Gemstone seedless viticulture at Shandong Academy of Grape, and each treatment selected 1 vine with the same growth potential, total 6 vines. Each vine was sprayed with 50mg/L hyperoside or treated with the buffer as a control. The process was sprayed twice, first on June 21^st and second on July 9^th. The spraying site was more than 2 lines of all leaves. Hyperoside reagent and the buffer was used to spray Gemstone seedless grapes from the early fruit stage to the ripening stage. The sample processing period was split into four stages (20 days at each stage for a total of 80 days) as follows: July 16 (B2, pre-color inversion stage), August 6 (B3, color reversal stage), August 26 (B4, early maturation stage), and September 16 (B5, complete ripening stage). Three copies of the experiment were carried out and the treated plants were immediately used for tissue sampling. Three fruits from the same plant were used as a treatment and each treatment sample was stored at -80°C until it was used.

Anthocyanin was extracted using the previous report (Fang et al., 2019), with minor modifications. Each sample and each repeat of the different treatments and different periods were powdered in liquid nitrogen. Afterwards, 10 ml methanol: acetic acid (99:1) solution was added with the ground-up sample completely immersed in the solution (pump if needed), blocked, and the sample was extracted in darkness for 24 hours (shaking could be repeated). 1 ml of the supernatant from the combination was combined with 4 ml of KCl buffer (pH=1.0) after the mixture was centrifuged for 10 minutes at 8,228 g. The measurements of OD levels were taken at 530 and 700 nm. Afterwards, 1 ml of the aforementioned extract was then mixed with 4 ml of a NaAc buffer (pH=4.5) and the mixture was incubated for 15 minutes at ambient temperature. The anthocyanin content was calculated as follows: C=ΔA ×5×0.005×1000×449.2/(26900×0.5); ΔA = (A530-A700) (pH = 1.0) - (A530-A700) (pH = 4.5). Where C is the anthocyanin content and A530 and A700 are the optical density values of sample solutions. Triplicates of each biological sample were created.

To assure the use of certified samples for transcription sequencing, the purity, concentration, and integrity of RNA samples are detected using cutting-edge molecular biology equipment. After the detection of qualified samples, the cDNA library was constructed as follows: (1) Oligo magnetic beads (dT) were added to eukaryotic mRNA; (2) mRNA was randomly aborted using the fragmentation buffer; (3) using mRNA as a model, the first and second strands of cDNA were then synthesized sequentially; and (4) the purified bacterial cDNA was repaired at the end, the A-tails were added, and the DNA was sequenced. (5) PCR enrichment was used to create the cDNA library. The library was built, and the RT-qPCR technique was utilized to precisely measure the real concentration of the library (> 2 nM) in order to ensure quality. Following a qualified inspection of the database, different libraries shall conduct pooling based on the target off-line data quantity. The sequencing was taken using the BMKCloud (https://www.biocloud.net).

The Gene Ontology Consortium created the Gene Ontology(GO) database in 2000 as a structured, standardized biomonitoring system (Ashburner et al., 2000). It intends to generate a standardized knowledge vocabulary on genes and their products, which applies to various species. Based on the non-central hypergeometric distribution of Wallenius, GOseq R software packages carried out GO enrichment analysis of differentially expressed genes (DEGs).

Different genetic products work together to carry out biological processes in organisms, and pathway annotation study of DEGs may assist in better understanding the role of the genes. KEGG (Kyoto Encyclopedia of Genes and Genomes) (Kanehisa et al., 2004) is a database for systematically analyzing gene function and genomic information, making it easier to better understand genomes. The information about genes and expression is usually investigated as a complete network. An enrichment analysis of the significance of pathways was performed to determine those pathways that were significantly enriched in DEGs in relation to the overall genome background using a hypergeometric test based on Pathway in the KEGG database. The statistical enrichment of DEGs was evaluated using the KOBAS program in the paths of KEGG (Bao et al., 2019). Methods with a reliable enrichment value (p-value < 0.05) were chosen for the analysis.

For the RT-qPCR analysis, the total RNA of each sample was isolated from both treatments of the non-core gemstone. Each RNA sample’s residual DNA was extracted, and each cDNA sample was produced using Oligo dT in accordance with the manufacturer’s instructions (SuperScript III, Invitrogen, USA). The cDNA was diluted and used as a model. The gene-specific primers were made with Primer3 (https://primer3.org/). Supplementary Table S2 presents the primer sequences. Quantitative fluorescence experiments were conducted with SuperReal PreMix (Probe) (Tianroot Biology, Beijing, China) and CFX was used (Bio-Rad, West Berkeley, California, USA) to assist in the analysis (Du et al., 2021). The delta CT methodology was used for the computation of gene expression results. All samples were split into three biological replicates of RT-qPCR.

Flavonoid pathway-related DEGs were selected, such as VvPAL (phenylalanine lyase), VvCHS (Chalone synthase), VvF3’5’H (Flavonoid 3,5—hydroxylase), and VvMYB62 expressed during grape development. Using this website (http://www.bioinformatics.com.cn/), the heat map was produced.

Particular primers were employed to isolate the presumed VvMYB62 gene. The PCR fragment was purified and cloned in the pROKII cloning vector (Song et al., 2022). The vector and the PCR-amplified product were mixed and ligated, and processed into Escherichia coli (E. coli) DH5α competent cells in accordance with the manufacturer’s instructions. DNAMAN software was used to compare gene sequences with the selected gene sequences with transcriptome sequences.

The embryogenic callus of the grape was transformed in a transitory manner (Bao et al., 2017). The coding sequence (CDS) of the VvMYB62 gene was reconstituted into a pROKII vector with a GFP marker sequence to build the recombinant plasmid 35S:VvMYB62-GFP. The recombinant plasmid was converted into Agrobacterium tumefaciens EHA105 cells. After thawing, 200 ul of the bacterial solution was absorbed by the pipette and placed in 5 ml of Yep, Rif, and Kanamycin solution at 180 R/min on a shaker and 28°C for 20 hours, centrifuged at 5000 R/min for 10 minutes at room temperature. The supernatant was subsequently settled and the solution (4.44 g/L MS, 60 g/L sucrose, and 200 umol/L acetosyringone) was suspended again for two hours. The callus of a similar growth state (one month old) was chosen and the outer surface was transformed. The callus was submersed in the reassortment solution and sealed in a petri dish by means of using a vacuum pump for 20 minutes. The grape callus was cultured in the dark at 24°C for 48 hours on a solid medium (MS with 60 g/L sucrose, 7 g/L agar, and 200 umol/L acetosyringone). Grape callus transformed with an empty vector was used as control. In this experiment, the grape calli were pre-treated with a strain with overexpression VvMYB62 and empty vector, then vacuumed, and then placed into a growing dish containing 50mg/L hyperoside solution for subsequent operations. One biological replicate was created from the grape calli treated in the same petri dish, and the analysis required at least three biological replicates. Calli were then collected, instantly frozen in liquid nitrogen, and then kept at -80°C for subsequent use.

The fruits of Gemstone seedless grape were submerged in an Agrobacterium solution containing an empty vector and a vector for VvMYB62 overexpression in accordance with the rapid transformation of the grape fruit technique (Yao, 2021). Grape fruits were placed in a box containing MS and sucrose during a photoperiod of 16 hours of light and 8 hours of darkness after overnight incubation at 26°C in the dark. After temporary metamorphosis, the grape epidermis’ color was seen. Anthocyanin concentration and gene expression levels were assessed between treatments.

The error bars data were used to display the findings after each test was run in triplicate. Utilizing the GraphPad Prism 9 tool, the T-test approach was used to statistically analyze the data. P < 0.05 was represented by * and P < 0.001 by ***.

To fully understand the effect of hyperoside treatment on anthocyanin content, we performed phenotypic analysis of grape fruits under two treatments.The changes in the color of grape fruit were observed at every stage.No obvious color differences were detected between the control and hyperoside-treated fruits at the color-turning and the early color-turning stages (stages B2 and B3). However, in the later stages of color alteration and fruit ripening (stages B4 and B5), the fruit color of hyperoside-treated samples was much paler than the control ( Figure 1A ). However, the fruit shape index did not significantly change from one period to the next ( Figure 1D ).

In view of the fact that treatment with hyperoside could significantly impact the antioxidant function of the fruit, the changes in anthocyanin content were further measured in both treatments. Phenotypic data showed that anthocyanin levels of both treatments increased substantilly at the B2-B5 stages with fruit development. Additionally, there were noticeable variations in anthocyanin content between both treatments at subsequent stages of color transfer and maturity (stages B4 and B5). Anthocyanin levels in hyperoside-treated fruit were significantly less than that of the control group ( Figures 1B, C ). These results suggest that the application of hyperoside could dramatically reduce the anthocyanin content in Gemstone seedless grapes. Transcriptome analysis was carried out on samples from the control and treatment groups at each of the five period groups in light of these phenotypic findings to identify the underlying molecular pathways of hyperoside-induced anthocyanin accumulation.

Gemstone seed-free grape cDNA samples were sequenced with the Illumina sequencing platform. The selected clean reads were mapped using the grape reference genome (ftp://ftp.ensemblgenomes.org/pub/plants/release25/fasta/vitis_vinifera/). After a thorough analysis of the quality control data analysis, a total of 6.58 GB of clean data were obtained for each sample with a baseline Q30 percentage of 92.64% or more. Clean reads of each sample were compared sequentially with the reference genome of the grape and the effectiveness of the comparison varied from 83.91 to 94.87% ( Supplementary Table S1 ). Based on the comparison results, DEGs were identified based on their levels of expression in various samples and functional annotations and enrichment analyses were conducted.

The DEGs were screened between various treatments at the same developmental stage and explored the effects of hyperoside application on grape color transformation after the viability of the quality control data was confirmed. The following criteria were provided to see if the same gene is expressed differently in the two treatments: p-value<0.05 and the absolute value of Fold Change≥1.5. The DEGs were displayed using a Venn diagram ( Figure 2A ). Data revealed that there were 2,278 DEGs (two-fold up or down-regulated genes) in the B2CK_vs_B2HY group, of which 842 genes were up-regulated and 1,436 genes were down-regulated. There were 164 DEGs in the B3CK_vs_B3HY group, including 71 up-regulated and 93 down-regulated genes. In addition, 483 DEGs, including 134 up-regulated and 349 down-regulated genes, were found in the B4CK_vs_B4HY group. 513 up-regulated and 255 down-regulated genes showed differential expression at this stage compared to the previous phase ( Figure 2B ). Heat maps are used to depict DEG expression patterns in a more understandable way ( Figure 2C ).

The three primary GO categories of “molecular function”, “biological processes”, and “cellular components” were applied to all DEGs ( Figure 3 ). Four periods of GO enrichment analysis were completed. There were many terms in the categories “biological process,” “process,” “metabolism,” “the organization’s process,” “adjust,” “stimulus-response”, “the organization of cells or biological”, “localization”, “development”, as well as the “signal” and “process” of a multicellular organism. The term “extracellular area” was the most prevalent in the category “cell component”, which also included the terms “cell”, “cell part”, “organelle”, “membrane” and “organelle part”. The words “combination” and “catalytic activity”, “transport activity”, “nucleic acid combined with the activity of transcription factors”, “activists sensor”, “molecular sensor signal activity”, “cell”, “antioxidant activity” and “activity of transcription factors and protein activity” were some of the most frequently used ones in the category “molecular functions”.

To further identify the biological functions of DEGs, a pathway analysis was performed using the KEGG database ( Figure 4 ). The DEGs between B2_CK_vs_B2_HY groups were significantly enhanced in ten pathways. Both “ABC transporters” and “flavonoid biosynthesis” were closely linked to anthocyanin biosynthesis. DEGs between B3_CK_vs_B3_HY groups were significantly enhanced in four pathways, of which “anthocyanin biosynthesis” was the most relevant for this study. Four pathways in B4_CK_vs_B4_HY groups were linked to anthocyanin biosynthesis, including “phenylalanine metabolism”, “flavone and flavonol biosynthesis”, “flavonoid biosynthesis”, and “phenylpropanoid biosynthesis”.In the B5_CK_vs_B5_HY group, “flavone and flavonol biosynthesis”, “phenylalanine metabolism”, and “flavonoid biosynthesis” were correlated to anthocyaninbiosynthesis.

The flavonoid biosynthesis pathway’s known genes were enumerated in the aforementioned four categories of comparison in order to better understand which genes are involved in the creation of color. Significantly more down-regulated genes were present than up-regulated ones. The B4 stage had the most DEGs that were down-regulated ( Figure 5A ). Statistics of the DEGs were presented in Figure 5B . The expression levels of the control and hyperoside treatments, as well as the B4 stage, were substantially different for a total of 22 DEGs involved in the synthesis of anthocyanin. The B5 stage has greater levels of gene expression than the B2 stage ( Figure 5C ).

In-depth explanations were provided for the metabolic pathway diagram used in this work. PAL catalyzes the formation of 4-coumaryl CoA from phenylalanine. Afterward, 4-coumaryl CoA is catalyzed by CHS to produce yellow chalcone, followed by CHI to form dihydroflavonol, and ultimately anthocyanin is synthesized. These three genes(VvPAL, VvCHS, and VvF3’5’H) with high expression were further examined using RT-qPCR verification. The outcomes demonstrated that the trend of fluorescence quantification-verified above gene expression was compatible with the transcriptome data, demonstrating the validity of the transcriptome data ( Figure 5D ).

An investigation of the co-expression patterns of all DEGs and the three aforementioned enzymes was carried out in order to learn more about how the treatment with hyperoside affects the biosynthesis of anthocyanin. The analysis of the fourth phase DEGs and the co-expression of all transcription factors revealed a separation into three clusters ( Figure 6A ). According to the rule of growth and development, heat map research revealed that several transcription factors had the same expression pattern as the enzymes ( Figure 6B ). The promoter sites of the aforementioned three enzymes were found to be largely localized in the family of MYB transcription factors using promoter homeogenic elements ( Figure 6C ). Three transcriptionfactors,VvMYB62,VvWRKY28,and VvTIFY9,were chosen for further verification( Figure 6D ). VvMYB62 was chosen for overexpression investigation as a result.

In plants, transient transformation is a reasonably quick method for detecting target gene expression (Sparkes et al., 2006). Grape calli were genetically altered using Agrobacterium tumefaciens to demonstrate that VvMYB62 may respond to hyperoside to control anthocyanin accumulation. Phenotypic alterations were seen once the VvMYB62 transcription factor was chosen for over-expression confirmation ( Figure 7A ). There were red callus lines where VvMYB62 was overexpressed. Calli overexpressing VvMYB62, however, displayed reduced redness in a hyperoside-induced media. The control’s calli were white. The measurement of anthocyanin content ( Figure 7C ) and RT-qPCR analysis ( Figure 7B ) corroborated the alterations in anthocyanin in the callus. According to the spectrophotometric measurement, VvMYB62-OE increased anthocyanin accumulation above the control. However, the capacity of VvMYB62-OE to promote anthocyanin accumulation was diminished after exogenous hyperoside therapy. RT-qPCR was used to assess the transcriptional levels of the genes involved in the anthocyanin biosynthesis pathway. According to these findings, VvMYB62, VvPAL, VvCHS, and VvF3’ 5’H were all up-regulated to varying degrees in VvMYB62-OE, while they were all down-regulated in VvMYB62-OE was hyperoside-mediated.

The VvMYB62 gene was transiently inserted into Gemstone seedless grape using Agrobacterium-mediated transfer in order to better investigate the function of the gene and its alterations after hyperoside treatment. Significant changes in fruit color between the treatment and control groups occurred after around 4 days of cultivation ( Figure 8A ). The trend difference of VvMYB62, VvPAL, VvCHS, and VvF3’5’H between the control and treatment groups was further established by RT-qPCR ( Figures 8B, C ). This study also showed that hyperoside treatment may minimize the coloring impact while VvMYB62 might increase the coloring of grape fruits.

It is generally recognized that both exogenous and internal variables, such as light, temperature, plant growth regulators, and nutritional status, have an impact on the production of flavonoids (Naik et al., 2022). This study showed a connection between endogenous anthocyanin accumulation and hyperoside administration externally. Hyperoside is a 3-O-galactoside of quercetin that encourages the flowering and seed formation of Abelmoschus esculentus (Naik et al., 2022). In order to monitor fruit growth, we sprayed each vine in this trial with 50mg/L hyperoside and buffer as a control. The findings show that Gemstone seedless grape anthocyanin content may be considerably reduced by exogenous hyperoside, delaying the stage of color change ( Figure 1 ).