A yellow seed coat is a desirable agronomic trait in the seeds of oilseed-type Brassica crops. In this study, we identified a candidate gene for seed coat color in Dahuang, a landrace of Brassica rapa. A previous study of Dahuang mapped the seed coat color gene Brsc1 to a 2.8-Mb interval on chromosome A9 of B. rapa. In the present study, the density of the linkage map for Brsc1 was increased by adding simple sequence repeat (SSR) markers, and the candidate region for Brsc1 was narrowed to 1.04 Mb. In addition, whole-genome resequencing with bulked segregant analysis (BSA) was conducted to identify candidate intervals for Brsc1. A genome-wide comparison of SNP profiles was performed between yellow-seeded and brown-seeded bulk samples. SNP index analyses identified a major candidate interval on chromosome A9 (A09:18,255,838–18,934,000, 678 kb) containing a long overlap with the target region recovered from the fine mapping results. According to gene annotation, Bra028067 (BrTT1) is an important candidate gene for Brsc1 in the overlapping region. Quantitative reverse transcription (qRT)-PCR revealed that BrTT1 mainly functions in the seed. Point mutations and small deletions in BrTT1 were found between yellow- and brown-seeded Dahuang plants. Collectively, the expression and sequence analysis results provide preliminary evidence that BrTT1 is a candidate gene for the seed coat color trait in Dahuang.
Citation: Wang Y, Xiao L, Guo S, An F, Du D (2016) Fine Mapping and Whole-Genome Resequencing Identify the Seed Coat Color Gene in Brassica rapa. PLoS ONE 11(11): e0166464. https://doi.org/10.1371/journal.pone.0166464
Editor: Maoteng Li, Huazhong University of Science and Technology, CHINA
Received: May 11, 2016; Accepted: October 28, 2016; Published: November 9, 2016
Copyright: © 2016 Wang et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the paper and its Supporting Information files.
Funding: The work was funded by National Key Research and Development Plan of China (2016YFD0101300), Industry Technology System of Rapeseed in China (CARS-13), National Natural Science Foundation of China (31560398) and National High-level personnel of special support program.
Competing interests: The authors have declared that no competing interests exist.
Yellow seed coat is typically a desirable trait in rapeseed breeding. Compared to dark-seeded varieties, yellow rapeseeds have a variety of desirable agronomic traits, including a thinner seed coat, lower husk proportion, lower crude oil pigment content and 5–7% higher oil content [1–3]. Brassica rapa L. (2n = 20, AA) is a major vegetable and oilseed crop in China, India and Bangladesh and is a parent species of Brassica napus L. (2n = 38, AACC) [4,5]. Yellow-seeded varieties of oilseed-type Brassica crops (such as ‘Dahuang’ and ‘Yellow Sarson’ in B. rapa) have many advantages over their dark-seeded (black or brown) counterparts [6–8]. Therefore, the analysis of genes that induce yellow-seeded germplasm in B. rapa is important for breeding and research. However, the molecular mechanism of seed coat color formation in B. rapa is not well understood. One or two genes are thought to be involved in the control of seed coat color in B. rapa [9–11]. A single gene was reported to control the difference between the brown-seeded Indian ‘Toria’ lines and the yellow-seeded ‘Yellow Sarson’ lines , and similar results were found for other yellow-seeded varieties of B. rapa [8,12]. Genetic maps constructed using various molecular markers and large mapping populations have been used to locate yellow-seed genes in B. rapa. Several major genes or QTLs for seed coat color have been found in multiple yellow-seeded lines of B. rapa [13–16].
Bulked segregant analysis (BSA) provides a simple and effective alternative strategy for identifying molecular markers that are linked to target genes by genotyping only a pair of bulked DNA samples from two sets of individuals with distinct or extremely opposite phenotypes . With the rapid development of next-generation sequencing (NGS) technologies, new strategies have been proposed to take advantage of the power of BSA and high-throughput genotyping; these strategies are useful for identifying QTLs or candidate genes. Lee et al.  increased the density of a previous genetic map in Brassica oleracea by adding new derived cleaved amplified polymorphic sequences (dCAPs) and SNP markers developed by the whole-genome resequencing of two cabbage parental lines and genome-wide SNP identification. One major QTL and three minor QTLs for black rot were identified using the new genetic map . Lu et al.  identified the candidate gene Ef1.1 for early flowering in cucumber by BSA and whole-genome resequencing. Abe et al.  and Takagi et al.  identified the major loci for important agronomic and resistance-related traits by the whole-genome resequencing of pooled DNA from segregating populations of rice.
Dahuang, a yellow-seeded landrace of B. rapa, originates from Huangyuan County in Qinghai Province. Dahuang not only has the desirable traits of most yellow-seeded cultispecies of B. rapa but also shows self-compatibility and a higher 1,000-grain weight (6–7 g vs. 5 g) . Therefore, the study of seed coat color in Dahuang is very important for B. rapa breeding programs. Based on previous studies that mapped the seed coat gene Brsc1 in Dahuang , researchers have developed new molecular markers [22,23]. In this study, we added additional simple sequence repeat (SSR) markers with the aim of delimiting the Brsc1 gene within a shorter interval. Fine mapping, whole-genome resequencing and quantitative reverse transcription (qRT)-PCR were performed to screen strong candidate genes for Brsc1.
Materials and Methods
Two B. rapa lines, Dahuang (bright yellow seeds) and 09A-126 (brown seeds), were self-pollinated for more than 8 generations, thus yielding stable seed coat coloration. Using 09A-126 as the donor parent and Dahuang as the recurrent parent, a near-isogenic line known as the BC4 population was constructed after 4 consecutive selective backcrosses. These offspring were then divided into yellow- and brown-seeded groups. Several yellow- and brown-seeded individuals from the BC4 population were chosen to obtain homozygous offspring by selfing for two generations. The homozygous yellow- and brown-seeded offspring were termed BC4F2-Y and BC4F2-B plants, respectively. The field trial was performed in the test field of Qinghai University and no specific permissions were required for these activities. The field studies did not involve endangered or protected species.
Fine mapping of the yellow-seed gene Brsc1
The BC4 population, which contained 1,739 individuals, was used to finely map the yellow-seed gene Brsc1 in Dahuang. Total DNA was extracted from leaves using the CTAB method . The seed coat colors of BC4 population individuals were determined after the seed matured. Two yellow-seeded gene pools were constructed using 12 randomly selected yellow-seeded plant DNA extracts from BC4, and 2 brown-seeded gene pools were similarly constructed. These 4 gene pools were used to screen molecular markers in subsequent steps. New SSR markers were developed according to the reference genome sequence near Brsc1 in the BRAD database (http://brassicadb.org/brad/). Existing SSR markers around the interval where Brsc1 is located were also downloaded from BRAD. Markers with tight linkage to Brsc1 were then used to screen recombinant individuals in the BC4 population to calculate their genetic distances using the Kosambi function . The addition of these new markers increased the density of the genetic map near Brsc1. A physical map of Brsc1 was also constructed by synteny analysis between molecular markers and the genome of B. rapa in the BRAD database. With these new markers, we expected to narrow the known interval where Brsc1 is located on the chromosome, thereby allowing the identification of candidate genes for seed color from genomic data.
Localization of Brsc1 by whole-genome resequencing
Whole-genome resequencing was performed on the parents and on 2 bulked samples (bulks) that were constructed by separately mixing the DNA of 20 yellow-seeded plants and 20 brown-seeded plants from the BC4 population. Sequencing was performed as follows. First, DNA was isolated using the CTAB method as described above. Next, DNA-seq libraries were generated using the TruSeq Nano DNA HT Sample Preparation Kit (Illumina, USA). Briefly, DNA samples were fragmented by sonication to 350 bp, and the DNA fragments were blunt end-repaired, A-tailed, and ligated with full-length adapters for PCR amplification. Finally, the PCR products were purified (AMPure XP) and sequenced on an Illumina HiSeq 2500. The reference genome of Brassica rapa was downloaded from the Ensembl Genomes database (//ftp.ensemblgenomes.org/pub/plants/release-24/fasta/brassica_rapa/dna/Brassica_rapa.IVFCAASv1.24.dna_sm.toplevel.fa.gz). After quality control, sequence data from the 4 samples were aligned to the reference genome using the Burrows-Wheeler Aligner (parameters: mem -t 4 -k 32 –M), and repetitive sequences were removed using SAMTOOLS (parameter: rmdup) . SNP calling was performed with the UnifiedGenotyper in GATK . The homozygous SNPs between the 2 parents were extracted, and 09A-126 was chosen as the reference. Read depth information for the above homozygous SNPs in the offspring pools was used to calculate the SNP index, which is derived from the ratio of the number of different alleles to the total number of reads for each SNP. Using a sliding window (with a window size of 1 Mb and a step size of 1 kb), the chromosomal distribution of SNP indices for the 2 offspring pools was calculated. Differences between the SNP indices of the 2 pools were calculated as the ΔSNP index. Candidate regions for Brsc1 were extracted from the chromosomes with 95% confidence intervals . Resequencing and candidate interval determination were performed by Novogene (Beijing, China). Based on the location of Brsc1, candidate genes within the candidate interval were selected for annotation according to linear comparative analysis in BRAD and homologous gene analysis in A. thaliana (TAIR) database (http://www.arabidopsis.org/). The gene that was most closely related to seed coat color was chosen for further analysis.
Expression profiling analysis of the candidate gene
Seeds from BC4F2-Y and BC4F2-B parents at different developmental stages (7, 14, 21, 28, 35, 42, and 49 days after pollination, DAP) were chosen for total RNA extraction using the EasySpin Plant RNA Kit (Biomed, Beijing, China). RNA was also extracted from other plant tissues, such as roots, stems, leaves, flowers and buds, at 7 DAP. RNA (1 μg) was used to synthesize first-strand cDNA using the PrimeScript™ RT Reagent Kit (Takara, Dalian, China). Candidate gene primers for qRT-PCR were designed to span two exons per the reference sequence in BRAD, and the amplified fragment length was limited to approximately 300 bp. The gene 25s (F: GATTTCTGCCCAGTGCTCTGAA, R: TCTGCCAAGCCCGTTCCCTT) was chosen as the reference gene for the relative quantification of the candidate gene. qRT-PCR was performed in 20-μl reactions that included 10 μl of SYBR® Premix Ex Taq™ II (Takara, Dalian, China), 8 μl of template cDNA (12.5 ng/μl), and 1 μl of each primer (10 ng/μl). The PCR conditions were as follows: 95°C for 2 min, followed by 40 cycles of 95°C for 10 s and 60°C for 30 s. A melting curve analysis was also performed. PCR was conducted in a LightCycler® 480 II (Roche). Three individual plants from each line (BC4F2-Y and BC4F2-B) were chosen for quantitative analysis, and the average relative expression levels were calculated. T tests were performed to test the significance of differences in expression levels among different samples.
Candidate gene sequencing and reference sequence analysis
A homologous sequence of the candidate gene was taken from the BRAD database, and the sequence of an orthologous gene was also downloaded from the TAIR database. Primers to amplify the coding region of the candidate gene (including from cDNA and gDNA templates) were designed using the homologous gene sequence taken from BRAD. Mixtures of cDNA and gDNA from each parent and from each BC4F2 offspring group were used as templates for the amplification of the target gene. To amplify gDNA, 0.5 μg of total genomic DNA was used as a template in a standard 50-μl reaction. To amplify cDNA, the template was substituted with 0.3 μg of cDNA. PCR conditions were as follows: pre-denaturation at 94°C for 4 min, followed by 34 cycles of amplification [94°C for 30 s, 55°C for 30 s and 72°C for 2 min] and a final extension at 72°C for 10 min. PCR products were visualized on 1% agarose gels. Target bands were recovered using a gel extraction kit (Sangon Biotech, Shanghai, China), and purified PCR products were incorporated into the pMD18-T vector (Takara). Recombinant vectors were transferred into competent DH5α cells using cold-shock treatment. After blue-white colony screening on solid LB medium, positive clones were chosen and propagated in LB liquid culture before being subjected to PCR amplification. Five positive bacterial solutions for each gene fragment were fully sequenced by Sangon Biotech. The amino acid sequence of the candidate gene was imputed from the coding sequence using the software Primer Premier 5. Sequence comparisons were performed using ClustalW (available at http://www.ebi.ac.uk/clustalw).
Whole-genome resequencing analysis
After filtering, 30.783 G of clean data was obtained from 31.275 G of raw data. The GC content ranged between 37.26% and 42.45%. The quantity and quality (Q20≥93.97% and Q30≥86.88%, respectively) of the data were sufficient for further analysis. The average depth was >15× for the four samples. After alignment to the reference genome, 1,660,186 SNPs were identified. Based on the 780,061 homozygous SNPs between the two parents, SNP indices for the two BC4 segregant populations’ pools were calculated. A total of 2,429 SNPs were eliminated by applying the following two conditions. First, loci were removed if the SNP indices were <0.3 with an SNP depth <7 in both offspring pools. Second, SNPs were discarded if the SNP indices were missing for either offspring. A ΔSNP index graph was generated from the SNP indices of the two offspring pools (Fig 1). Three candidate intervals (17.83–18.93 Mb; 20.53–20.99 Mb; 22.57–26.09 Mb) that exceeded the threshold value were identified for Brsc1 on chromosome A9 (Fig 2a).
(a) SNP index Manhattan plot of the brown-seeded pool from the BC4 population; (b) SNP index Manhattan plot of the yellow-seeded pool from the BC4 population; (c) ΔSNP index Manhattan plot. The blue line indicates the threshold value.
(a.) Details of the ΔSNP index on chromosome A9. Dentate lines indicate the threshold value; the two parallel black lines delineate the interval from 17.8 to 21.0 Mb on chromosome A9. (b.) Genetic linkage map of linkage markers around Brsc1 on chromosome A9. The figures on the left indicate the corresponding genetic distances (cM) between markers and Brsc1. Some markers are from previous studies (Y05 and Y06: ; Y14 and B089L03-3: ; BrA05, BrA06 and BrA09: . (c.) Physical map of linkage markers for Brsc1 and BrTT1 (Bra028067). Figures on the left represent the physical locations (Mb) of linkage markers on chromosome A9. The shaded region indicates the candidate region for Brsc1. (d.) Fragment of A. thaliana chromosome 1. Orthologs of B. rapa genes are shown on the left, including TT1 (AT1G34790).
Mapping of Brsc1 using molecular markers
BC4 individuals segregated in a 1:1 (yellow-seeded:brown-seeded) ratio. SSR assays were combined with BSA to examine molecular markers that were linked to the Brsc1 gene. Primers distinguishing brown-seeded DNA bulks and yellow-seeded DNA bulks were believed to be potentially linked to Brsc1. Ten SSR primers that revealed polymorphisms between yellow- and brown-seeded bulks were used to amplify gDNA from 12 yellow-seeded plants and 12 brown-seeded plants. This screening eventually identified 8 polymorphic markers, including 7 newly developed markers (designated BrA10-BrA16) and 1 previously available SSR marker (BrID10685) from the BRAD database (Table 1). To more precisely map Brsc1, 2 previously identified PCR-based markers (Y14 and B089L03-3)  were used to detect recombinants in the BC4 population, which consisted of 1,739 plants. A total of 48 recombinants between Y14 and Brsc1 and 54 recombinants between B089L03-3 and Brsc1 were identified. All of the recombinants were then subjected to genotyping with the eight newly developed SSR markers to evaluate their genetic distance from Brsc1. BrA11-BrA15 cosegregated with Brsc1, and BrA10 and BrID10685, the two closest flanking markers, were 0.57 and 0.11 cM from the target gene, respectively. A genetic linkage map surrounding the Brsc1 gene was then constructed (Fig 2b). Three SSR markers (BrA05, BrA06, BrA09)  and two AFLP (Amplified Fragment Length Polymorphism) markers (Y05, Y06)  from previous studies were also incorporated. According to a homologous comparison, Brsc1 was located between BrA10 (A09:18,255,987–18,255,838) and BrID10685 (A09:19,342,792–19,342,739) (Fig 2c), representing a shorter target interval than previously defined.
Dissection of the target region of Brsc1
From a linkage map based on SSR and AFLP markers, Brsc1 was located on chromosome A9 from 18.26 to 19.34 Mb; meanwhile, an adjacent interval from 17.83 to 18.93 Mb for Brsc1 was also detected by whole-genome resequencing. Combining the results from fine mapping and whole-genome resequencing, the target region of Brsc1 was narrowed to a 678-kb interval (18.26–18.93 Mb), which contains 46 predicted genes in the B. rapa reference genome (S1 Table). Most of the predicted genes were annotated using information from homologous A. thaliana genes in TAIR. Among the predicted genes, we found that the gene Bra028067 (BrTT1) (Fig 2) was highly similar to AT1G34790 (Transparent Testa 1, TT1), which encodes a zinc-finger protein and is involved in flavonoid biosynthesis in A. thaliana. TT1 mutants of A. thaliana have transparent testa (TT) due to a deficiency in pigment deposition in the endothelium of the seed coat. Therefore, we assumed that the gene Bra028067 (BrTT1) was the most important candidate gene for Brsc1 and chose it for further study.
Expression pattern analysis of BrTT1
We investigated the expression pattern of BrTT1 using qRT-PCR in the two offspring groups (BC4F2-B, BC4F2-Y) (Fig 3). A BrTT1 primer pair for qRT-PCR was designed in Primer Premier 5 using the reference sequence in BRAD (qBrTT1-F: GCTCTTGGCTGAGTCCCTTT; qBrTT1-R: TTGTCACTTACGATGCCGGA). Expression differences in BrTT1 between BC4F2-B and BC4F2-Y were slight in some organs, such as the flowers, stems, buds, leaves and roots, while those differences in the seeds were notable exceptions. The overall expression in seeds was significantly higher than that in other organs, which indicates that BrTT1 is specifically expressed in seeds. The expression level of BrTT1 in BC4F2-B was markedly higher than that in BC4F2-Y during different stages of seed development, especially at 21 DAF and 28 DAF, which represent the early stages before the color staining of the seed coat (Fig 4). The expression level of BrTT1 peaked at 21 DAF relative to the other stages. A remarkable decline in BrTT1 expression was detected at 28 DAF, which was more severe in yellow-seeded plants than in brown-seeded plants. The most important stage for the formation of seed coat color may occur at 28 DAF in Dahuang. However, the BrTT1 expression patterns in yellow-seeded and brown-seeded plants were similar, suggesting that BrTT1 functions in the same way in the progeny of Dahuang with different seed coat colors. In conclusion, it is reasonable to postulate that BrTT1 is a potential candidate gene for the seed coat color locus in Dahuang.
DAP: days after pollination. The sampling time of different organs (bud, leaf, root, flower, stem) was 7 DAP. **, extremely significant difference between BC4F2-B and BC4F2-Y plants; *, significant difference between BC4F2-B and BC4F2-Y plants.
A, B, C, D, E, F: seeds of BC4F2-B plants; G, H, I, J, K, L: seeds of BC4F2-Y plants; A, G: seeds at 21 DAP; B, H: seeds at 28 DAP; C, I: seeds at 35 DAP; D, J: seeds at 42 DAP; E, K: seeds at 49 DAP; F, L: mature seeds; DAP: days after pollination.
Sequence analysis of BrTT1
The cDNA and gDNA sequences of the region containing the BrTT1 coding sequence in the parents and BC4F2 offspring plants were successfully obtained. PCR primers were designed according to the reference sequence in BRAD using Primer Premier 5 (cDNA, F: ATGTTTTCATCACTCTCAAACCACT; R: TTAAAAGAGCGTTTCAGAGACATAAG; gDNA, F: ACTTCAGACATGGGCTCTTTCAAAGATTC; R: AGGGTCGAGAAGGTTACACGTTGCATAT). No sequence differences were found between a given parent and the corresponding BC4F2 offspring with the same seed coat color when subjected to sequence alignment analysis. Reference sequences of BrTT1 (Bra028067) and AtTT1 (At1g34790) from BRAD and TAIR, respectively, were also collected. ClustalW analysis of the gDNA sequences from BC4F2-B and BC4F2-Y plants showed 3 different deletions (of 114 bases, 2 bases and 9 bases) in introns, in addition to single-base changes in the exons of yellow-seeded lines. Two deletions (of 114 bases and 9 bases) were also found in AtTT1 introns (S1 Fig). Amino acid sequences were predicted from cDNA sequences using Primer Premier 5. Compared with brown-seeded plants, yellow-seeded plants showed three amino acid changes, while most other amino acids were conserved (98.3%). ClustalW showed that 93.67% and 92% of the amino acid sequences of the reference BrTT1 were identical to sequences of BrTT1 from BC4F2-B and BC4F2-Y plants, respectively (Fig 5). By comparing the initiation codons of BrTT1 in Dahuang to those of the reference BrTT1, we found two closely linked initiation codons in Dahuang, the second of which was concordant with the reference BrTT1. The alternative start codon added an additional 13 amino acids in Dahuang, but whether the added amino acids affect BrTT1 function in Dahuang remains unknown. Additionally, BLAST alignment showed that the CDS similarity between BrTT1 and AtTT1 was as high as 81.71%, while BrTT1 encodes a protein that shares 81.18% identity with the AtTT1 protein (Fig 5).
Molecular markers have played an increasingly important role in the fine mapping of target genes over the past few decades. AFLPs and SSRs are two of the most commonly used types of molecular markers [28–30]. To accurately locate target genes, a high-resolution genetic map should be constructed with many tightly linked markers, while a large segregation population is required to determine the genetic distances between markers and the target gene. As more individuals are screened, the representation of the marker location becomes more accurate. In previous mapping results for Brsc1 in Dahuang, increasing the density of the linkage map for Brsc1 using SSR markers narrowed the candidate region to chromosome A9. However, we hypothesized that a more precisely defined region was possible. The rapid development of NGS techniques set the stage for whole-genome sequencing. High-throughput sequence data and reduced costs have enabled genotyping by sequencing (GBS), which is a faster and more effective method for detecting target regions with a large number of SNPs [31,32]. Combined with the BSA method, partially coincident location results from whole-genome resequencing verified the mapping results from molecular markers while also improving the accuracy of the latter. Three significant regions for Brsc1 were detected by resequencing, covering a total of 5.1 Mb on chromosome A9. The best way to identify the region most likely containing Brsc1 among the three intervals was to analyze the information from linked molecular markers or from the annotation of related genes. A combination of GBS and traditional molecular marker technology may generate more accurate location results [19,21]. Our research provided more evidence for BSA with whole genome resequencing being an effective and rapid method to locate the major QTLs or genes for important traits in diploid plants.
The observed diversity of seed coat color in rapeseed is caused by the differential accumulation of pigments in the endothelium [9,33], which is similar to the TT phenotype in A. thaliana [34,35]. As A. thaliana is an important model plant, a detailed study of a series of TT genes in A. thaliana has been conducted [36,37]. TT genes in A. thaliana are mainly involved in the flavonoid biosynthetic pathway and may be divided into three categories: structural genes, regulatory genes and transport genes. Defects in different TT genes result in different degrees of variation in testa color . Moreover, considering the close phylogenetic relationship between Arabidopsis and Brassica, orthologs of TT genes may play similar roles in Brassica species; quantitative analysis of gene expression supports this inference [33,38]. In A. thaliana, monogenic mutations may cause remarkable deficiencies in pigment deposition. Meanwhile, some yellow-seeded varieties of B. rapa and B. juncea may also be ascribed to mutations in a single major gene, such as TT8 [9,39], TTG1 . In B. napus, however, some TT genes are likely involved in the staining of seed coats, but there is no direct evidence indicating that one dominant TT gene controls seed coat color in these species [33,40,41].
In our research, one ortholog of AtTT1 was found directly in the target region of Brsc1. In A. thaliana, TT1 encodes a WIP-type zinc finger protein that functions during the construction of the endothelium and the accumulation of proanthocyanidin pigments as a regulatory gene that interacts with the R2R3-MYB protein TT2. A series of TT1 mutations leads to the TT phenotype [42,43]. We postulate that BrTT1 is responsible for the lack of pigment in the seed coats of yellow-seeded Dahuang. Further expression profiling and sequence comparison analyses have partially confirmed this hypothesis, but we cannot yet definitively conclude that this gene is responsible for seed coat color until a detailed functional analysis is performed. Furthermore, TT genes are generally single-copy in Arabidopsis, whereas multiple copies are found in Brassica species . Additionally, multiple orthologs of TT1 were detected in the B. rapa genome according to BLAST analysis in the BRAD database, which may complicate its function. As such, more in-depth research on BrTT1 in Dahuang is warranted. Different kinds of yellow-seeded B. rapa varieties are cultivated and researched throughout the world. Comparative sequencing of BrTT1 orthologous genes in different yellow-seeded and brown-seeded B. rapa varieties may help to identify the mutation sites responsible for the expression differentiation in different B. rapa varieties. Functional complementation analysis is also required to verify BrTT1 function in Dahuang.
S1 Fig. The gDNA sequence alignment between BrTT1 in B. rapa and AtTT1 in A. Thaliana.
The gDNA sequence of AtTT1 and BrTT1 are downloaded from TAIR and BRAD database separately; "Brown" indicates the gDNA sequence of BrTT1 in 09A-126 and BC4-B population; "Yellow" indicates the gDNA sequence of BrTT1 in Dahuang and BC4-Y population.
We greatly appreciate the technical advice from Dr. Xiaolin Dun (Oilcrops Research Institute, Chinese Academy of Agricultural Sciences) and Dr. Xinhua Zeng (Oilcrops Research Institute, Chinese Academy of Agricultural Sciences) and assistance with the amplified sequence experiment from Zhangsheng Tao (Oilcrops Research Institute, Chinese Academy of Agricultural Sciences). We also thank Dr. Ku Lu (Southwest University) and Dr. Zhen Huang (College of Agronomy, Northwest A&F University) for valuable suggestions regarding the gene studied. We thank Dr. Baolong Liu (Northwest Plateau Institute of Biology; The Chinese Academy of Sciences) for his valuable suggestions during the preparation of this manuscript. We thank Huiyan Zhao (Southwest University) and Qiang Xin (Huazhong Agricultural University) for their assistance with the expression experiment.
- Conceptualization: DD LX.
- Data curation: YW LX.
- Formal analysis: YW.
- Funding acquisition: DD LX.
- Investigation: YW FA.
- Methodology: YW LX.
- Project administration: DD LX.
- Resources: DD LX.
- Supervision: DD FA.
- Validation: YW SG.
- Visualization: SG FA.
- Writing – original draft: YW.
- Writing – review & editing: YW.
- 1. Marles M, Gruber MY. Histochemical characterisation of unextractable seed coat pigments and quantification of extractable lignin in the Brassicaceae. J Sci Food and Agric. 2004;84: 251–262.
- 2. Shirzadegan M, Röbbelen G. Influence of seed color and hull proportion on quality properties of seeds in brassica napus L. Fette, Seifen, Anstrichmittel. 1985;87: 235–237.
- 3. Stringam G, McGregor D, SH P, Chemical and morphological characteristics associated with seedcoat color in rapeseed. In: Proceedings of the 4th International Rapeseed Conference. Germany: Giessen; 1974.
- 4. Rahman M, Mamidi S, McClean P. Quantitative trait loci mapping of seed colour, hairy leaf, seedling anthocyanin, leaf chlorosis and days to flowering in F2 population of Brassica rapa L. Plant Breed. 2014;133: 381–389.
- 5. Zhao J, Wang X, Deng B, Lou P, Wu J, Sun R, et al. Genetic relationships within Brassica rapa as inferred from AFLP fingerprints. Theor Appl Genet. 2005;110: 1301–1314. pmid:15806345
- 6. Xiao L, Zhao Z, Du D, Yao Y, Xu L, Tang G. Genetic characterization and fine mapping of a yellow-seeded gene in dahuang (a Brassica rapa landrace). Theor Appl Genet. 2012;124: 903–909. pmid:22120455
- 7. Ahmed S, Zuberi M. Inheritance of seed coat color in Brassica campestris L. variety Toria. Crop Science. 1971;11: 309–310.
- 8. Chen B, Heneen W. Inheritance of seed colour in Brassica campestris L. and breeding for yellow-seeded B. napus L. Euphytica. 1992;59: 157–163.
- 9. Li X, Chen L, Hong M, Zhang Y, Zu F, Wen J, et al. A Large insertion in bHLH transcription factor BrTT8 resulting in yellow seed coat in Brassica rapa. PloS One. 2012;7: e44145. pmid:22984469
- 10. Rahman M. Production of yellow-seeded Brassica napus through interspecific crosses. Plant Breed. 2001;120: 463–472.
- 11. Zaman M. Inheritance of seed colour in Brassica campestris. Sveriges Utsadesförenings Tidskrift. 1989;99: 205–207.
- 12. Lou P, Zhao J, Kim JS, Shen S, Del Carpio DP, Song X, et al. Quantitative trait loci for flowering time and morphological traits in multiple populations of Brassica rapa. J Exp Bot. 2007;58: 4005–4016. pmid:18048374
- 13. Bagheri H, Pino-Del-Carpio D, Hahnart C, Bonnema G, Keurentjes J, Aarts M. Identification of seed-related qtl in Brassica rapa. Span J Agric Res. 2013;11: 1085–1093.
- 14. Kebede B, Cheema K, Greenshields DL, Li C, Selvaraj G, Rahman H. Construction of genetic linkage map and mapping of QTL for seed color in Brassica rapa. Genome. 2012;55: 813–823. pmid:23231600
- 15. Rahman M, McVetty PB, Li G. Development of SRAP, SNP and multiplexed SCAR molecular markers for the major seed coat color gene in Brassica rapa L. Theor Appl Genet. 2007;115: 1101–1107. pmid:17846742
- 16. Zhang J, Lu Y, Yuan Y, Zhang X, Geng J, Chen Y, et al. Map-based cloning and characterization of a gene controlling hairiness and seed coat color traits in Brassica rapa. Plant Mol Biol. 2009;69: 553–563. pmid:19039665
- 17. Michelmore RW, Paran I, Kesseli RV. Identification of markers linked to disease-resistance genes by bulked segregant analysis: a rapid method to detect markers in specific genomic regions by using segregating populations. Proc Natl Acad Sci U S A. 1991;88: 9828–9832. pmid:1682921
- 18. Lee J, Izzah NK, Jayakodi M, Perumal S, Joh HJ, Lee HJ, et al. Genome-wide SNP identification and QTL mapping for black rot resistance in cabbage. BMC Plant Biol. 2015;15:32. pmid:25644124
- 19. Lu H, Lin T, Klein J, Wang S, Qi J, Zhou Q, et al. QTL-seq identifies an early flowering QTL located near Flowering Locus T in cucumber. Theor Appl Genet. 2014;127: 1491–1499. pmid:24845123
- 20. Abe A, Kosugi S, Yoshida K, Natsume S, Takagi H, Kanzaki H, et al. Genome sequencing reveals agronomically important loci in rice using MutMap. Nat Biotechnol. 2012;30: 174–178. pmid:22267009
- 21. Takagi H, Abe A, Yoshida K, Kosugi S, Natsume S, Mitsuoka C, et al. QTL-seq: rapid mapping of quantitative trait loci in rice by whole genome resequencing of DNA from two bulked populations. Plant J. 2013;74: 174–183. pmid:23289725
- 22. Li X, Xiao L, Du D. Fine mapping of Brsc1 gene and map integration in Dahuang rape (Brassica rapa L.). Acta Agron Sin. 2015;41: 1039–1046.
- 23. Zhao H, Xiao L, Zhao Z, Du D. Development of molecular markers and map integration for seed color traits in Dahuang rape (Brassica rapa. L). Acta Agro Sin. 2014;40: 965–972.
- 24. Doyle J. Isolation of plant DNA from fresh tissue. Focus. 1990;12: 13–15.
- 25. Kosambi DD. The geometric method in mathematical statistics. Am Math Mon. 1944;51: 382–389.
- 26. Li H, Durbin R. Fast and accurate short read alignment with burrows-wheeler transform. Bioinformatics. 2009;25: 1754–1760. pmid:19451168
- 27. McKenna A, Hanna M, Banks E, Sivachenko A, Cibulskis K, Kernytsky A, et al. The genome analysis toolkit: a mapreduce framework for analyzing next-generation DNA sequencing data. Genome Res. 2010;20: 1297–1303. pmid:20644199
- 28. Appleby N, Edwards D, Batley J. New technologies for ultra-high throughput genotyping in plants. Methods Mol Biol. 2009;513: 19–39. pmid:19347650
- 29. Miah G, Rafii MY, Ismail MR, Puteh AB, Rahim HA, Islam Kh N, et al. A review of microsatellite markers and their applications in rice breeding programs to improve blast disease resistance. Int J Mol Sci. 2013;14: 22499–22528. pmid:24240810
- 30. Grover A, Sharma PC. Development and use of molecular markers: past and present. Crit Rev Biotechnol. 2016;36: 290–302. pmid:25430893
- 31. He J, Zhao X, Laroche A, Lu ZX, Liu H, Li Z. Genotyping-by-sequencing (GBS), an ultimate marker-assisted selection (MAS) tool to accelerate plant breeding. Front Plant Sci. 2014;5: 484. pmid:25324846
- 32. Yang H, Tao Y, Zheng Z, Li C, Sweetingham MW, Howieson JG. Application of next-generation sequencing for rapid marker development in molecular plant breeding: a case study on anthracnose disease resistance in Lupinus angustifolius L. BMC Genomics. 2012;13: 318. pmid:22805587
- 33. Qu C, Fu F, Lu K, Zhang K, Wang R, Xu X, et al. Differential accumulation of phenolic compounds and expression of related genes in black- and yellow-seeded Brassica napus. J Exp Bot. 2013;64: 2885–2898. pmid:23698630
- 34. Debeaujon I, Nesi N, Perez P, Devic M, Grandjean O, Caboche M, et al. Proanthocyanidin-accumulating cells in arabidopsis testa: regulation of differentiation and role in seed development. Plant Cell 2003;15: 2514–2531. pmid:14555692
- 35. Yu CY. Molecular mechanism of manipulating seed coat coloration in oilseed Brassica species. J Appl Genet. 2013;54: 135–145. pmid:23329015
- 36. Appelhagen I, Thiedig K, Nordholt N, Schmidt N, Huep G, Sagasser M, et al. Update on transparent testa mutants from arabidopsis thaliana: characterisation of new alleles from an isogenic collection. Planta. 2014;240: 955–970. pmid:24903359
- 37. Lepiniec L, Debeaujon I, Routaboul JM, Baudry A, Pourcel L, Nesi N, et al. Genetics and biochemistry of seed flavonoids. Annu Rev Plant Biol. 2006;57: 405–430. pmid:16669768
- 38. Liu X, Lu Y, Yuan Y, Liu S, Guan C, Chen S, et al. De novo transcriptome of Brassica juncea seed coat and identification of genes for the biosynthesis of flavonoids. PloS One. 2013;8: e71110. pmid:23990927
- 39. Padmaja LK, Agarwal P, Gupta V, Mukhopadhyay A, Sodhi YS, Pental D, et al. Natural mutations in two homoeologous TT8 genes control yellow seed coat trait in allotetraploid Brassica juncea (AABB). Theor Appl Genet. 2014;127: 339–347. pmid:24247234
- 40. Jiang J, Shao Y, Li A, Lu C, Zhang Y, Wang Y. Phenolic composition analysis and gene expression in developing seeds of yellow- and black-seeded Brassica napus. J Integr Plant Biol. 2013;55: 537–551. pmid:23445079
- 41. Stein A, Wittkop B, Liu L, Obermeier C, Friedt W, Snowdon RJ. Dissection of a major QTL for seed colour and fibre content in Brassica napus reveals colocalization with candidate genes for phenylpropanoid biosynthesis and flavonoid deposition. Plant Breed. 2013;132: 382–389.
- 42. Appelhagen I, Lu GH, Huep G, Schmelzer E, Weisshaar B, Sagasser M. TRANSPARENT TESTA1 interacts with R2R3-MYB factors and affects early and late steps of flavonoid biosynthesis in the endothelium of Arabidopsis thaliana seeds. Plant J. 2011;67: 406–419. pmid:21477081
- 43. Sagasser M, Gui-Hua L, Klaus H, Bernd W. A. thaliana transparent testa 1 is involved in seed coat development and defines the wip subfamily of plant zinc finger proteins. Genes Dev. 2002;16:138–149. pmid:11782451
- 44. Guo N, Cheng F, Wu J, Liu B, Zheng S, Liang J, et al. Anthocyanin biosynthetic genes in Brassica rapa. BMC Genomics. 2014;15: 426. pmid:24893600