Transcriptome and Metabolome Analysis Unveil Anthocyanin Metabolism in Pink and Red Testa of Peanut (Arachis hypogaea L.)

Peanut (Arachis hypogaea L.) is an important source of oil and food around the world, and the testa color affects its appearance and commercial value. However, few studies focused on the mechanism of pigment formation in peanut testa. In this study, cultivars Shanhua 15 with pink testa and Zhonghua 12 with red testa were used as materials to perform the combined analysis of transcriptome and metabolome. A total of 198 flavonoid metabolites were detected, among which petunidin 3-O-glucoside and cyanidin O-acetylhexoside in Zhonghua12 were 15.23 and 14.72 times higher than those of Shanhua 15 at the R7 stage, revealing the anthocyanins underlying the red testa. Transcriptome analysis showed that there were 6059 and 3153 differentially expressed genes between Shanhua 15 and Zhonghua 12 in different growth periods, respectively. These differentially expressed genes were significantly enriched in the flavonoid biosynthesis, biosynthesis of secondary metabolites, and metabolic pathways. Integrated analysis of transcriptome and metabolome indicated CHS gene (arahy.CM90T6), F3′H genes (arahy. 8F7PE4 and arahy. K8H9R8), and DFR genes (arahy. LDV9QN and arahy. X8EVF3) may be the key functional genes controlling the formation of pink and red testa in peanut. Transcription factors MYB (arahy.A2IWKV, arahy.US2SKM, arahy.SJGE27, arahy.H8DJRL, and arahy.PR7AYB), bHLH (arahy.26781N, arahy.HM1IVV, and arahy.MP3D3D), and WD40 (arahy.L6JJW9) in the biosynthetic pathway of anthocyanin were significantly upregulated in Zhonghua 12 which may be the key regulatory genes in testa pigment formation. This is a comprehensive analysis on flavonoid metabolites and related genes expression in peanut testa, providing reference for revealing the regulatory mechanism of pigment accumulation in peanut testa.


Introduction
Anthocyanin is a natural hydrosoluble pigment widely found in plant leaves, stems, flowers, fruits, and seeds. It is mainly synthesized on the surface of the endoplasmic reticulum and accumulated in vacuoles. Anthocyanins have a wide range of biological functions, making plant organs blue, pink, red, or purple [1]. In many plants, anthocyanins can protect them from drought stress, cold, ultraviolet radiation, and microorganisms. Besides, they make flowers and fruits colorful and attract animals and insects, thus promoting pollination and seed transmission [2][3][4][5]. In addition to the natural character, anthocyanins also have high nutritional value and antioxidant capacity. They can scavenge pathogenic free radicals in the human body [6], inhibit the oxidation of lowdensity lipids, and prevent cardiovascular and cerebrovascular diseases, cancer, and other chronic diseases [1,7].
Peanut (Arachis hypogaea L.) is an important oil and economic crop in the world, and the testa color affects its commercial and nutritional value. There are many colors of peanut testa, including pink, red, purple, black, white, and multicolor, the most common of which are pink and red. However, there are only a few reports on the genetic mechanism of testa color in peanut. For example, the AhTc1 gene, encoding a R2R3-MYB transcription factor, which is located in the 4.7Mb region on chromosome A10, is responsible for the purple testa [19], while a single dominant gene located in the 0.905cM region on chromosome 3 controls the red seed coat, and the upregulation of anthocyanin synthesis genes may lead to red testa [20]. The genes PAL, C4H, CHS, CHI, and homologs of AtMYB111 in the flavonoid biosynthesis pathway may be the main regulators for pink pigment accumulation in peanut seed testa [21]. The difference of testa color is closely related to the expression of anthocyanin synthesis-related genes and the accumulation of metabolites. However, there is few research on the mechanism of different color formation in peanut seed by combining transcriptome and metabolome methods. At present, the gene regulation network of testa color formation in peanut is still unclear.
In the present study, seed pigment accumulation differences were investigated between two varieties, Shanhua 15 (pink testa) and Zhonghua 12 (red testa), during the growth period of R4 and R7, based on a previous classification of peanut pod developmental stages [22]. Meanwhile, the expression of flavonoid-related metabolites and corresponding genes was also analyzed by combined transcriptome and metabolome methods. The results will not only provide a new strategy for the study of gene function and the relationship between genes and metabolites in peanut but also provide reference information for revealing the formation and regulation mechanism of different testa colors in peanut.

Materials and Methods
2.1. Plant Materials. Cultivars Shanhua 15 with pink testa and Zhonghua 12 with red testa were used as materials, and they were planted in the test field of the agricultural experimental station of Shandong Agricultural University (36.15°N, 117.15°E), Tai'an, China. According to a previous study, peanut pod development can be divided into nine stages, R1 to R9 [22]. In this study, seeds with identical size were selected from Shanhua 15 and Zhonghua 12 from R2 and R7 stages, respectively, and the testa was peeled off the seeds, frozen immediately in liquid nitrogen, and stored in a low-temperature refrigerator at -80°C for follow-up analysis. The peanut seeds begin to pigment deposition in testa within the R4 stage, and they are nearly mature with the almost completed coloration in the R7 stage; thus, testa samples of Shanhua 15 and Zhonghua 12 at R4 and R7 stage, named S4 and S7 and Z4 and Z7, respectively, were used for transcriptome and metabolome analysis.

Total Anthocyanin Content Determination.
For extraction of anthocyanin, the method described by Rabino and Mancinelli [23] with some optimization was as followed. Testa samples were powdered in a mortar by adding liquid nitrogen; then, 0.1 g powder was mixed with 5 ml ethanol mixture (85 : 15 95% ethanol : 1.5 M HCl, v/v) and stored overnight at 4°C. The mixture was centrifuged at 5000 × g for 6 mins, and the supernatant containing anthocyanin was used to absorbance measurement at 530, 620, and 650 nm. Total anthocyanin content was calculated by the following formula: anthocyanin content ðmmol : g −1 FWÞ = ½ðA1 -A2Þ − 0:1 × ðA3 -A2Þ/4:62 × 10 4 × V/m × 106. A1, A2, and A3 represent the absorbance at 530, 620, and 650 nm, respectively, and V represents the extract volume, and m represents sample weight. The value 4:62 × 10 4 represents the extinction coefficient of anthocyanin at 530 nm. Three replicates were analyzed for each sample.

Flavonoids-Metabolites Detection and Multiple Reaction
Monitoring (MRM). Flavonoids-metabolites identification and quantification was carried out by Metware Biotechnology Co., Ltd. (Wuhan, China) using LC-ESI-MS/MS system (HPLC, Shim-pack UFLC Shimadzu CBM30A system, http://www.shimadzu.com.cn/; MS, Applied Biosystems 6500Q TRAP, http://www.appliedbiosystems.com.cn/). Detailed operation procedures were performed according to the methods described by Dong et al. [1]. Flavonoidsmetabolites identification was adopted by Partial least square discriminant analysis (PLS-DA). The variable importance in projection (VIP) of the first principal component represents the contribution of the differential metabolites between two groups, and the differential metabolites were screened using thresholds of fold change (FC) ≥2 or ≤0.5 and VIP >1.

Transcriptome Analysis.
Total RNA extraction, concentration, and cDNA library construction were performed according to methods described by Cao et al. [24]. With three biological replicates for each sample, a total of twelve libraries were constructed for samples S4, S7, Z4, and Z7. Then, the library products were sequenced on the HiSeq 4000 platform (Illumina, San Diego, CA, USA). After removing reads containing the adapter and low-quality sequences, the resulting high-quality clean data was mapped to the peanut reference genome (https://www.peanutbase.org/data/ public/Arachis_hypogaea/).
The fragments per kb per million reads (FPKM) method was used to calculate gene expression [25]. Differential expression analysis of testa samples was performed by the DEGseq R package, and the P values were adjusted using q values [26]. Differentially expressed genes (DEGs) were screened by the thresholds of log2 (FC) ≥1 and False Discovery Rate (FDR) ≤0.05. Multiple databases were used for gene function annotation, including Kyoto Encyclopedia of Genes and Genomes (KEGG), Gene Ontology (GO), Clusters of Orthologous Groups (COG) of proteins database, NCBI non-redundant protein sequences (Nr), a manually annotated and reviewed protein sequence database Swiss-Prot, and database of TrEMBL that contains all translations of EMBL nucleotide sequence entries.

Validation of Transcriptome Profile.
To validate gene expression results, nine flavonoid-related genes were randomly selected for quantitative real-time PCR (qRT-PCR) reactions, and the specific primers (Table S1) were designed by Primer Premier 5 software. The reaction was carried out according to the instructions of the SuperReal PreMix Plus kit (TIANGEN, Beijing, China), and the reaction system included 10 μL SYBR PreMix Plus, 1μL each primer (10 μM), 2 μL cDNA template, and 6 μL RNase-free water. It was performed on the ABI 7500 PCR instrument (Applied Biosystems), with the following protocol: 95°C for 30 s, followed by 40 cycles at 95°C for 10 s, 55°C for 20 s, 72.0°C for 20 s, and 75.0°C for 5 s. Peanut Actin gene was selected as the housekeeping gene [27], and relative gene expression was analyzed by the 2 -△△Ct method. Significant differences between samples were performed by standard deviation.

Testa Anthocyanin Content Changes during Seed
Development. As shown in Figure 1(a), the testa color of Shanhua 15 and Zhonghua 12 got deeper gradually from R2 to R7, especially in R7, when the seeds were colored completely and entered the mature stage. Testa of Shanhua 15 is pink, while it is red in Zhonghua 12. Anthocyanin content determination also showed that the total testa anthocyanin content increased gradually with the seed development and reached the highest level at the R7 stage, which were 0.026 mmol.g -1 and 0.087 mmol.g -1 in Shanhua 15 and Zhonghua 12, respectively (Figure 1(b)). The total testa     International Journal of Genomics anthocyanin content of Zhonghua 12 was significantly higher than that of Shanhua 15 at each stage (p < 0:01), with approximately three times higher on average (Figure 1(b)).
Therefore, it is suggested that anthocyanin accumulation difference is cultivar specific and eating peanuts with red testa is more beneficial to health.

Testa Metabolites Differences between Shanhua15 and
Zhonghua12. In order to further study the composition of peanut testa, samples of Shanhua 15 (S4 and S7) and Zhonghua 12 (Z4 and Z7) testa at the R4 and R7 stage were used for flavonoid-related metabolite profiles analysis. A total of 198 flavonoid-related metabolites were detected in all samples (Table S2). Principal component analysis (PCA) showed that S4, S7, Z4, and Z7 were obviously separated in the PC1 × PC2 score chart (Figure 2(a)), and obvious differences were showed between both cultivars and growth periods ( Figure 2(b)). The volcano plot also indicated that there were significant differences in metabolites content between the two varieties at R4 (Figure 2(c)) and R7 (Figure 2(d)) stages. The significantly changed metabolites (SCMs) could be divided into eight categories, including anthocyanins, proanthocyanidins, flavanone, flavone, flavonoid, flavonol, isoflavone, and polyphenol. Compared with S4, 68 SCMs were detected in Z4, 37 of which were upregulated and 31 were downregulated (Figure 2(e)), respectively, while 63 SCMs were identified in Z7 when comparing to those in S7, among which 34 SCMs were upregulated and 29 were downregulated, respectively ( Figure 2(f)). The category with the most SCMs was flavone, followed by flavonol.
Anthocyanin plays an important role in the coloration of seed coat and fruit skin [12,21]. Eleven anthocyanins were identified from all samples. The content of cyanidin O-diacetyl-hexoside-O-glyceric acid in Z4 was 7.87 times higher than that in S4, while pelargonin content decreased significantly (Table S2). In Z7 testa, the contents of petunidin 3-O-glucoside and cyanidin O-acetylhexoside increased by 15.23 times and 14.72 times compared to S7, respectively (Table S2). The results suggested that the content of cyanidin O-diacetyl-hexoside-O-glyceric acid possibly leads to the color difference of peanut testa in the R4 stage, but petunidin 3-O-glucoside and cyanidin O-acetylhexoside possibly play an important role in the testa color formation in the R7 stage.

Transcriptome Analysis of Shanhua15 and Zhonghua12.
In order to further analyze the regulation mechanism of flavonoids and anthocyanin biosynthesis in the process of testa color formation, transcriptome sequencing was performed at the R4 and R7 stages using testa samples from Shanhua 15 and Zhonghua 12. With three biological replicates for each sample, twelve cDNA libraries were constructed and sequenced on the HiSeq4000 platform. The original sequencing data were stored in the NCBI Short Read Archive database (SRA, http://www.ncbi.nlm.nih.gov/ Traces/sra_sub/sub.cgi) with the BioSample accession number SRP271546 (BioProject ID: PRJNA638812). After removing the adaptors and low-quality reads, 57140101, 59405717, 56610333, and 48707327 clean reads were obtained from S4, S7, Z4, and Z7 libraries, respectively. Q20 value, proportion of nucleotides with quality value >20, was above 97% (Table S3), indicating that the RNA-Seq results were of high quality and suitable for follow-up analysis. Finally, 51299 genes were subsequently assembled from the four libraries.
In the S7-vs-Z7 group, ontology analysis showed that 7198, 3104, and 7284 genes were assigned to cell component, molecular functional, and biological process class, respectively ( Figure S1). Based on enrichment results, 58 DEGs were enriched in the secondary metabolite biosynthetic process (GO:0044550), including 32 upregulated and 26 downregulated genes, and it was the largest group (Table S4). From COG annotation, 1781 DEGs in the S7vs-Z7 group were classified into 25 categories, and the largest class was the general functional cluster prediction only (318 genes, 17.85%), followed by signaling mechanism (219 genes, 12.30%), secondary metabolites biosynthesis, transport, and catabolism (156 genes, 8.76%) ( Figure S2).
The enriched metabolic pathways analysis showed 1349 DEGs from the S4-vs-Z4 group were enriched into 280 KEGG pathways, while 679 DEGs from S7-vs-Z7 were enriched into 268 pathways. Five categories, cellular processes, environmental information processing, genetic infor-mation processing, metabolism, and organismal systems, were revealed from KEGG classification analysis. Pathways of biosynthesis of secondary metabolites, flavonoid, and cell cycle-yeast were significantly changed at the R4 stage, while circadian rhythm-plant, biosynthesis of flavonoid, secondary metabolites, and phenylpropanoid were the significantly changed pathways at the R7 stage (Table 1). Table S2, a large number of flavonoids were detected in the testa of Shanhua15 and Zhonghua12. We finally rearranged 8 major compounds of flavonoids (Figures 2(e) and 2(f)). A total of 34 DEGs detected in S4-vs-Z4 were associated with flavonoid and anthocyanin biosynthesis, and 16 related DEGs were detected in S7-vs-Z7 (Table S5).

Differential Expression Genes Related to Flavonoid and Anthocyanin Biosynthesis. As shown in
Compared to Shanhua15, 24 upregulated and 10 downregulated genes were identified in Zhonghua12 at the R4 stage, indicating that flavonoids and anthocyanidins biosynthesis pathways were enhanced by upregulated gene LGAM8W), and FNS gene (arahy.Q2BYIT) were downregulated. Interestingly, many related genes had no significant difference in expression at the R7 stage, including genes C4H, CHI, F3H, ANS, ANR, and LAR. Nine genes were detected to be upregulated at the R7 stage, including CHS, F3′H, DFR, FLS, FNS, and four 4CL genes (Table S5). Among these upregulated genes, CHS (arahy. CM90T6) was upregulated by 4.97fold, and it catalyzes the transformation of p-coumaryl-coA to both naringenin chalcone and isoliquiritigenin.
Then, isoliquiritigenin is finally catalyzed to formononetin 7-O-glucoside, and this may be the reason for its high accumulation in Z7 testa of Zhonghua12. The F3 ′ H genes (arahy. 8F7PE4 and arahy. K8H9R8) catalyze the conversion of dihydrotricetin to dihydromyricetin and finally to petunidin 3-O-glucoside, and they have high expression in Z4 testa. The DFR enzyme catalyzes different types of substrates to synthesize leucodelphinidin and leucocyanidin, and the corresponding genes (arahy. LDV9QN and arahy. X8EVF3) showed 7.48-and 1.71-fold increments in Z4 compared to S4 (Figure 4, Table S5).

Transcription Factors Involved in Flavonoids and
Anthocyanin Biosynthesis. Except for functional genes, transcription factors (TFs) play an important role in regulating    International Journal of Genomics most metabolic pathways. Studies have shown that the biosynthesis of flavonoid and anthocyanin is regulated by three types of TFs, including R2R3-MYB factor, basic helix-loophelix (bHLH) proteins, and WD40 proteins [18]. Compared with Shanhua15, 425 and 237 TFs genes were found to be significantly changed in Zhonghua12 testa at R4 and R7 stages, respectively. Meanwhile, compared with R4, 354 differentially expressed TFs genes were detected in Shanhua15 at the R7 stage and 671 for Zhonghua12 ( Table 2, Table S6). MYB and bHLH have been confirmed to play important roles in regulating the expression of genes involved in flavonoid and anthocyanin biosynthesis in many plants [28][29][30]. In the S4-vs-Z4 group, 51 MYB TF genes were significantly changed, among which 35 genes were upregulated and 16 were downregulated in Zhonghua12. In the S7-vs-Z7 group, 30 MYB TF genes were differentially expressed in Zhon-ghua12, including 18 upregulated and 12 downregulated genes. As the seed developing from R4 to R7, 46 (10-up and 36-down) and 70 (15-up and 55-down) MYB TF genes were significantly changed in Shanhua15 and Zhonghua12 testa, respectively (Table 2). Moreover, 41 bHLH genes showed expression differences at the R4 stage, including 24 highly upregulated and 17 downregulated genes in Zhon-ghua12, while 29 bHLH DEGs were identified at the R7 stage, including 10 upregulated and 19 downregulated genes. With the development of peanut testa in one cultivar, 37 bHLH DEGs were identified in Shanhua 15, of which 8 were highly upregulated at the R7 stage, while 59 bHLH DEGs were detected in Zhonghua12, including 15 upregulated and 44 downregulated genes at the R7 stage (Table 2).
We further recruited five R2R3-MYBs, three bHLH, and one WD40 DEGs ( Figure 5). The five MYBs-genes arahy.A2IWKV, arahy.US2SKM, arahy.SJGE27, arahy.H8DJRL, and arahy.PR7AYB were annotated in seed coat development, anthocyanin regulatory, and flavonol biosynthetic process. The expression of all five genes was highly upregulated at the R4 stage in Zhonghua12. Protein sequence comparison revealed that arahy.US2SKM and arahy.SJGE27 was homologous (99%) to GmMYB58 from Glycine max, arahy.H8DJRL, and arahy.PR7AYB clustered together with Glycine max GmMYB12b, while arahy.A2IWKV was closely related to Arabidopsis thaliana AtMYB5 (Figure 5(b)). Protein sequence alignment showed five genes were identified as typical R2R3 MYB proteins ( Figure 5(a)), which might play an important role in peanut testa pigmentation regulation. The bHLH gene arahy.HM1IVV was clustered with AtTT8 and BjTT8 ( Figure 5(d)). Three bHLH genes contain N-terminal MYB interaction region, while arahy.26781 N and arahy.HM1IVV contain a basic helix1-loop-helix ( Figure 5(c)). The protein sequence of WD40 gene-arahy.L6JJW9 had a higher sequence similarity to Physcomitrella patens PpWD ( Figure 5(f)), which contained four highly conserved WD-repeat motifs ( Figure 5(e)). The transcript showed that three types of TF genes have high expression levels at the R4 stage of seed development, resembling the expression pattern of most structural genes. That

10
International Journal of Genomics indicated MYB forms a ternary complex with bHLH and WD40 to activate genes related to anthocyanin biosynthesis.
3.6. qRT-PCR Validation of Gene Expression. A total of nine DEGs (Table S1) involved in flavonoid biosynthesis, seven functional genes, and two TF genes were randomly selected for qRT-PCR validation. As shown in Figure 6, the relative gene expression of nine test genes was all consistent with the RNA-seq results; thus, we believe that the transcriptome data is of high accuracy and reliability.

Discussion
Peanut contains many important nutrients, and the testa color affects its nutritional and commercial value. The antioxidant capacity of peanut testa is much higher than that of cotyledons [31]. For example, proanthocyanidin B is widely distributed in peanut testa, and it is a novel allosteric AKT inhibitor, which not only has antioxidant and antiinflammatory properties but also has a strong antitumor efficacy [32]. In recent years, consumers prefer peanuts with dark testa due to their health benefits from the higher content of anthocyanins and other flavonoids [33]. In addition, aflatoxin resistance is also related to the biosynthesis pathway of flavonoids and phenylpropyl [34]. At present, the testa of most peanut varieties on the market is pink, and some varieties are red. Previous studies on peanut testa pig-mentation were limited to genome and transcriptome level, for example, the differences of testa transcriptome in three growth stages of pink peanut varieties were analyzed by RNA-Seq [21]. In this paper, the combined analysis of transcriptome and metabolome provided a new approach to reveal the molecular mechanism of pigment formation in peanut testa. The anthocyanin content and composition are important factors for different colors of plant tissues or organs. We detected the total anthocyanin content of testa of two peanut varieties during seed development, and the results showed that the total anthocyanin content of two varieties both increased from R2 to R7 stage, suggesting that testa pigments gradually accumulated with seed development. Moreover, the total anthocyanin content of red testa variety Zhonghua 12 was significantly higher than that of pink testa variety Shanhua 15 (Figure 1). Anthocyanins are considered to be the most important pigments in most plants [35]. However, previous studies have shown that different plants have different anthocyanins for pigmentation. During the process of red pigment formation in jujube peel, delphinidin, malvidin 3-O-glucoside, and delphinidin 3-O-glucoside were considered to be the key anthocyanins [36], but cyanidin O-malonylhexoside was the first anthocyanin identified in purple mutant figure [12]. When comparing different asparagus varieties, peonidin, cyanidin, and their glycoside derivatives were the major anthocyanins [1]. The strong

11
International Journal of Genomics red color of longan was due to the abundant contents of cyanidin 3-O-glucoside, cyanidin 3-O-6 ″ -malonyl-glucoside, and cyanidin O-syringic acid found in its pericarp [37]. In order to clarify the pigmentation mechanism of peanut testa and analyze the metabolite basis of the difference between pink and red testa, the composition and content of testa metabolites at the R4 and R7 stage of Shanhua 15 and Zhonghua 12 were determined using the UPLC-MS-/MS-based approach. The metabolome analysis showed that a total of 198 flavonoid-related SCMs were identified, and the main pigment components of peanut red testa were petunidin 3-O-glucoside and cyanidin O-acetylhexoside. This study further illustrates the primary anthocyanins that affect seed color in plant are petunidin derivatives and cyanidin derivatives.
Flavone plays an important role in the yellow pigment formation of many plants [38,39]. Interestingly, the contents of tricin O-malonylhexoside, naringenin C-hexoside, and apigenin 7-O-glucoside were extremely higher in the Shan-hua15 S7 sample, and they were all over 1:00 + E04 times higher than that of Zhonghua12 Z7, especially for tricin Omalonylhexoside, which was more than 1:00 + E06 times higher. Then, we noticed that the endotesta of Shanhua15 was yellow, while it was white in Zhonghua12, suggesting that high content of Tricin O-malonylhexoside, naringenin C-hexoside, and apigenin 7-O-glucoside maybe give a yellow coloration in peanut endotesta. In this study, changes of many other metabolites in peanut testa were also investigated, many of which were not directly related to testa pigmentation, but correlated with human health. For example, formononetin 7-O-glucoside (Ononin) can inhibit the proliferation of tumor cells and promote apoptosis and reduce cell invasion and migration [40], and its content was significantly higher in Zhonghua12 testa with more than 2000-fold than Shanhua15. Besides, persicoside content was very high at both the R4 and R7 stages of Zhonghua12 testa, and it has potential radical scavenging activity [41]. To sum up, peanut with red testa contains more ingredients that are beneficial to human health.
Based on the identification of SCMs, the gene expression profile of peanut testa was performed between two varieties at R4 and R7 stages. Compared with Shanhua 15 S4 and S7, thousands of significant DEGs were identified in Zhonghua 12, and many of them were enriched in the secondary  Figure 6: qRT-PCR verification of DEGs between S4, Z4, S7, and Z7. Transcript levels of 9 genes, which were involved in flavonoid and anthocyanin biosynthesis. The y-axis shows the relative gene expression levels analyzed by qRT-PCR and RNA-Seq. The x-axis shows the testa sample of developmental stages. In this figure, gray columns correspond with expression data of qRT-PCR, while red lines denote data of RNA-Seq. S4, Shanhua15 R4 pod period; Z4, Zhonghua12 R4 pod period; S7, Shanhua15 R7 pod period; Z7, Zhonghua12 R7 pod period. 12 International Journal of Genomics metabolite biosynthetic process and flavonoid biosynthesis pathway through GO and KEGG analysis, suggesting that these genes are likely to be related to pigmentation in pink and red testa. The pigment of peanut testa accumulated gradually with seed development, and the coloring was basically completed at the R7 stage. It was found that the genes involved in flavonoid and anthocyanidin biosynthesis, PAL, C4H, CHS, F3H, F3 ′ H, DFR, ANS, ANR, LAR, and FLS, were significantly upregulated in Zhonghua12 testa at the R4 stage, but most of them were not significantly changed at the R7 stage. These results suggested that changes in gene expression occurred much earlier than phenotypic changes.
In the early stage of peanut seed development, although the seed coat has not been completely stained, the genes related to flavonoid biosynthesis have been actively expressed, similar to studies on blueberry and pear that genes involved in anthocyanin and flavonoid synthesis were also highly expressed in the early stages of fruit development [42,43]. In addition, affected by the active expression of related genes, a large number of flavonoid molecules have been accumulated in the white-period of jujube peel, and with the fruit ripening, the early actively expressed genes gradually become silenced [36]. Therefore, genes PAL, C4H, F3H, F3 ′ H, DFR, ANS, ANR, LAR, and FLS are likely to play important roles in the early anthocyanin accumulation. Besides, CHS has been verified to be related to the red coloration in crabapple varieties [44], and it was significantly inhibited in whiteflowered mustards, while other genes in the anthocyanin biosynthesis pathway were not significantly different from colored individuals [45]. The high expression of CHS was also related to higher anthocyanin content in peach, and the results were also verified in 30 peach varieties [24]. In our study, CHS (arahy. CM90T6) had higher expression in Zhonghua12 and was upregulated by 4.97-fold in Z7, indicating that the CHS gene may be related to testa pigmentation. The expression of F3 ′ H genes determined which component of anthocyanin will be synthesized in grape and rose [46,47]  13 International Journal of Genomics arahy. K8H9R8) have a high expression level in Z4 testa, which catalyze the conversion of dihydrotricetin to dihydromyricetin or dihydrokaempferol to dihydroquercetin. The DFR enzyme can catalyze different types of substrates to synthesize leucodelphinidin and leucocyanidin. DFR genes (arahy. LDV9QN and arahy. X8EVF3) showed 7.48-and 1.71-fold increments in Zhonghua12 compared to Shan-hua15 in the R4 period. These DEGs may be the critical factor for the high accumulation of dihydromyricetin, cyanidin O-acetylhexoside, and petunidin 3-O-glucoside in Zhon-ghua12 (Figure 7).
In addition to those functional genes, TFs have also been confirmed to play a critical role in the regulation of anthocyanin biosynthesis, including MYB, bHLH, and WD40 [44,48]. For example, MdMYB1 and MdMYB3 were the main regulators for anthocyanin biosynthesis and fruit coloring in apple [49,50], and PpMYB10 and PpMYB114 regulate anthocyanin biosynthesis in pear [38]. In peanut, an R2R3-MYB gene was identified previously, regulating purple testa formation [19]. In the present study, many MYBs were detected to be differentially expressed between Shanhua 15 and Zhonghua 12 (Table S5). Among them, five MYB genes were annotated in seed coat development, anthocyanin regulatory, and flavonol biosynthetic process. The expression of all five genes was highly upregulated at the R4 stage in Zhonghua12. Protein sequence comparison revealed that the genes clustered together with GmMYB58, AtMYB5, and GmMYB12b, which have a major role in regulating testa developmental process in Glycine max and Arabidopsis thaliana [51][52]. The MYBs had similar expression patterns to structural genes, and their expression trends were closely related to the contents of petunidin 3-O-glucoside and cyanidin O-acetylhexoside. We also identified three differentially expressed bHLH (arahy.26781N, arahy.HM1IVV, and arahy.MP3D3D) and WD40 (arahy.L6JJW9) transcription factor genes that may be involved in anthocyanin biosynthesis in peanut testa. The expression patterns of MYB, bHLH, and WD40 was consistent with flavonoid synthesis DEGs (Figure 7). This implies that control of the biosynthesis of the pigments in peanut testa may occur through transcriptional regulation by the MBW complex. The detailed functional analysis needs further verification.

Conclusions
In summary, flavonoid metabolites and related genes in pink and red testa of peanut were identified through combined transcriptome and metabolome analysis. The accumulation of petunidin 3-O-glucoside and cyanidin O-acetylhexoside in the late stage of seed development is the main reason for the red testa appearance. In addition, the genes PAL, C4H, CHS, F3H, F3 ′ H, DFR, ANS, LAR, and ANR in the flavonoid biosynthesis pathway were highly expressed in the early stage, playing an important role in anthocyanin accumulation. CHS gene arahy.CM90T6, F3 ′ H genes (arahy. 8F7PE4 and arahy. K8H9R8), and DFR genes (arahy. LDV9QN and arahy. X8EVF3) may be the key functional genes controlling the formation of pink and red testa in peanut. Transcriptomic analysis indicated MYB, bHLH, and WD40 in the biosynthetic pathway of anthocyanin may be the key regulatory genes in pink and red pigment formation. This is a comprehensive analysis on flavonoid metabolites and related genes expression in peanut testa, providing a reference for revealing the regulatory mechanism of pigment accumulation in peanut testa.

Data Availability
The data used to support the findings of this study are included within the article. The data used to support the findings of this study are included within the supplementary information files.