Integrated QTL mapping and candidate gene analysis for yield-related traits and salt tolerance in a rice RIL population

Rice is a globally critical staple crop, and enhancing its yield and stress resilience is essential for food security. In this study, we employed a recombinant inbred line (RIL) population derived from cultivars Liangxiang5 and 03GY28 to dissect the genetic basis of yield-related traits, leaf color, and germination stage salt tolerance. A high-density genetic map was constructed using 1, 101 bin markers, spanning 1, 132.95 cM with an average marker interval of 1.03 cM. Over two-year field trials, we identified 16 quantitative trait loci (QTLs) for nine agronomic traits distributed across chromosomes 3, 5, 6, 7, 9, and 11. These QTLs accounted for 5.48%-19.03% of phenotypic variance (PVE), with LOD scores ranging from 2.52 to 8.93. Notably, a major-effect QTL, qLeafColor9.1, explaining 19.03% of variance and was mapped to chromosome 9 and co-localized with the known senescence-associated gene OsSGR, which was significantly upregulated under salt stress. Additionally, QTL hotspots on chromosomes 9 and 11 governed multiple yield-related traits-including panicle branching number, filled grain number, and total grain number-including pleiotropy or tight gene linkages. Additionally, two salt tolerance-related QTLs (qRSDR6.1 and qRSDR7.1) were identified, and candidate genes responsive to abiotic stress were annotated within these intervals. These findings advance our understanding of the genetic architecture underlying rice agronomic traits and provide actionable targets for marker-assisted breeding to improve yield, stress tolerance, and grain quality.

1 Introduction

Rice (Oryza sativa L.) is one of the most important staple crops globally, serving as a primary food source for over half of the world’s population (Guo et al., 2025; Ma X. et al., 2025). With the increasing challenges posed by climate change, population expansion, and diminishing arable land, improving rice yield and stress tolerance has become a central goal in modern breeding programs. Grain yield (Ma Z. et al., 2025; Yang et al., 2025), leaf color (Yuan et al., 2025), and germination stage salt stress tolerance (Li Y. et al., 2022) are key agronomic traits. They are especially significant because they directly influence productivity, resource-use efficiency, and overall crop resilience.

Grain yield is a complex quantitative trait controlled by multiple genes and significantly influenced by environmental conditions. Over the past decades, numerous quantitative trait loci (QTLs) associated with yield components such as panicle number, grain number per panicle, and grain weight have been identified in rice (Yun et al., 2024; Zhao H. et al., 2024; Zhao X. et al., 2024). However, the genetic basis of yield variation remains only partially elucidated, primarily due to the polygenic control and significant genotype-by-environment interactions. Thus, dissecting the genetic architecture of grain yield through high-resolution mapping populations is essential for marker-assisted selection and facilitating gene cloning efforts.

Leaf color, often associated with chlorophyll content and photosynthetic efficiency, is not only an indicator of plant health but also a key determinant of biomass accumulation and yield potential (Qiu et al., 2024). Variations in leaf coloration often reflect underlying differences in nutrient uptake, light utilization efficiency, and adaptive responses to environmental stresses. Although several genes involved in chlorophyll biosynthesis and degradation have been cloned, the genetic control of natural variation in leaf color, especially under complex field conditions, is still not well understood (Shin et al., 2020; Xie et al., 2025). A deeper understanding of the genetic mechanisms regulating leaf color could contribute to the development of rice varieties with improved photosynthetic performance and stress adaptability.

Salt stress is one of the most severe abiotic stresses limiting global rice production and the expansion of arable land (Wu et al., 2025). It is particularly detrimental during the seed germination stage, which is highly vulnerable to salinity. The ability of seeds to successfully germinate and establish robust seedlings in saline soils directly determines the seedling establishment rate, population uniformity, and ultimately the final yield in direct-seeding rice systems. Therefore, enhancing salt tolerance at the germination stage is of paramount importance for ensuring food security.

Conventional breeding efforts have been hindered by the polygenic nature of salt tolerance and the scarcity of efficient, scalable phenotyping protocols. However, recent progress in genomics and high-throughput phenotyping has accelerated the discovery of genetic loci associated with this trait. Using both linkage analysis and association mapping, several QTLs conferring stable germination-stage salt tolerance across diverse environments have been identified (Lin et al., 2004; Li et al., 2021; Goto et al., 2022; Geng et al., 2024; Wei et al., 2024; Xu et al., 2024). Furthermore, a series of key genes have been successfully cloned, including the Na⁺ transporter gene OsHKT1;5 (Ren et al., 2005; Wang et al., 2020), the cytokinin signaling gene OsRR22 (Liu et al., 2023), and the transcription factors OsGRF7 (Chen et al., 2024) and OsWRKY53 (Yu et al., 2023). These genes enhance salt tolerance by regulating ion homeostasis, root system development, and stress-responsive transcriptional networks, respectively.

These findings have greatly advanced our understanding of the molecular basis of salt tolerance and have offered functional gene targets for molecular breeding. Nevertheless, the number of salt tolerance genes that have been successfully deployed in breeding programs and demonstrated significant phenotypic effects remains limited. The homozygosity and genetic stability of RIL populations, combined with high-density resequencing, provide a powerful foundation for high-resolution QTL mapping and candidate gene identification (Li F. et al., 2024; Wang F. et al., 2024; Hou et al., 2025; Li Z. et al., 2025; Shi et al., 2025; Wang et al., 2025). This study was designed to address the genetic sources of the pronounced phenotypic divergence between Liangxiang5 and 03GY28 in key agronomic traits. We specifically sought to: construct a high-density bin map from a resequenced RIL population; systematically map QTLs for yield-related traits, leaf color, and salt tolerance; and evaluate the potential of identified favorable alleles for marker-assisted selection.

2 Materials and methods

2.1 Plant materials

A RIL population consisting of 207 lines was developed from a cross between the rice cultivars Liangxiang5 (female parent) and 03GY28 (male parent). The population was advanced through eight generations of continuous selfing using the single-seed descent (SSD) method. Significant phenotypic differences between the two parental lines were observed for multiple agronomic traits, including: primary branch number (PBN), filled grain number per panicle (FGN), total grain number per panicle (TGN), 1000-grain weight (TGW), length/width ratio (LWR), mean grain length (MGL), mean grain width (MGW), leaf color, relative salt damage rate (RSDR).

2.2 Phenotypic data collection and statistical analysis

A field trial was conducted at the experimental station of the Wensu Rice Experimental Station, Crop Research Institute of Xinjiang Uygur Autonomous Region Academy of Agricultural Sciences during the 2020–2021 growing seasons (41°16’ N, 80°12’ E). The region is characterized by an average annual sunshine duration of 2800–3000 hours, an average annual precipitation of 75 mm, the mean annual evapotranspiration is 1, 200-1, 500 mm. and a frost-free period ranging from 200 to 220 days. It experiences cold winters and hot summers, with a mean annual temperature of 10.0 °C. The soil at the experimental site is clayey, with the following physicochemical properties (0–30 cm depth): pH 8.0, electrical conductivity 2.48 mS·cm⁻¹, organic matter 35.90 g·kg⁻¹, total nitrogen 2.29 g·kg⁻¹, available nitrogen 168.50 mg·kg⁻¹, available phosphorus 48.00 mg·kg⁻¹, and available potassium 156.00 mg·kg⁻¹. A total of 207 RILs along with their parental lines were planted using a completely randomized block design (CRBD) with three replicates. Seedlings were raised on April 2 and manually transplanted on May 9. Each variety planted in two rows with row measuring 2.00 m in length. Plant spacing was 15 cm × 30 cm. Border rows were established around the experimental plots to minimize edge effects. Field management practices were consistent with those applied in general production fields.

Phenotypic traits of the RIL population and parental lines were recorded at the maturity stage to ensure data consistency and comparability. All measurements were performed using standardized protocols. With reference to the Descriptors and Data Standard for Rice Germplasm (Han and Wei, 2006), seven quantitative traits were investigated, including PBN, FGN, TGN, TGW, LWR, MGL, and MGW. The stay-green trait was evaluated at the mature stage based on visual observation of the whole plant. Based on the criterion of leaf color retention, plants were designated as “green” if they maintained a deep green color, or “yellow” if they displayed yellow or light green coloration.

2.3 Salt stress treatment at germination stage

Fifty plump and uniform seeds of each accession were selected, surface-sterilized with 75% ethanol for 15 minutes, and rinsed three times with sterile water. The seeds were then placed in Petri dishes lined with filter paper. Two treatments were established: a control group (distilled water) and a salt stress treatment group (NaCl solution), with three replicates per treatment. For the treatment group, 15 mL of 200 mM sodium chloride (NaCl) solution was added to each dish. All dishes were covered and placed in a constant-temperature incubator at 30 °C for germination. Seed germination was observed daily, and the solution and salt concentration were maintained throughout the experiment. On the 7th day, the germination rate of each material was recorded. Salt tolerance at the germination stage was evaluated based on the relative salt damage rate:

Relative Salt Injury Rate (%) = (Germination Rate of Control – Germination Rate of Treatment)/Germination Rate of Control × 100

The collected data were initially recorded and processed using Microsoft Excel 2019. Descriptive statistics, correlation analysis, and frequency distribution histograms were generated using the R software (version 4.0).

2.4 RNA sequencing

When the seedlings of Nipponbare reached the three-leaf stage, Nipponbare seedlings having similar growth vigor were selected and treated for 0 h, 12 h and 24 h with 200 mM NaCl. The leaves were clipped and collected for transcriptome at the corresponding treatment time. Three biological replicates were performed for each sample, the raw transcriptome sequencing data is sourced from NCBI (PRJNA1037192).

2.5 DNA extraction and sequencing

At the tillering stage, young leaves were collected from the parents and progeny of the RIL population. The leaf samples were immediately frozen in liquid nitrogen and stored at −80 °C until DNA extraction. Total genomic DNA was extracted from each sample using a Plant DNA Extraction Kit, following the manufacturer’s instructions. The concentration and purity of the extracted DNA were measured using a NanoDrop 2000 spectrophotometer. Only DNA samples with an OD260/280 ratio between 1.8 and 2.2, a concentration of at least 40 ng/μL, and a total amount of no less than 2 μg were used for subsequent library construction.

DNA libraries were prepared using the TruSeq Library Construction Kit according to the standard protocol. The qualified libraries were then sequenced on the Illumina HiSeq platform to generate 150 bp paired-end reads.

The whole-genome re-sequencing of the parents and the RIL population and RNA-seq were performed by Novogene Bioinformatics Technology Co., Ltd. (Tianjin, China).

2.6 Detection and annotation of genome variation

Prior to alignment, the raw sequencing reads were filtered to remove low-quality reads containing ≥10% unidentified bases (N’s) or >50% of bases with a Phred quality score < 5 (Bolger et al., 2014). The cleaned reads were then aligned to the Nipponbare reference genome from the Rice Genome Annotation Project (MSU7, https://rice.uga.edu/download_osa1r7.shtml) using BWA (version 0.7.15-r1140) (Li and Durbin, 2009). The resulting alignment files were converted to BAM format using SAMtools (version 1.3.1) (Li et al., 2009). Only uniquely mapped reads were retained for subsequent SNP detection.

SNPs and InDels were identified across all samples using the Genome Analysis Toolkit (GATK, version 4.0) (McKenna et al., 2010). Candidate SNPs that were homozygous in both parents and the RIL population were selected for further analysis. To ensure high-quality variant calls, the following filtering criteria were applied: genotype missing rate ≤ 50%, heterozygosity rate ≤ 15%, and minor allele frequency (MAF) ≥ 20%. Finally, functional annotation of the variant sites was performed using SnpEFF (Cingolani et al., 2012).

2.7 Genetic linkage map construction based on bin map

A chi-square test was performed to assess the segregation ratio of markers. Markers showing significant segregation distortion (P < 0.001) were excluded from subsequent analyses. Genotyping was conducted using a sliding window approach with a window size of 15 consecutive non-missing SNPs. Within each window, if the number of SNPs originating from either parent was ≥11, the individual was assigned a homozygous genotype; otherwise, it was classified as heterozygous. Consecutive 100 kb genomic intervals with identical genotypes across the entire RIL population were grouped into a single recombination bin. These bins were utilized as genetic markers to construct a genetic linkage map using JoinMap 4.0 software (Ooijen, 2006), with a recombination frequency threshold of <0.4 and a minimum LOD (logarithm of odds) score of 6.

The integrated phenotypic and genotypic datasets were subjected to QTL mapping analysis using the inclusive Composite Interval Mapping (ICIM) method implemented in QTL IciMapping v4.2 (Zhang et al., 2022), with the Grouping module set to LOD > 2.5. QTL analysis was performed using inclusive composite interval mapping (ICIM) with a step size of 1.0 cM and a probability threshold for entering variables (PIN) set at 0.001. Genetic distances were calculated using the Kosambi mapping function, and the genetic linkage map was visualized using R software (version 4.0). A LOD threshold of ≥2.5 was used to declare significant QTLs for each trait. Based on the QTL mapping results, candidate genes within the QTL regions were annotated using rice genome functional information, and expression data were integrated to prioritize potential candidate genes. QTL nomenclature followed the McCouch (Mccouch et al., 1997) naming convention.

3 Results

3.1 Phenotypic variation of the parents and RIL individuals

From 2020 to 2021, a two-year field investigation was conducted to evaluate the phenotypic performance of seven yield-related traits in the parental lines and a RIL population (Table 1). The parental lines exhibited distinct differences in these traits. Overall, Liangxiang5 demonstrated superiority in grains per panicle, grain length, and length-to-width ratio, whereas 03GY28 possessed higher thousand-grain weight and grain width (Supplementary Table S1).

Table 1

Table 1. Phenotypic statistics of seed traits in parental and RIL populations across two environments.

In 2020, leaf color at the maturity stage was recorded. Liangxiang5 exhibited yellow-green leaves, whereas 03GY28 maintained dark green leaves. In 2021, germination rates under salt stress were assessed for both parents and the RIL population. The relative salt damage rate (RSDR) of 03GY28 was 35.66%, while that of Liangxiang5 was 84.03%.

Both yield traits and RSDR in the RIL population showed significant phenotypic variation with continuous distributions (Figures 1A-H), as evidenced by broad value ranges and moderate to high coefficients of variation (CV%) (Table 1). These results suggest that the traits are quantitatively inherited. FGN displayed the widest variation range (69.22–101.00) and the highest CV% (13.40–21.06), indicating significant segregation and considerable potential for genetic improvement. In comparison, LWR and PBN showed relatively low CV% (8.44–11.69), suggesting higher stability across different environments. The mean values of all traits in the RIL population fell between those of the two parents, while the minimum and maximum values exceeded the parental ranges. The absolute values of skewness and kurtosis for all traits were less than 1, indicating that these traits followed an approximately normal distribution. Additionally, significant segregation in leaf color at maturity was also observed within the RIL population (Figure 1I).

Figure 1

Figure 1. Phenotypic variation of RIL populations and parents. (A) primary branch number (PBN). (B) filled grain number (FGN). (C) total grain number (TGN). (D) 1000-grain weight (TGW, g). (E) length/width ratio (LWR). (F) mean grain length (MGL, mm). (G) mean grain width (MGW, mm). (H) relative salt damage rate (RSDR). (I) leaf color.

3.2 Correlation analysis

Correlation analysis revealed significant relationships among key agronomic traits in the rice RIL population (Figure 2, Table 2). The strongest positive correlation was observed between TGN and FGN (r = 0.91), indicating that genotypes with higher total grain production tend to produce more filled grains. TGW was also moderately correlated with FGN (r = 0.59) and TGN (r = 0.35), suggesting a synergistic effect between grain yield components. Notably, most morphological traits, such as LWR, MGL and MGW, exhibited weak or non-significant correlations with yield-related traits, highlighting their potential as independent indicators for grain quality evaluation. These findings provide valuable insights into trait interactions and support future QTL mapping and marker-assisted selection for rice improvement.

Figure 2

Figure 2. Correlation analysis of nine traits in RIL population.

Table 2

Table 2. Statistical significance of phenotypic trait correlations.

3.3 Sequencing data analysis and construction of genetic map

To construct a high-density genetic linkage map in rice, whole-genome resequencing was performed on 207 F₆ individuals and their parental lines. The parental lines, Liangxiang5 and 03GY28, generated 7.76 Gb and 8.46 Gb of clean reads, respectively (Supplementary Table S2). A total of 741.77 Gb of clean reads were obtained from the 207 RILs. After aligning the clean reads to the reference genome, the parental alignment rates exceeded 99.80%, with average coverage depths greater than 16.53X, 1X coverage above 97.02%, and 4X coverage above 93.55%. For the progeny, the average alignment rate was 99.79%, with an average coverage depth of 7.41X and mean 1X coverage over 96.14% (Supplementary Table S3).

Based on the parental genotyping results, polymorphic markers between the parents were developed. After filtering out loci with missing parental information, a total of 697, 294 polymorphic loci were identified (Supplementary Table S4). Among these, 268, 279 homozygous SNPs (“aaxbb” type) were detected between the two parents (Figure 3A). Following standard genotyping filtering in the progeny, 60, 100 SNPs were retained for bin marker generation in the F₆ population. SNPs showing significant segregation distortion (P < 0.001) were excluded from subsequent analyses.

Figure 3

Figure 3. Marker screening and genetic map construction. (A) The number of Marker types. (B) Map of marker distribution in linkage groups. (C) Distribution of bin lengths. (D) Collinearity between genetic map and physical map.

An improved sliding window approach was employed to identify recombination breakpoints in the RILs. Adjacent bins with identical genotypes were merged into a single recombination bin, resulting in the construction of a high-density genetic map comprising 1, 101 recombination bin markers (Figure 3B). The cumulative physical length of all bins was 365 Mb, with individual bin sizes ranging from 580 bp to 5.33 Mb and an average size of 331.72 kb (Figure 3C). A total of 12 linkage groups were generated. The total genetic distance of the bin map was 1132.95 cM, with an average distance of 1.03 cM between adjacent bin markers (Supplementary Table S5). Notably, 97.28% of the intervals between adjacent bin markers were less than 5 cM (Supplementary Table S6). Collinearity analysis was performed by comparing the physical positions of the markers on the genome with their genetic locations on the linkage map (Figure 3D). The high degree of collinearity observed indicates that the marker order is accurate and the map is of high quality.

3.4 QTL mapping for eight traits

In this study, a total of 16 QTLs associated with seven important agronomic traits were identified across five chromosomes (Chr 3, 5, 6, 7, 9, and 11) (Table 3, Figures 4A-D). These QTLs exhibited LOD (Logarithm of Odds) scores ranging from 2.52 to 8.93 and individually explained phenotypic variance (PVE) from 5.48% to 19.03%.

Table 3

Table 3. Summary of QTL mapping for nine traits.

Figure 4

Figure 4. Locations and LOD of QTL for nine traits. (A) Distribution of QTLs for seven yield-related traits in 2020. (B) Distribution of QTLs for seven yield-related traits in 2021. (C) Distribution of QTLs for the average values of seven yield-related traits. (D) Distribution of QTLs for RSDR and Leaf Color.

Several QTLs for distinct panicle-related traits were mapped to overlapping genomic regions, suggesting the presence of either pleiotropic genes or tightly linked gene clusters (Table 3, Figures 4A-C). A major co-localization hotspot was identified on chromosome 9 (~80 cM), where QTLs for PBN (qPBN9.1), filled grain number (qFGN9.1), and total grain number (qTGN9.1, qTGN9.2) were consistently detected. This indicates that this genomic region harbors a key locus or loci that synergistically regulate panicle architecture and grain yield components. Another significant co-localization region was found on chromosome 11 (~61 cM), controlling qPBN11.1, qFGN11.1, and qTGN11.1, further reinforcing the genetic linkage between these traits.

The stability of QTLs across different environments or developmental stages is a critical factor for their utility in breeding programs. In this study, several QTLs demonstrated high stability by being detected in both individual years and in the analysis of mean phenotypic values. Notably, the QTLs for TGN on chromosomes 9 and 11 (qTGN9.1, qTGN9.2, and qTGN11.1) were highly stable, suggesting their genetic effects are robust against environmental fluctuations. Similarly, qPBN11.1, qFGN9.1, qLWR3.1, and qMGL3.1 were consistently identified, marking them as reliable targets for marker-assisted selection (MAS).

Additionally, we localized two salt tolerance-related QTLs (qRSDR6.1 and qRSDR7.1) using the phenotype of RSDR (Table 3, Figure 4D). Among all the QTLs identified, qLeafColor9.1 stood out as a major-effect QTL. It was mapped to chromosome 9 with an exceptionally high LOD score of 8.93 and explained a substantial 19.03% of PVE, which is considerably higher than the other QTLs (Figure 4D). This indicates that qLeafColor9.1 is a primary genetic determinant of leaf color variation in the studied population and represents a prime candidate for further fine-mapping and gene cloning.

3.5 Candidate genes predicting

Key QTLs controlling yield-related traits, namely the co-localized QTLs qPBN9.1/qFGN9.1/qTGN9.1 and qPBN11.1/qFGN11.1/qTGN11.1, were fine-mapped to two distinct genomic hotspots (Table 4). The first cluster, qPBN9.1/qFGN9.1/qTGN9.1, resides in a 356.50 kb region on chromosome 9 (Chr9: 12, 570, 311-12, 926, 814), while the second, qPBN11.1/qFGN11.1/qTGN11.1, spans an 809.57 kb interval on chromosome 11 (Chr11: 8, 414, 531-9, 224, 102). To elucidate the genetic basis underlying these QTLs, we conducted functional annotation and expression profiling of all genes within these intervals. Within the 356.50 kb region on chromosome 9, 49 genes were identified. Notably, five of these genes (LOC_Os09g20900, LOC_Os09g20980, LOC_Os09g21110, LOC_Os09g21210, and LOC_Os09g20990) exhibited specific expression in panicles or grains (Table 5), suggesting their direct involvement in panicle development and grain formation. In the larger 809.57 kb region on chromosome 11, 116 genes were annotated. Expression analysis revealed that only eight genes (LOC_Os11g16280, LOC_Os11g16370, LOC_Os11g16550, LOC_Os11g14950, LOC_Os11g16410, LOC_Os11g16430, LOC_Os11g16540 and LOC_Os11g16580) were expressed in panicles or grains (Table 5, Supplementary Figure S1). These spatiotemporally expressed genes represent the most promising candidates for regulating panicle traits, providing clear targets for future functional validation and molecular breeding applications.

Table 4

Table 4. Physical positions of candidate QTLs.

Table 5

Table 5. Summry of candidate genes within QTL intervals in rice.

Two QTLs associated with salt resistance were identified based on RSDR. qRSDR6.1 was mapped to a 187.16 kb interval on chromosome 6 (Chr6:4, 744, 946-4, 932, 103), which contains 24 predicted genes. Expression profiling revealed that four of these genes, LOC_Os06g09560, LOC_Os06g09630, LOC_Os06g09660, and LOC_Os06g09540, were significantly responsive to drought stress (Table 5 and Figure 5), showing either marked suppression or induction. The other QTL, qRSDR7.1, was located in a 500.80 kb region on chromosome 7 (Chr7:6, 148, 080-6, 648, 883) harboring 74 predicted genes. Notably, despite the larger number of genes within this interval, expression analysis identified only a single candidate gene that was significantly induced under drought stress: a cytochrome P450 family gene (Table 5 and Figure 5).

Figure 5

Figure 5. Expression of candidate genes under salt stress. (A) LOC_Os06g09560. (B) LOC_Os06g09630. (C) LOC_Os06g09660. (D) LOC_Os06g09540. (E) LOC_Os07g11739. (F) LOC_Os09g36200 (OsSGR).

qLeafColor9.1 is a QTL located on chromosome 9 of rice, spanning a physical interval from 20, 847, 909 bp to 21, 516, 552 bp, with a length of approximately 668.60 kb and a remarkably high LOD score of 19.03 (Table 4). These data strongly indicate that this locus is closely associated with leaf color variation and the senescence process. Notably, the physical position of this QTL perfectly coincides with the OsSGR gene (Shin et al., 2020), further corroborating the central role of OsSGR in regulating leaf senescence and chlorophyll degradation in rice. Given that natural variations in the promoter region of the OsSGR gene significantly influence the rate of senescence and grain yield, this gene can serve as a key target for molecular marker-assisted selection and gene-editing breeding strategies. Further, we observed that the expression of the OsSGR gene is significantly upregulated under salt stress conditions (Figure 5). It provides a solid theoretical foundation and valuable genetic resources for the development of novel rice varieties with delayed senescence, enhanced photosynthetic capacity, salt tolerance and increased grain yield.

4 Discussion

In this study, we successfully constructed a high-density genetic linkage map using a RIL population derived from a cross between Liangxiang5 and 03GY28, two rice cultivars with contrasting phenotypes in yield-related traits, salt tolerance, and leaf senescence. Through comprehensive QTL mapping, we identified 16 QTLs associated with nine key agronomic traits. Notably, we discovered major QTL hotspots controlling panicle architecture, novel QTLs for salt tolerance, and a major-effect QTL, qLeafColor9.1, which co-localizes with the OsSGR gene (Shin et al., 2020). We performed high-depth resequencing on the parents and identified one SNP (03GY28:C, Liangxiang5:T) in the promoter region of the OsSGR gene (Chr9:20867148). This finding is consistent with the results reported by Shin et al., which indicate that the T-to-C variation in the OsSGR promoter leads to delayed senescence, resulting in the stay-green phenotype in plants. These findings provide profound insights into the genetic basis of these complex traits and offer valuable genetic resources for future rice breeding programs.

4.1 Genetic dissection of yield-related traits and identification of QTL hotspots

Grain yield is a complex quantitative trait controlled by multiple genes. In this study, we identified several stable QTLs for yield components, including PBN, FGN, and TGN. A significant finding was the identification of two robust QTL hotspots on chromosomes 9 and 11, where QTLs for PBN, FGN, and TGN were consistently co-localized. The co-localization of qPBN9.1, qFGN9.1, and qTGN9.1 on chromosome 9, and qPBN11.1, qFGN11.1, and qTGN11.1 on chromosome 11, strongly suggests the presence of either pleiotropic genes or tightly linked gene clusters that synergistically regulate panicle development and grain set. This phenomenon is consistent with previous studies, which have also reported QTL clusters for yield-related traits on these chromosomes (Wei et al., 2021; Zhao X. et al., 2024; Ma Z. et al., 2025). The high stability of these QTLs across different environments (2020, 2021, and averaged data) underscores their reliability as targets for marker-assisted selection (MAS). The candidate genes identified within these intervals, such as the trehalose-6-phosphate synthase gene (LOC_Os09g20990) in the Chr9 hotspot, are particularly compelling. Trehalose-6-phosphate is a crucial signaling molecule in carbon metabolism and has been implicated in regulating plant development and stress responses (Li Z. et al., 2022; Li M. et al., 2024; Li J. et al., 2025), making it a strong candidate for influencing panicle traits.

4.2 Unraveling the genetic basis of salt tolerance

Salt stress is a major abiotic constraint to rice production (Jiang et al., 2025; Wu et al., 2025). We identified two QTLs, qRSDR6.1 and qRSDR7.1, associated with the relative salt damage rate (RSDR). The correlation analysis revealed a strong positive correlation between RSDR and grain number traits (FGN and TGN), suggesting a potential trade-off: genotypes with higher grain yield potential might be more susceptible to salt stress at the seedling stage. This intriguing finding highlights the complexity of breeding for high-yielding and salt-tolerant varieties and suggests that these traits may need to be pyramided carefully. Within the qRSDR6.1 interval, we identified four candidate genes (LOC_Os06g09560, LOC_Os06g09630, LOC_Os06g09660, and LOC_Os06g09540) that were significantly responsive to salt stress. Given the well-known crosstalk salt stress responses (often involving osmotic adjustment and oxidative stress) (Zhou et al., 2025), these genes represent promising candidates for conferring salt tolerance. Notably, LOC_Os06g09660 encodes an auxin response factor (ARF), a key transcriptional regulator in plant growth and development. ARFs have been shown to play critical roles in modulating root system architecture (Wang C. et al., 2024; Lu et al., 2025), which is a primary determinant of water and nutrient uptake under stress conditions. The functional characterization of these genes will be crucial to validate their role in salt tolerance.

4.3 qLeafColor9.1: a major QTL for delayed senescence with immense breeding potential

The most striking finding of this study is the identification of qLeafColor9.1, a major-effect QTL controlling leaf color at the maturity stage. This QTL explained an exceptionally high proportion (19.03%) of the phenotypic variance, far exceeding the PVE of other QTLs identified. The physical position of qLeafColor9.1 perfectly overlaps with the OsSGR (STAY-GREEN) gene, a well-characterized central regulator of chlorophyll degradation and leaf senescence in rice (Shin et al., 2020). Natural allelic variation in OsSGR is known to be responsible for the “stay-green” phenotype, which is associated with delayed senescence, extended photosynthetic activity, and ultimately, increased grain filling and yield. Our results not only confirm the pivotal role of OsSGR but also provide a precise genetic location (qLeafColor9.1) and a robust marker for this valuable trait. Furthermore, our observation that OsSGR expression is significantly upregulated under salt stress provides a novel link between senescence regulation and abiotic stress response. This suggests that the OsSGR might be a key integrator of developmental and stress-induced senescence pathways. Therefore, qLeafColor9.1 represents a prime target for molecular breeding. Developing functional markers based on the causal polymorphism within OsSGR or using gene-editing technologies (CRISPR/Cas9) to create favorable alleles could enable the efficient development of novel rice varieties with the highly desirable “stay-green” trait, leading to enhanced photosynthetic capacity, improved stress resilience, and higher grain yield.

4.4 Strengths, limitations, and future perspectives

The primary strength of this study lies in the use of a high-density genetic map constructed via whole-genome resequencing, which provided high resolution and accuracy for QTL mapping. The multi-environment phenotyping over two years enhanced the reliability of the detected QTLs. However, some limitations remain. The physical intervals for the QTLs, especially the hotspots, are still relatively large (809.57 kb for the Chr11 hotspot), containing dozens of candidate genes. Fine-mapping through developing near-isogenic lines (NILs) or using a larger population is necessary to narrow down these intervals and identify the causal genes. Furthermore, while our expression analysis under salt stress provided valuable clues, the candidate genes’ functions require rigorous validation through genetic approaches, such as overexpression, RNA interference, or gene knockout.

5 Conclusion

In summary, this study provides a comprehensive genetic analysis of yield, salt tolerance, and senescence-related traits in rice. The identified stable QTLs, QTL hotspots, and candidate genes, especially the major-effect qLeafColor9.1 associated with OsSGR, serve as a solid foundation for future molecular dissection and breeding applications. The pyramiding of favorable alleles from qLeafColor9.1 (for delayed senescence), the yield-related QTL hotspots (for high yield potential), and the salt tolerance QTLs (qRSDR6.1 and qRSDR7.1) holds great promise for developing elite rice varieties capable of maintaining high productivity under suboptimal conditions, thereby contributing to global food security.

Data availability statement

The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found in the article/Supplementary Material.

Author contributions

YZ: Project administration, Writing – original draft, Writing – review & editing. YW: Software, Writing – original draft, Writing – review & editing. XD: Data curation, Writing – original draft. XW: Methodology, Writing – original draft. MK: Data curation, Writing – review & editing. TH: Validation, Writing – review & editing. FT: Investigation, Writing – review & editing. YQ: Investigation, Writing – original draft. ZZ: Methodology, Writing – review & editing. QY: Data curation, Formal Analysis, Writing – review & editing. BA: Investigation, Writing – review & editing. DL: Resources, Writing – review & editing. FW: Conceptualization, Writing – review & editing. JY: Project administration, Writing – review & editing.

Funding

The author(s) declared that financial support was received for this work and/or its publication. This work was supported by the Project of Fund for Stable Support to Agricultural Sci-Tech Renovation (xjnkywdzc-2023001-19, xjnkywdzc-2025001-14), the National Natural Science Foundation of China (32460478), Xinjiang Uygur Autonomous Region’s “Tianshan Talent” Training Program-High-level Talents for Development of System Agriculture and Animal Husbandry Industry (2024SNGGGCC035), the Earmarked Fund for China Agriculture Research System (CARS-01-74).

Conflict of interest

The authors declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Generative AI statement

The author(s) declare that no Generative AI was used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fpls.2025.1711018/full#supplementary-material

Supplementary Table 1 | Phenotypic statistics of seed traits in parental.

Supplementary Table 2 | Summary of whole-genome resequencing of 207 recombinant inbred lines and their parents.

Supplementary Table 3 | Statistics on sequencing depth and coverage.

Supplementary Table 4 | The number of Marker types.

Supplementary Table 5 | The number of Marker types.

Supplementary Table 6 | Statistics on genetic map gap information.

Supplementary Figure 1 | Expression of 8 genes in Panicles and Seeds.

References

Bolger, A., Lohse, M., and Usadel, B. (2014). Trimmomatic: a flexible trimmer for Illumina sequence data. Bioinformatics 30, 2114–2120. doi: 10.1093/bioinformatics/btu170

PubMed Abstract | Crossref Full Text | Google Scholar

Chen, Y., Dan, Z., and Li, S. (2024). GROWTH REGULATING FACTOR 7-mediated arbutin metabolism enhances rice salt tolerance. Plant Cell. 36, 2834–2850. doi: 10.1093/plcell/koae140

PubMed Abstract | Crossref Full Text | Google Scholar

Cingolani, P., Platts, A., Wang le, L., Coon, M., Nguyen, T., Wang, L., et al. (2012). A program for annotating and predicting the effects of single nucleotide polymorphisms, SnpEff: SNPs in the genome of Drosophila melanogaster strain w1118; iso-2; iso-3. Fly (Austin) 6, 80–92. doi: 10.4161/fly.19695

PubMed Abstract | Crossref Full Text | Google Scholar

Geng, L., Zou, T., Zhang, W., Wang, S., Yao, Y., Zheng, Z., et al. (2024). Integration linkage mapping and comparative transcriptome analysis to dissect the genetic basis of rice salt tolerance associated with the germination stage. Int. J. Mol. Sci. 25, 10376. doi: 10.3390/ijms251910376

PubMed Abstract | Crossref Full Text | Google Scholar

Goto, I., Neang, S., Kuroki, R., Reyes, V. P., Doi, K., Skoulding, N. S., et al. (2022). QTL analysis for sodium removal ability in rice leaf sheaths under salinity using an IR-44595/318 F₂ population. Front. Plant Sci. 13, 1002605. doi: 10.3389/fpls.2022.1002605

PubMed Abstract | Crossref Full Text | Google Scholar

Guo, D., Li, Y., Lu, H., Zhao, Y., Kurata, N., Wei, X., et al. (2025). A pangenome reference of wild and cultivated rice. Nature 642, 662–671. doi: 10.1038/s41586-025-08883-6

PubMed Abstract | Crossref Full Text | Google Scholar

Han, L. and Wei, X. (2006). Descriptors and Data Standard for Rice (Oryza sativa L.) (Beijing: China Agriculture Press).

Google Scholar

Hou, L., Chen, D., Pan, X., Jiang, S., Liu, J., Li, Q., et al. (2025). 9311 allele of OsNAR2.2 enhances nitrate transport to improve rice yield and nitrogen use efficiency. Plant Biotechnol. J. 23, 2501–2511. doi: 10.1111/pbi.70073

PubMed Abstract | Crossref Full Text | Google Scholar

Jiang, Y., Wang, M., Wu, L., Dong, G., Chen, F., Zhang, Y., et al. (2025). FKBP12-EBP1-mediated regulation of salinity tolerance in rice. Plant Cell Environ. 48, 6285–6300. doi: 10.1111/pce.15583

PubMed Abstract | Crossref Full Text | Google Scholar

Li, H. and Durbin, R. (2009). Fast and accurate short read alignment with Burrows-Wheeler transform. Bioinformatics 25, 1754–1760. doi: 10.1093/bioinformatics/btp324

PubMed Abstract | Crossref Full Text | Google Scholar

Li, J., Guo, H., Lou, Q., Zeng, Y., Guo, Z., Xu, P., et al. (2025). Natural variation of indels in the CTB3 promoter confers cold tolerance in japonica rice. Nat. Commun. 16, 1613. doi: 10.1038/s41467-025-56992-7

PubMed Abstract | Crossref Full Text | Google Scholar

Li, H., Handsaker, B., Wysoker, A., Fennell, T., Ruan, J., Homer, N., et al. (2009). 1000 genome project data processing subgroup. The sequence alignment/map format and SAMtools. Bioinformatics 25, 2078–2079. doi: 10.1093/bioinformatics/btp352

PubMed Abstract | Crossref Full Text | Google Scholar

Li, Z., Li, X., Liu, S., Mai, S., Qin, Y., Wang, S., et al. (2025). Identification and validation of quantitative trait loci for seven quality-related traits in common wheat (Triticum aestivum L.). Theor. Appl. Genet. 138, 57. doi: 10.1007/s00122-025-04851-w

PubMed Abstract | Crossref Full Text | Google Scholar

Li, M., Li, H., Zhu, Q., Liu, D., Li, Z., Chen, H., et al. (2024). Knockout of the sugar transporter OsSTP15 enhances grain yield by improving tiller number due to increased sugar content in the shoot base of rice (Oryza sativa L.). New Phytol. 241, 1250–1265. doi: 10.1111/nph.19411

PubMed Abstract | Crossref Full Text | Google Scholar

Li, C., Liu, J., Bian, J., Jin, T., Zou, B., Liu, S., et al. (2021). Identification of cold tolerance QTLs at the bud burst stage in 211 rice landraces by GWAS. BMC Plant Biol. 21, 542. doi: 10.1186/s12870-021-03317-7

PubMed Abstract | Crossref Full Text | Google Scholar

Li, Z., Wei, X., Tong, X., Zhao, J., Liu, X., Wang, H., et al. (2022). The OsNAC23-Tre6P-SnRK1a feed-forward loop regulates sugar homeostasis and grain yield in rice. Mol. Plant 15, 706–722. doi: 10.1016/j.molp.2022.01.016

PubMed Abstract | Crossref Full Text | Google Scholar

Li, F., Wu, L., Li, X., Chai, Y., Ruan, N., Wang, Y., et al. (2024). Dissecting the molecular basis of the ultra-large grain formation in rice. New Phytol. 243, 2251–2264. doi: 10.1111/nph.20001

PubMed Abstract | Crossref Full Text | Google Scholar

Li, Y., Zhou, J., Li, Z., Qiao, J., Quan, R., Wang, J., et al. (2022). AND ABA RESPONSE ERF1 improves seed germination and salt tolerance by repressing ABA signaling in rice. Plant Physiol. 189, 1110–1127. doi: 10.1093/plphys/kiac125

PubMed Abstract | Crossref Full Text | Google Scholar

Lin, H., Zhu, M., Yano, M., Gao, J., Liang, Z., Su, W., et al. (2004). QTLs for Na⁺ and K⁺ uptake of the shoots and roots controlling rice salt tolerance. Theor. Appl. Genet. 108, 253–260. doi: 10.1007/s00122-003-1421-y

PubMed Abstract | Crossref Full Text | Google Scholar

Liu, Y., Peng, X., Ma, A., Liu, W., Liu, B., Yun, D. J., et al. (2023). Type-B response regulator OsRR22 forms a transcriptional activation complex with OsSLR1 to modulate OsHKT2;1 expression in rice. Sci. China Life Sci. 66, 2922–2934. doi: 10.1007/s11427-023-2464-2

PubMed Abstract | Crossref Full Text | Google Scholar

Lu, K. K., Yang, H., Liao, C. Y., Song, R. F., Hu, X. Y., Ren, F., et al. (2025). A transcriptional recognition site within SOS1 coding region controls salt tolerance in Arabidopsis. Dev. Cell. 60:2626–2642.e5. doi: 10.1016/j.devcel.2025.05.010

PubMed Abstract | Crossref Full Text | Google Scholar

Ma, Z., Ming, Y., Li, J., Li, H., Wang, H., Zhu, X., et al. (2025). Natural variation in GNP3 determines grain number and grain yield in rice. Nat. Commun. 16, 5892. doi: 10.1038/s41467-025-61326-8

PubMed Abstract | Crossref Full Text | Google Scholar

Ma, X., Wang, H., Yan, S., Zhou, C., Zhou, K., Zhang, Q., et al. (2025). Large-scale genomic and phenomic analyses of modern cultivars empower future rice breeding design. Mol. Plant 18, 651–668. doi: 10.1016/j.molp.2025.03.007

PubMed Abstract | Crossref Full Text | Google Scholar

Mccouch, S., Cho, Y., Yano, M., Paul, E., Blinstrub, M., Morishima, H., et al. (1997). Report on QTL nomenclature. Rice Genteic Newsletter 14, 11–13.

Google Scholar

McKenna, A., Hanna, M., Banks, E., Sivachenko, A., Cibulskis, K., Kernytsky, A., et al. (2010). The Genome Analysis Toolkit: a MapReduce framework for analyzing next-generation DNA sequencing data. Genome Res. 20, 1297–1303. doi: 10.1101/gr.107524.110

PubMed Abstract | Crossref Full Text | Google Scholar

Ooijen, J. W. (2006). “JoinMap^® 4.0, Software for the calculation of genetic linkage maps,” in Plant Research International (Wageningen, the Netherlands).

Google Scholar

Qiu, T., Wei, S., Fang, K., Zhang, M., Li, Y., Feng, Y., et al. (2024). The atypical Dof transcriptional factor OsDes1 contributes to stay-green, grain yield, and disease resistance in rice. Sci. Adv. 10, eadp0345. doi: 10.1126/sciadv.adp0345

PubMed Abstract | Crossref Full Text | Google Scholar

Ren, Z., Gao, J., Li, L., Cai, X., Huang, W., Chao, D., et al. (2005). A rice quantitative trait locus for salt tolerance encodes a sodium transporter. Nat. Genet. 37, 1141–1146. doi: 10.1038/ng1643

PubMed Abstract | Crossref Full Text | Google Scholar

Shi, H., Li, P., Yun, P., Zhu, Y., Zhou, H., Wang, L., et al. (2025). A DOF transcription factor GLW9/OsDOF25 regulates grain shape and tiller angle in rice. Plant Biotechnol. J. 23, 2367–2382. doi: 10.1111/pbi.70064

PubMed Abstract | Crossref Full Text | Google Scholar

Shin, D., Lee, S., Kim, T. H., Lee, J. H., Park, J., Lee, J., et al. (2020). Natural variations at the Stay-Green gene promoter control lifespan and yield in rice cultivars. Nat. Commun. 11, 2819. doi: 10.1038/s41467-020-16573-2

PubMed Abstract | Crossref Full Text | Google Scholar

Wang, C., Chai, S., Li, S., Liu, D., Han, H., Wu, Y., et al. (2025). Genetic dissection of foxtail millet bristles using combined QTL mapping and RNA-seq. Theor. Appl. Genet. 138, 33. doi: 10.1007/s00122-025-04820-3

PubMed Abstract | Crossref Full Text | Google Scholar

Wang, C., Li, X., Zhuang, Y., Sun, W., Cao, H., Xu, R., et al. (2024). A novel miR160a-GmARF16-GmMYC2 module determines soybean salt tolerance and adaptation. New Phytol. 241, 2176–2192. doi: 10.1111/nph.19503

PubMed Abstract | Crossref Full Text | Google Scholar

Wang, J., Nan, N., Li, N., Liu, Y., Wang, T. J., Hwang, I., et al. (2020). A DNA methylation reader-chaperone regulator-transcription factor complex activates OsHKT1;5 expression during salinity stress. Plant Cell. 32, 3535–3558. doi: 10.1105/tpc.20.00301

PubMed Abstract | Crossref Full Text | Google Scholar

Wang, F., Zhao, T., Feng, Y., Ji, Z., Zhao, Q., Meng, Q., et al. (2024). Identification of candidate genes and genomic prediction of soybean fatty acid components in two soybean populations. Theor. Appl. Genet. 137, 211. doi: 10.1007/s00122-024-04716-8

PubMed Abstract | Crossref Full Text | Google Scholar

Wei, X., Qiu, J., Yong, K., Fan, J., Zhang, Q., Hua, H., et al. (2021). A quantitative genomics map of rice provides genetic insights and guides breeding. Nat. Genet. 53, 243–253. doi: 10.1038/s41588-020-00769-9

PubMed Abstract | Crossref Full Text | Google Scholar

Wei, H., Wang, X., Zhang, Z., Yang, L., Zhang, Q., Li, Y., et al. (2024). Uncovering key salt-tolerant regulators through a combined eQTL and GWAS analysis using the super pan-genome in rice. Natl. Sci. Rev. 11, nwae043. doi: 10.1093/nsr/nwae043

PubMed Abstract | Crossref Full Text | Google Scholar

Wu, Y., Hou, J., Xiao, H., Ye, S., Tu, D., Qiu, R., et al. (2025). OsHDAC1 deacetylates the aldehyde dehydrogenase OsALDH2B1, repressing OsGR3 and decreasing salt tolerance in rice. Plant Physiol. 198, kiaf149. doi: 10.1093/plphys/kiaf149

PubMed Abstract | Crossref Full Text | Google Scholar

Xie, W., Xue, X., Wang, Y., Zhang, G., Zhao, J., Zhang, H., et al. (2025). Natural mutation in Stay-Green (OsSGR) confers enhanced resistance to rice sheath blight through elevating cytokinin content. Plant Biotechnol. J. 23, 807–823. doi: 10.1111/pbi.14540

PubMed Abstract | Crossref Full Text | Google Scholar

Xu, S., Zheng, J., Du, H., Du, X., Li, C., Duan, Y., et al. (2024). GWAS combined with linkage analysis reveals major QTLs and candidate genes of salt tolerance in Japonica rice seedlings. Front. Plant Sci. 15, 1462856. doi: 10.3389/fpls.2024.1462856

PubMed Abstract | Crossref Full Text | Google Scholar

Yang, L., He, W., Zhu, Y., Lv, Y., Li, Y., Zhang, Q., et al. (2025). GWAS meta-analysis using a graph-based pan-genome enhanced gene mining efficiency for agronomic traits in rice. Nat. Commun. 16, 3171. doi: 10.1038/s41467-025-58081-1

PubMed Abstract | Crossref Full Text | Google Scholar

Yu, J., Zhu, C., Xuan, W., An, H., Tian, Y., Wang, B., et al. (2023). Genome-wide association studies identify OsWRKY53 as a key regulator of salt tolerance in rice. Nat. Commun. 14, 3550. doi: 10.1038/s41467-023-39167-0

PubMed Abstract | Crossref Full Text | Google Scholar

Yuan, X., Luan, Y., Liu, D., Wang, J., Peng, J., Zhao, J., et al. (2025). The SUMO-conjugating enzyme OsSCE1a from wild rice regulates the functional stay-green trait in rice. Plant Biotechnol. J. 23, 615–631. doi: 10.1111/pbi.14524

PubMed Abstract | Crossref Full Text | Google Scholar

Yun, P., Zhang, C., Ma, T., Xia, J., Zhou, K., Wang, Y., et al. (2024). Identification of qGL4.1 and qGL4.2, two closely linked QTL controlling grain length in rice. Mol. Breed 44, 11. doi: 10.1007/s11032-024-01447-y

PubMed Abstract | Crossref Full Text | Google Scholar

Zhang, L., Wang, X., Wang, K., and Wang, J. (2022). GAHP: An integrated software package on genetic analysis with bi-parental immortalized heterozygous populations. Front. Genet. 13, 1021178. doi: 10.3389/fgene.2022.1021178

PubMed Abstract | Crossref Full Text | Google Scholar

Zhao, H., Shan, J., Ye, W., Dong, N., Kan, Y., Yang, Y., et al. (2024). A QTL GN1.1, encoding FT-L1, regulates grain number and yield by modulating polar auxin transport in rice. J. Integr. Plant Biol. 66, 2158–2174. doi: 10.1111/jipb.13749

PubMed Abstract | Crossref Full Text | Google Scholar

Zhao, X., Xu, Z., Chen, Y., Du, Y., Li, M., Huang, B., et al. (2024). Development of introgression lines and mapping of qGW2, a novel QTL that confers grain width, in rice (Oryza sativa L.). Mol. Breed 44, 10. doi: 10.1007/s11032-024-01453-0

PubMed Abstract | Crossref Full Text | Google Scholar

Zhou, J., Fang, S., Liu, X., Luo, L., Liu, Y., and Zhang, H. (2025). OsERF2 acts as a direct downstream target of OsEIL1 to negatively regulate salt tolerance in rice. Rice (N Y) 18, 29. doi: 10.1186/s12284-025-00787-3

PubMed Abstract | Crossref Full Text | Google Scholar