Development of the PARMS Marker of the Dominant Genic Male Sterility (DGMS) Line and Its Utilization in Rapeseed (Brassica napus L.) Breeding

The 8029AB line is a dominant genic male sterility (DGMS) two-type line in Brassica napus L., which can be used in a three-line approach for the seed production of rapeseed hybrids. Genetic analyses have demonstrated that the sterility of 8029A is controlled by a single dominant nuclear gene (BnMS5e) interacting with one recessive gene (BnMS5c). Six pairs of penta-primer amplification refractory mutation system (PARMS) markers were designed according to the sequence of BnMS5a, BnMS5c and BnMS5e. Two pairs of these PARMS markers were successfully identified and validated. The PARMS markers MS5-1Fc/MS5-1Ft/MS5-1R12 could distinguish BnMS5c from BnMS5a/BnMS5e, and the PARMS markers MS5-2Ft/MS5-2Fa/MS5-1R12 could genotype BnMS5a and BnMS5c/BnMS5e. The combination of these two pairs of PARMS markers could be used to identify the presence or absence of BnMS5a/BnMS5c/BnMS5e effectively. Consequently, marker-assisted selection can be carried out in the early generation to shorten the breeding period and improve the breeding efficiency.


Introduction
Rapeseed is the second most important oilseed crop in the world [1]. Heterosis utilization has become an important strategy to increase the productivity of rapeseed [2,3]. Remarkable heterosis had been well demonstrated in rapeseed [4]. Male sterility has been proved to be an economical and effective way to utilize heterosis in crops. At present, the main types of male sterility used worldwide in Brassica napus L. (oilseed rape) are the male sterility Lembke (MSL) in Europe and North America [5], Plant Genetic Systems (PGS) in North America [6], Ogura-Inra in North America and Europe [7], and Polimar cytoplasmic male sterility (Pol CMS) in China [8]. Compared with cytoplasmic male sterility (CMS), genic male sterility (GMS) has many advantages in heterosis utilization, including the facts that it represents stable and complete male sterility, it is easy to take advantage of the strong combination, it has a broad spectrum of restoration, there are no negative cytoplasmic effects and it includes diverse cytoplasmic sources [9]. GMS has a great potential in heterosis utilization, and some GMS systems have been used widely. However, recessive GMS (RGMS) has obvious disadvantages, i.e., it is difficult to obtain a whole sterile population, and 50% of fertile plants in the sterile line must be artificially removed during reproduction and seed production. The question of how to remove 50% of the fertile plants in the mother line with an economical and effective method is the main problem that has to be solved in the widespread application of nuclear sterility.
As an effective alternative to RGMS, dominant genic male sterility (DGMS) has been studied in B. napus. To date, some spontaneous DGMS mutants (Yi3A, 609A and TE5A) were systematically characterized. Genetic analyses confirmed that the sterility of Yi3A is controlled by one gene with three different alleles, which are MS5 a (corresponding to the previous Rf ), MS5 b (corresponding to the Ms) and MS5 c (corresponding to the ms), respectively [10,11]. Individuals with the homozygous MS5 b allele are male sterile, and the sterility is maintained by crossing with the lines carrying the homozygous MS5 c . Another allele, BnMS5 d , was derived from BnMS5 a by mutagenesis and confers male sterility when it is homozygous [12]. Specifically, the BnMS5 a allele is dominant over BnMS5 b or BnMS5 d , but BnMS5 c is recessive to BnMS5 b or BnMS5 d . BnMS5 a BnMS5 a lines are referred to as restorer lines because they restore fertility when they crossed with BnMS5 b BnMS5 b /BnMS5 d BnMS5 d lines. BnMS5 c BnMS5 c lines are called temporary maintainer lines, because BnMS5 b BnMS5 c /BnMS5 c BnMS5 d plants are male sterile [11][12][13].
Our previous studies showed that 8029AB is a population with male fertile and sterile individuals, which can be segregated. It is a new type of DGMS sterile material whose fertility is controlled by a pair of multiple alleles (BnMS5 a /BnMS5 c /BnMS5 e ). The fertility was normal when BnMS5 c was homozygous and BnMS5 a was homozygous or heterozygous. Individuals with genotypes BnMS5 a and BnMS5 c BnMS5 c were fertile, while plants with genotypes BnMS5 e BnMS5 e and BnMS5 c BnMS5 e were sterile. This material can simulate genic male sterility to achieve three lines of seed production: a two-type sterile line (AB line, genotype: BnMS5 e BnMS5 e and BnMS5 a BnMS5 e ), a maintainer line (genotype: BnMS5 c BnMS5 c ) and a restorer line (genotype: BnMS5 a BnMS5 a ). The sterile line crosses with the maintainer line to produce a 100% sterile line; then, the 100% sterile line crosses with the restorer line to produce hybrids. Consequently, the problem that 50% fertile individuals must be removed from the parent line will be solved in three-line DGMS hybrid seed production.
The application of DGMS is still relatively limited because of the low efficiency of identifying complex genotypes with the desired and complicated agronomic traits, which is laborious and time consuming in the traditional breeding process. It is beneficial to develop molecular markers tightly linked to the sterility gene, maintainer gene and restorer gene for timely genotype identification in a DGMS three-line breeding system. Many AFLP markers linked to BnMS5 have been identified; some of them were converted into SCAR markers [10,14,15]. Given the loose linkage relationship of some markers with BnMS5, the flanking SCAR markers BE10 and SCD8 restrict BnMS5 to a 21 kb region containing six predicted functional genes [11]. Nevertheless, it remains unclear whether these markers are informative enough for molecular marker-assisted selection (MAS) in a DGMS three-line breeding system. Markers tightly linked with the sterility gene, maintainer gene and restorer gene, respectively, have not been developed. In this study, we report the genetics analysis of a new type of DGMS 8029A. We developed and validated two PARMS markers that were tightly linked to the sterility gene, maintainer gene and restorer gene, respectively. The application of these markers in high-throughput MAS of a new DGMS three-line hybrid system is demonstrated.

Plant Materials
8029AB is a newly bred, dominant genic male sterile (DGMS) two-type line in Brassica napus (Figure 1). The 8029A line (genotype: BnMS5 e BnMS5 e ) was the sterile line, Zhongshuang11 (genotype: BnMS5 c BnMS5 c ) was the maintainer line and 6449 (genotype: BnMS5 a BnMS5 a ) was the restorer line. All the materials were procured from the Oil Crop Research Institute of the Chinese Academy of Agricultural Sciences. The seed setting of 8029A was normal under the conditions of artificially assisted pollination or natural pollination, indicating that the seed setting rate was not affected by the male sterile gene ( Figure 1).

Genetic Analysis
One hybrid population (F 1 -1) was derived from the cross between the homozygous sterile plants of 8029A and the maintainer line Zhongshuang11. The BC 1 -1 line was obtained by backcrossing F 1 -1 to Zhongshuang11. The phenotype of F 1 -1 and the segregation ratio of the BC 1 -1 population were identified in the field to detect the genetic pattern of 8029A.
Furthermore, to analyze the genetic pattern of the restorer line 6449, F 1 -2, F 2 -2 and BC 1 -2 populations were constructed. F 1 -2 was obtained by crossing the homozygous sterile plants of 8029A with the restorer line 6449. The F 2 -2 population was derived from selfpollination of the F 1 -2 plants. BC 1 -2 was derived by crossing the homozygous sterile plants of 8029A with the F 1 -2 plants. The F 1 -2 phenotype and the segregation ratios of the F 2 -2 and BC 1 -2 populations were identified directly in the field.

DNA Extraction and PCR
The genomic DNA of young leaves, which were collected from individual plants at the seedling stage, was extracted by using the CTAB method, with some modifications. The DNA concentration was measured using a NanoDrop 2000 spectrophotometer (Thermo, Wilmington, Delaware, USA) and adjusted to 50 ng/µl.
The BnMS5 gene of these B. napus lines was amplified via PCR. The master mix for the PARMS markers was purchased from Gentides Biotech Co., Ltd. (Wuhan, China). The 10 µl PCR reaction system contained 2× PARMS PCR reaction mix (containing two common fluorescent primers, PCR buffer, dNTP, Taq enzymes and internal standard ROX), 150 nM of each allele-specific primer, 400 nM of locus-specific primer, 50 ng of DNA template and 3.3 µl of ddH 2 O. The reaction conditions of PCR were optimized with parameters as follows: 95°C for 5 min, followed by 10 cycles of 95°C for 20 s and 65°C (−0.8°C per cycle) for 1 min, and 32 cycles of 95°C for 20 s and 57°C for 1 min and 72°C for 5 min. The well plate was read using a TECAN infinite M1000 plate reader; SNP calling and plots were carried out using the online software snpdecoder (http://www.snpway.com/snpdecoder/ accesed on 17 June 2021) combining manual modification.

PARMS Markers Design and Markers Analysis
According to the sequence of BnMS5 homologs, BnMS5 e was amplified in 8029A. BnMS5 e was sequenced and compared with BnMS5 a using BLAST. BnMS5 e differs from BnMS5 a by a few bases. Then, BnMS5 e was transferred to the restorer line 6449. Part of transgenic plants were male sterile.
According to the sequence of BnMS5 a (KX223882), BnMS5 c (KX223881) and BnMS5 e (Supplementary Materials Additional File S1), the SNP sites were excavated, and SNP markers were designed using Primer Premier 5.0. The PARMS molecular marker MS5-1F was designed based on the SNP variations from AC (MS5 c ) to GT (MS5 a /MS5 e ). Additionally, MS5-2F was designed based on the SNP variations from T (MS5 c /MS5 e ) to A (MS5 a ). The difference loci were placed in the position of the second base from the 3' end of the primers; then, the fluorescent markers were added. MS5-1F and MS5-2F share a pair of reverse primers MS5-1R12. Two common primers which are consistent with the underlined parts of the two forward primers had different fluorescent markers at the tail. The underlined sequence was annealed to the oligonucleotides labeled with the FAM/HEX fluorophore ( Table 1). The primers were commercially synthesized by Gentides Biotech Co., Ltd (Wuhan, China).

Genetics Analysis of the Sterile Line 8029A and the Restorer Line 6449
The phenotype of the populations was thoroughly investigated in the field (Table 2). F 1 -1 individuals from the cross between 8029A and Zhongshuang11 were all male sterile, which were subsequently backcrossed with Zhongshuang11, resulting in a BC 1 -1 population that comprised 238 individuals (116 male sterile: 122 normal fertile). The segregation ratio of sterile individuals to fertile individuals showed the expected Mendelian inheritance ratio of 1:1 (χ 2 = 0.15, p > 0.05). The data indicate that the sterility trait of 8029A is controlled by a single dominant nuclear allelic gene (BnMS5 e ).

Validation of the PARMS Markers
Thirty-six B. napus materials with genotypes BnMS5 a BnMS5 a , BnMS5 c BnMS5 c or BnMS5 c BnMS5 e were genotyped using the BnMS5 functional molecular fluorescence markers MS5-1Fc/MS5-1Ft/MS5-1R12 and MS5-2Ft/MS5-2Fa/MS5-1R12, respectively (Figure 2  To further validate these two pairs of PARMS markers in the detection of BnMS5 a , BnMS5 c and BnMS5 e , the BC 1 -1, F 2 -2 and BC 1 -2 populations were selected to detect the polymorphism using MS5-1Fc/MS5-1Ft/MS5-1R12 and MS5-2Ft/MS5-2Fa/MS5-1R12. The amplification results of each plant were classified according to fluorescent signals. The genotypes of plants were classified based on a scatter diagram. The scatter diagram intuitively classified the corresponding genotype of each material. A comparison of the statistical data revealed that the phenotypic differentiation of the marker MS5-1Fc/MS5-1Ft/MS5-1R12 and MS5-2Ft/MS5-2Fa/MS5-1R12 was consistent with the field phenotypes statistics, indicating that the fluorescent molecular markers detected the SNP variations quickly and accurately.

Discussion
Several DGMS mutants were found in B. napus, and the inheritance of DGMS has been studied intensively [16][17][18][19][20]. However, due to their genetic complexity and low efficiency in the traditional breeding process, these DGMS materials have rarely been applied to large-scale commercial hybrid seed production. The 8029AB line is a new DGMS two-type line whose fertility is controlled by a pair of multiple alleles (BnMS5 a /BnMS5 c /BnMS5 e ). The sterility of 8029A was complete, and the seed setting rate was not affected by male sterile gene. Thus, this material can be used in a three-line approach for the seed production of rapeseed hybrids: a two-type sterile line (AB line, genotype: BnMS5 e BnMS5 e and BnMS5 a BnMS5 e ), a maintainer line (genotype: BnMS5 c BnMS5 c ) and a restorer line (genotype: BnMS5 a BnMS5 a ). The sterile line crosses with the maintainer line to produce 100% sterile population, and then, the 100% sterile line and restorer line are used to produce hybrids.
The development of new DGMS lines as well as maintainer lines and excellent restorer lines through conventional backcrossing and selfing is time consuming and labor intensive. To facilitate the three-line breeding process, many attempts have been made to develop a molecular marker linked to BnMS5. Two AFLP markers were firstly identified at either side of the BnMS5 b allele, at a distance of 3.7 and 5.9 cM, respectively [10]. Subsequently, several SCAR markers linked to BnMS5 a were developed, which followed a dominant inheritance mode, and it is impossible to distinguish between heterozygous and homozygous individuals at the BnMS5 a site [14]. Meanwhile, using DGMS line 609AB, the marker linked to BnMS5 a was conversed into a marker linked to BnMS5 b , providing proof of allelism between previously Ms and Mf genes [21]. Later, two closer markers with respective distances of 0.1 and 1.2 cM to the BnMS5 locus were reported [22]. Given the loose linkage relationship of some markers with BnMS5, the flanking SCAR markers BE10 and SCD8 restrict BnMS5 to a 21 kb region containing six predicted functional genes [11]. Nevertheless, it remains unclear whether these markers are informative enough for MAS in a DGMS three-line breeding system. What is more, most molecular markers are random DNA markers which might be lost during recombination [23].
Gel-electrophoresis-based markers such as SSR (simple sequence repeat), AFLP (amplified fragment length polymorphism), SCAR (sequence-characterized amplified region) and CAPS (cleaved amplified polymorphic sequence) have been used in molecular breeding [24]. However, electrophoresis is time consuming, laborious and has a low throughput. To break this bottleneck, the gel-free and automated SNP marker system with high throughput, such as sequencing, was extensively developed. SNP markers have been developed and employed in rapeseed, maize, rice and bean [25][26][27]. Nevertheless, it is too expensive for most breeders to afford [18,28]. In contrast, PARMS (Gentides, China) detects SNPs based on fluorescence detection. It is an easy, robust and low-cost genotyping system which is adaptable to rapeseed, rice, maize and wheat in mapping and molecular breeding [29][30][31]. In our study, we successfully developed the PARMS molecular markers of BnMS5, which are highly efficient in genotyping for BnMS5 ( Figure 2). Hence, they will greatly accelerate DGMS three-line breeding of BnMS5 in B. napus.

Conclusions
In this study, we identified PARMS markers, which were designed according to the sequence of BnMS5 a , BnMS5 c and BnMS5 e . The PARMS markers were tightly linked to BnMS5 a , BnMS5 c and BnMS5 e , respectively, which can predict the presence or absence of BnMS5 a /BnMS5 c /BnMS5 e effectively. Consequently, marker-assisted selection can be carried out in the early generation to shorten the breeding period and improve the breeding efficiency, so as to realize the large-scale commercial production of DGMS three-line hybrid seed in rapeseed.

Supplementary Materials:
The following are available online at https://www.mdpi.com/article/10 .3390/plants11030421/s1, Additional File S1: cDNA of BnMS5e, Additional File S2: Classification data of PARMS primer sets for genotyping according to the fluorescent signals.