Multi-Omics Analysis Revealed the AGR-FC.C3 Locus of Brassica napus as a Novel Candidate for Controlling Petal Color

Variations in the petal color of Brassica napus are crucial for ornamental value, but the controlled loci for breeding remain to be unraveled. Here, we report a candidate locus, AGR-FC.C3, having conducted a bulked segregant analysis on a segregating population with different petal colors. Our results showed that the locus covers 9.46 Mb of the genome, harboring 951 genes. BnaC03.MYB4, BnaC03.MYB85, BnaC03.MYB73, BnaC03.MYB98, and BnaC03.MYB102 belonging to MYB TFs families that might regulate the petal color were observed. Next, a bulk RNA sequencing of white and orange-yellow petals on three development stages was performed to further identify the possible governed genes. The results revealed a total of 51 genes by overlapping the transcriptome data and the bulked segregant analysis data, and it was found that the expression of BnaC03.CCD4 was significantly up-regulated in the white petals at three development stages. Then, several novel candidate genes such as BnaC03.ENDO3, BnaC03.T22F8.180, BnaC03.F15C21.8, BnaC03.Q8GSI6, BnaC03.LSD1, BnaC03.MAP1Da, BnaC03.MAP1Db, and BnaC03G0739700ZS putative to controlling the petal color were identified through deeper analysis. Furthermo re, we have developed two molecular markers for the reported functional gene BnaC03.CCD4 to discriminate the white and orange-yellow petal colors. Our results provided a novel locus for breeding rapeseed with multi-color petals.


Introduction
B. napus is an important oilseed crop, supplying a primary source of edible vegetable oils and high-protein animal feeds [1][2][3].The petal color is an essential indicator of agronomic traits in rapeseed, and flowers with rich colors have played a vital role in attracting insects and other pollinators for adequate pollination [4,5].Additionally, the petal color in B. napus also holds ornamental value, and economically beneficial flowers with specific colors have been obtained [6][7][8][9].
Flower coloration is primarily determined by the content of carotenoids, flavonoids, and betalains [4].Carotenoids are responsible for developing yellow, orange, and red flower colors [4,10].The enzymes responsible for cleaving specific double-bond regions of carotenoids are known as carotenoid cleavage oxygenases (CCOs) or carotenoid-cleaving dioxygenases (CCDs) [5].For instance, the CmCCD4a gene is expressed explicitly in chrysanthemum petals, degrading carotenoids into colorless compounds and changing petal color from yellow to white [11].Similarly, the flesh was yellow due to carotenoid accumulation Plants 2024, 13, 507 2 of 19 in the PpCCD4 mutant of peaches [12].Mutations in the CrtR-b gene, which explicitly expresses β-carotene hydroxylase in tomato flowers, significantly decrease carotenoid levels and result in white petals [13].The carotenoid isomerase (CRTISO) is crucial in carotenoid synthesis by converting the prolycopene to all-trans lycopene [14]; it was found that the pro-lycopene was accumulated in the mutation of CmCRTISO and led to the petal color changing from yellow to light yellow in melon [15].The function of BrCRTISO1 has also been confirmed to control the yellowish color of petals in B. rapa [16].A significant reduction in total carotenoid content resulted in a creamy-white phenotype in petals through the site-directed mutagenesis of BnaCRTISO in B. napus [17].
Flavonoids play an important role in forming petal color [18].Anthocyanins (a subclass of flavonoids) are widespread in vascular plants and are responsible for various colors in flowers, such as orange, red, purple, and blue [4].Manipulating or modifying genes that are associated with anthocyanin synthesis or its precursors has been successfully used in altering the petal color in several plant species, for example, in Torenia hybrida, Petunia hybrida, Nicotiana tabacum, and Eustoma grandiflorum (Raf.)Shinn [19][20][21][22][23].In red-flowered rapeseed, various colors ranging from raspberry red to beige red and zinc yellow can be achieved via RNA interference to target the anthocyanidin synthase (ANS) BnaA03.ANS gene at different levels [24].
Other regulation factors have also been uncovered to influence petal color.In B. juncea, the stable storage of carotenoids was disrupted in the mutation of BjA02.PC1 and BjB04.PC2 (belonging to the esterase/lipase/thioesterase family of acyltransferases) led to a white-flowering phenotype [25].Additionally, a specific rapeseed line with a pair of Orychophragmus violaceus chromosome M4 exhibited reddish petals [26], and further analysis showed that the OvPAP2 transcript of O. violaceus is expressed specifically in rapeseed petals [27].
Bulked segregant analysis (BSA) is a commonly used method that identifies the molecular markers associated with a specific gene or genomic region [28].For example, CCD4, which had functions for controlling the petal color, was identified by combining BSA and positional cloning in B. napus [29].Additionally, a transcriptomic and metabolomic analysis revealed that the reduced lutein and zeaxanthin are the primary reason for the appearance of white-flowered rapeseed [29], and the related genes (BnaA09.ZEP and BnaC09.ZEP) that encoded zeaxanthin epoxidase (ZEP) were identified using BSA in B. napus [30].
The molecular mechanisms underlying flower coloration are still not fully understood.In this study, F 1 was first obtained by hybridizing the B. napus with orange-yellow petals (YP) and white petals (WP).Then, an F 2 segregation population with a series of petal colors was obtained.The extreme material pools with WP and YP in F 2 were selected for BSA-Seq and RNA-Seq analyses, and the associated genomic region (AGR) of AGR-FC.C3 that might control the petal colors and some key genes involved in carotenoid pathways were identified.These findings could enhance the breeding process of ornamental rapeseed and provide a theoretical foundation for future breeding with white and orange-yellow flowers in B. napus.

BSA for Petal Colors in B. napus
The reciprocal hybridization between B. napus 16214 and 16287 was performed (Figure 1A).It was revealed that all F 1 plants, no matter 16214 × 16287 or 16287 × 16214, had light-orange petals.The F 1 seeds were subsequently sown in a field to yield the F 2 generation.Further observation revealed that five distinct petal colors (white, creamy-white, light-orange, yellow, and orange-yellow) existed in the F 2 generation (Figure 1B).Further analysis showed that the content of carotenoid in YP was significantly higher than in WP, particularly in S3, which was approximately three-fold more elevated than that of the WP (Figure 1C).analysis showed that the content of carotenoid in YP was significantly higher than in WP, particularly in S3, which was approximately three-fold more elevated than that of the WP (Figure 1C).To investigate the genetic basis underlying the color difference, BSA was conducted on extreme pools with WP and YP in F2 lines, respectively.A total of 147.32 GB of clean data was obtained with a Q30 of 94.48% (Table S1).Each sample showed an average alignment rate of 93.50% to the reference genome, with an average 28× coverage depth and genome coverage of 94.90%, which satisfied the subsequent analysis requirements (Table S2).Genome-wide, a total of 668,993 single nucleotide polymorphisms (SNPs) and 668,993 insertion-deletions (InDels) between 16214 and 16287 were identified (Figure 2, Tables S3  and S4).It was revealed that a significant genomic region of 9.46 Mb (63.03-72.49Mb) on ChrC03 was associated with petal color (Figures 2 and 3A).Since the Darmor_bzh genome has a gap in this region, this region in the Darmor_bzh was also mapped to the genome of Zhongshuang 11 (ZS11) (Figure 3B).Further analysis showed that a significantly To investigate the genetic basis underlying the color difference, BSA was conducted on extreme pools with WP and YP in F 2 lines, respectively.A total of 147.32 GB of clean data was obtained with a Q30 of 94.48% (Table S1).Each sample showed an average alignment rate of 93.50% to the reference genome, with an average 28× coverage depth and genome coverage of 94.90%, which satisfied the subsequent analysis requirements (Table S2).Genome-wide, a total of 668,993 single nucleotide polymorphisms (SNPs) and 668,993 insertion-deletions (InDels) between 16214 and 16287 were identified (Figure 2, Tables S3 and S4).It was revealed that a significant genomic region of 9.46 Mb (63.03-72.49Mb) on ChrC03 was associated with petal color (Figures 2 and 3A).Since the Darmor_bzh genome has a gap in this region, this region in the Darmor_bzh was also mapped to the genome of Zhongshuang 11 (ZS11) (Figure 3B).Further analysis showed that a significantly associated region called AGR-FC.C3 with a ∆(SNP-index) > 0.715 and p-value < 0.01 within this genomic region was detected (Figure 3C).associated region called AGR-FC.C3 with a Δ(SNP-index) > 0.715 and p-value < 0.01 within this genomic region was detected (Figure 3C).A total of 951 genes were found within the interval of AGR-FC.C3, and 71 were classified as TFs, in which the bHLH, Dof, MYB, and bZIP families were found to be the most abundant (Figure S1).TFs BnaC03.MYB4, BnaC03.MYB73, BnaC03.MYB98, BnaC03.MYB102, and BnaC03.MYB85, which regulate the biosynthesis of carotenoids and flavonoids, were observed, which might play important roles in petal color formation (Table S5).Additionally, the 951 genes within the ChrC03 interval were performed the Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis, and several pathways, including the cysteine and methionine metabolism, RNA polymerase, taurine, and hypotaurine metabolism, and the citrate cycle were enriched (Figure S2).A total of 951 genes were found within the interval of AGR-FC.C3, and 71 were classified as TFs, in which the bHLH, Dof, MYB, and bZIP families were found to be the most abundant (Figure S1).TFs BnaC03.MYB4, BnaC03.MYB73, BnaC03.MYB98, BnaC03.MYB102, and BnaC03.MYB85, which regulate the biosynthesis of carotenoids and flavonoids, were observed, which might play important roles in petal color formation (Table S5).Additionally, the 951 genes within the ChrC03 interval were performed the Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis, and several pathways, including the cysteine and methionine metabolism, RNA polymerase, taurine, and hypotaurine metabolism, and the citrate cycle were enriched (Figure S2).

RNA-Seq and Differentially Expressed Genes (DEGs) Analysis between the YP and WP in B. napus
To elucidate the underlying mechanisms responsible for color variation in petals, the YP and WP of S1, S2, and S3 were subjected to transcriptomic analysis.A total of 509.62 Gb of raw reads were obtained, with all samples exhibiting Q30 scores exceeding 93% (Table S6).The quality-controlled clean reads were then aligned to the ZS11 genome, with each sample achieving a mapping ratio exceeding 83% (Table S7).The expression levels (log 10 (FPKM + 1)) of each sample were mainly distributed between 0 and 1 (Figure S3).Pearson correlation coefficients indicated a significant correlation between YP and WP at the same developmental stage (Figure S4A).Substantial differences were observed between petals at different developmental stages (Figure S4A).Principal component analysis (PCA) further revealed differences between YP-S3 and WP-S3.In contrast, no significant differences were observed during S1 and S2 (Figure S4B).Six DEGs were randomly selected for reverse transcription quantitative real-time PCR (RT-qPCR) analysis to verify the accuracy of the RNA-Seq.The consistent expression tendency existed with the transcriptomic results (Figure S5).A total of 8058 DEGs obtained in S1, S2, and S3 were performed for trend analysis, and six significant trends were found.Interestingly, regardless of the color of the petals, whether orange-yellow or white, the expression patterns were identical in cluster 2, cluster 4, and cluster 6 (Figure 4C).The gene expression levels in cluster 2 were relatively higher in S1 and gradually decreased with the petal's development.However, somewhat higher expression levels were observed in WP-S1, which indicated a gradual decline in gene expression with the petal's development (Figure 4C).Similarly, the gene expression trend in cluster 4 resembled that of cluster 2, except higher expression levels were observed in YP-S1.Moreover, the gene expression trend in cluster 6 was consistent in both the WP and YP A total of 8058 DEGs obtained in S1, S2, and S3 were performed for trend analysis, and six significant trends were found.Interestingly, regardless of the color of the petals, whether orange-yellow or white, the expression patterns were identical in cluster 2, cluster 4, and cluster 6 (Figure 4C).The gene expression levels in cluster 2 were relatively higher in S1 and gradually decreased with the petal's development.However, somewhat higher expression levels were observed in WP-S1, which indicated a gradual decline in gene expression with the petal's development (Figure 4C).Similarly, the gene expression trend in cluster 4 resembled that of cluster 2, except higher expression levels were observed in YP-S1.Moreover, the gene expression trend in cluster 6 was consistent in both the WP and YP of S1, S2, and S3 (Figure 4C).
A Gene Ontology (GO) enrichment analysis of these 51 candidate genes revealed that the significantly enriched terms were related to N-terminal protein amino acid modification, metalloexopeptidase activity, and polysomal ribosome (Figure 5C).BnaC03.MAP1Da and BnaC03.MAP1Db were enriched in the N metabolism and metalloexopeptidase activity, while BnaC03.RPL18AB, BnaC03.RPS12C, and BnaC03.RPL8C were enriched in polysomal ribosome.Moreover, a KEGG enrichment analysis indicated that the significantly enriched pathways were the Vitamin B6 metabolism, ribosome biogenesis, selenocompound metabolism, and carotenoid biosynthesis (Figure 5D).

The Carotenoid Pathway Influences the Petal Colors in B. napus
Based on the analysis of WP and YP, the genes related to TFs, the phenylalanine pathway, carotenoid pathway, flavonoid pathway, 2-C-methyl-D-erythritol 4-phosphate (MEP) pathway, and xanthophylls pathway were investigated to construct a regulation network.MYB4 and MYB85 might be at the center of regulating petal color by affecting the phenylalanine, carotenoid, and flavonoid pathways (Figure 6).BnaC03.MYB4 and BnaC03.MYB85 were located in the interval of AGR-FC.C3 and were considered valuable candidates.In the network, it was shown that MYB4 and MYB85 might regulate the expressions of CCD4, nine-cis-epoxy carotenoid dioxygenase2 (NCED2), nine-cis-epoxy carotenoid dioxygenase3 (NCED3), beta-carotene hydroxylase (β-HOASE), and phytoene synthase 1 (PSY1) to influence petal color (Figure 6).Previous studies have reported that the carotenoid pathway plays a crucial regulatory role in petal color formation.Therefore, the expressions of carotenoid pathway genes in YP and WP were analyzed in this study.This revealed the alterations in the expression of multiple genes encoding enzymes involved in carotenoid biosynthesis (Figure 7).The PSY can promote the geranylgeranyl diphosphate (GGPP) synthesis of phytoene, and it was shown that the BnaA10.PSY (BnaA10G0196500ZS) was up-regulated in YP-S2, which means more phytoene may be synthesized in YP.Multiple DEGs were found in the βcarotene degradation pathway, such as the expressions of BnaA08.CCD4 (BnaA08G0113600ZS) and BnaC03.CCD4 in WP being much higher than those of YP in S1, S2, and S3, suggesting that the carotenoid degradation was the reason for the petal color turning from yellow to white.BnaC09.β-Hoase(BnaC09G0345900ZS) could inhibit the conversion of β-carotene to zeaxanthin, and it was found to be down-regulated in YP-S3.Furthermore, the expression of NCED involved in zeaxanthin generation was also detected; BnaA01.NCED2 (BnaA01G0092600ZS) was down-regulated in YP-S2 and up-regulated in YP-S3, while BnaA03.NCED2 (BnaA03G0447100ZS) and BnaA03.NCED3 (BnaA03G0338300ZS) were up-regulated in YP-S3, and BnaA01.NCED3 (BnaA01G0353200ZS) was upregulated in YP-S2 (Figure 7).In addition, no obvious DEGs related to the flavonoid pathway of BnaANS, dihydroflavonol-4-reductase (BnaDFR), flavanone 3-hydroxylase (BnaF3H), UDP-glycosyltransferase (BnaUFGT), and UDP-glucose: flavonoid 3-o-glucosyltransferase (BnaUF3GT) were detected in the YP or WP (Figure S8).Previous studies have reported that the carotenoid pathway plays a crucial regulatory role in petal color formation.Therefore, the expressions of carotenoid pathway genes in YP and WP were analyzed in this study.This revealed the alterations in the expression of multiple genes encoding enzymes involved in carotenoid biosynthesis (Figure 7).The PSY can promote the geranylgeranyl diphosphate (GGPP) synthesis of phytoene, and it was shown that the BnaA10.PSY (BnaA10G0196500ZS) was up-regulated in YP-S2, which means more phytoene may be synthesized in YP.Multiple DEGs were found in the β-carotene degradation pathway, such as the expressions of BnaA08.CCD4 (BnaA08G0113600ZS) and BnaC03.CCD4 in WP being much higher than those of YP in S1, S2, and S3, suggesting that the carotenoid degradation was the reason for the petal color turning from yellow to white.BnaC09.β-Hoase(BnaC09G0345900ZS) could inhibit the conversion of β-carotene to zeaxanthin, and it was found to be down-regulated in YP-S3.Furthermore, the expression of NCED involved in zeaxanthin generation was also detected; BnaA01.NCED2 (BnaA01G0092600ZS) was down-regulated in YP-S2 and up-regulated in YP-S3, while BnaA03.NCED2 (BnaA03G0447100ZS) and BnaA03.NCED3 (BnaA03G0338300ZS) were up-regulated in YP-S3, and BnaA01.NCED3 (BnaA01G0353200ZS) was upregulated in YP-S2 (Figure 7).In addition, no obvious DEGs related to the flavonoid pathway of BnaANS, dihydroflavonol-4-reductase (BnaDFR), flavanone 3-hydroxylase (BnaF3H), UDPglycosyltransferase (BnaUFGT), and UDP-glucose: flavonoid 3-o-glucosyltransferase (BnaUF3GT) were detected in the YP or WP (Figure S8).

Application of Linkage Marker of BnaC03.CCD4 in B. napus
The SNPs between BnaC03.CCD4 in the WP and YP were further compared, and several SNPs in the exon region were found, in which eight SNPs were found in the second exon (including two C-to-T changes, two G-to-A changes, two A-to-T changes, and A-to-G and C-to-G changes) (Figure S9).
To further investigate the sequence variations in BnaC03.CCD4 between 16214 and 16287, the whole gene length sequence of BnaC03.CCD4 was amplified from the genomic DNA for Sanger sequencing.A 7613 bp deletion and 50 bp insertion were identified in the first and second intron of BnaC03.CCD4 in 16214, which might influence the mRNA splicing (Figure 8A).A 700 bp insertion was detected in the promoter of BnaC03.CCD4 in 16214, resulting in a higher expression in WP than that in YP (Figures 8A and S5).
The SNPs between BnaC03.CCD4 in the WP and YP were further compared, and several SNPs in the exon region were found, in which eight SNPs were found in the second exon (including two C-to-T changes, two G-to-A changes, two A-to-T changes, and A-to-G and C-to-G changes) (Figure S9).
To further investigate the sequence variations in BnaC03.CCD4 between 16214 and 16287, the whole gene length sequence of BnaC03.CCD4 was amplified from the genomic DNA for Sanger sequencing.A 7613 bp deletion and 50 bp insertion were identified in the first and second intron of BnaC03.CCD4 in 16214, which might influence the mRNA splicing (Figure 8A).A 700 bp insertion was detected in the promoter of BnaC03.CCD4 in 16214, resulting in a higher expression in WP than that in YP (Figure 8A, Figure S5).
To rapidly identify B. napus individuals with different petal colors, a pair of primers was designed according to the sequence variations in BnaC03.CCD4 between 16214 and 16287, which could act as molecular markers (Figure 8B, Table S8).A band with 951 bp was found in 16214, the WP homozygote (BB), and the heterozygote when using primers of PF1/PR1, while no band was found in 16287 and the YP homozygote (AA) (Figure 8B).When using primers of PF2/PR2, a band with 695 bp was found in 16287, the YP homozygote (AA), and the heterozygote, while no band in 16214 and the WP homozygote (BB) (Figure 8B).The molecular markers of 16214.PF1/16214.PR1 and 16287.PF2/16287.PR2 were implied to distinguish the WP and YP (Figure 8B, Table S8).To rapidly identify B. napus individuals with different petal colors, a pair of primers was designed according to the sequence variations in BnaC03.CCD4 between 16214 and 16287, which could act as molecular markers (Figure 8B, Table S8).A band with 951 bp was found in 16214, the WP homozygote (BB), and the heterozygote when using primers of PF1/PR1, while no band was found in 16287 and the YP homozygote (AA) (Figure 8B).When using primers of PF2/PR2, a band with 695 bp was found in 16287, the YP homozygote (AA), and the heterozygote, while no band in 16214 and the WP homozygote (BB) (Figure 8B).The molecular markers of 16214.PF1/16214.PR1 and 16287.PF2/16287.PR2 were implied to distinguish the WP and YP (Figure 8B, Table S8).

Discussion
Flower color is one of the most critical traits in B. napus, and many researchers have attempted to explore the genetic mechanisms of flower color formation in recent years [25,[29][30][31].However, the molecular mechanisms affecting flower color are still insufficient.
To further understand the regulatory mechanism of petal color, QTL, BSA, transcriptomics, and metabolomics have been used to identify the key factors that affect the petal color formation previously.The candidate gene of BnaPAP2.A7b, located within 219.81 kb on chromosome A07, was identified through a combination of BSA and QTL fine mapping in B. napus and the ectopic expression of the OvPAP2 in Orychophragmus violaceus could form red-flowered rapeseed [27].It was found that a 211 bp insertion in the promoter of BnaPAP2.A7b could cause a change in flower color from red to orange in rapeseed [32].In this study, a significant locus, AGR-FC.C3 on ChrC03, which covers 9.46 Mb, was obtained through a BSA on WP and YP.Moreover, the number of DEGs between the WP and YP varied greatly in different developmental stages.A co-analysis of BSA and transcriptomic results showed that 51 genes were obtained, as well as some candidate genes, such as BnaC03.ENDO3, BnaC03.T22F8.180,BnaC03.F15C21.8,BnaC03.Q8GSI6, BnaC03.LSD1, BnaC03.MAP1Da, BnaC03.MAP1Db, and BnaC03G0739700ZS that have not been reported to be associated with flower color.BnaC03.ENDO3 is orthologous to the Arabidopsis gene ENDONUCLEASE 3, which encodes a putative endonuclease and expressed in the floral meristem and stamen development; this gene positively regulates the anthocyanin metabolism [33].BnaC03.T22F8.180 is orthologous to the Arabidopsis gene AT4G39280, which is encoded in the phenylalanyl-tRNA synthetase [34].BnaC03.F15C21.8 is orthologous to the Arabidopsis gene AT1G36280, which encodes an L-Aspartase-like family protein [35].BnaC03.Q8GSI6 is a TRAM, LAG1, and CLN8 (TLC) lipid-sensing domaincontaining protein, which is orthologous to Arabidopsis gene AT4G19645.However, there are no reports about the gene BnaC03.Q8GSI6 and its Arabidopsis orthologous gene AT4G19645.BnaC03.MAP1Da and BnaC03.MAP1Db, orthologous to AT4G37040, encode a methionine aminopeptidase, and are enriched in the N metabolism and metalloexopeptidase activity [36].BnaC03.LSD1, orthologous to AT4G20380, contains zinc-finger motifs and negatively regulates a basal defense pathway [37].These results indicated the complex pathways for flower color regulation in B. napus.
Carotenoids have been considered one of the main factors that affect petal colors [17,31,38].The previous research showed that the total carotenoid content in yellowflowered rapeseed was much higher than that in the white-flowered inbred lines [39].In this study, KEGG analysis also indicated that the carotenoid biosynthesis was enriched, and the total content of carotenoids in the YP was significantly higher than that in WP at the S2 and S3 stages, consistent with the previous studies [17,38].The co-analysis of metabolomic and transcriptomic results revealed that γ-carotene might be the substance that influenced the pigment deposition in petals of Liriodendron tulipifera, and it indicated that the CRTISO and ε-lycopene cyclase (ε-LCY) were the core enzymes that encoded the specific orange pigment deposition in petals [40].However, in the present study, the expression of genes that encode the CRTISO and ε-LCY enzymes between the WP and YP has not changed.Through map-based cloning, it was shown that the absence of BnaC09.ZEP and a 1695 bp deletion in BnaA09.ZEP could lead to a flower color change from yellow to orange [30].It was also shown that the mutation of BnaCRTISO could cause significant down-regulations in the essential genes of BnaPSY and BnaC4H that are involved in carotenoid and flavonoid synthesis.Still, the expressions of some other genes that relate to carotenoid and xanthophyll synthesis (such as BnaPDS3, BnaZEP, and BnaBCH1) were up-regulated [17].In the present study, the expression of BnaA10.PSY was up-regulated in YP-S2, which could lead to a higher production of phytoene through the promotion of carotenoid synthesis.
CCD4 has been shown to play an important role in flower color formation [11,29,38].It was found that the BnaC03.CCD4 is preferentially expressed in the petals of white-flowered rapeseed, and a CACTA-like transposable element 1 (TE1) insertion into its coding region led to the reduced expression in yellow-flowered rapeseed [29].In the present results, we found that the BnaC03.CCD4 is detected in the interval of AGR-FC.C3, and the expression in the WP was significantly higher than that in YP.The highly expressed BnaC03.CCD4 could promote the degradation of carotenoid in petals, and then result in a white flower color.In addition to CCD4, other studies reported that NCED also plays an important role in flower color formation [39,41].It was shown that the expression of BnNCED4b in white petals is significantly higher than that in yellow petals in rapeseed, which indicated that it might also play an important role in white petal formation [39].The present study shows no difference in the NCED between WP-S1 and YP-S1.However, in S2, BnaA01.NCED2 was downregulated in the WP, while BnaA01.NCED3 was up-regulated in WP.In S3, BnaA01.NCED2, BnaA03.NCED2, and BnaA03.NCED3 were up-regulated in WP.The expression profiles of BnaA01.NCED3, BnaA03.NCED2, and BnaA03.NCED3 were consistent with those in the previous report [41].
Eight SNPs were found to exist in the second exon of BnaC03.CCD4 in the WP and YP, these variations might lead to the transformation from WP to YP.In addition, compared to 16214 with white petals, a completely different promoter region with a 700 bp insertion of BnaC03.CCD4 was identified.Interestingly, a 7613 bp deletion and 50 bp insertion in the intron of BnaC03.CCD4 were also detected in 16214.These results indicated that the different promoters of BnaC03.CCD4 might promote its expression in 16214 and lead to the degradation of carotenoids.It was found that the CmGATA4 could lead to the low expression of CmCCD4a-5 by directly binding to the promoter of CmCCD4a-5 and then promoting the accumulation of carotenoids in the orange-flowered chrysanthemum mutant [38].A pair of molecular markers were then successfully developed based on the allelic variation of BnaC03.CCD4, providing an ideal molecular marker for flower color breeding in B. napus.
The flavonoid pathway is also involved in the formation of petal color [18, 24,42], and it was reported that the flavonols, flavones, and anthocyanidins were the main contributors to petal coloration [42].A KEGG functional annotation and RT-qPCR analysis revealed that the expression of BnaA10g23330D (BnF3 ′ H) could affect the synthesis of downstream peonidin and delphinidin in B. napus [18].The RNA-Seq analysis showed that genes involved in the anthocyanin biosynthesis pathway were primarily regulated by ANS, DFR, and UF3GT in B. napus petals [24].The genetic analysis showed that the yellowish-white trait of B. napus was controlled by a single recessive gene, BnaA08.PDS3, and truncating this gene could lead to a reduction in carotenoid biosynthesis [31].In the present study, no significant changes were found in BnaANS, BnaDFR, BnaF3H, BnaUFGT, and BnaUF3GT (the critical genes of the flavonoid pathway).
The MYBs that regulate the anthocyanin metabolism are mainly R2R3-MYB and R3-MYB [32].R2R3-MYB-like are widely involved in a variety of primary and secondary metabolic processes in the plant phenylpropane-flavonoid metabolic pathway [27,43], which play an important role in the formation of final color in tissues and organs [44].The BnaA07.PAP2 In-184-317 activates the expressions of genes related to anthocyanin synthesis and promotes the accumulation of anthocyanins in rapeseed petals [43].Some novel TFs, such as BnaC03.MYB4, BnaC03.MYB73, BnaC03.MYB98, BnaC03.MYB102, and BnaC03.MYB85 that belong to MYB TF families were found in the present research.MYB4 and MYB85 might co-regulate the expressions of genes in the carotenoid pathway, such as CCD4, PSY1, β-HOASE, NCED2, and NCED3, which influence the flower color formation.A network analysis implicated MYB4 and MYB85 in regulating the phenylalanine pathway, flavonoid pathway, xanthophylls pathway, and carotenoids pathway.These different pathways may work together through associations to influence the formation of flower colors.In Arabidopsis, AtPAP1, AtPAP2, AtMYB113, and AtMYB114 are involved in the flavonoid metabolic pathway by regulating the genes of the flavonoid metabolic pathway [27,45].The AtTT2 encodes the R2R3-MYB transcription factor that can form a ternary complex (MBW complex) with TT8 (bHLH) and TTG1 (WD40 protein), which regulates the accumulation of proanthocyanidins to some extent [45].BnaC03.MYB4, orthologous to the Arabidopsis gene ATMYB4, encodes a R2R3 MYB protein, which plays a role in abiotic stress [46] and lignin synthesis pathways [47].BnaC03.MYB85, orthologous to the Arabidopsis gene ATMYB85, directly activates lignin biosynthesis genes and phenylalanine biosynthesis genes during secondary wall formation [48].However, how the TFs MYB4 and MYB85 regulate the genes of the carotenoid pathway and affect flower color remains to be explored in B. napus.

Plant Materials
All plant materials were planted in an experimental field (Wuhan, Hubei Province of China), and the standard management described before was followed [55].The WP B. napus 16214 was provided by the Hybrid Rape Research Center of Shaanxi Province, and the YP B. napus 16287 was discovered in our routine field trials.A reciprocal cross was conducted between the B. napus of 16214 (WP) and 16287 (YP), and F 1 hybrids were obtained.The phenotype of F 2 individuals was also inspected, and the materials with a series of petal colors (including orange-yellow, yellow, light-orange, creamy-white, and white) were observed.

BSA Analysis of Two Pools of WP and YP in the F 2 Population
A total of 217 F 2 individuals were investigated, of which 30 individuals with WP and YP were selected for BSA.The genomic DNA of selected individuals was extracted using a Plant Genome DNA Extraction Kit (Tiangen, Beijing, China) and balanced-mixed to construct the two pools of WP and YP.The WP and YP then underwent next-generation sequencing (NGS, MGISEQ-2000RS, MGI Tech Co., Ltd., Shenzhen, China) together with the genomic DNA of 16214 and 16287, respectively.
The raw data were filtered and mapped to the Darmor_bzh B. napus genome using Bowtie2 software [56].The significant loci were mapped to the ZS11 genome with a higher sequenced quality ZS11 genome released [57].The SNPs and InDels between 16214 and 16287 were filtered using Genome Analysis Toolkit software [58], and the SNP index as well as the ∆(SNP-index) of the WP and YP were calculated subsequently.The AGR was defined according to the average value of ∆(SNP-index), the sliding window was set as 1 Mb, and the step size was set as 1 kb.A threshold significance of 0.99 was selected.

Identification of Petal Color-Related Candidate Genes and Potential Regulatory Network
The genetic linkage map can be aligned to the ZS11 genome using BLAST according to the previous study [59].Based on the collinearity relationship between the genetic linkage map and the ZS11 genome, genes within the AGR were regarded as potential candidate genes.
The orthologous of potential candidate genes in AGR and their annotations were obtained using BLASTN based on the A. thaliana database (http://www.arabidopsis.org/)that accessed on 1 November 2023.The interaction network was analyzed and generated using String_V12.0 (http://string-db.org/) that accessed on 1 November 2023 and visualized using Cytoscape_V3.8.2 [60].

RNA-Seq and DEGs Analysis of the WP and YP
The WP and YP at three developmental stages, four days before flowering (S1), two days before flowering (S2), and the day of flowering (S3), were collected for transcriptomic analysis, and three biological replicates were conducted.YP-S1, YP-S2, and YP-S3 represent the RNA sequencing results of S1, S2, and S3 in the YP, respectively.WP-S1, WP-S2, and WP-S3 represent the RNA sequencing results of S1, S2, and S3 in the WP, respectively.
The total RNA was extracted from the petals of WP-S1, WP-S2, WP-S3, YP-S1, YP-S2, and YP-S3 using a RNAprep Pure Plant Plus Kit (TIANGEN, Beijing, China, DP441) with three biological replicates, respectively.The mRNA was isolated from total RNA through an A-T base pairing of magnetic beads with Oligo (dT) to ployA.The mRNA was randomly broken and small fragments of about 300 bp were isolated via magnetic bead screening.The first strand of cDNA was synthesized in the M-MuLV reverse transcriptase system, and then RNaseH was used to degrade the RNA strand.The second strand of cDNA was synthesized in the DNA polymerase I system.The purified double-stranded cDNA was end-repaired, A-tailed, and connected to sequencing adapters.AMPure XP beads (BECKMAN, USA) were used to screen cDNA of about 250-300 bp, PCR amplification was performed, and AMPure XP beads were used again to purify the PCR products finally to obtain a library.The library was constructed using a NEBNext ® Ultra ™ RNA Library Prep Kit (USA) for Illumina ® .After the library passed the inspection, different libraries were pooled according to the effective concentration and target off-machine data volume requirements, and then Illumina sequencing was performed [61].Initially, raw data were subjected to quality control to eliminate the low-quality reads using the fastp software [62].To determine the genes responsible for the randomly fragmented mRNA fragments, the quality-controlled clean reads were aligned to the ZS11 genome.This alignment was carried out according to a previous methodology using the HISAT2 software [63].Following the alignment, gene expression quantification was performed using StringTie and Ballgown software [63].This step yielded expression levels for each sample in terms of fragments per kilobase of transcript per million mapped reads (FPKM).Genes were considered differentially expressed if they met the criteria of a False Discovery Rate (FDR) < 0.01 and a log 2 (Fold Change) ≥ 1.The R package clusterProfiler was used to conduct GO and KEGG enrichment analyses [64].

The Expression Validation of DEGs via RT-qPCR
A HiScript III 1st Strand cDNA Synthesis Kit (Vazyme, Nanjing, China, R312) was used to synthesize the first strand of cDNA from the RNA.Each PCR reaction was carried out three technical times, and the reaction system was performed following the instructions with SYBR qPCR mix (Vazyme, Q712).BnaActin7 was used as a reference for normalization and analyzed using the QuantStudio 3 Real-Time PCR System (Thermo Fisher, USA), and each reaction was performed in triplicate.The method of 2 −∆CT was used to measure the gene expression according to a previous study [65].Primers for RT-qPCR are shown in Table S9.

The Method for Carotenoid Measurement
Petals of white, creamy-white, and orange-yellow flowers were freeze-dried, and 0.02 g of petal tissue was taken for carotenoid content measurements.The carotenoid extractions and measurements were performed using the Plant Carotenoid Content Assay Kit (Solaibio, Beijing, China, BC4330).The absorbance of petal extracts was measured at 440 nm.A two-way ANOVA analysis of variance was used to assess the differences in carotenoid content.The GraphPad Prism_V9 was used for this analysis.

The Molecular Marker Design and the Detection of B. napus with WP and YP
The whole-length sequence of BnaC03.CCD4, including its 2 kb promoter, was amplified from the genomic DNA of 16214 and 16287 for Sanger sequencing.Two pairs of primers were designed within the differential sequence of BnaC03.CCD4 between 16214 and 16287, which were subsequently applied to PCR using the genomic DNA of 16214 and 16287, respectively.A single band was detected in the PCR production of the genomic DNA of the WP, while no band was detected in that of the YP when using primers of PF1/PR1, while the opposite is the case for the primers of PF2/PR2.Thus, 16214.PF1/16214.PR1 and 16287.PF2/16287.PR2 were selected as molecular markers in this study.

Figure 1 .
Figure 1.Phenotypic variations in orange-yellow and white flowers of B. napus.(A) Phenotypes of 16287 and 16214 in S1, S2, and S3.Bar, 1 cm.(B) The number of individuals with different petal colors in the F1 generation.(C) The determination of carotenoid content.Triplicate experiments with error bars represent a standard deviation.Nine biological replicates for each sample.** p < 0.01, *** p < 0.001, **** p < 0.0001; ns, no significant.

Figure 1 .
Figure 1.Phenotypic variations in orange-yellow and white flowers of B. napus.(A) Phenotypes of 16287 and 16214 in S1, S2, and S3.Bar, 1 cm.(B) The number of individuals with different petal colors in the F 1 generation.(C) The determination of carotenoid content.Triplicate experiments with error bars represent a standard deviation.Nine biological replicates for each sample.** p < 0.01, *** p < 0.001, **** p < 0.0001; ns, no significant.

Figure 3 .
Figure 3. BSA analysis of WP and YP in B. napus.(A) Δ(SNP-index) of 19 chromosomes on the Dar-mor_bzh genome.(B) Alignment of the AGR hotspot region on the ChrC03 from the Darmor_bzh genome to the ZS11 genome.(C) Δ(SNP-index) of the AGR hotspot region on the ChrC03 of the ZS11 genome post mapping.The threshold is 0.715.

Figure 3 .
Figure 3. BSA analysis of WP and YP in B. napus.(A) ∆(SNP-index) of 19 chromosomes on the Darmor_bzh genome.(B) Alignment of the AGR hotspot region on the ChrC03 from the Darmor_bzh genome to the ZS11 genome.(C) ∆(SNP-index) of the AGR hotspot region on the ChrC03 of the ZS11 genome post mapping.The threshold is 0.715.

Figure 5 .
Figure 5. Combined analysis of BSA and transcriptome.(A) Venn diagram illustrating genes from significant regions on the ChrC03 selected through BSA and DEGs in the "YP-S1 vs. WP-S1", "YP-S2 vs. WP-S2", and "YP-S3 vs. WP-S3" groups.(B) Heatmap of 51 common genes with the fold change.Red indicates up-regulated, and blue indicates down-regulated.GO (C) and KEGG (D) enrichments of the 51 common genes.

Figure 5 .
Figure 5. Combined analysis of BSA and transcriptome.(A) Venn diagram illustrating genes from significant regions on the ChrC03 selected through BSA and DEGs in the "YP-S1 vs. WP-S1", "YP-S2 vs. WP-S2", and "YP-S3 vs. WP-S3" groups.(B) Heatmap of 51 common genes with the fold change.Red indicates up-regulated, and blue indicates down-regulated.GO (C) and KEGG (D) enrichments of the 51 common genes.

Figure 6 .
Figure 6.The network of DEGs and related pathways of phenylalanine, flavonoid, carotenoids, MEP, xanthophyll, and TF regulations.Orange ovals represent genes that are differentially expressed in WP and YP.

Figure 6 .
Figure 6.The network of DEGs and related pathways of phenylalanine, flavonoid, carotenoids, MEP, xanthophyll, and TF regulations.Orange ovals represent genes that are differentially expressed in WP and YP.

Figure 7 .
Figure 7.The biosynthesis pathway of carotenoid.This figure depicts the MEP pathway.The compounds produced within the carotenoid pathway are represented in blue rounded rectangles, while enzymes catalyzing the reactions are italicized.The samples from left to right represent WP-S1, WP-S2, WP-S3, YP-S1, YP-S2, and YP-S3.The color bar of the heatmap indicates log2FPKM, with FPKM values being average.The orange indicates high expression, while the blue indicates low expression.The orange triangles in the heatmap represent genes showing differential expressions in YP and