Localization of salt-tolerant QTL in rice germination stage under different salinity concentrations

Salt stress is an important abiotic stress, which has seriously affected the reproductive development of rice in many parts of the world. Therefore, it is particularly important to understand the genetic mechanism of salt tolerance in rice. In this study, we preliminarily located some quantitative trait loci (QTL) for root length, bud length, and survival percent under different salinity conditions (0, 100, 200 and 400 mM NaCl), using a population of chromosome segment substitution lines (CSSLs) constructed by Nipponbare and 9311. A total of 18 QTLs were identified, which explained the phenotypic variation of 4.76–37.59%, among which 13 QTLs were detected under salt stress condition. These salt tolerance related QTLs were divided into two categories, QTL expressed both under control and salt stress conditions (qSP3, qBL8), and QTL expressed just under salt stress condition (qBL3, qRL3, qSP4, qRL5, qSP7, qSP9, qRL10, qBL11-1, qRL11-1, qRL11, qSP12). qSP3, qSP4, qRL5, qSP7, qRL10, qRL10-1, qRL11, qRL11-1, qBL11, qBL11-1 and qSP12 were reported to be related with salt stress for the first time. These QTLs identified under salt stress may be valuable genetic factors for improving salt tolerance of rice by molecular markers technology, which will help to further understand the genetic mechanism of salt tolerance of rice.


Introduction
Rice is the staple grain crop in the world, feeding half of the world's population (Lai et al. 2016;Pires et al. 2015;Shi et al. 2017). Salinization of land is a serious problem in agriculture, which is one of the major abiotic stress (Kumari et al. 2018). Due to unsustainable farming methods, poor irrigation methods (He et al. 2019;Li et al. 2019), sea level rise and improper use of fertilizers (Punyawaew et al. 2016;Yang and Guo 2018), most of the land and irrigated fields in the world are affected by salt stress (Pires et al. 2015), which severely affects the growth and development of crops (Shi et al. 2017). Rice is a glycophyte with different sensitivity to salt stress at different development stages (Rao et al. 2018;Wang et al. 2011). Salt stress has a serious effect on seed germination, Abstract Salt stress is an important abiotic stress, which has seriously affected the reproductive development of rice in many parts of the world. Therefore, it is particularly important to understand the genetic mechanism of salt tolerance in rice. In this study, we preliminarily located some quantitative trait loci (QTL) for root length, bud length, and survival percent under different salinity conditions (0, 100, 200 and 400 mM NaCl), using a population of chromosome segment substitution lines (CSSLs) constructed by Nipponbare and 9311. A total of 18 QTLs were identified, which explained the phenotypic variation of 4.76-37.59%, among which 13 QTLs were detected under salt stress condition. These salt tolerance related QTLs were divided into two categories, QTL expressed both under control and salt stress conditions (qSP3, qBL8), and QTL expressed just under salt stress condition (qBL3,qRL3,qSP4,qRL5,qSP7,  seedlings and reproductive development in the rice (Ganie et al. 2019). Therefore, the breeding of salttolerant rice varieties is an important target for breeders and also an effective method to reduce salt stress damage in rice (Lekklar et al. 2019;Shi et al. 2017).
The previous research showed that the salt tolerance of rice was controlled by many genes (Lekklar et al. 2019), and the related traits of salt tolerance were complex (Li et al. 2019). Mining and utilizing salt-tolerant genes/QTL is not only beneficial to the cultivation of salt-tolerant rice (Lin et al. 2004), but also have great significance for understanding the genetic mechanism of salt tolerance in rice (He et al. 2017;Wang et al. 2012b). With the development of molecular marker technology, genetic mapping has become a powerful tool to identify QTL/genes that control important complex agronomic traits (Mardani et al. 2014;Pandit et al. 2010). In rice, many QTLs of salt tolerance have been identified by genetic map, most of which were located on chromosomes 1, 2, 6 and 7, and a few are located on chromosomes 10 and 11 (Ammar et al. 2009;Zheng et al. 2015), but so far only a few salt tolerance genes have been cloned (Jahan et al. 2020). SKC1 is the first salt-tolerant gene successfully isolated by map-based cloning, which is located on chromosome 1 (Ren et al. 2005). SKC1 encodes a sodium transporter of HKT family , which regulates Na + /K + homeostasis under salt stress (He et al. 2019). Another salt-tolerant gene, DST, was obtained from salt-tolerant mutants by map-based cloning and located on chromosome 3 (Li et al. 2019). DST encodes a new zinc finger transcription factor, which negatively regulates the drought and salt tolerance of rice (Huang et al. 2009). HST1 is a newly identified salt-tolerant gene, which encodes a B-type response regulatory protein OsR22. HST1 may be involved as a transcription factor in regulating the expression of osmotic or ion transport related genes (Takagi et al. 2015). Although many QTLs for salt tolerance have been identified in rice, there are few studies on QTLs for salt tolerance under different salinity concentrations (Wang et al. 2012b), and the regulation mechanism of salt tolerance in rice is still unclear.
In this study, a set of chromosome segment substitution lines (CSSLs), included 118 lines, derived from indica 9311 and japonica Nipponbare, were used to map and analyze the QTL for root length (RL), bud length (BL) and survival percent (SP) under different salinity concentrations at germination stage in rice. The aim of this study was to explore the genetic mechanism for salt stress tolerance and provide QTL for salt-tolerance rice varieties breeding by molecular-assisted selection (MAS) (Lai et al. 2016;Mardani et al. 2014).

Plant materials
A set of CSSL population, included 118 lines, derived from 9311 and Nipponbare. 9311, an indica variety, was used as the recipient parent; Nipponbare, an elite japonica variety, was used as the donor.

Stress treatment and evaluation
Fourty filled and healthy seeds of parents and CSSLs were sterilized in 10% sodium hypochlorite solution for 15 min and then rinsed with distilled water for three times (Mardani et al. 2014;Shi et al. 2017). The seeds were soaked in distilled water for 3 days to germinate Wang et al. 2012b). Finally, 30 uniform germinated seeds were selected and placed in a petri dish with single-layer filter paper ). In the experimental group, the seeds were treated with 100 mM, 200 mM and 400 mM sodium chloride solution. In the control group, the seeds were treated with distilled water (0 mM NaCl condition). Each treatment has three replications (Wang et al. 2011). The treated seeds were cultured in an artificial climate chamber, maintaining a 14-h light/10-h dark cycle (27 °C/25 °C) and 80% relative humidity (Shi et al. 2017;Wu et al. 2020). The solutions were replaced everyday ensure that the concentration of sodium chloride solution and the volume of distilled water remain unchanged (Mardani et al. 2014). The RL, BL and SP of each line were measured and collected on seventh day (Basu et al. 2017).

Statistical analyses
Statistical analysis and QTL mapping of CSSLs population treated with different concentrations of sodium chloride were carried out by QTL IciMapping version 4.2 (Meng et al. 2015). The correlation Page 3 of 10 78 Vol.: (0123456789) analysis between RL, BL and SP were conducted by SPSS.statistics.22 software.

Identification of QTL
Taking RL, BL and SP under different concentrations of sodium chloride treatment as indicators, QTL mapping for salt tolerance in rice germination stage was carried out on CSSLs population by QTL IciMapping version 4.2 (Meng et al. 2015), and LOD > 2.5 was selected as threshold to determine whether QTL existed (Zheng et al. 2015). QTL nomenclature refers to the method proposed by McCouch (Lai et al. 2016;Wang et al. 2011).

Phenotypic variation of parents and CSSLs population
The values of RL, BL and SP of parents and CSSLs population under different salinity conditions are shown in Table 1. There is a significant difference in RL between 9311 and Nipponbare under 0 and 100 mM NaCl conditions, and the length of 9311 is significantly higher than that of Nipponbare (Fig. 1); There is a significant difference in BL between 9311 and Nipponbare under 0, 100 and 400 mM NaCl conditions, and the length of 9311 is significantly lower than that of Nipponbare (Fig. 1). There was no significant difference in SP under different salinity conditions. RL, BL and SP of CSSLs population showed continuous frequency distribution and transgressive segregation, which were consistent with the genetic characteristics of quantitative traits (Fig. 2).

Phenotypic correlation
Pearson correlation coefficients of three salt-tolerant traits RL, BL and SP under different salinity conditions are shown in Table 2. RL, BL and SP were significantly correlated at 200 and 400 mM NaCl (p < 0.01), although there was no correlation between RL and SP at 200 mM NaCl. There is no significant correlation between RL, BL and SP under control and 100 mM NaCl, although there is a correlation between RL and BL under 100 mM NaCl (p < 0.05).

QTL analysis
Under different salinity conditions, QTLs for three salt-tolerant traits are shown in Table 3, and the positions of these QTLs on chromosomes are shown in Fig. 3.

QTLs for root length
Seven QTLs were detected for RL (Fig. 3, Table 3). Under controlled condition, qRL1 and qRL10-1 were located on chromosome 1 and 10, with LOD values of 3.46 and 3.88, which explained 10.41% and 11.96% of phenotypic variation, respectively. Under the 100 mM NaCl condition, qRL10 was

QTLs for bud length
Four QTLs were detected for BL (Fig. 3, Table 3). Under controlled condition, qBL8 and qBL11 were located on chromosome 8 and 11, with LOD values of 5.13 and 2.92, which explained 15.84% and 8.63% of phenotypic variation, respectively. Under the 100 mM NaCl condition, there was no QTL detected. Under the 200 mM NaCl condition, QTLs qBL8 and qBL11-1 were located on chromosome 8 and 11, with LOD values of 2.56 and 3.58, which explained the phenotypic variation of 7.99% and 11.41%, respectively. Under the 400 mM NaCl condition, qBL3 was located on chromosome 3, with LOD value of 3.43, which explained 12.51% of phenotypic variation.

QTLs for survival percent
Seven QTLs were detected for SP (Fig. 3, Table 3). Under controlled condition, qSP2, qSP3 and qSP7-1 were located on chromosome 2, 3 and 7, with LOD values of 3.69, 4.31 and 15.31, which explained 7.07%, 8.46% and 37.59% of phenotypic variation, respectively. Under the 100 mM NaCl condition, qSP12 was located on chromosome 12, with LOD value of 6.26, which explained 19.61% of phenotypic variation. Under the 200 mM NaCl condition, QTLs qSP7 and qSP9 were located on chromosome 7 and 9, with LOD values of 6.64 and 2.54, which explained the phenotypic variation of 22.55% and 7.95%, respectively. Under the 400 mM NaCl condition, qSP3 and qSP4 was located on chromosome 3 and 4, with LOD value of 5.96 and 3.50, which explained 18.38% and 10.28% of phenotypic variation.

Discussion
For breeders, it is a feasible way to cultivate salttolerant rice by aggregating salt-tolerant QTL/genes (Ganie et al. 2019). The detection of QTL is greatly promoted by using multiple related traits under different salinity concentrations (Wang et al. 2012b). In order to reveal the genetic control of salt tolerance at rice germination stage, 18 QTLs were identified on 12 chromosomes by using the newly constructed genetic map under different salinity conditions (0, 100, 200 and 400 mM NaCl).

Comparison of the detected QTLs
By comparing the chromosome positions of QTLs detected in our study with previously identified genes (QTLs) (http:// qtaro. abr. affrc. go. jp/ cgi-bin/ gbrow se/ Oryza_ sativa/), we found that three QTLs in this study were close to the positions of several genes (QTLs) related to salt tolerance that have been mapped. For example, qBL8 located on chromosome 8 has the same chromosome interval as OsCPK21, which is involved in the positive regulation of abscisic acid and salt stress signal pathway (Asano et al. 2011). The qSP9 on chromosome 9 has the same chromosome interval as OsRNS4 (Zheng et al. 2014). The qRL3 located on chromosome 3 has the same chromosome interval as OsSUT1 and OsJAZ9. OsJAZ9 is involved in regulating potassium homeostasis, affecting Na + /K + homeostasis and improving salt tolerance of rice (Siahpoosh et al. 2012;Wu et al. 2015). These coincidences indicate that QTL mapping results are reliable and accurate. Moreover,14 QTLs (qRL1,qSP2,qBL11,qSP3,qSP4,qRL5,qSP7,qRL10,qRL11, were detected for the first time, and the chromosome segment substitution lines harboring these QTLs also could serve as candidates for future fine mapping and positional cloning projects. qSP3 and qBL8 were located in Chr3-bin146 and Chr8-bin428, respectively (Fig. 3, Table 3). These QTLs are highly repetitive and have been detected under two different salinity conditions. These QTLs with large phenotypic variation can be further studied. Although qSP3 and qBL8 were detected at different concentrations, the other QTLs were rarely colocated under different salinity concentrations. qRL3 and qBL3 controlling different traits were detected at the same chromosomal position, which may be related to the pleiotropy of QTLs, that is, QTLs of a certain segment on the chromosome act on multiple traits at the same time, which is common in rice (Lin et al. 2004). These QTL regions located in the same place are very useful for improving multiple salt stress traits at the same time (Lai et al. 2016).

QTLs related with salt stress
In this study, a total of 18 QTLs were located, of which 5 QTLs were detected just under control conditions and the other 13 QTLs were detected under salt stress. These 13 QTLs identified under the salt stress were connected with salt tolerance, and could be divided into two categories: first, QTLs were expressed in both control and salt stress conditions, but the expression level were different, such as qSP3 and qBL8, suggesting that the salt treatment significantly affected the expression level of the genes underlying these QTLs; second, QTLs were detected just under salt stress condition, such as qBL3, qRL3,qSP4,qRL5,qSP7,qSP9,qRL10,qRL11 and qSP12, which indicated that the genes underlying these QTLs were induced significantly after salt stress, when they were not normally expressed or expressed very lowly under control condition. The salt tolerance of rice was improved via a design-breeding approach according to the different expression ways of genes underlying these salt stress related QTLs.

Candidate genes for salt tolerance QTLs
For the new major QTLs responsible for salt tolerance, we conducted further candidate genes analysis of qSP3, qSP4, qRL10, qBL11-1, qRL11 and qSP12, whereas the interval for each QTL was less than 500 kb. The result showed that there were 198 putative genes underlying the six QTLs. Homologous analysis showed that 10 of the 198 putative candidate genes were closely related to previously characterized salt-tolerant genes (Asano et al. 2011;He et al. 2019;Huang et al. 2008;Siahpoosh et al. 2012;Toda et al. 2013;Wadekar et al. 2013;Wang et al. 2012a;Wu et al. 2015;Zhou et al. 2013) (Fig. 4, Table 4). The predicted candidate genes in this study will provide some reference for cloning salt-tolerant genes, although these candidate genes may need to be verified in the future. Interesting, the candidate gene LOC_Os12g02200 of qSP12 detected in our study codes for calcineurin-like phosphatase β subunit interacting protein kinase. Previous studies have shown that calcineurin β subunit interacting protein kinase family gene OsCIPK31 is involved in salt stress in rice (Piao et al. 2010). This result further indicates the rationality and reliability of candidate gene analysis, thus, the chromosome segment substitution lines harboring qSP12 represented a good candidate for LOC_Os12g02200 cloning under salt stress in the future.

Potential implication in rice salt tolerance breeding
Most of the QTLs previously located are based on recombinant inbred lines or backcross inbred line population Wang et al. 2012b), and few of them use chromosome segment substitution lines. In this study, the population of CSSLs comprised of 118 lines, and each line of CSSLs population is homozygous with good stability. As the background of most receptor parents in multi-generation backcross is gradually covered by recurrent parents, the interference of genetic background is eliminated, and the accuracy of QTL detection is improved (Bian et al. 2010). Therefore, CSSLs carrying QTLs detected under salt stress is an effective resource for improving salt tolerance of rice (Bian et al. 2010). These markers are closely linked with rice salt tolerance QTLs (Lin et al. 2004), which will provide reference for rice salt tolerance breeding, contribute to the polymerization of rice salt tolerance QTLs, and realize high level of salt tolerance of rice.   HAD superfamily phosphatase, putative, expressed DST LOC_Os03g21820 Cell wall relaxation protein OsCPK21 LOC_Os03g21400 Cytochrome p450 SRWD1 LOC_Os03g21390 None OsHKT1 qRL10 LOC_Os10g26190 LOC_Os11g41890 RNA recognition motif containing protein, putative, expressed RSS3 qSP12 LOC_Os12g02200 Calcineurin-like phosphatase β subunit interacting protein kinase SRWD4 LOC_Os12g02170 Retrotransposon protein, putative, SINE subclass, expressed SRWD1