Efficient molecular markers for the selection of rapeseed genetic materials with high seed oil content and ideal fatty acid (FA) composition are preferred by rapeseed breeders. Recently, we reported the molecular mechanism of TRANSPARENT TESTA 2 (TT2) in inhibiting seed FA biosynthesis in Arabidopsis. However, evidence showing the association of rapeseed TT2 homologs and seed FA production are still insufficient. In this study, we collected 83 rapeseed (Brassica napus L.) landraces from different geographical backgrounds to conduct association mapping of BnaC.TT2.a in relation to seed coat color and FA biosynthesis. Population background was corrected by 84 pairs of SSR markers that were uniformly distributed among the linkage groups of the Tapidor-Ningyou-7 DH population. A single copy of BnaC.TT2.a for single nucleotide polymorphism (SNP) assay was cloned by a pair of previously reported specific primers. From the analysis of BnaC.TT2.a allelic variations using GLM+Q model, four SNPs on intron 1 of BnaC.TT2.a that were associated with seed FA were discovered. Moreover, an InDel at position 738 on exon 3 of BnaC.TT2.a indicated a change of protein function that was significantly associated with seed coat color, linoleic acid (C18:2), and total FA content. These findings revealed the role of BnaC.TT2.a in regulating the seed color formation and seed FA biosynthesis in rapeseed, thereby suggesting effective molecular markers for rapeseed breeding.
Citation: Zhou L, Li Y, Hussain N, Li Z, Wu D, Jiang L (2016) Allelic Variation of BnaC.TT2.a and Its Association with Seed Coat Color and Fatty Acids in Rapeseed (Brassica napus L.). PLoS ONE 11(1): e0146661. https://doi.org/10.1371/journal.pone.0146661
Editor: Maoteng Li, Huazhong university of Science and Technology, CHINA
Received: October 24, 2015; Accepted: December 20, 2015; Published: January 11, 2016
Copyright: © 2016 Zhou 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.
Funding: The work of the authors’ lab was sponsored by the National Key Basic Research Project (abbreviated as 973 project, Code No. 2015CB150205), Natural Science Foundation of China (abbreviated as NSFC, Code No. 31371542), and Jiangsu Collaborative Innovation Center for Modern Crop Production (abbreviated as JCIC-MCP). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Rapeseed (Brassica napus L.) is a globally important oilseed crop ranked third after soybean and oil palm. Rapeseed breeders deal with both quantitative and qualitative traits for genetic improvement. However, most of the traits relating to seed yield and seed quality are quantitative. Numerous research efforts have been invested into obtaining favorable agronomic traits, such as high total seed fatty acids (FA) and high proportion of unsaturated 18C FAs (i.e., oleic and linoleic acids), simultaneously. Quality degrading traits, such as seed erucic acid, thioglycoside, color substances, and crude fiber need to be reduced .
At present, the majority of commercially available B. napus seeds vary in color from brown to dark brown or black [1–3]. However, yellow-seeded varieties have become interesting to researchers because of their decreased thickness, lignin content, and defective proanthocyanidin (PA) biosynthesis. The superior characteristics of yellow seeds result in low pigment deposition in the seed coats, high proportion of seed oil, and synthesis of meal protein [4, 5]. At present, Arabidopsis mutants with several defective flavonoid pathway genes have been identified and correlated with yellow to pale brown seed coat color. These mutants are referred to as transparent testa (tt), which includes tt1, tt2, tt4, tt5, tt6, tt7, tt8, tt10, and tt12. These mutants contain higher seed oil than the wild type . In the studies that aimed to elucidate the metabolic pathway of flavonoids in A. thaliana, the complex of TT2, TT8, and TTG1 could regulate the expressions of TT3 (DFR), TT18 (DLOX), TT12 (MATE), and ANR, which are the key factors affecting the seed coat color [7–10]. More recently, it was found that TT2 and TT8 could target FUS3, LEC1, LEC2, and CDS2 to regulate seed development and/or oil biosynthesis [6, 11]. In the past, key studies on the transparent testa gene family in Brassica species include the cloning of BnTT2 and BnTTG1 [12–15] in B. napus and the alteration in seed color through insertion of the BrTT8 transcription factor in B. rapa . These studies provide a basis for the association between these genes and seed coat color/FA in rapeseed. The stability of yellow-seeded phenotypes in diploid Brassica species, such as B. rapa and B. oleracea, is not rare. However, yellow-seeded types in B. napus do not occur naturally ; these types are developed from interspecific crosses with related diploid progenitor species, such as B. rapa, B. oleracea, B. juncea, and B. carinata [17, 18]. Hence, to get insight into the molecular mechanism regulating seed color is important for the successful breeding of stable yellow-seeded cultivars of B. napus.
In Brassica, mapping of quantitative trait loci (QTL) in segregating populations is well documented and has been used to analyze and identify QTL for seed coat color [19–21] and other important traits [22–25]. Although QTL mapping helps identify and estimate the contributive effects of loci on a quantitative trait, it still has several inherent disadvantages. QTL mapping still requires a specific mapping population and can only calculate the allelic diversity for the cross between two parents with different alleles controlling a trait. Therefore, QTLs cannot be identified for the parents both have same allele . Moreover, the limitation in number of recombination events in a segregating population could impose restriction to the resolution of QTL mapping. This limitation leads to variation in confidence intervals for the QTL positions, ranging from a few cM up to several tens of cM [27, 28].
Linkage Disequilibrium (LD) mapping, which is also called association mapping or linkage mapping, is a potentially alternative approach to QTL mapping. LD mapping probes the associations between marker(s) and trait(s) by studying the degree of LD between markers and functional polymorphisms across diverse germplasm [29, 30]. The main advantages of LD mapping over the QTL mapping technique include improved mapping resolution, reduced research time, and greater allele number . Hence, the use of modern genetic technologies and methods could help researchers to use natural diversity in large gene pools of plant species and locate valuable genes in the genome [29, 30, 32]. LD measures the extent of non-random association between alleles at different loci or regions of the genome that are inherited together at a frequency higher than expected based on recombination . Single nucleotide polymorphism(SNP), with moderate or high LD frequency, may extend over the length of genes or gene clusters that could possibly determine the SNP haplotypes. Identification of the trait(s) associated with SNPs using LD decay in Arabidopsis could provide a powerful data set for evolutionary and genetic association studies . Moreover, LD assists breeders to screen large germplasm with potentially powerful and fast-tracked breeding program methods for crop improvement .
TT2 regulates PA biosynthesis, and/or tannin accumulation in plant seeds. Moreover, TT2 could play a vital role in seed development and regulate the expression of numerous genes involved in the FA biosynthesis pathway by targeting FUSCA3 in Arabidopsis . In association analyses between markers and traits in canola type rapeseed, LD decay resulted in significantly higher resolution than QTL analyses in segregating populations, as shown by interval mapping . So far, in B. napus only a few studies have been conducted regarding the association of phenotypes and candidate genotype governing genes. For instance, it was found there are 26 and 12 allelic polymorphisms at two loci, namely BnaX.VTE3.a, BnaA.PDS1.c, respectively, responsible for changes of seed tocopherol content . Moreover, association analysis of BnaA.FRI.a in 248 B. napus accessions revealed significant correlation between six SNPs sites and variation of flowering time .
Given the functional attributes of TT2 in flavonoids and seed oil synthesis in Arabidopsis, we evaluated the allelic variation of BnaC.TT2.a in a diverse collection of rapeseed (B. napus) landraces, and associated its genetic polymorphism with seed color and FA content using association mapping technique.
Materials and Methods
Ethics statement and plant materials
Ethics statement is “N/A” as the study was conducted on plants. The field used for the experiment was Huajiachi Agricultural Farm which is the property of Zhejiang University, Hangzhou, P.R. China and professors of the university are allowed to use the land for their research work. Thus, no specific permissions were required. Moreover, a collection of 83 rapeseed genotypes (B. napus L.) with various genetic backgrounds were obtained mainly from the Department of Plant Breeding and Genetics, The University of Agriculture, Peshawar, Pakistan (49, the majority of which are local varieties), 28 Chinese local varieties were provided by Prof. Wu Xiaoming of Institute of Oil Crops, Chinese Academy of Agricultural Sciences. The genotypes with genetic backgrounds from Germany, Canada, and France were obtained from the genebank of the IPK Gatersleben, Germany (Table 5). Furthermore, we confirm that the field studies did not involve endangered or protected species. A field experiment in a randomized complete block design with three replications was carried in the growing season 2010–2011. Initially, an inflorescence of each genotype was randomly tagged and covered with bags for selfing. The seeds from the upper raceme of the main branch or from the uppermost side branch were collected for the analysis of FAs.
Seed coat color assessment
The seed coat color was interpreted according to the Color Index for Red Grapes (CIRG) color criteria, following the method of Shen et al.,  and by using Minisan XE PLUS (USA Hunter Associate Laboratory, Inc.). These color traits were expressed in terms of tristimulus indices, namely, L*, a*, and b*. L* indicates lightness (100 = white and 0 = black), a* shows redness-greenness (positive = red), and b* presents yellowness-blueness (positive = yellow). Moreover, the chroma (C) value shows color intensity or saturation, and is calculated as C = (a*2 + b*2)1/2, whereas hue angle is calculated as Ho = tan-1(b*/a*). Seed coat color data was summarized using MS-Excel Windows software and the CIRG index presented by the Quantile-quantile (QQ) plot package in the R program.
Fatty acid analysis
FA was analyzed following the methods described by Zou et al.,  and Zhu et al.,  with slight modifications. Briefly, 50 mg of the milled seed meal was homogenized with 2 mL of solution containing chloroform/isopropanol (2:1, v/v) in a 12 mL screw-top glass tube. Samples were kept in the dark for 2 h at room temperature and vigorous vortexing was done for 30 s per 30 min. Samples were then centrifuged at 2500 rpm for 5 min. The 400 μL supernatant was collected in a new tube and 2 mL of 1% MeOH/H2SO4 (v/v) was added to each tube. The tubes were warmed for 1 h in a water bath at 80°C. Furthermore, 2 mL of 0.9% NaCl was added to each sample after cooling the tubes at room temperature. Hexane was added thrice (1 mL each time) and samples were vortexed for the extraction of FA, followed by centrifugation at 2500 rpm for 2 min. Supernatant of about 700 μL was collected in GC vials, from which 2 μL were auto-injected into the gas chromatograph machine (SHIMADZU, Kyoto, Japan, GC-2014). The chromatograph machine was equipped with a flame ionization detector (FID) and a column (Supelco was-10, Schnelldorf, Germany) with length × inner diameter × liquid membrane thickness of 30 m × 0.25 mm × 0.5 μm, respectively. Temperature programming was configured as follows: an initial column temperature of 160°C for 1 min, which was then raised to 240°C at the rate of 4°C per min, and finally held for 16 min to end the analysis time set per sample. The peaks of FA species were identified according to their respective retention times. Moreover, the concentration of the individual peaks was normalized and quantified against methyl heptadecanoate, which was used as an internal standard and/or control. Total FA content was measured as the sum of all 11 FA species. Data was summarized using MS-Excel software and seed oil content presented by the QQ plot package in the R program.
The population structure of 83 B. napus genotypes was shaped from 84 pairs of SSR markers, which were distributed uniformly in B. napus Tapidor-Ningyou-7 DH population. At least four pairs of markers were located at each linkage group. B. napus is an allopolyploid with a complex genome. Thus, one pair of SSR markers could result in multiple loci bands of a gene and eventually lead to non-targeted genotyping. Moreover, the reading of the bands obtained from SSR markers was based on their overall shape. The Q value of the population was obtained using the software STRUCTURE 2.3.4 (http://pritch.bsd.unchicago.edu/structure.html) . Initially, we assessed the K values from 1 to 10, five times for each value, with length of burn-in period at 10,000 and Markov chain Monte Carlo at 100,000. The most appropriate groups were divided based on the two consecutive change rates of the Ln p(D) value. The K index corresponding to the peak Δk value was considered the number of taxa divisions .
DNA of all the 83 genotypes was extracted according to the CTAB method , and the concentration was set to 50 ng/μL. The sequence of candidate gene BnaC.TT2.a was obtained from the National Center for Biotechnology Information (NCBI) website. This gene was selected for association mapping based on its differential expression between yellow-seeded and black-seed rapeseed phenotypes, as well as its high similarity with Brassica TT2 genes in the sequence database . Gene-specific primer pairs were selected to amplify the single-gene product using PrimeSTAR HS DNA Polymerase (Takara Biotechnology [Dalian] Co., Ltd.). The amplification product was then purified by using 1% agarose gel electrophoresis. Amplification products of all 83 samples were subjected to Sanger sequencing (Shanghai Sunny Biotechnology Co., Ltd.). Sequences of all the 83 DNA samples were compared by the sequence alignment using CLASTALW2 software. The sequence alignment results were then subjected to DNA polymorphism analysis at the locus BnaC.TT2.a among the 83 genotypes using TASSEL software (Version 3.0). To identify significant SNPs/indel, threshold value of the minor allele frequencies (MAF) was set to 5%. Moreover, the genotype based haplotype groups evaluation was conducted according to the diversity of SNP observed for each genotype.
Using the TASSEL 3.0 software, DNA polymorphism was further analyzed to identify the association between color traits, including CIRG, L*, a*, b*, C, and H°, seed FA composition, and total FA . Analysis for association mapping was performed using the GLM + Q model and the population structure were estimated using the Q-matrix function of the STRUCTURE software. Moreover, the level of significance was set to (Bonferroni correction) P≤0.05 for the correlation analysis of the characteristics. Significance of the linkage between the polymorphismic sites and the traits were interpreted and characterized based on P and R2 values, respectively.
Distribution of seed coat color and seed fatty acids
Phenotypic diversity in terms of seed color and FA contents in all 83 genotypes was observed and measured. QQ plot analysis revealed normal data distribution from the analysis of seed coat color traits (Fig 1A) and seed FA content (Fig 1B). We investigated the relationship between seed coat color and seed FA. Color measurements, expressed as tristimulus parameters, revealed that CIRG had a significantly negative correlation with FA composition (C18:0, C18:1, and C18:2) and total seed FA (Table 1, Fig 2A). Moreover, the lightness (L*) of the seed coat had a significantly positive correlation with seed FA composition (C18:0 and C18:2) and total seed FA (Table 1, Fig 2B). The other significantly positive correlations found were a* with C18:2, b* with C18:1 and C18:2, and C with C18:0, C18:1, and C18:2. A non-significant correlation was observed among H°, total FA, and FA composition (Table 1).
(A) Shows the normal distribution and QQ plot configuration of the CIRG, whereas, (B) shows the normal distribution and QQ plot configuration of seed total FA, respectively.
Population structure (Q) and phenotype of the subpopulation
Population (POP) structure of 83 B. napus accessions was analyzed using 84 SSR markers. These SSR markers were distributed across the genetic map of B. napus and divided the population into two subpopulations (POP1 and POP2) based on their genetic diversity, as shown in the bar colors (Fig 3A). Red bars showing POP1 included 58 genotypes, whereas POP2 included 25. Moreover, the number of subpopulations suitable for association analysis was estimated by applying the Δk criterion. The distribution of Δk relative coefficients revealed significant variation in the likelihood of 83 B. napus accessions, ranging from 1 to 10 (Fig 3B). The most evident and highest likelihood for a subpopulation was observed with k = 2 and Ln p(D) = -10,523.7, with a variance of 897.26, using the STRUCTURE software (Fig 3B, Table 2). Furthermore, data on tristimulus parameters (color traits), and FA analysis revealed that POP2 had greater mean lightness (L*), redness (a*), and chroma (C) values than POP1. On the other hand, POP1 had higher yellowness (b*) and Ho values than POP2. Overall, POP1 had higher seed oil content than POP2, but this change was statistically non-significant. POP2 had higher C18 FAs, such as C18:0, C18:1, C18:2, and C18:3; and lower long-chained but mostly undesirable FA content, including C20:0, C20:1, C22:0, C22:1, and C24:0 than POP1 (Table 3).
The accessions are represented by bars which could be divided into two categories based on their colors. X-axis indicates the number of accessions and Y-axis shows the percentage of group membership.
Nucleotide polymorphism and haplotype diversity
To determine the nucleotide(s) variation of BnaC.TT2.a in a panel of the 83 genotype, a single primer pair, namely, BnaC.TT2-L 5ʹ-TTGATAGCTGGGAGGCTTCCAGG-3ʹ and BnaC.TT2-R 5ʹ-CCAAACCATCAAAGCCCATTAA-3ʹ , was used to amplify the coding sequence of the BnaC.TT2.a gene. SNPs were detected using the alignment of sequencing results, and the entire population was categorized into various haplotypes. Sequence of BnaC.TT2.a gene with accession number “DQ778645.1” was used as a reference to categorize the population into haplotypes. We found seven haplotype groups (H) based on the SNPs detected across the entire population of B. napus (Table 4). Analyses revealed that among the population of 83 B. napus genotypes, haplotype group H3 presented the most (34 genotypes) members of the population, followed by H0, H4, H2, H1, H5, and H6 with 29, 8, 6, 3, 2, and 1 numbers of genotypes, respectively (Table 5). SNPs detected were both monomorphic as in the case of H1 and H2, as well as polymorphic as shown by H0, H3, H4, H5, and H6.
Phenotypic differences among different haplotypes
Individual comparison among different haplotypes based on seed coat color and FA content and/or composition, revealed some interesting findings. For instance, the H4 haplotype has a relatively high L*, b*, C*, and H0 index values except for a*, suggesting the plausible functional attributes of TT2 gene in changing the seed coat color (Fig 4A). Conversely, haplotypes H5 and H6 have FA composition with higher C22:1, but with lower C18 derivatives and total FA contents compared with other haplotypes. Thus, these genotypes have relatively low seed oil content. We found that in C22:1, which may negatively affect human health, H0 and H1 haplotypes have relatively low C22:1, but higher C18:0, C18:1, and C18:2 FA derivatives compared with other haplotypes. Thus, these are a candidate germplasm pool for oilseed breeding programs (Fig 4B).
Linkage disequilibrium at the BnaC.TT2.a locus
All the 83 B. napus accessions were genotyped at BnaC.TT2.a locus by PCR amplification from genomic DNA with single primer pair and subsequently subjected to sequencing. The primer covers the gene from the start codon to 890 bp. To analyze the degree of LD within BnaC.TT2.a, R2 values for each pair of polymorphism was determined using TASSEL software (v3.0) . Only SNPs with frequencies ≥ 0.05 were considered for LD decay analysis. In total, five SNPs and one Indel were identified. Of these six polymorphic sites, four were located within introns and two within the coding sequence. P<0.001 was observed between SNP 166 and SNP 226 for the BnaC.TT2.a polymorphism (Fig 5), spanning four introns. Moreover, the population was structured into seven different haplotypes, which were further categorized based on the significant outcomes from LD analysis. From the four SNPs (gene position 166, 188, 222, and 226) that have significant phenotypic to genotypic trait associations depicted from the LD map (Fig 5), we grouped the haplotypes H0, H1, and H2 as G1. The haplotypes H3, H4, H5, and H6, having four common and significant linkage SNPs at these polymorphic sites, were grouped as G2. We found that genotypes of G2 have comparatively lower oil content (approx. 8.4%) than those of G1 (Fig 6A). Moreover, the oil quality of G2 was inferior to G1 in terms of higher components of C22:1 and lower components of C18:2 and C18:1 (Fig 6B). Non-significant variations were observed in the seed coat color of both G1 and G2 (Fig 6C).
Association analysis between genotypic and phenotypic traits
Polymorphism observed in DNA samples were analyzed for association with seed color and FA traits using the GLM + Q model. Polymorphic sites with minor allele frequencies greater than 5% were used for the association analysis. In total, 28 significant associations (P<0.05) were observed for six polymorphic sites (Table 6) in BnaC.TT2.a locus. We found that C18:1, C18:2, C22:0, and C22:1 and total seed FA were significantly associated with SNPs including T/A, G/A, C/A, and T/C at positions 166, 188, 222, and 226, respectively, on gene locus BnaC.TT2.a (Fig 5). All these changes were observed in intron (1). There is a significant association of the four SNPs on introns of BnaC.TT2.a with C18:1 and C18:2 as well as with the long chain FA C22:0 and C22:1, respectively, indicating the underlying role of BnaC.TT2.a in regulating the unsaturated FA composition of Brassica seeds. Moreover, significant associations (P value = 0.0188, R2 = 0.09459) of an Indel (insertion) at position 738 on Exon 3 of BnaC.TT2.a locus were observed with a*, C18:2, and total FA. These associations indicate that alteration or mutation (SNP) at this site could possibly trigger a functional change in the regulation of seed coat color and composition (C18:2) and total seed oil content in BnaC.TT2.a (Table 6). However, T/G change at position 790 of Exon 3 was significantly correlated with seed color traits b*, L, H°, CIRG, and C18:2. The mutation at position 790 (T/G) caused significant changes in the amino acid (Phe/Ser) that could be significantly associated with seed coat color (Table 6). In contrast, we did not find any significant association between this position of BnaC.TT2.a and total seed oil content.
Commercial rapeseed varieties differ in seed color from brown to dark brown or black. Breeders have been interested in yellow-seeded genotypes for decades because of their low anti-nutritive elements and higher accumulation of seed FA and protein . Arabidopsis TT2 prevent FA biosynthesis from developing embryos by directly binding to FUSCA3 at its regulatory region and mediating the expression profile of several genes involvedin the FA biosynthesis . These downstream genes include BCCP2, CAC2, MOD1, and KASII, which regulate the initial steps of FA chain synthesis. Moreover, FAD2 and FAD3 are mainly involved in FA desaturation, whereas, FAE1 catalyzes the chain elongation process of FA. The quantity of PAs in the Arabidopsis seed coat negatively correlates with the accumulation of FA in the embryo. Arabidopsis and Brassica belong to the same family (Cruciferae). Thus, a relationship between the allelic variation of a BnTT2 locus and the rapeseed phenotypic traits, such as seed coat color, FA content, and composition, is speculated. Our results clearly suggest that BnaC.TT2.a associates with seed color and seed FA accumulation in a collection of 83 rapeseed germplasm.
Rapeseed has an amphidiploid genome that originated from interspecific hybridization between B. rapa and B.oleracea. Normally, 2 to 6 homologous copies of a gene are located in different sub-genomes (A or C), and different nucleotide sequences exist among various homologous copies. We aligned the B. napus TT2 sequence with the sequences in the B. napus genome database, and found its best match in C08g07960D, which is allocated to chrC08 from 11760227 bp to 11761328 bp. The match was referred to as BnaC.TT2.a following the standardized gene nomenclature for B. napus by Østergaard and King (2008) . The gene consists of three exons and two introns (Ref: http://www.genoscope.cns.fr/blat-server/cgi-bin/colza/webBlat).
Direct Sanger sequencing of allopolyploid species such as B. napus often results in insufficient sequence quality for SNP detection because of existing homologous genes [36, 46]. Fortunately, we used a pair of published primers that was efficient enough to amplify a single PCR fragment and could generate a high quality sequence chromatograph . We successfully cloned a single copy of BnaC.TT2.a to ensure the accuracy of SNP detection. Successful application of this approach has also been reported in previous studies [46–48]. Sequencing data revealed significant polymorphism at the BnaC.TT2.a locus among the 83 B. napus genotypes, which were categorized into seven haplotype groups based on the accession “DQ778645.1” (Table 4). The SNPs that we detected were both monomorphic and polymorphic. This information enabled us to link the seed phenotypic traits, such as seed coat color and FA, to the genotypic variations evaluated through the DNA sequence alignment of each specimen from the entire germplasm population. We conducted association mapping, for which LD was determined. Phenotypic traits were associated with genotypic variations by applying the GLM + Q model. The blend of phenotypic and genotypic trait associations provided a high number of markers that helped us to accurately investigate the genome-wide diversity and the extent of LD in rapeseed. For instance, previously 845 AFLP markers were used to analyze the extent of LD in 85 winter rapeseed lines. The markers revealed that the highest LD extension in evaluating canola-quality rapeseed was at 2 cM . Moreover, winter rapeseed genotypes were successfully differentiated from other genotypes among 509 inbred lines based on their release dates, levels of erucic acid, and glucosinolates, by analyzing the extent of LD with 89 SSR markers . Furthermore, LD decay within 0.5–1 cM at the genome level was analyzed using 451 SSR markers. The analysis resulted in considerable categorization of a worldwide collection of 192 inbred lines of Brassica based on population size, genetic background, and genetic drift . Similarly, associative transcriptomics of 53 B. napus lines using >60 K SNPs were analyzed by LD. LD analysis identified the transcription factor HAG1 (At5g61420), which regulates the biosynthesis of aliphatic glucosinolate in A. thaliana . Association mapping is conducted based on the LD when regions of the genome are inherited together at a frequency higher than expected based on recombination . In our study, LD analysis resulted in the classification of the population into haplotypes, which we further categorized into two groups (G) based on SNP configurations obtained. The groups included haplotypes H0, H1, and H2 as G1 and haplotypes H3, H4, H5, and H6 as G2. G2 genotypes were observed to have lower oil content and inferior oil quality compared with G1 genotypes. The higher C22:1 level and lower C18:2 and C18:1 composition resulted in higher oil quality in G2 genotypes than that of G1. These properties provide the basis for the selection of genotypes with desirable traits for efficient rapeseed breeding.
R2 is the most relevant LD measurement used to identify the significant association between SNPs or haplotypes and phenotypic trait variations. Typically, R2 values of 0.1 or 0.2 are often used to describe LD decay. However, we only considered SNPs with frequencies ≥ 0.05 for LD decay analysis to avoid ambiguity. We observed six SNPs and one Indel, four of which were located within the first intron and two within the coding sequence (Exon 3) of BnaC.TT2.a. To that end, 33 significant associations (P < 0.05) were observed for six polymorphic sites (Table 6) on BnaC.TT2.a locus. Out of all those associations, an insertion at position 738 on Exon 3 of BnaC.TT2.a was associated with the color index for redness (a*), C18:2, and total FA. Interestingly, the T/G change at position 790 of Exon 3 was significantly correlated with seed color traits b*, L, H°, CIRG, and FA composition C18:2. This mutation at position 790 resulted in an amino acid (Phe/Ser) change and could significantly be associated with multiple seed traits, particularly the yellowness (b*) of the seed coat (Table 6). The amino acid conversion resulted in the functional change of the protein. This result indicates that BnaC.TT2.a is involved in the regulation of seed oil and seed coat color traits, thereby providing a promising significance for rapeseed breeding. Recently, researchers have focused on applying association mapping to evaluate a broad range of plant species in terms of gene(s) identification that could be responsible for the variation in complex quantitative traits with agricultural and evolutionary perspective. Identification of a gene’s function could be done using transgene technology. We successfully developed molecular markers to improve seed oil quality and regulate seed color using BnaC.TT2.a as a candidate gene in B. napus.
Regulating the biosynthesis of color-inducing phenolic compounds, such as PA, is widely studied in A. thaliana. However, the amphidiploid or allotetraploid nature of B. napus and its extensive gene duplication pose a great challenge in laying out the map of candidate genes responsible for seed pigmentation. Despite these limitations, association mapping shows promising features that could help elucidate the genetic basis of complex traits, which are of qualitative, economic, and ecological importance. This is the first study to investigate the candidate genes behind the B. napus seed color and oil traits by association mapping. Our results could broaden the scope of research on seed coat color. Our study showed that quantitative genetic approaches, such as association mapping, could improve map-based cloning of key genes for targeted (desirable or undesirable) seed metabolites or compounds in B. napus.
We thank Ms. Mei Li and Xianyin Zhang for their technical assistance, and EnPaper company for language improvement.
Conceived and designed the experiments: LJ NH. Performed the experiments: LZ YL. Analyzed the data: LZ YL ZL DW. Contributed reagents/materials/analysis tools: LJ LZ. Wrote the paper: LZ NH.
- 1. Nesi N, Delourme R, Brégeon M, Falentin C, Renard M. Genetic and molecular approaches to improve nutritional value of Brassica napus L. seed. Comptes Rendus Biologies. 2008;331(10):763–771. pmid:18926490
- 2. Theodoridou K, Yu P. Effect of processing conditions on the nutritive value of canola meal and presscake. Comparison of the yellow and brown-seeded canola meal with the brown-seeded canola presscake. J Sci Food Agric. 2013;93(8):1986–1995. pmid:23255288
- 3. Yu CY. Molecular mechanism of manipulating seed coat coloration in oilseed Brassica species. J Appl Genet. 2013;54(2):135–145. pmid:23329015
- 4. Marles MAS, Gruber MY. Histochemical characterisation of unextractable seed coat pigments and quantification of extractable lignin in the Brassicaceae. J Sci Food Agr. 2004;84(3):251–262.
- 5. Rahman MH, Joersbo M, Poulsen MH. Development of yellow-seeded Brassica napus of double low quality. Plant Breeding. 2001;120(6):473–478.
- 6. Wang Z, Chen M, Chen T, Xuan L, Li Z, Du X, et al. TRANSPARENT TESTA2 regulates embryonic fatty acid biosynthesis by targeting FUSCA3 during the early developmental stage of Arabidopsis seeds. Plant J. 2014;77(5):757–769. pmid:24397827
- 7. Baudry A, Heim MA, Dubreucq B, Caboche M, Weisshaar B, Lepiniec L. TT2, TT8, and TTG1 synergistically specify the expression of BANYULS and proanthocyanidin biosynthesis in Arabidopsis thaliana. Plant J. 2004;39(3):366–380. pmid:15255866
- 8. Debeaujon I, Peeters AJ, Leon-Kloosterziel KM, Koornneef M. The TRANSPARENT TESTA12 gene of Arabidopsis encodes a multidrug secondary transporter-like protein required for flavonoid sequestration in vacuoles of the seed coat endothelium. Plant Cell. 2001;13(4):853–871. pmid:11283341
- 9. 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
- 10. Nesi N, Jond C, Debeaujon I, Caboche M, Lepiniec L. The Arabidopsis TT2 gene encodes an R2R3 MYB domain protein that acts as a key determinant for proanthocyanidin accumulation in developing seed. Plant Cell. 2001;13(9):2099–2114. pmid:11549766
- 11. Chen MX, Xuan LJ, Wang Z, Zhou LH, Li ZL, Du X, et al. TRANSPARENT TESTA8 inhibits seed fatty acid accumulation by targeting several seed development regulators in Arabidopsis. Plant Physiol. 2014;165(2):905–916. pmid:24722549
- 12. Chai YR, Lei B, Huang HL, Li JN, Yin JM, Tang ZL, et al. TRANSPARENT TESTA 12 genes from Brassica napus and parental species: cloning, evolution, and differential involvement in yellow seed trait. Mol Genet Genomics. 2009;281(1):109–123. pmid:19018571
- 13. Deng W, Chen GQ, Peng F, Truksa M, Snyder CL, Weselake RJ. Transparent Testa16 plays multiple roles in plant development and is involved in lipid synthesis and embryo development in canola. Plant Physiol. 2012;160(2):978–989. pmid:22846192
- 14. Lu J, Li JN, Lei B, Wang SG, Chai YR. Molecular cloning and characterization of two Brassica napusTTG1 genes reveal genus-specific nucleotide preference, extreme protein-level conservation and fast divergence of organ-specificity. Genes Genom. 2009;31(2):129–142.
- 15. Wei YL, Li JN, Lu J, Tang ZL, Pu DC, Chai YR. Molecular cloning of Brassica napus TRANSPARENT TESTA 2 gene family encoding potential MYB regulatory proteins of proanthocyanidin biosynthesis. Mol Biol Rep. 2007;34(2):105–120. pmid:17115250
- 16. 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(9):e44145. pmid:22984469
- 17. Rahman MH. Production of yellow-seeded Brassica napus through interspecific crosses. Plant Breeding. 2001;120(6):463–472.
- 18. Rashid A, Rakow G, Downey RK. Development of yellow seeded Brassica napus through interspecific crosses. Plant Breeding. 1994;112(2):127–134.
- 19. Fu FY, Liu LZ, Chai YR, Chen L, Yang T, Jin MY, et al. Localization of QTLs for seed color using recombinant inbred lines of Brassica napus in different environments. Genome. 2007;50(9):840–854. pmid:17893725
- 20. Padmaja KL, Arumugam N, Gupta V, Mukhopadhyay A, Sodhi YS, Pental D, et al. Mapping and tagging of seed coat colour and the identification of microsatellite markers for marker-assisted manipulation of the trait in Brassica juncea. Theoretical and Applied Genetics. 2005;111(1):8–14. pmid:15902399
- 21. Yan XY, Li JN, Fu FY, Jin MY, Chen L, Liu LZ. Co-location of seed oil content, seed hull content and seed coat color QTL in three different environments in Brassica napus L. Euphytica. 2009;170(3):355–364.
- 22. Delourme R, Falentin C, Huteau V, Clouet V, Horvais R, Gandon B, et al. Genetic control of oil content in oilseed rape (Brassica napus L.). Theor Appl Genet. 2006;113(7):1331–1345. pmid:16960716
- 23. Qiu D, Morgan C, Shi J, Long Y, Liu J, Li R, et al. A comparative linkage map of oilseed rape and its use for QTL analysis of seed oil and erucic acid content. Theoretical and Applied Genetics. 2006;114(1):67–80. pmid:17033785
- 24. Zhao JY, Becker HC, Zhang DQ, Zhang YF, Ecke W. Conditional QTL mapping of oil content in rapeseed with respect to protein content and traits related to plant development and grain yield. Theoretical and Applied Genetics. 2006;113(1):33–38. pmid:16614833
- 25. Zhao JY, Dimov Z, Becker HC, Ecke WG, Mollers C. Mapping QTL controlling fatty acid composition in a doubled haploid rapeseed population segregating for oil content. Mol Breeding. 2008;21(1):115–125.
- 26. Yamamoto E, Iwata H, Tanabata T, Mizobuchi R, Yonemaru J, Yamamoto T, et al. Effect of advanced intercrossing on genome structure and on the power to detect linked quantitative trait loci in a multi-parent population: a simulation study in rice. BMC Genet. 2014;15:50. pmid:24767139
- 27. Darvasi A, Weinreb A, Minke V, Weller JI, Soller M. Detecting marker-QTL linkage and estimating QTL gene effect and map location using a saturated genetic-map. Genetics. 1993;134(3):943–951. pmid:8349116
- 28. Van Ooijen JW. Accuracy of mapping quantitative trait loci in autogamous species. Theor Appl Genet. 1992;84(7–8):803–811. pmid:24201478
- 29. Yang R, Li H, Fu L, Liu Y. An efficient approach to large-scale genotype-phenotype association analyses. Brief Bioinform. 2014;15(5):814–822. pmid:23990269
- 30. Zhu CS, Gore M, Buckler ES, Yu JM. Status and prospects of association mapping in plants. Plant Genome-Us. 2008;1(1):5–20.
- 31. Yu JM, Buckler ES. Genetic association mapping and genome organization of maize. Curr Opin Biotech. 2006;17(2):155–160. pmid:16504497
- 32. Hall D, Tegstrom C, Ingvarsson PK. Using association mapping to dissect the genetic basis of complex traits in plants. Brief Funct Genomics. 2010;9(2):157–165. pmid:20053815
- 33. Atwell S, Huang YS, Vilhjalmsson BJ, Willems G, Horton M, Li Y, et al. Genome-wide association study of 107 phenotypes in Arabidopsis thaliana inbred lines. Nature. 2010;465(7298):627–631. pmid:20336072
- 34. Cowling WA, Balazs E. Prospects and challenges for genome-wide association and genomic selection in oilseed Brassica species. Genome. 2010;53(11):1024–1028. pmid:21076518
- 35. Ecke W, Clemens R, Honsdorf N, Becker HC. Extent and structure of linkage disequilibrium in canola quality winter rapeseed (Brassica napus L.). Theor Appl Genet. 2010;120(5):921–931. pmid:19953219
- 36. Fritsche S, Wang XX, Li JQ, Stich B, Kopisch-Obuch FJ, Endrigkeit J, et al. A candidate gene-based association study of tocopherol content and composition in rapeseed (Brassica napus). Front Plant Sci. 2012;3.
- 37. Wang NA, Qian W, Suppanz I, Wei LJ, Mao BZ, Long Y, et al. Flowering time variation in oilseed rape (Brassica napus L.) is associated with allelic variation in the FRIGIDA homologue BnaA.FRI.a. Journal of Experimental Botany. 2011;62(15):5641–5658. pmid:21862478
- 38. Shen Y, Jin L, Xiao P, Lu Y, Bao JS. Total phenolics, flavonoids, antioxidant capacity in rice grain and their relations to grain color, size and weight. J Cereal Sci. 2009;49(1):106–111.
- 39. Zou J, Katavic V, Giblin EM, Barton DL, MacKenzie SL, Keller WA, et al. Modification of seed oil content and acyl composition in the brassicaceae by expression of a yeast sn-2 acyltransferase gene. Plant Cell. 1997;9(6):909–923. pmid:9212466
- 40. Zhu YN, Cao ZY, Xu F, Huang Y, Chen MX, Guo WL, et al. Analysis of gene expression profiles of two near-isogenic lines differing at a QTL region affecting oil content at high temperatures during seed maturation in oilseed rape (Brassica napus L.). Theoretical and Applied Genetics. 2012;124(3):515–531. pmid:22042481
- 41. Pritchard JK, Stephens M, Donnelly P. Inference of population structure using multilocus genotype data. Genetics. 2000;155(2):945–959. pmid:10835412
- 42. Evanno G, Regnaut S, Goudet J. Detecting the number of clusters of individuals using the software STRUCTURE: a simulation study. Mol Ecol. 2005;14(8):2611–2620. pmid:15969739
- 43. Stewart CN Jr., Via LE. A rapid CTAB DNA isolation technique useful for RAPD fingerprinting and other PCR applications. Biotechniques. 1993;14(5):748–750. pmid:8512694
- 44. Bradbury PJ, Zhang Z, Kroon DE, Casstevens TM, Ramdoss Y, Buckler ES. TASSEL: software for association mapping of complex traits in diverse samples. Bioinformatics. 2007;23(19):2633–2635. pmid:17586829
- 45. Ostergaard L, King GJ. Standardized gene nomenclature for the Brassica genus. Plant Methods. 2008;4:10. pmid:18492252
- 46. Westermeier P, Wenzel G, Mohler V. Development and evaluation of single-nucleotide polymorphism markers in allotetraploid rapeseed (Brassica napus L.). Theor Appl Genet. 2009;119(7):1301–1311. pmid:19756476
- 47. Durstewitz G, Polley A, Plieske J, Luerssen H, Graner EM, Wieseke R, et al. SNP discovery by amplicon sequencing and multiplex SNP genotyping in the allopolyploid species Brassica napus. Genome. 2010;53(11):948–956. pmid:21076510
- 48. Ganal MW, Altmann T, Roder MS. SNP identification in crop plants. Curr Opin Plant Biol. 2009;12(2):211–217. pmid:19186095
- 49. Bus A, Korber N, Snowdon RJ, Stich B. Patterns of molecular variation in a species-wide germplasm set of Brassica napus. Theor Appl Genet. 2011;123(8):1413–1423. pmid:21847624
- 50. Xiao Y, Cai D, Yang W, Ye W, Younas M, Wu J, et al. Genetic structure and linkage disequilibrium pattern of a rapeseed (Brassica napus L.) association mapping panel revealed by microsatellites. Theor Appl Genet. 2012;125(3):437–447. pmid:22437490
- 51. Harper AL, Trick M, Higgins J, Fraser F, Clissold L, Wells R, et al. Associative transcriptomics of traits in the polyploid crop species Brassica napus. Nat Biotechnol. 2012;30(8):798–802. pmid:22820317
- 52. Collins AR. Linkage disequilibrium and association mapping: an introduction. Methods Mol Biol. 2007;376:1–15. pmid:17984534