Genetic Diversity Analysis of Hybrid Rice Parental Lines and Genetic Purity Assessment of Hybrid Seeds of China

Thirty-five pairs of SSR primers were used for genetic diversity analysis and DNA fingerprinting of 31 hybrid rice core parental lines developed in centraland southern-China using one japonica rice line and three inbred rice lines as the check varieties. The average number of alleles (Na) per SSR locus was 4.02, with a range of two to eight, the effective number of alleles (Ne) was 83.16 with a mean of 2.38, ranging from 1.19 to 4.66. The polymorphic information content (PIC) ranged from 0.16 to 0.79, with an average number of 0.52. The results of the cluster analysis indicated that the check varieties viz., one japonica rice and three inbred rice, were clustered into two groups with similarity coefficients of 0.62 and 0.71 respectively indicating their relatedness. Thirty-one hybrid rice parental lines were clustered into 6 groups according to their different types, pedigrees and regions of development with similarity coefficients of approximately 0.76. The highest genetic similarity coefficient (0.94) was observed between Y58S and C815S, and the lowest (0.63) was observed between Quan9311A and Peiai64S. The purity of one hybrid rice cultivar was tested using characteristic marker and the field test, and it was demonstrated that the purities obtained using the two methods were similar. This research will be helpful for rice breeding, new cultivar registration and seed production.


Introduction
Rice is the consumed staple food by more than half of the people in the world (Zhang et al., 2007). Approximately 70 percent in China is dependent on rice, that necessitated more attention to its varietal improvement. The development of new rice varieties in China has been continuously increased in recent years. For example, 565 varieties were developed in 2008, and 943 varieties were developed in 2018, among which indica rice accounted for 77 percent, and hybrid rice accounted for 72 percent (http://www.ricedata.cn/variety/index.htm). Due to rapid increase in quantity of developed hybrids, a better method for variety identification and genetic diversity analysis is required.
Four basic requirements should be met in crop variety identification: stability of the environment, ability to identify existing variation among varieties, minimum variation within varieties and reliability of the results. The DNA fingerprinting, defined as the electrophoretic profiles reveal differences among individual organisms, not only has the characteristics mentioned above but also has the advantages of saving time and labor with reliability in results obtained. Therefore, this method is currently playing a more important role in rice breeding (Joshi et al., 2001). The simple sequence repeat (SSR) markers are widely distributed in the genome and exhibit the advantages of codominance, better repeatability, higher polymorphism, stable amplification, ease of handling (McCouch et al., 1997;Cook et al., 2002). Thus, they are widely used in rice fingerprinting and genetic diversity analysis (Zhang et al., 2015;Meng et al., 2018;Mnasri et al., 2018;Satturu et al., 2018;Bhattacharjee et al., 2019).
International Union for the Protection of New Varieties of Plants (UPOV) evaluated different DNA marker parameters for use in the DUS test to examine plant varieties (Bredemeijer et al., 2002;Röder et al., 2002;UPOV-BMT, 2002), and a similar policy was also released by the Ministry of Agriculture in China.
The DNA fingerprinting can also be effectively used for the identification of seed authenticity, especially for those that are difficult to distinguish morphologically. Few elite parental lines are used frequently in modern breeding practices. As a result, the genetic background and morphological traits become increasingly narrow and small, but it is easier to distinguish these varieties based on genomic differences using molecular markers (Weising et al., 1991;Akkaya et al., 1992;Beyermann et al., 1992) and this method is widely used in rice now (Ramakishana et al., 1995;Nandakumar et al., 2004;Moorthy et al., 2011). The DNA fingerprinting can also be used in seed purity assessment. Purity should be guaranteed across seed production and marketing. When compared with the field test, the molecular marker test is more suitable for seed purity assessment (Akagi et al., 1997).
With the aid of fingerprinting using molecular markers, genetic similarity analysis and cluster analysis can be performed on parental lines used in breeding programs to determine varietal genetic diversity, relatedness, and genetic differences to help predict heterosis, choose appropriate parental lines to make crosses, and improve breeding efficiency.
In this paper, 35 pairs of SSR primers were used to fingerprint 4 check varieties and 31 hybrid rice core parental lines. Based on the benefits stated above, genetic diversity was analyzed using cluster analysis, and in the end, the feasibility was evaluated by assessing seed purity using characteristic markers. This research will provide guidance for rice breeding, variety identification and seed production at the molecular level.

Plant Materials
There were two types of hybrid rice with similar cultivated area in China, three-line system and two-line system. For the former, the maintainer lines were needed to propagate cytoplasmic male sterile (CMS) lines, and then hybrid seeds were produced by crossing CMS lines with restorer lines which contain restorer genes. And for the letter, the fertility transformation of the male sterile (MS) lines depended on photoperiod and temperature, so no maintainer line was needed, and nearly all varieties be able to restore them. Thirty-five rice varieties were used in this study, including 31 hybrid rice core parental lines and 4 check varieties (Table 1). The parental lines of hybrid rice consisted of 4 three-line CMS lines, 7 two-line MS lines and 20 restorer lines. One japonica rice and 3 inbred lines of rice were used as the check varieties. Seed purity assessment was performed using the hybrid rice Guangliangyou476, with Guangzhan63S as the female parent and R476 as the male parent. All the varieties of seeds and hybrid seeds were provided by the Food Crops Institute, Hubei Academy of Agricultural Sciences, and all the seeds were planted in Wuhan, Hubei province.

DNA Extraction, PCR Amplification and PCR Products Detection
For each variety, the same quantity of leaves from five sample plants was mixed, and then the CTAB method was employed to extract DNA (Doyle et al., 1978). The DNA concentration was adjusted to 30 ng/μL for PCR amplification using a NanoDrop. The DNA amplification was carried out in a 20 μL mixture containing 2.0 μL template DNA, 2.0 μL 25 mmol/L buffer, 1.0 μL primers (10 mmol/L for both forward and reverse primers), 200 µM each dNTP, 0.5 units of Taq DNA polymerase, and 13.8 μL ddH 2 O. The PCR consisted of an initial denaturation of the template DNA at 94 °C for 5 min; followed by 35 cycles of denaturation at 94 °C for 1 min; annealing at 55 °C for 1 min; and extension at 72 °C for 1 min; The last cycle consisted of extension at 72 °C for 10 min.
Each reaction was mixed with 4 μL loading buffer, and then, 2 μL of the mixture was analyzed on a 6 percent polyacrylamide gel. A modified silver-staining procedure was employed after electrophoresis as follows: the gel was separated from the plate, placed into a staining solution containing 2 percent (w/v) silver nitrate, and shaken gently for 8 min. The gel was then immersed in distilled water for 10 s and transferred to a developing solution containing 15 percent (w/v) sodium hydroxide and 3 percent formaldehyde until the band was visible. Finally, the gel was imaged using a lamp box with light emitting diode (LED) light source.

SSR Markers
The SSR markers used in this paper were selected from a protocol for the identification of rice varieties using the SSR marker method (NY/T1433-2014), which was released by the Ministry of Agriculture of China. Together, there are four sets of makers with 12 markers per set. Thirty-five markers with good polymorphism and clear electrophoresis bands were used for rice fingerprinting and genetic analysis (Table 2). Note. The paramters in columns 5, 6, 7 labelled as Na, Ne and Pic indicate the number of alleles, effective number of alleles and polymorphism index content.

Variety Fingerprinting Identity Numbers
The SSR markers used in rice fingerprinting were ranked according to their group and chromosome number, as shown in Table 2. For the electrophoretic profile of each marker, each type of band was given a number according to its mobility; for example, the band with the lowest mobility scored one, and so on for the other bands. Each variety received a number from each marker, and all 35 numbers from the 35 SSR markers were used as the variety fingerprinting identity numbers.

Data analysis
The number of alleles (Na) is the sum of all the multiple loci. The formula for the effective number of alleles (Ne) is Ne = 1/Σ(Pi) 2 , the formula for the polymorphism index content (PIC) is PIC = 1 -Σ(Pi) 2 , and Pi is the gene frequency for the SSR marker at the i locus. The pairwise similarity coefficient (F) was calculated by F = 2Nxy/(Nx + Ny), where Nxy is the number of polymorphic bands shared by rice varieties X and Y, and Nx and Ny are the number of polymorphic bands scored for rice varieties X and Y, respectively. The SSR products were numbered as "0" or "1" according to the electrophoretic profile. At the same mobility loci for a variety, if there is a band, the product was assigned "1"; otherwise, the product was assigned "0". Then, the data matrix was obtained. Using the statistical program NTSYS-pc (Rohlf, 2000), cluster analysis was performed with the data matrix using an unweighted pair group method with arithmetic mean (UPGMA) method.

Polymorphisms of the SSR Markers
Thirty-five SSR markers with stable amplification, good polymorphism and even distribution in the genome were used in the genetic diversity analysis. Altogether, there were 141 alleles in 35 varieties with an average of 4.03 for each marker, ranging from two to eight. The effective number of alleles was 83.16 with a mean of 2.38, ranging from 1.19 to 4.66. The PIC values ranged from 0.16 to 0.79, with an average of 0.52 (Table 2).

SSR Fingerprinting Identity Numbers and Variety Characteristic Markers
Thirty-five SSR markers were used for PCR analysis of 35 rice varieties. The electrophoretic bands of the PCR products were numbered from small to large according to their mobility, and each variety was given a number according to the PCR product for each marker. All 35 SSR markers consisted of a variety fingerprinting identity number (Table 3). Fourteen varieties had unique electrophoresis bands among all the varieties with 10 SSR markers, so these markers could be used as characteristic markers for the corresponding variety. By combining two markers, the other 21 varieties also obtained their characteristic markers (Table 3). With the aid of these characteristic markers, the corresponding variety could be distinguished from the others.

Cluster Analysis
Based on the genetic similarity coefficient, cluster analysis was performed, and a dendrogram was generated ( Figure 1). The check variety Nipponbare (subgroup VIII) was clustered into one group with a genetic similarity of 62 percent with the other varieties in remaining clusters. With a similarity coefficient of 0.71, 3 inbred rice varieties were clustered into subgroup VII. With a similarity coefficient of approximately 0.76, all the other varieties were clustered into 6 different subgroups according to the variation existed among these lines, pedigrees and released regions, including three-line CMS line subgroups I and V, two-line sterile line subgroup IV, restorer line subgroups II, III and VI.
The pairwise genetic similarity coefficient indicated that among 31 hybrid rice parental lines, the highest genetic similarity coefficient (0.94) was observed between Y58S and C815S, and the lowest (0.63) was observed between Quan9311A and Peiai64S, with an average of 0.74 for all the plant materials (Table 4).

Discussion
In this paper, DNA fingerprinting and varietal characteristic markers for sterile lines and restorer lines that are widely used in China were reported. The cluster analysis and genetic diversity analysis were conducted based on DNA fingerprinting. Finally, the genetic purity of the seeds of a hybrid rice variety was determined using parental characteristic markers.
The DUS test has been indispensable in the process of rice variety registration to identify varietal characteristics that could distinguish new varieties from released varieties. Traditionally, the DUS tests of varieties depended on the field morphological descriptions observed during entire crop growth duration, which is time-consuming. In addition, the results of the DUS test are not exact because some descriptors rely on the experience, and the interaction between genotype and environment that may result in an unstable phenotype (Satturu et al., 2018). Furthermore, with the rapid development of the breeding program, some elite main parents are frequently used in breeding practice, which results in high genetic similarity among the developed varieties, so it is not suitable to distinguish them based on morphological traits. In addition, variety identification based on morphological traits cannot meet the requirements of crop breeding and crop improvement.
There are great advantages for variety identification at the genome level using DNA fingerprinting. For example, the use of SSR markers for fingerprinting has the advantages of codominance, high polymorphism, good genetic stability and so on, so they could be used in DUS tests and variety identification, especially for varieties with high morphological similarity (Akkaya et al., 1992;Beyermann et al., 1992;Nandakumar et al., 2004;Moorthy et al., 2011). The use of DNA fingerprinting in the DUS test has been extensively considered by the Biochemical and Molecular Techniques (BMT) Working Group of UPOV and the Ministry of Agriculture in China (UPOV-BMT, 2002). In this research, using SSR fingerprinting, 10 characteristic SSR markers were identified that could distinguish 14 varieties from other varieties. In addition, by combining two SSR markers, the other 21 varieties could also be distinguished. Therefore, SSR markers have great advantages in variety identification.
As the number of developed varieties in the market increases, especially varieties with similar phenotypes, some disputes about variety rights arise. For example, the very popular restorer line Huazhan, 148 developed hybrid rice cultivars using it as a male parent in 2008-2018, and some of these varieties also had female parents with high genetic similarity, so these varieties behaved similarly in the field. For some popular varieties, there will be imitation varieties with similar phenotypes for sale. Therefore, with the aid of varietal characteristic markers, these varieties can be easily distinguished to protect the rights of the variety owner.
The genetic purity of hybrid seeds is an essential requirement across their production and marketing because seed contamination would lead to yield loss and poor grain quality, resulting in disputes between farmers and seed sellers (Nandakumar et al., 2004). There are many means of hybrid seed contamination in production plots, such as pollen shedders, fertility restoration of two-line fertile lines, mechanical mixing during harvest and the following post-harvest handling procedures (Bora et al., 2016;Nethra et al., 2016). Currently, in China, approximately half of the hybrid rice varieties are two-line hybrid rice varieties, and fertility conversion is the main factor leading to seed contamination of these varieties. Both low temperature during the booting stage and genetic shift of the sterile line during its reproduction would lead to unstable fertility, and fertility restoration would then require to occur. Therefore, more attention should be paid to the genetic purity of two-line hybrid rice.
The results of the seed genetic purity assessment of a two-line hybrid rice in this study indicated that the purity was almost the same as that obtained by the field test and the varietal characteristic marker method, and all the mixed seeds were sterile lines. Therefore, the characteristic markers in this research could be effectively used in the genetic purity assessment of hybrid seeds.
Crop variety can be divided into different genetic pools according to their relatedness using cluster analysis based on the genetic similarity coefficient matrix, which will provide great help to breeders in choosing the appropriate parental lines to be crossed. Melchinger et al. (1990a) reported that with the aid of genetic distance based on the DNA markers, the yield of crosses between lines could be predicted. Genetic diversity and heterosis analysis of hybrid rice parental lines based on PCR markers indicated that there was a significantly positive correlation between yield, potential heterosis and genetic distance within japonica rice or indica rice (Xiao et al., 1996), which was also proven in other crops (Corbellini et al., 2002).
The dendrogram (Figure 1) of the hybrid rice parental lines in this research also indicated that the two-line sterile line, three-line sterile line and restorer line were all clustered into a single subgroup, and the similarity coefficient among the subgroups was approximately 0.76. However, the average similarity coefficient was higher within subgroups, such as 0.84 in the restorer line (III) and 0.79 in the two-line sterile line (subgroup IV), which are the two largest subgroups in the dendrogram. Thus, a wide genetic distance between parental lines is a pre-requisite for heterosis of hybrid rice.
Although the genetic distance between the inbred rice subgroup and sterile line subgroup was wider (with a lower similarity of approximately 0.71), no hybrid rice variety was developed using either of these lines as the male parent. It was likely that all these inbred rice belong to high-grade good quality varieties with long slender grains though their yield was generally lower compared with that of the restorer lines. Zhang et al. (1996) reported that yield-related quantitative trait loci (QTLs) linked with markers determined the specific combining ability between the parental lines to be crossed, so the lack of yield-related QTLs may be the reason why inbred rice could not be used as male parents for hybrid rice.

Conclusion
In this paper, 35 pairs of SSR primers were used for genetic diversity analysis and DNA fingerprinting of 31 hybrid rice core parental lines developed in central-and southern-China using one japonica rice line and three inbred rice lines as the check varieties.
The average number of alleles (Na) per SSR locus was 4.02, with a range of two to eight, the effective number of alleles (Ne) was 83.16 with a mean of 2.38, ranging from 1.19 to 4.66. The polymorphic information content (PIC) ranged from 0.16 to 0.79, with an average number of 0.52. The results of the cluster analysis indicated that 31 hybrid rice parental lines were clustered into 6 groups with similarity coefficients of approximately 0.76. The highest genetic similarity coefficient (0.94) was observed between Y58S and C815S, and the lowest (0.63) was observed between Quan9311A and Peiai64S. The purity of one hybrid rice variety was tested and the results were reconfirmed with similarity in observations using characteristic markers and the field test methods. This research will be helpful for rice breeding, new cultivar registration and seed production.