A multiplex PCR system for detection of wide compatibility allele S 5n and erect panicle allele dep 1 in rice

Wide compatibility and erect panicles are important traits for heterosis and ideal plant type breeding in rice and accordingly are targets for rice yield improvement. In this study, a multiplex PCR system based on functional markers was developed to simultaneously identify genotypes of the wide compatibility allele S5-n and the erect panicle allele dep1. Forty-nine rice varieties in the Huang-Huai-Hai region of China were identified using this system. Thirty-three varieties contained the erect panicle allele dep1 and two varieties contained the wide compatibility allele S5-n. The frequency of dep1 was obviously higher than that of S5-n. The multiplex PCR method was also tested using an F2 segregating population and was found to be simple, efficient, and reliable. It will be useful in marker-assisted selection based on the wide compatibility allele S5-n and erect panicle allele dep1 to breed super rice varieties using indica and japonica heterosis.


INTRODUCTION
Rice is one of the most important food crops worldwide.The continuous improvement of rice yield is an important way to ensure food security (Liang et al. 2014).According to Chen et al. (2001), rice yield improvement in the future will depend on super high-yield rice breeding that combines ideal plant type and heterosis.Indica and japonica, two subspecies of Asian cultivated rice (Oryza sativa L.), show strong F 1 heterosis, and accordingly have the potential to produce super hybrid rice plants (Konishi et al. 2006, Qian et al. 2016).However, owing to the high genetic distance between indica and japonica, the F 1 hybrids generally have low fertility, which is a major obstacle in the utilization of inter-subspecific heterosis.Accordingly, overcoming the problem of hybrid sterility between indica and japonica to improve yield is an important topic in rice breeding research.Ikehashi and Araki (1986) proposed the wide compatibility theory and mapped the S5 locus for indica-japonica hybrid sterility to chromosome 6.There are three alleles at the S5 locus, i.e., the indica allele S5-i, the japonica allele S5-j, and the neutral allele S5-n (referred to as the wide-compatibility allele).Plants with the S5-n/S5-j (or S5-i) genotype are fully fertile, while plants with the S5-i/S5-j genotype have low fertility.Chen et al. (2008) cloned the S5 gene from the wide-compatibility variety 02428 using a map-based cloning approach and characterized its function.The S5 of indica and japonica differ by two nucleotides, resulting in an amino acid difference and thus leading to hybrid F 1 sterility.A 136-bp deletion in S5 including the translation start site in the wide-compatibility variety results in the functional loss of aspartic protease, leading to a lack of a reproductive barrier between indica and japonica.The successful cloning of the wide compatibility allele S5-n facilitates gene exchange and the utilization of indica and japonica heterosis.
An erect panicle is an ideal plant type trait in high-yield rice (Qian et al. 2016, Zhao et al. 2016).Erect panicle varieties with short, dense panicles, straight leaves, and a small leaf angle could improve the rice population structure and thus increase the efficiency of photosynthesis and the accumulation of photosynthetic products (Xu et al. 1995, Cheng et al. 2011).Huang et al. (2009) cloned the DEP1 from the super rice variety Shennong 265 using a map-based cloning method.The DEP1, including 5 exons and 4 introns, encodes amino acids that are functionally similar to the phospholipid diethanolamide binding protein.The erect panicle allele dep1 has a 625-bp deletion in the fifth exon of DEP1 resulting in the premature termination of transcription.The mutated dep1 allele could enhance rice yield by improving panicle density, branch number, grain number per panicle, and nitrogen uptake and metabolism.
DNA marker-assisted breeding combines molecular markers and crop breeding.The technique has many advantages, such as its insensitivity to environmental conditions, high reliability, and time-efficiency (Dudley et al. 1993, Ribaut et al. 1998, Schuster 2011, Xu 2013).A functional marker is designed according to polymorphic sites within gene sequences affecting phenotypic variation (Andersen and Lübberstedt et al. 2003).Compared with the traditional marker-assisted selection, functional markers have apparent advantages for the detection of the target trait allele in both artificial and natural selection populations (Andersen andLübberstedt 2003, Varshney et al. 2005).The multiplex PCR functional marker system has various benefits, including its favorable accuracy, repeatability, and efficiency (Varshney et al. 2005, Ramkumar et al. 2010, Salgotra et al. 2011).In this study, a functional marker-based multiplex PCR assay for the simultaneous detection of S5-n and dep1 was developed and was used to detect 49 rice varieties in the Huang-Huai-Hai region of China and a F 2 segregating population derived from a cross between indica and japonica.The results indicated that the system could be applied to the rapid detection of S5-n and dep1 in rice germplasm resources and to molecular marker-assisted selection in indica-japonica hybrid breeding.

Plant material
Forty-nine japonica varieties in the Huang-Huai-Hai region of China and one F 2 population derived from a cross between japonica Huaidao 6 and indica 9311 were used to evaluate the multiplex PCR system.Information about the 49 varieties is provided in Online Resource 1 and additional information can be found at the China Rice Data Center (http://www.ricedata.cn/index.htm).The wide-compatibility variety 02428 was used as the positive control for S5-n and Nipponbare, an incompatible japonica variety with the S5-i genotype, was used as the negative control for S5-n.Wuyunjing 8, an erect and dense panicle variety, was used as the positive control for dep1 and Nipponbare, a non-erect and non-dense panicle variety, was used as the negative control.From fresh leaves of 3-week-old seedlings, DNA was extracted using the modified CTAB method (Rogers et al. 1989).

Development of functional markers
One deletion of 136 bp near the upstream and downstream regions of the translation initiation site (ATG) differentiates the wide compatibility allele S5-n and the incompatibility alleles S5-j or S5-i and leads to a loss of function of the S5 (Os06g0213100).The primer pair S5-1 for S5 was designed according to the flanking sequence of the 136-bp deletion (Figure 1).For DEP1 (Os09g0441900), a 625-bp deletion in the fifth exon distinguishes the erect panicle variety from the non-erect panicle variety.The primer pair DEP1-1 for DEP1 was designed according to the sequences upstream and downstream of the fifth exon (Figure 2).Primers were synthesized by Sangon Biotech Co., Ltd.(Shanghai, China).

Multiplex PCR amplification and detection
The multiple PCR reaction system for S5-n and dep1 used a reaction volume of 20 μL, containing 10 μL of 2×Es Taq Master Mix (including 0.1 U μL -1 Es Taq DNA polymerase, 2×Es PCR Buffer, 3 mM MgCl 2 , and 0.4 mM dNTP mix, provided JZ Li et al. by Beijing ComWin Biotech Co., Ltd.), 2 μL (40-60 ng μL -1 ) of template DNA, 1 μL (0.4 μmo1 L -1 ) of forward primer for each gene, 1 μL (0.4 μmo1 L -1 ) of reverse primer for each gene, and 4 μL of ddH 2 O.The PCR was performed using a T100 thermal cycler (Bio-Rad, Hercules, CA, USA).The PCR conditions were as follows: 95 °C for 5 min; 35 cycles of 94 °C for 30 s, 56 °C for 30 s, and 72 °C for 1 min, 30 s; 72 °C for 8 min.PCR products were then separated on 1% agarose gel containing DNA green in 1× TAE buffer at 120 V for 30 min.The resulting bands were visualized under ultraviolet light and recorded using Gel Doc TM EZ imager (BIO-RAD).

RESULTS AND DISCUSSION
Based on the comprehensive consideration of factors that influence PCR for the two genes, e.g., PCR product length, annealing temperature, and PCR byproducts, sequence polymorphisms were used to design functional markers for S5-n and dep1.The primer pair S5-1 was designed according to the 136-bp deletion in S5 of the wide-compatibility variety compared to the incompatible variety.The primers amplified a 321-bp fragment in the wide-compatibility variety and a 457-bp fragment in the incompatible variety (Figure 1).The primer pair DEP1-1 was designed according to the 625-bp deletion in DEP1 in the erect panicle variety compared to the non-erect panicle variety.The primers amplified a 1235-  JZ Li et al. bp fragment in the erect panicle variety and a 1860-bp fragment in the non-erect panicle variety (Figure 2).Finally, two pairs of primers, S5-1 and DEP1-1, yielding clear and differential PCR products with similar annealing temperatures were obtained.
By the continuous optimization of the annealing temperature, number of reaction cycles, and DNA template concentration, the optimum PCR reaction conditions were obtained, i.e., an annealing temperature of 56 °C, 35 cycles, and a template concentration of 4-6 ng μL -1 .Using agarose gel electrophoresis, four amplification products of 321, 457, 1235, and 1860 bp for S5-n in the wide-compatibility variety, S5-(i or j) in the incompatible variety, dep1 in the erect panicle variety, and DEP1 in the non-erect panicle variety were clearly observed.The multiple PCR system was used to detect the distribution of S5-n and dep1 genotypes in 49 rice varieties recently grown in the Huang-Huai-Hai region of China (Figure 3).Huaidao 6 and Handao 277 had the same 321-bp band as that observed for the wide compatibility control variety 02428, indicating that 4.08% were wide-compatibility varieties.With respect to dep1, 33 of 49 varieties had the same 1235-bp band as the erect panicle control variety Wuyunjing 8, indicating that 67.3% were erect panicle varieties.Only Huaidao 6 had both wide compatibility allele S5-n and erect panicle allele dep1 (Table 1).The multiplex PCR system was applied to an F 2 population derived from Huaidao 6 and 9311.Twenty individual plants were randomly selected from the F 2 population for detection (Figure 4).Two individuals (2 and 6) were homozygous for S5-n and dep1.Three individuals (10, 12, and 20) were homozygous for S5-i and DEP1.Six individuals (1, 5, 8, 15, 16, and 19) were heterozygous for S5 and DEP1.Other individuals had one homozygous gene and one heterozygous gene.The results indicated that the multiplex PCR system can be used to detect different S5 and dep1 genotypes in F 2 segregating populations.Using molecular marker technology, a large number of functional alleles controlling important agronomic traits in rice have been successfully located or cloned, providing an impetus for super rice breeding.The combination of heterosis and ideal plant type is an important breakthrough in super rice breeding.The wide compatibility allele S5-n and erect panicle allele dep1 are important for heterosis and ideal plant type breeding in rice.Conventional breeding methods to select plants based on S5-n and dep1 require tedious, time-consuming, and inefficient work related to phenotyping and determining offspring fertility.Thus, a simple and efficient detection system for S5-n and dep1 is required for super rice breeding.Multiplex PCR technology can be used to construct an effective diagnostic system (Chen et al. 2015).Recently, multiplex PCR was successful applied to detect several genes in rice, such as xa13, Xa21, and fgr (Salgotra et al. 2011), fgr and Wx (Cheng et al. 2015), andXa4, xa5, Xa7, xa13, andXa21 (Yap et al. 2016).Here, the multiplex PCR system based on the functional markers was a reliable and sensitive method for the simultaneous detection of S5-n and dep1.The system could precisely detect plants with S5-n and dep1 by one-time conventional PCR amplification and agarose gel electrophoresis.All results demonstrated that the method is rapid, has high repeatability, and is efficient for identifying S5-n and dep1 in germplasm and for gene-based selection in rice breeding programs, irrespective of plant growth stage, tissue type, and variety.
Compared with conventional PCR, the establishment of a multiplex PCR system is more challenging; it requires a comprehensive analysis and repeated optimization (Chen et al. 2015).Based on the characteristics of the target gene and PCR product, it is essential to design appropriate primer combinations and establish a suitable PCR reaction system.In our opinion, primer design is the most critical step.The annealing temperature should be the same or similar and the length differences between the PCR product of multiplex primers should be distinct.Other factors, such as PCR cycles, template DNA concentration, Taq DNA polymerase, and dNTP concentrations should also be repeatedly tested and modified.In this study, after the repeated optimization procedure, four specific bands of 321, 457, 1235, and 1860 bp for S5-n in the wide-compatibility variety, S5-(i or j) in the incompatible variety, dep1 in the erect panicle variety, and  In this study, the erect panicle allele dep1 was substantially more frequent than the wide-compatibility allele S5-n in 49 varieties from the Huang-Huai-Hai region of China.Thirty-three varieties had the erect panicle allele dep1, two varieties had the wide-compatibility allele S5-n, and only one variety had both alleles.These results indicated that DEP1 is widely used in the Huang-Huai-Hai region of China and most rice varieties have erect panicles, as observed in the field.In contrast, wide-compatibility varieties were very rare and S5-n is seldom used in rice breeding in the Huang-Huai-Hai region of China, limiting the utilization of japonica and indica heterosis.S5 is a key gene involved in indica-japonica hybrid sterility.Wide-compatibility varieties with S5-n are able to overcome reproductive barriers.Thus, they produce highly fertile hybrids when crossed with indica or japonica varieties, which promotes gene flow and heterosis.The DEP1 gene is pleiotropic; it is involved in erect panicles, number of grains per panicle, and nitrogen uptake and metabolism (Huang et al. 2009), and could control the efficiency of photosynthesis and nitrogen-use efficiency (Sun et al. 2014).The dep1 alleles have effects on vascular bundle-and panicle-related traits of rice in both indica and japonica genetic backgrounds (Xu et al. 2015).According to Li et al. (2016), a mutation of DEP1 could increase panicle density in japonica Zhonghua 11, using the CRISPR/Cas9 system.An elite indica restorer line with dense and erect panicles was selected by marker-assisted selection of dep1 alleles from backcross populations (Cheng et al. 2011).Therefore, using S5-n and DEP1 to breed a super rice variety based on heterosis and ideal plant type is highly significant.Based on the distributions of S5-n and dep1 in the Huang-Huai-Hai region of China, the utilization of S5-n needs to increase to develop more widecompatibility lines.Selection based on S5-n and dep1 to breed elite rice lines with wide compatibility and erect panicle traits may also be useful for super rice breeding by the combined application of heterosis and ideotype.

CONCLUSIONS
A multiplex PCR system for simultaneously detecting genotypes of the wide compatibility allele S5-n and the erect panicle allele dep1 was developed.This method was found to be simple, efficient, and reliable for 49 rice varieties in the Huang-Huai-Hai region of China and one F 2 segregating population.The dep1 allele has already been widely exploited in japonica rice breeding, unlike the rather scarce allele S5-n, which needs to be included more frequently in breeding, to obtain more incompatible varieties in the Huang-Huai-Hai region of China.

Figure 1 .
Figure 1.Location of the primers and deletion sequence in S5-n.The dashed line indicates the deletion in S5-n; arrows indicate the primers; red letters indicate the translation start site; S5 indicates Os06g0213100.

Figure 2 .
Figure 2. Location of the primers and deletion sequence in DEP1.The dashed line indicates the deletion in DEP1; dots indicate common bases between the two varieties, which were omitted; arrows indicate the primers; DEP1 indicates Os09g0441900.

Figure 3 .
Figure 3. Pattern of PCR amplification obtained using the multiplex PCR system for 49 rice varieties.M indicates the DNA marker; ck1 indicates the positive control for S5-n; ck2 indicates the incompatible japonica variety with the S5-i allele as the negative control for S5-n; ck3 indicates the non-erect and non-dense panicle variety with the DEP1 allele as the negative control for dep1; ck4 indicates the erect and dense panicle variety with the dep1 allele as the positive control.Samples 1-49 are the 49 rice varieties described in Supplemental Table S1; S5 indicates Os06g0213100; DEP1 indicates Os09g0441900.

Figure 4 .
Figure 4. Pattern of PCR amplification obtained using the multiplex PCR system for an F 2 population derived from a cross between japonica Huaidao 6 and indica 9311.M indicates the DNA marker; ck1 indicate the positive control for S5-n; ck2 indicate the incompatible japonica variety with the S5-i allele as the negative control for S5-n; ck3 indicate the erect and dense panicle variety with the dep1 allele as the positive control; ck4 indicate the non-erect and non-dense panicle variety with DEP1 allele as the negative control for dep1.Samples 1-20 are the 20 individual plants that were randomly selected from the F 2 population; S5 indicates Os06g0213100; DEP1 indicates Os09g0441900.

Table 2 .
The detection results for the multiplex PCR system for 49 rice varieties."+" indicates the positive detection of the wide-compatibility allele S5-n or the erect panicle allele dep1, "-" indicates otherwise.S5 indicates Os06g0213100; DEP1 indicates Os09g0441900 Online

Table 1 .
Name, approval date, and pedigree of 49 japonica rice varieties DEP1 in the non-erect panicle variety were clearly observed.The results demonstrated the feasibility of our method for multiplex PCR design.