Fine mapping of the QTL cqSPDA2 for chlorophyll content in Brassica napus L.

Background Chlorophyll is the most important factor enabling plants to absorb, transfer and transform light energy and plays an important role in yield formation. Brassica napus is one of the most important oil crops. Breeding Brassica napus for high light efficiency by improving photosynthetic efficiency has considerable social and economic value. In Brassica napus, there have been studies of the initial location of chlorophyll in seed embryos and pericarps, but there are few reports on the fine mapping of chlorophyll QTLs. We constructed near-isogenic lines (NIL), fine-mapped a chlorophyll locus, and evaluated the effect of this dominant locus on agronomic traits. Results The cqSPDA2 locus was mapped to an interval of 21.87–22.91 Mb on the chromosome A02 of Brassica napus using doubled haploid (DH) lines. To fine-map cqSPDA2, we built NIL and designed Indel primers covering the mapping interval. The 469 individuals in the BC3F2 population were analyzed using these indel primers. Among these indel primers, 15 could narrow the mapping interval to 188 kb between Indel3 and Indel15. Next, 16 indel primers and 19 SSR primers were designed within the new narrower mapping interval, and 5 of the primer-amplified fragments were found to be polymorphic and tightly linked to the cqSPDA2 locus in the BC4F2 population. The mapping interval was narrowed to 152 kb on A02 between SSR2 and Indel15. By gene expression analysis, we found three annotated genes in the mapping interval, including BnaA02g30260D, BnaA02g30290D and BnaA02g30310D, which may be responsible for chlorophyll synthesis. Conclusions The locus cqSPDA2, a dominant QTL for chlorophyll content in Brassica napus, was fine-mapped to a 21.89–22.04 Mb interval on A02. Three annotated genes (BnaA02g30260D, BnaA02g30290D and BnaA02g30310D) that may be responsible for chlorophyll synthesis were found. Supplementary Information The online version contains supplementary material available at 10.1186/s12870-020-02710-y.


Background
The material basis of crop yield formation is derived from photosynthesis. Studying crop high yield from photosynthesis is becoming a new breeding hotspot [1]. Chlorophyll is the most important factor enabling plants to absorb, transfer and transform light energy and plays an important role in the growth and development of plants [2]. Maintaining a high level of chlorophyll content in leaves is an important factor in increasing photosynthetic activity [3]. In a certain range, there is a positive correlation between chlorophyll content and photosynthetic rate, which directly determines the yield [4,5]. Therefore, chlorophyll content plays an important role in yield formation [6,7]. The seedling development of Brassica napus leads to a higher yield stability and has a high importance for plant breeders [8]. Chlorophyll content is a quantitative characteristic that is primarily controlled by nuclear genes and has high heritability. In recent years, researchers have analyzed QTLs of chlorophyll content in seedling leaves of different crops with different populations from different perspectives and made considerable progress, establishing a foundation for future research attempting to elucidate the molecular genetic mechanism of chlorophyll [9][10][11][12][13][14][15].
The completion of the whole genome sequencing of Brassica napus indicates that research on the Brassica napus genome has entered a new era. In recent years, with the rapid development of molecular marker technology, it has become possible to construct a high-density molecular marker genetic map of Brassica napus. Therefore, e cient light breeding of Brassica napus with chlorophyll QTLs is an important breakthrough in improving the yield potential and direction of high-yield breeding of Brassica napus in the future. Qinghai Province is located in the Qinghai-Tibet Plateau in northwestern China. The solar radiation and its intensity during the growth period of crops are higher than those in the interior of China, and the light duration is longer. Under this light condition, performing high light e ciency breeding of Brassica napus from the perspective of improving photosynthetic e ciency has strong social and economic value. At present, studies on the initial location of chlorophyll content in seed embryos [16] and pericarps [17] of winter Brassica napus have been conducted, QTL loci have been identi ed under drought and salt stress, and even candidate genes related to salt tolerance have been predicted [18][19][20]. However, studies on the ne mapping of chlorophyll content QTLs in Brassica napus have rarely been reported.
In a previous study, we discovered a dominant QTL named cqSPDA2 located in 21.87-22.91 Mb on A02 by DH lines from a cross between Zhongshuang11 (ZS11) and QU (under review). In this study, the near isogenic line (NIL) of cqSPDA2 was constructed by anking markers closely linked to cqSPDA2 and backcross selection. The near isogenic lines were analyzed using a molecular marker to further narrow the cqSPDA2 to the range of 150 kb.
This study established a foundation for future research investigating the cloning of chlorophyll genes controlling photosynthetic function and provided a theoretical basis for improving germplasm resources and selecting new high-yield varieties by molecular markers.

Phenotypic and genetic analysis
The rst fully developed leaves counting from the top of the 2061 individuals in the BC 4 F 2 populations at the six-leaf stage were measured by SPAD (SPAD 502, Japan). The Chi-square test showed that the segregation pattern of the chlorophyll content trait was in keeping with the expected Mendelian segregation ratio of 3:1 (χ 2 = 2.53) (SPAD = 43, high chlorophyll content vs. low chlorophyll content) (Fig. 1). Among the random selection of 198 individuals in the BC 6 F 1 population, the markers SSR2 and Indel100 were used to validate the effect of cqSPDA2. The result of a Chi-square test was in keeping with a 1:1 (χ 2 = 1.46; χ 2 = 1.82) (AA:Aa) Mendelian ratio (Additional le 1: Table S1).

Fine mapping of cqSPDA2
To ne map the cqSPDA2 locus and identify the candidate genes, 87 primer pairs of Indel markers were designed to uniformly cover the preliminary mapping interval 21.87-22.91 As a result, 28 polymorphic markers were detected with the two parental lines and some DH lines. Twenty-three of these markers were found to be cosegregated in the DH lines. The 469 individuals in the BC 3 F 2 population were analyzed using these Indel primers. The linkage map constructed using the Indel data and corresponding chlorophyll content phenotypes showed that 15 Indel primers were tightly linked with the cqSPDA2 locus (Additional le 2: Table S2). The cqSPDA2 locus was delimited to an interval of 5.2 cM between Indel3 and Indel15 (Additional le 3: Fig. S1). The fragments of the primers indel3, indel6, indel15 and indel 17 near the QTL locus were recovered. TA clone analysis was performed with the PMD18T vector, and the physical location of the region was found to be within the 188-kb range of 21.88-22.07 Mb (Additional le 4: Figure S2).
Next, 16 Indel and 19 SSR primers were designed within the new narrow mapping interval, and 7 primers were polymorphic and tightly linked to the cqSPDA2 locus (Additional le 5: Table S3). These new primers helped to narrow the interval for 250 individuals in the BC 4 F 2 population. As a result, the cqSPDA2 locus was mapped to a 152-kb interval between SSR2 and Indel15 ( Fig. 2). BSNP88 and BSNP90 are SNP markers developed in the interval (Additional le 6: Table S4). SSR2 is a codominant marker and closely linked with cqSPDA2. Twenty plants with low, medium and high chlorophyll phenotypes in BC 4 F 2 were selected, and the corresponding genotypes were identi ed by SSR2. The results showed that the three groups of different phenotypes can be divided into three genotypes: AA, Aa and aa. It is suggested that SSR2 was closely linked with cqSPDA2 and could be effectively used in MAS breeding (Fig. 3).

Quantitative RT-PCR of genes in the mapping interval
According to the Brassica napus genome annotation database (http://www.genoscope.cns.fr/brassicanapus/), twenty-seven genes were identi ed in the targeted mapping interval 21.89-22.04 Mb on A02 (Additional le 7: Table S5). The melting curve and ampli cation curves of twenty-seven genes were analyzed, and the results showed that 24 primers can be used to analyze gene expression (Additional le 8: Table S6). The twenty-four genes and the housekeeping gene Actin7 were quanti ed by qRT-PCR (Additional le 9: Table S7). The results showed that the expression levels of three genes (BnaA02g30260D, BnaA02g30290D and BnaA02g30310D) were all higher in ZS11 and BC 4 F 2:3 (AA) than in QU and BC 4 F 2:3 (aa) at the three stages. There were signi cant differences between BC 4 F 2:3 (AA) and BC 4 F 2:3 (aa) at the 6-leaf stage in BnaA02g30290D and BnaA02g30310D (p < 0.05), and there was a highly signi cant difference in BnaA02g30260D (p < 0.01) (Fig. 4). There was no consistent expression of other genes in QU, BC 4 F 2:3 (AA), ZS11 and BC 4 F 2:3 (aa) at the three stages. Therefore, BnaA02g30260D, BnaA02g30290D and BnaA02g30310D were likely candidate genes for cqSPDA2.

Agronomic traits analysis
To investigate the effect of cqSPDA2 on agronomic traits, 50 plants with the AA genotype (high chlorophyll content) and 50 plants with the aa genotype (low chlorophyll content) were selected from the BC 4 F 2 population by molecular marker and SPAD. We investigated plant height, silique length, number of seeds per silique, number of siliques per plant, 1000-seed weight, and individual plant yield. The results showed that plant height, number of seeds per silique, number of siliques per plant and individual plant yield were highly signi cantly different between the AA genotype plants and the aa genotype plants (P < 0.01). There was a difference in 1000 grain weight but no difference in silique length (P < 0.05) (Additional le 10: Table S8).

Discussion
Leaf is the main photosynthetic organ, and chlorophyll content is an important agronomic trait for plant yield. Ninety to ninety-ve percent of plant dry matter is produced by photosynthesis, and crop yield is primarily derived from the photosynthetic products of leaves [21]. Chlorophyll is an important pigment involved in photosynthesis in chloroplasts, which can absorb and transform light energy. Chlorophyll is also an important index to evaluate the photosynthetic capacity of leaves [22]. Increasing crop yield by increasing chlorophyll content is one of the important breeding objectives of high light e ciency breeding [23]. Chlorophyll content is a quantitative characteristic that is primarily controlled by nuclear genes and has high heritability. At present, research on chlorophyll QTLs has been performed on various crops, such as rice [8,10,12,13,15], wheat [9,11], beans [24,25], and cabbage [14], especially rice [13]. There have been studies on the location of chlorophyll in the embryo [16] and pericarp [17] of winter Brassica napus. Under drought and salt stress conditions, QTLs were also detected, and even a candidate gene related to salt tolerance was predicted [18][19][20]. However, there is no report on the ne location of the chlorophyll QTL locus in Brassica napus. This report is the rst to describe ne mapping of cqSPDA2 on Brassica napus. In this study, the NILs of cqSPDA2 regions were constructed using recurrent parent ZS11 with anked markers. Indel and SSR markers were used to scan the populations of BC 3 F 2 and BC 4 F 2 , and the mapping interval was reduced to 152 kb. This strategy for mapping the target genes is reasonable, inexpensive, and highly e cient. According to the BC 4 F 2 phenotype and BC 6 F 1 genotype analysis, the strategy is consistent with the Chi-square test. The results indicated that cqSPDA2 is the main QTL locus. A positive correlation between cqSPDA2 and agronomic characteristics, such as yield, was determined through the analysis of NIL.
BnaA02g30260D, BnaA02g30290D and BnaA02g30310D were identi ed as good candidate genes of qSPADA2 among the twenty-nine genes annotated from Brassica napus genomes in the mapping interval according to the qRT-PCR analysis. BnaA02g30260D, which is a disease-resistance protein family, has transmembrane receptor activity, nucleoside-triphosphatase activity, nucleotide binding, and ATP binding function and is involved in signal transduction, defense response, apoptosis, and innate immune response according to the annotations. Further study is needed to determine whether BnaA02g30260D affects chlorophyll synthesis. BnaA02g30290D is FK506-and rapamycin-binding protein 15 kD-2 (FKBP15-2) and has peptidyl-prolyl cis-trans isomerase activity related to protein folding. Luan et al. [26] found that AtFKBP15-1 and AtFKBP15-2 had the highest homology to FKBP13 and encoded functional homologs of FKBP13. AtFKBP13 was reported to be associated with Rieske protein, both before and after the import of the proteins into the chloroplast stroma. AtFKBP13 can play a role in the downregulation of Rieske protein accumulation. Rieske is a subunit of the cytochrome b 6 f complex, which is one of the four complexes of the photosynthetic electron transport chain [27]. It was also reported that ScFKBP12 was transferred into Arabidopsis, chloroplast formation was inhibited and the expression of genes related to chloroplast formation was inhibited [28]. In this study, the expression levels of BnaA02g30290D (AtFKBP15-2) in NILs (aa) and ZS11 were all higher than those in NILs (AA) and QU at the three stages, which inhibited the formation of chlorophyll and was consistent with the above mentioned results. BnaA02g30310D is homologous to GCH-1 in Arabidopsis thaliana. GCH-1 is the rst enzyme for tetrahydrobiopterin (BH4) biosynthesis [29]. BH4 is an essential coenzyme for all three kinds of nitric oxide synthase (NOS) [30]. AtNOA1 (AtNOS1) is located in Arabidopsis chloroplasts, and OsNOA1 (OsNOS1) is also located in rice chloroplasts [31][32][33]. Yang et al. [33] found that the chlorophyll content decreased with increasing OsNOA1 at a low temperature (22 °C). He [34] suggested that OsNOA1 directly regulates the chloroplast self-coding protein by affecting the function of the chloroplast ribosome and then transmits the signal to the nucleus through the chloroplast retrograde signal pathway mediated by Mg protoporphyrin IX, further affecting the expression of chloroplast protein encoded by the nuclear gene. Qinghai Province is located in the Qinghai-Tibet Plateau. The average temperature during crop growth is lower, and the expression level of BnaA02g30310D (GCH-1) in NILs (aa) and ZS11 was more strongly expressed at three stages in the study, especially at the six-leaf stage, which is consistent with the results obtained by He [33,34].
In addition, more research, such as a transgenic complementary test, CRISPR/Cas9, VIGS and RNAi, is warranted to examine whether BnaA02g30260D, BnaA02g30290D and BnaA02g30310D are the target genes of cqSPDA2. Analysis of the regulatory network for chlorophyll synthesis will facilitate Brassica napus molecular breeding for high yield.

Conclusions
In the study, we built near-isogenic lines and narrowed the interval of cqSPDA2 to 152 kb on A02 between SSR2 and Indel15. According to the Brassica napus genome annotation database, there were twentyseven genes in the targeted mapping interval. BnaA02g30260D, BnaA02g30290D and BnaA02g30310D were identi ed as good candidate genes of cqSPAD A2 according to the qRT-PCR analysis, which may be responsible for chlorophyll synthesis were found. The dominant locus cqSPDA2 has positive effects on agronomic traits.

Plant materials
The leaves of ZS11 have low chlorophyll content, and the leaves of QU have high chlorophyll content (Additional le 11: Fig. S3). To investigate the genetic control regulation mechanism for the leaf chlorophyll content trait, we crossed ZS11 with QU to produce F 1 populations. In a previous study, the main effect QTL cqSPDA2 was detected. To obtain a relatively simple genetic background and to ne map cqSPDA2, we constructed the near-isogenic line (NIL). The F 1 line with the QU genotype in the cqSPDA2 region was selected and backcrossed with ZS11 for three generations. BC 3 F 1 individuals were selfed to generate BC 3 F 2 mapping populations backcrossed with ZS11. The markers Indel1 and Indel87 were used for marker-assisted selection (MAS) of each generation among the segregating progenies.
The BC 4 F 1 individuals with a QU genetic background in the cqSPDA2 region selected with the markers indel3 and indel15 were selfed to generate BC 4 F 2 populations for ne mapping of the cqSPDA2 locus.
The detailed process of population development is illustrated in Additional le 12: Fig. S4. BC 4 F 2:3 (AA genotype with cqSPDA2 and aa genotype without cqSPDA2) were detected for qRT-PCR analysis. The BC 3 F 2 and BC 4 F 2 populations were grown at the same density in elds in Yunnan and Xining, respectively. BC 4 F 2:3 and BC 6 F 1 populations were grown in a greenhouse at the Academy of Agricultural and Forestry Sciences, Qinghai University. Spacing was maintained at 30 cm between rows and 15 cm between plants. Standard crop management practices were followed.

Phenotypic trait and data analysis
The testing targets were every plant of populations that eliminated diseases and insect pests. According to previous research, we measured the rst fully developed leaves counted from the top at the six-leaf stage by SPAD. Each measurement was repeated three times. Statistical analysis was calculated by Excel. Chi-square tests were performed on the segregation data to determine the genetic regulation of the chlorophyll content.

DNA extraction and development of molecular markers
Total DNA was extracted from fresh leaves using the CTAB method [35]. PCR was performed in a 20-µL reaction solution containing 2 µL DNA, 2 µL 2 mM dNTPs, 2 µL 10 × PCR buffer, 1 µL Taq, 1 µL of 2 µM forward and reverse primers and 12 µL of ddH 2 O. The PCR program was carried out according to Yang's method with minor modi cations [36]. The PCR products were separated on 6% nondenatured polyacrylamide gels and detected by silver staining [37]. Indel (insertion/deletion) markers were developed from the target region to determine recombination sites and the genotype of recombinant progenies based on a previous study. The sequences of SSR markers were designed using SSR Hunter

TA clone
The speci c markers closely linked to cqSPDA2 were sequenced by NIL population scanning. Speci c fragments were collected according to Yi et al. [38]. The product was connected to the pDM18-T vector (Takara), and the transformed clone was detected with primers M13. Six positive clones were randomly selected and sequenced by Sangon Biotech (Shanghai) Co., Ltd. [39].
Genes in the mapping interval

RNA extraction and qRT-PCR analysis
Total RNA was isolated from leaves (4 leaf stage, 6 leaf stage and squaring stage) of BC 4 F 2:3 and parents using TRNzol-A + Total RNA Reagent (Takara, Dalian, China) according to the manufacturer's protocol. RNA integrity was monitored using 1% agarose gel electrophoresis. cDNA was obtained via reverse transcription of total RNA using the PrimeScript RT reagent Kit (Takara, Dalian, China) and following the manufacturer's instructions.
We performed a quantitative reverse transcription-PCR (qRT-PCR) analysis to determine the genes in the mapping interval. Real-time PCR was conducted using LightCycler 480 II 96-Well PCR Plates (Roche, Rotkreuz, Switzerland). The utilized reaction system contained 10 µL of 2 × SG Fast qPCR Master Mix (B639271, BBI), 2 µL cDNA, and 10 µM gene-speci c primers in a nal volume of 20 µL. The thermal cycling conditions used were 95 °C for 30 min, followed by 45 cycles at 95 °C for 5 s and 60 °C for 30 s followed by a nal extension stage. The housekeeping gene Actin7 was used as a reference gene for calculating the relative expression levels of each gene.

Agronomic traits
To evaluate the agronomic e ciency of cqSPDA2, 100 individuals (50 AA and 50 aa genotype plants) were sampled using the markers indel3 and indel15 referencing chlorophyll content from the BC 4 F 2 population, respectively. The agronomic traits investigated were as follows: plant height, total siliques per plant, silique length, seeds per silique, 1000-seed weight, and yield per plant. The mean values, standard deviations and signi cant differences analysis of all the agronomic traits were compared between AA and aa genotype plants by Minitab16 and Excel2010.  Genetic and physical maps of the cqSPDA2 gene locus and candidate gene analysis.