obtain the longest
   UniGene cluster sequences for further analysis.

2.5. Expression Quantification and Screening of Differentially Expressed
Genes (DEGs)

   The acquired sequence number was normalized to the FPKM value; gene
   expression levels were determined by normalizing the FPKM value. Using
   the DESeq2 plug‐in of TBtools and the negative binomial distribution
   model, DEGs were identified. The following screening parameters were
   used to identify DEGs in the mature apricot fruits: FDR < 0.05 and
   |log2Fold Change| ≥ 1.

2.6. Functional Annotation of UniGene Clusters

   After the amino acid sequences of the UniGene clusters were predicted,
   sequence comparison was carried out using the Pfam database and HMMER
   software to obtain the functional annotation information of the
   corresponding UniGene clusters. UniGene cluster sequences were compared
   using the Kyoto Encyclopedia of Genes and Genomes (KEGG), NCBI
   nonredundant (NR), Swiss‐Prot, Gene Ontology (GO), Eukaryotic
   Orthologous Groups (KOG), and Trembl databases via BLAST software (E
   value ≤ 10^−5).

2.7. Coexpression Network Construction and Identification of Hub Genes

   The DEGs identified in 30 samples of ‘WYX’ and ‘LZ’ were subjected to
   WGCNA. After initial DEG filtering, the samples were subjected to WGCNA
   with a soft threshold of β‐value = 14, a minimum module gene number of
   50, and a similar module merging threshold of 0.25. The IQR filter
   threshold was 0.5. To identify key transcription factors in metabolic
   pathways, the target module gene coexpression network was visualized
   using maximum clique centrality (MCC) topology analysis, with a degree
   value ≥ 1.

2.8. Validation of RNA‐Seq Data via RT–qPCR

   To confirm the accuracy of the RNA‐seq data, RT–qPCR was performed to
   measure the expression levels of the 11 DEGs in the two apricot
   varieties. As described in section 2.4.1, RNA was extracted and
   utilized as a template. Next, DNA was produced via reverse
   transcription using the PrimeScript FAST RT reagent kit with gDNA
   Eraser, and spare reagents were stored in a refrigerator at −20°C.

   RT–qPCR amplification was carried out using TB Green Premix Ex TaqTM II
   (Tli RNaseH Plus) and a Light Cycler 480 device. The total volume of
   the reaction mixture was 20 μL and was composed of 10 μL of TB Green
   Premix Ex Taq II (Tli RNaseH Plus) (2X), 1 μL of sample cDNA, 1 μL of
   each of the upstream and downstream primers (20 μmol/L), and 7 μL of
   ddH2O. The reaction procedure was the following: predenaturation (95°C,
   30 s, 1 cycle); amplification (95°C, 5 s; 60°C, 30 s, 40 cycles);
   melting (95°C, 5 s; 60°C, 1 min; 95°C, 1 cycle); cooling (50°C, 30 s,
   1 cycle); and storage at 4°C. Primer5 was utilized in the construction
   of the primers, and the internal reference gene was the 18S gene from
   the almond genome (Lin [61]2023). Four technical duplicates were set
   up. For qRT‐PCR, each gene was measured in three biological replicates,
   and data were analyzed using the 2^−ΔΔCt method to ensure the
   statistical reliability of the results (Livak and Schmittgen [62]2001).

2.9. Data Analysis

   SIMCA14.1 software was employed to analyze the principal component
   analysis (PCA) results; Origin Pro 2021 software was used to analyze
   data and generate Venn diagrams, histograms, differentially abundant
   metabolite heatmaps, KEGG bubble plots, and hierarchical cluster
   diagrams. Metabolic pathway diagrams were generated using PowerPoint
   and Adobe Illustrator CC 2017SP. GraphPad Prism 8 v8.0.2, TBtools, and
   Adobe Illustrator 2020 were used to create heatmaps depicting the
   expression of DEGs. The WGCNA package of R language software was used
   to generate and analyze weighted gene coexpression networks for DEGs,
   and Cytoscape 3.9.1 was employed to visualize the gene interactions
   within the target module.

3. Results

3.1. Soluble Sugar Content in the Two Apricot Varieties During Fruit
Development

   In this experiment, the changes in the sugar contents of ‘WYX’ and ‘LZ’
   across five stages of fruit growth (S1: young fruit stage; S2:
   fast‐growing stage; S3: nodule stage; S4: color‐transitioning stage;
   and S5: commercial ripening stage) were analyzed.

   The sucrose content of ‘WYX’ increased significantly throughout the
   five periods (Figure [63]1a), peaking at 44.45 and 118.73 mg/g in S4
   and S5, respectively. These findings suggest that the final
   developmental stages of ‘WYX’ are characterized by a notable increase
   in sucrose accumulation. Moreover, the sucrose content of ‘LZ’ also
   increased during the five periods, except for a brief decrease in S4,
   when it peaked at 93.76 mg/g. The sucrose levels of ‘WYX’ were much
   greater than those of ‘LZ’ in the late stages, particularly in S5.
   Moreover, the increase in the sucrose content of ‘LZ’ in S5 was not
   significant overall. Finally, ‘WYX’ had a sucrose content that was
   almost 25% greater than that of ‘LZ’.

FIGURE 1.

   FIGURE 1
   [64]Open in a new tab

   Soluble sugar and ascorbic acid contents of apricot fruits at different
   developmental stages: (a) Sucrose, (b) sorbitol, (c) glucose, (d)
   fructose, (e) ascorbic acid.

   From S1 to S3, the sorbitol content of ‘WYX’ increased progressively
   and peaked at 12.76 mg/g (Figure [65]1b) before declining in S4 and S5.
   This pattern revealed that sorbitol was required for the growth of
   ‘WYX’ in the early stages. The sorbitol content of ‘LZ’ increased
   progressively throughout the five periods, peaking at S4 and S5, where
   it reached 21.33 and 17.49 mg/g, respectively. These findings indicated
   that more sorbitol accumulated in the late stage of ‘LZ’ growth than in
   the other stages. Particularly in S4 and S5, sorbitol levels were much
   greater in ‘LZ’ than in ‘WYX’, and in S4, the sorbitol concentration in
   ‘LZ’ was almost 70% greater than that in ‘WYX’.

   The glucose content of ‘WYX’ increased significantly between S1 and S3,
   peaking at 37.40 mg/g (Figure [66]1c). Then, it stabilized in S4 and
   S5. The glucose content of ‘LZ’ also increased significantly from S1 to
   S3, reaching a peak of 42.25 mg/g before beginning to decrease in S4
   and S5. Although ‘LZ’ had a greater glucose content in S3 than ‘WYX’,
   the general changes were similar. Glucose accumulation in both
   varieties peaked at S3, and the glucose content of ‘LZ’ was nearly 13%
   higher than that of ‘WYX’.

   Throughout the five stages, the ‘WYX’ fructose content progressively
   increased, reaching 13.86 and 28.08 mg/g in X4 and X5, respectively
   (Figure [67]1d). These findings suggested that fructose levels
   significantly increased in the later stages of ‘WYX’ development, which
   was related to its role in different stages of growth and development.
   Throughout the five periods, the ‘LZ’ fructose level varied very
   little, peaking at 32.26 mg/g in X5. During the S5 period, the ‘LZ’
   fructose content increased, but overall, the concentration was quite
   constant.

   The ascorbic acid concentrations of ‘WYX’ were 28, 36, 35, 30, and
   25 mg/100 g throughout the five periods (Figure [68]1e). Overall,
   during S2, the ascorbic acid concentration peaked at 36 mg/100 g and
   then progressively decreased. This finding suggested that ascorbic acid
   synthesis occurred more in the earlier stages of growth in ‘WYX’ than
   in the later stages. Throughout the five periods, the ‘LZ’ ascorbic
   acid concentration varied less: it was 68.5, 32, 31, 25, and
   19 mg/100 g in the five stages. During S1, the amount of ascorbic acid
   reached a maximum value of 68.5 mg/100g and subsequently steadily
   decreased. This finding suggested that although ascorbic acid
   production in ‘LZ’ was lower in the latter phases, it was still much
   higher than that in ‘WYX’ in the early stages. During S1, the ‘LZ’
   ascorbic acid level was approximately 145% higher than that of ‘WYX’.

3.2. Transcriptome Sequencing and Gene Expression Analysis of Two Apricot
Varieties in Different Developmental Stages

   Thirty samples from the two apricot varieties were sequenced at five
   critical stages to investigate the mechanism of soluble sugar and
   ascorbic acid accumulation in apricot fruits. A total of 214.65 Gb of
   clean data were acquired; each sample's clean data exceeded 6 Gb, and
   93% or more of the Q30 bases were present. HISAT was used to compare
   the clean reads to the Prunus armeniaca F106 reference genome in the
   Genome Database for Rosaceae (GDR). Thirty libraries were sequenced
   from the clean reads, and the results revealed purities of 91.40% to
   95.39% and 95.37% to 98.37%.

   On the basis of the FPKM values of the sample genes, transcriptome data
   were utilized to perform a PCA of the 30 samples; the statistical
   findings are displayed in Figure [69]2a. A relatively significant
   correlation was found between biological duplicates of the same set of
   samples, indicating strong replicability among the chosen samples. The
   five growth periods of the two apricot varieties were roughly divided
   into two groups according to the results of the PCA (Figure [70]2b).
   The first three developmental stages (XA, XB, XC, LA, LB, and LC) of
   ‘WYX’ and ‘LZ’ were in one group, and there was a greater correlation
   between these stages. The commercial ripening stages of ‘WYX’ and ‘LZ’
   were in the other group, and ‘LZ’ color change stage was in its own
   group.

FIGURE 2.

   FIGURE 2
   [71]Open in a new tab

   Gene expression correlation in two apricot varieties at five stages.
   (a) The higher the correlation between two duplicate samples is, the
   closer the absolute value of the Pearson correlation coefficient (r) is
   to 1. (b) PCA plots illustrating the transcriptome clustering for each
   of the five developmental stages. (c) The five developmental stages of
   ‘WYX’ and ‘LZ’ are shown in horizontal coordinates, whereas the total
   numbers of upregulated and downregulated genes between and within the
   varieties are represented in vertical coordinates. (d) Venn diagrams
   illustrating the unique and shared genes between ‘WYX’ and ‘LZ’ in five
   developmental stages.

   The genes expressed in the two varieties during the five developmental
   stages were identified using the criteria of |log[2]Fold Change| ≥ 1
   and p < 0.05 to determine the genes involved in soluble sugar and
   ascorbic acid synthesis, as shown in Figure [72]2c. As shown in
   Figure [73]2b, 2063, 2224, 2371, 4444, and 2477 DEGs were
   differentially expressed among XA vs. LA, XB vs. LB, XC vs. LC, XD vs.
   LD, and XE vs. LE, respectively. Among these genes, 882/1181 (XA vs.
   LA), 1219/1005 (XB vs. LB), 1029/1342 (XC vs. LC), 2504/1940 (XD vs.
   LD), and 1265/1212 (XE vs. LE) genes were significantly differentially
   expressed in the apricots. A Venn diagram was used to analyze the data
   further, as shown in Figure [74]2d. Among the 339 shared DEGs, several
   DEGs were found to be shared between the two apricot varieties in each
   of the five growth periods (LA vs. XA (477), LB vs. XB (400), LC vs. XC
   (462), LD vs. XD (2169), and LE vs. XE (899)).

3.3. Dynamic Analysis of Transcriptome Data

   Using the scale function of the R package, the FPKMs of the DEGs were
   normalized to further investigate the expression of DEGs in the two
   apricot varieties at various stages of growth and development
   (Figure [75]3). A K‐means (k‐means clustering algorithm) clustering
   analysis was subsequently conducted. DEGs with similar functions
   presented comparable patterns of change in different developmental
   periods. There were 13,411 and 13,442 DEGs in ‘WYX’ and ‘LZ’,
   respectively. These genes were grouped into 10 distinct clusters. On
   the basis of the trends in soluble sugar and ascorbic acid
   concentrations, we were able to determine the clustering profiles
   (‘WYX’: Class2; ‘LZ’: Class6) that corresponded to the trends in sugar
   metabolism (‘WYX’: Class6; ‘LZ’: Class1) and DEGs involved in ascorbic
   acid synthesis. According to the results, there were 1546 DEGs in ‘WYX’
   and 2213 DEGs in ‘LZ’ involved in sugar metabolism; in the class
   corresponding to ascorbic acid, there were 2769 DEGs in ‘WYX’ and 3075
   DEGs in ‘LZ’; additionally, in the sugar metabolism clusters, the
   phenotypic changes in ‘WYX’: Class 2 and ‘LZ’: Class 2 were opposite of
   each other. Interestingly, the trends in ‘WYX’ glucose metabolism and
   ascorbic acid metabolism were similar.

FIGURE 3.

   FIGURE 3
   [76]Open in a new tab

   Cluster plot of DEGs using K–means algorithm for apricot gene
   expression (a) X2 (b) X6 (c) L1 (d) L2 (e) L6.

3.4. KEGG and GO Analyses Based on K‐Mean Clustering

   The top 15 and top 2 DEGs in ‘WYX’ and ‘LZ’ were chosen for GO
   biological process (BP), molecular function (MF) and cellular component
   (CC) enrichment analyses to further explore the functions of the DEGs
   in sugar metabolism and ascorbic acid production in the two apricot
   varieties (Figure [77]4).

FIGURE 4.

   FIGURE 4
   [78]Open in a new tab

   Bar graphs of GO enrichment and scatter plots of KEGG. (a) X2'GO; (b)
   X2′KEGG; (c) X6′GO (d) X6′: KEGG (e) L1′GO (f) L1′KEGG (g) L2′GO; (h)
   L2′KEGG; (i) X6′GO (j) L6′KEGG.

   The BP terms cellular processes and metabolic processes and the MF
   terms cellular anatomical entities and protein‐containing complexes
   were enriched in the DEGs.

   Additionally, the CC terms binding and catalytic activity and the
   containment complex were enriched in the DEGs. Furthermore, ascorbic
   acid synthesis was also enriched in the DEGs, which suggests that these
   genes are involved in the production of soluble sugars and ascorbic
   acid in these two apricot varieties.

   KEGG pathway enrichment analysis revealed that pathways involving
   ribosomes and motor proteins, the nucleotide sugar biosynthesis
   pathway, amino sugar and nucleotide sugar metabolic pathways, and
   α‐linolenic acid metabolic pathways were significantly enriched in the
   DEGs between the two apricot varieties in X2. K‐means analysis revealed
   that the DEGs in X2 that were involved in ascorbic acid synthesis
   exhibited similar expression trends, while the DEGs involved in sugar
   synthesis exhibited opposite expression trends. Ascorbic acid and
   alternate metabolism were also enriched in the DEGs in L6, along with
   other pathways such as those involving nucleosomes and motor proteins
   and those associated with glycosaminoglycan breakdown, monobasic
   monomer production, lysine biosynthesis, and purine metabolism.
   Ascorbic acid and alternative metabolism were also enriched in the DEGs
   in L2. A correlation study can be performed to determine whether there
   is a relationship between the two pathways.

   Propionic acid metabolism, homologous recombination, and DNA
   recombination were enriched in the DEGs in the remaining stages. Only
   glycerophospholipid, selenium compound, and purine metabolism were
   enriched in the DEGs in X6. The MAPK signaling pathway, phytohormone
   signaling, carbon metabolism, the synthesis of secondary metabolites,
   the metabolism of fatty acids, the metabolism of alanine, aspartate,
   and glutamate, and the breakdown of acetylleucine and oleic acid were
   all enriched in the DEGs in L1.

3.5. Identification of DEGs Associated With Gluconeogenesis and Ascorbic Acid
Synthesis Pathways

   DEGs related to the starch and sucrose synthesis were screened to
   determine the major genes that regulate sugar metabolism
   (Figure [79]5). Variations in the expression of these genes led to
   variations in the sugar content in the apricots in different
   developmental periods. We verified that the DEGs were involved in the
   synthesis of sugar by conducting a K‐means analysis, which revealed
   that their patterns of change matched those of fructose and sucrose.

FIGURE 5.

   FIGURE 5
   [80]Open in a new tab

   Expression patterns of DEGs involved in sugar metabolism.

   ‘WYX’ expressed 12 unique DEGs involved in the regulation of enzymes in
   the starch, sucrose, fructose, and glucose metabolic pathways,
   including MGAM (sucrose, PaF106G0400017358.01), HK
   (sucrose‐6‐phosphatase, PaF106G0700026522.01), TGG2, BGLU47 (cellulose,
   PaF106G0400015877.01, PaF106G0600022723.01), EGLC
   (glucanendo‐1,3‐β‐D‐glucosidase, PaF106G0400015157.01), GH9B3
   (endo‐glucanase, the PaF106G0800032867.01), PGM (phospho‐glucose
   translocase, PaF106G0100003912.01), otsB (alginate 6‐phosphate
   esterase, PaF106G0400018078.01, PaF106G0200010293.01.
   PaF106G0100001132.01), and AMY (β‐amylase, PaF106G0100003246.01,
   PaF106G0300011580.01). ‘LZ’ expressed four unique DEGs including
   maltase‐glucoamylase (PaF106G0700028769.01), dextran
   endo‐1,3‐β‐D‐glucosidase (PaF106G0600022982.01), sucrose synthase
   (PaF106G0600022982.01), and glucose‐1‐phosphate adenylyltransferase
   (PaF106G0700026815.01). Moreover, ‘WYX’ and ‘LZ’ exhibited high
   expression levels of SUS (sucrose synthase, PaF106G0700026815.01).
   These results imply that DEGs linked to sugar metabolism are crucial in
   controlling sugar accumulation in apricots.

   One of the key components of high‐quality apricots is ascorbic acid
   (Figure [81]6). The DEGs involved in the ascorbic acid synthesis
   pathway were analyzed, with notable variations in expression occurring
   in different growth periods. These genes included GMEs (GDP‐D‐mannose
   3′,5′‐differential anisomerase, in ‘WYX’ PaF106G0100003725.01), VTC2_5
   (GDP‐L‐galactose phosphorylase, PaF106G0200008910.01), VTC4 (inositol
   phosphatase/L‐galactose 1‐phosphate phosphatase, PaF106G0400018346.01),
   GalDH (L‐galactose dehydrogenase, PaF106G0100000284.01), AO, SKS5,
   SKU5, SKS3 (L‐ascorbic acid oxidase, PaF106G0400016721.01,
   PaF106G0100004207.01, PaF106G0500019378.01, and PaF106G0800030285.01),
   DHAR (glutathione dehydrogenase/transferase, PaF106G0200009695.01),
   MDAR6 (monodehydroascorbic acid reductase (NADH),
   PaF106G0500019749.01), and APX3 (L‐ ascorbic acid peroxidase,
   PaF106G0700027091.01). Several genes differed in expression between
   ‘LZ’ and ‘WYX’, and these genes were VTC4 (inositol
   phosphatase/L‐galactose 1‐phosphate phosphatase, PaF106G0400018346.01,
   PaF106G0800031491.01), GalDH (L‐galactose dehydrogenase,
   PaF106G0800030702.01), SKS4, SKS5 (L‐ascorbic acid oxidase,
   PaF106G0400018234.01, PaF106G0300012038.01), and APX4 (L‐ascorbic acid
   peroxidase, PaF106G0100006021.01). In addition, SKS5, SKU5 (L‐ascorbic
   acid oxidase, PaF106G0100004207.01, PaF106G0500019378.01), a downstream
   metabolite of ascorbic acid, was highly expressed in ‘WYX’ and ‘LZ’.

FIGURE 6.

   FIGURE 6
   [82]Open in a new tab

   Expression patterns of DEGs involved in ascorbic acid metabolism.

3.6. Identification of Coexpression Networks and Hub Genes Related to Sugar
Metabolism and Ascorbic Acid Synthesis

   We examined the gene regulatory networks linked to the synthesis of
   sucrose, glucose, fructose, and ascorbic acid and their expression
   patterns via WGCNA. Following DEG identification, 13,469 genes were
   analyzed via transcriptome sequencing, and the genes were further split
   into 16 expression modules (Figure [83]7a). The findings indicated
   (Figure [84]7b) that the red module had the strongest correlation with
   glucose synthesis (r = 0.81, p = 5.9e^−8), the blue module (2685) had
   the strongest correlation with fructose synthesis (r = 0.94,
   p = 1.3e^−14) and sucrose synthesis (r = 0.97, p = 9.8e^−19), and the
   turquoise (5002) module had the strongest correlation with ascorbic
   acid synthesis (r = 0, p = 0.0049). Interestingly, the expression of
   genes in the blue module decreased throughout the developmental stages
   and was significantly expressed in XE1, XE2, XE3, LE1, LE2, and LE3.
   Both the blue and turquoise modules followed the same expression trend.
   On the basis of high gene significance (GS) and high module affiliation
   (MM), we screened the genes in the blue module.

FIGURE 7.

   FIGURE 7
   [85]Open in a new tab

   Identification of hub genes in the coexpression network. (a) WGCNA
   hierarchical clustering tree of 16 modules of coexpressed genes
   (clustering dendrogram). Some genes always have similar expression
   changes in a physiological process or different tissues. These genes
   may be functionally related, and they can be defined as a module
   (module). For the upper half of the dendrogram, the longitudinal
   distances represent the distance between two nodes (between genes), and
   the horizontal distance is meaningless. (b) Heatmap of sugar and
   ascorbic acid contents; each row represents a module with different
   colors to indicate correlation values from −1 to 1. (c) Blue module
   coexpression network module (d) Turquoise module coexpression network
   module.

   With a total of 20 genes (Figure [86]7c), the gene coexpression network
   was built on the basis of the weighted values and the degree of linkage
   between the candidate genes and the DEGs involved in glucose
   metabolism. Notably, HSFB1 was identified as a key hub gene with 18
   linkage edges, and 6 structural genes related to glucose metabolism had
   the highest values of the 14 top‐ranked genes. FTIP4 was shown to be
   the primary hub gene in the turquoise module (Figure [87]7d). To
   construct the gene coexpression network, we chose the top 12 genes with
   the highest degree values and the top nine DEGs involved in ascorbic
   acid production.

3.7. qRT–PCR Analysis

   The relative expression of 11 candidate genes linked to the pathways
   involved in ascorbic acid synthesis and sugar metabolism in the two
   apricot varieties was confirmed via qRT–PCR to assess the accuracy of
   the RNA‐seq results (Figure [88]8). The qRT–PCR results demonstrated a
   steady expression trend, which was consistent with the RNA‐seq results.
   These results show the accuracy of the transcriptome analysis and the
   reliability of the data.

FIGURE 8.

   FIGURE 8
   [89]Open in a new tab

   The expression levels of 10 candidate DEGs were verified via qRT–PCR.
   The columns and lines represent the RNA‐seq and qRT–PCR data,
   respectively, of the candidate DEGs. 18S rRNA was used as the internal
   control. (a) F16F14.27, (b) PDCB3, (c) PMM, (d) GME, (e) TPS5, (f)
   HXK2, (g) SUS1, (h) SS2, (i) BAM1, (j) VTC2. The error bars indicate
   the SDs.

4. Discussion

   In apricots, sugar is a key metabolite that plays a critical role in
   maintaining proper fruit growth and development, facilitating the
   synthesis of nutrients, and promoting the accumulation, including
   fructose, glucose, and sucrose (Alajil et al. [90]2021; Tian
   et al. [91]^2023). We examined sugar accumulation in apricots during
   five developmental stages (the young fruit stage, rapid growth stage,
   nodule stage, color change stage, and commercial ripening stage). Under
   the combined action of photosynthesis and sugar metabolism in plant
   bodies, the contents of sucrose, glucose, fructose, and sorbitol
   dynamically change and are the main components of soluble sugars in
   Rosaceae.

   During the developmental stages, sucrose was the dominant sugar
   accumulated in both ‘WYX’ and ‘LZ’. The sugar accumulation in ripe
   fruits was 8 and 6 times higher in ripe fruits, respectively, which was
   consistent with the sucrose content findings in previous studies (Gou
   et al. [92]2023). At the color change and ripening stages, ‘WYX’ had
   higher sucrose and glucose contents than ‘LZ’ did, and from the
   fast‐growing stage onwards, ‘LZ’ had a significantly higher sorbitol
   content than ‘WYX’. From the fast‐growing stage onwards, the sorbitol
   content of ‘LZ’ was significantly greater than that of ‘WYX’. Since the
   sorbitol content increases with altitude (Sarıdaş et al. [93]2024),
   this may be due to ‘WYX’ being a native species of Hohhot, at an
   altitude of 1040 m above sea level, whereas ‘LZ’ is an imported species
   from Lanzhou city, Gansu Province, growing at an altitude of 1520 m
   above sea level. The variations in soluble sugar contents between the
   two apricot varieties may be attributed to differences in their growing
   locations. Glucose was the predominant sugar present during the early
   stages of fruit growth and development. However, with fruit maturation,
   glucose level decreased, while sucrose and sorbitol contents increased.
   The activities of sucrose synthase and succinate dehydrogenase are
   linked to the shift in sugar metabolism from catabolism to anabolism.
   Consequently, the total sugar and sucrose contents of the fruit
   increased (Chen et al. [94]2009; Zheng et al. [95]^2016;
   Weng [96]^2017).

   During the experiment, we discovered that the change in the AsA content
   of ‘WYX’ as inversely related to the sugar content. This could be due
   to the pH of the fruit, as pH has a significant impact on the stability
   of ascorbic acid. Additionally, a decrease in sugar content may be
   accompanied by an increase in pH, which could have an indirect effect
   on ascorbic acid metabolism (Liu and Xu [97]2015). Furthermore, as a
   potent antioxidant, ascorbic acid may exacerbate oxidative stress in
   response to increased fruit sugar buildup, necessitating increased
   ascorbic acid intake to preserve redox balance. However, if the
   antioxidant system is robust enough, the need for ascorbic acid may
   decline during ripening, which would result in a decrease in its level.
   Moreover, ethylene, a plant hormone essential for fruit ripening, may
   influence ascorbic acid metabolism and sugar accumulation. Ethylene may
   hinder the synthesis of ascorbic acid and promote sugar accumulation.
   To understand the precise link between sugar production and ascorbic
   acid metabolism, extensive experimental research is needed.

   In this study, 15 different sugar transporter genes, including MGAM,
   HK, TGG2, BGLU47, EGLC, GH9B3, PGM, otsA, otsB, AMY, SUS, glgA, glgC,
   and PYG were detected. Among these genes, HK (PaF106G0700026522.01),
   EGLC (PaF106G0600022982.01), SUS (PaF106G0700026815.01), PGM
   (PaF106G0100003912.01), and glgC (PaF106G0300012278.01) significantly
   influenced fruit sweetness. Among these, SUS (PaF106G0700026815.01) was
   highly expressed in both apricot varieties. Previous research (Aslam
   et al. [98]2019) also indicated that the SUS gene plays a critical role
   in sugar the accumulation and sweetness in apricots. In the metabolism
   of plants and animals, ascorbic acid is a essential antioxidant
   molecule and an enzyme cofactor (Fenech et al. [99]2018). The
   consumption of plants is a primary means through which humans can
   obtain their recommended daily intake of ascorbic acid. As ascorbic
   acid can be synthesized in fruiting plants via a variety of pathways,
   its biosynthesis is differentially regulated by the expression of
   important genes. Producing transgenic fruits and vegetables that are
   high in ascorbic acid can be achieved by better understanding of
   ascorbic acid anabolism as well as the role and expression of
   associated genes in fruiting plants. The growth trend of ‘WYX’
   initially showed an increase, followed by a decrease, mirroring the
   trend observed in the AsA levels of two different types of kiwi fruit.
   Similarly, ‘LZ’ and peach followed comparable trends, consistent with
   findings from previous studies (Bulley et al. [100]2009; Shu
   et al. [101]^2023; Imai et al. [102]^2009). The AsA concentration of
   ‘LZ’ was three times greater than that of ‘WYX’ during the S1 period
   and gradually decreased over the ripening stage (Ioannidi
   et al. [103]2009). However, this pattern didnot occur in tomatoes.
   During the S2 (fast‐growing) period, ‘WYX’ had the maximum amount of
   ascorbic acid, before gradually decreased. During the same period, the
   level of ascorbic acid in ‘LZ’ was higher than during ripening. GalLDH
   catalyzes the oxidation of the final substrate, L‐GalL, to AsA via the
   main l‐galactose or Smirnoff–Wheeler pathway for AsA biosynthesis in
   plant cells. It is also essential d‐galacturonide pathway (Hancock and
   Viola [104]2005), where GalDH is the critical enzyme that determines
   whether the plant synthesizes AsA via the L‐galactose pathway.
   Furthermore, young fruits have a greater capacity to convert GalL to
   AsA (Li et al. [105]2008), leading to higher AsA concentration during
   the S2 stage. Since AsA accumulation is mainly determined by the
   balance between DHA synthesis and oxidative loss in plant cells, the
   AsA content of both fruits began to decrease from S2 onwards. This
   occurred because the rate of AsA synthesis in young fruit was lower
   than the rate of oxidative loss during fruit development (Hancock and
   Viola [106]2005). The genome of T acutus contains genes that encoding
   glucose‐phosphate isomerase (PGI1), mannose‐6‐phosphate isomerase
   (PMI1), phosphoadenosine mutase (PMM), mannose pyrophosphorylase
   (GMP1), GDP mannose‐3′, 5′ epimase (GME1 and GME2), involved in the
   biosynthesis and accumulation of ascorbic acid via the L‐galactose
   pathway in immature fruits (Dos Santos et al. [107]2019). Four linear
   and cyclic pathways, involving numerous enzymes directly or indirectly
   involved in the metabolic reactions of AsA, regulate the synthesis of
   AsA (Davey et al. [108]2000; Wolucka and Van Montagu [109]^2003;
   Wheeler et al. [110]^1998). In this study, 10 regulators of AsA
   synthesis were identified: GME, VTC2_5, VTC4, DHAR, MDAR6, GalDH, APX4,
   SKS4, and SKS5. We deduced that APX4, SKS4, and SKS5 (L‐ascorbic acid
   oxidase, PaF106G0100004207.01, PaF106G0500019378.01), highly expressed
   in both varieties, may be crucial enzymes in the regulation of AsA
   synthesis. Different AsA synthesis routes and accumulation patterns in
   plants. For example, the l‐galactose and d‐galacturonic acid pathways
   exist in strawberries, and they each have different functions in
   maintaining AsA levels in immature and ripe strawberries (Cruz‐Rus
   et al. [111]^2011). In kiwifruit, the expression of the
   L‐glucuronolactone oxidase gene is strongly correlated with ascorbic
   acid content, the balance between AsA synthesis and catabolism is
   tightly regulated, and environmental factors regulate the metabolic
   pathway by influencing gene transcription (Dos Santos et al. [112]2019;
   Wang et al. [113]^2023).

   WGCNA revealed complex systemic gene relationships. We identified gene
   modules characterized by strong coexpression associated with the
   metabolism of sucrose and ascorbic acid. By analyzing the relationships
   between these modules and various physiological markers, we inferred
   their roles in metabolic regulation. The blue and turquoise modules
   were derived from WGCNA of the K‐means clustering results on the basis
   of sugar and ascorbic acid contents. HSFB1 and FTIP4 emerged as the
   top‐ranked hub genes based on their degree values. The top four hub
   genes (dl3250c, T5J17.30, ISTL3, CFIS2) and the top 11 hub genes
   (MRB17.4, OBP4, DALL3, LSH6, T1B9.2, CDFI, JAZ1, CIA2, NIK2, T29E15.15,
   ZFP7, KNAT7) were analyzed in coexpression studies with the DEGs
   involved in metabolic pathways. Gene coexpression analyses revealed
   that seven DEGs were associated with hub genes involved in
   gluconeogenesis and ascorbic acid synthesis.

   The role of HSFB1 in sugar metabolism differed from that of HSFB1 in
   sugar content. The production of a trait often involves the joint
   expression of multiple genes. Research has shown that the inhibition or
   expression of HSFB1 can increase plant tolerance to heat. For example,
   in Arabidopsis, HSFB1 represses transcription and negatively regulates
   the expression of heat‐induced stress resistance genes (Ikeda
   et al. [114]2011); in tomato leaves, HSFB1 knocked out improves heat
   tolerance (Paupière et al. [115]2020); and in grapes, HSFB1 was found
   to be a key regulator of heat tolerance (Ye et al. [116]2020). Plant
   heat tolerance has a positive effect on the accumulation of sucrose and
   glucose (Commisso et al. [117]2016; Liu et al. [118]^2019).

   We discovered that FTIP4, a key regulator of AsA synthesis, plays a
   crucial role in the regulation of AsA synthesis, as evidenced by its
   consistent correlation with AsA content. FTIP3/4 is a crucial
   regulatory factor controlling plant cell differentiation and tissue
   formation by influencing the intracellular transportation of the SHOOT
   MERISTEMLESS (STM) protein (Liu, Li, et al. [119]^2018). The STM
   protein operates on the plasma membrane of plant cells, facilitating
   the synthesis of cytokinin (CK) to prevent premature cell
   differentiation and assisting undifferentiated cells in the formation
   of self‐sustaining meristematic tissue, which is vital for plant growth
   and development (Scofield et al. [120]2014). In strawberry research,
   spraying cytokinins like 6‐benzylaminopurine (6‐BA) and kinetin (KT)
   effectively increased the vitamin C content in strawberries. These
   regulators affect the internal hormonal balance of the plant and
   promote the biosynthesis of vitamin C (Yuan [121]2019). Recent studies
   on the FTIP gene family indicate that FTIP1 is involved in the
   molecular mechanisms underlying the response of rice to drought stress,
   which has great implications for understanding the genetic basis of
   plant drought resistance (Chen et al. [122]2021). Given that genes
   within the same family may exhibit functional similarities (Liu, Liu,
   and Rajapakse [123]^2018) and considering the role of ascorbic acid as
   a potent antioxidant, FTIP1 may aid plants in mitigating reactive
   oxygen species generated under drought conditions, thereby protecting
   cells from oxidative damage and enhancing drought resistance (Davey
   et al. [124]2000; Mellidou and Kanellis [125]^2017; Fenech
   et al. [126]^2018). Based on the identified hub genes, we hypothesize
   that FTIP4 may be involved in the synthesis of ascorbic acid.

   These studies included metabolite analysis, gene expression analyses,
   and physiological and biochemical tests providing insights into ‘the
   optimization of fruit quality and nutrition’.

5. Conclusions

   We utilized transcriptome sequencing to investigate the gene expression
   associated with apricot fruit development and analyzed sugar and
   ascorbic acid levels in ‘WYX’ and ‘LZ’. We identified sugar transport
   genes, ascorbic acid synthesis regulatory factors, and content
   differences. At maturity, ‘WYX’ accumulated more sucrose, whereas ‘LZ’
   accumulated more sorbitol. In the early fruit stage, ‘LZ’ had a
   relatively high ascorbic acid content, which decreased as the fruit
   aged. Our identification of the genes FTIP4 and HSFB1 provides new
   insights into apricot fruit metabolism. However, importantly,
   transcriptome analysis represents only a fraction of metabolic
   regulation; further investigation is needed to identify other
   mechanisms. This study contributes to our understanding of apricot
   metabolism and aids in the establishment of germplasm resources and
   molecular breeding for improved apricot varieties.

Author Contributions

   Yanhong He: conceptualization (equal), data curation (equal), formal
   analysis (equal), funding acquisition (equal), investigation (equal),
   methodology (equal), project administration (equal), resources (equal),
   software (equal), supervision (equal), validation (equal),
   visualization (equal), writing – original draft (equal), writing –
   review and editing (equal). Yang Zhao: conceptualization (equal), data
   curation (equal), formal analysis (equal), funding acquisition (equal),
   investigation (equal), methodology (equal), project administration
   (equal), resources (equal), software (equal), supervision (equal),
   validation (equal), visualization (equal), writing – original draft
   (equal), writing – review and editing (equal). Tianxing Guo: resources
   (supporting). Wenbao Jia: software (supporting). Chun Tian: supervision
   (supporting). Yu‐e Bai: conceptualization (lead).

Ethics Statement

   The landowner allows access to the orchard for the purpose of
   collecting fruit.

Consent

   The authors have nothing to report.

Conflicts of Interest

   The authors declare no conflicts of interest.

Acknowledgments