Uncovering a stable QTL qSRI.A06 and candidate gene for rapeseed pod shatter resistance

Abstract

Rapeseed pods are prone to dehiscence, resulting in yield loss at maturity. In the present study, we investigated the shatter resistance index (SRI) of 280 doubled haploid (DH) lines derived from a cross between ZS11 (susceptible line) and R11 (resistant line). Based on the phenotypic data obtained from four environments and a high-density genetic map, a significant QTL (qSRI.A06) for shatter resistance on A06 chromosome were stably detected. This locus explained 4.80% − 15.00% of the phenotypic variation and the peak position covered a 664 Kb region. The effect of qSRI.A06 was verified in BC₃F₂ and BC₃F₃ populations and delimited in a 90.8 Kb region comprising 11 genes. Out of these genes, a differencially expressed gene, BnaA06g27900D, was identified to be involved in cell wall development by comparative transcriptome analysis. Regional association revealed four SNP/Indel variations in the promoter associated with pod shattering resistance. The highest Indels A06-41975887 showed suggestive association with SRI (p = 8.80E-06) with a TG allele variation. The stable locus qSRI.A06 and the candidate gene BnaA06g27900D will be helpful for understanding the resistance mechanism and improving shatter resistance in rapeseed.

Supplementary Information

The online version contains supplementary material available at 10.1007/s11032-025-01590-0.

Introduction

Rapeseed (Brassica napus L.) as one of the major oilseed crops globally, provides nutritious edible oil and valuable animal meal. However, rapeseed pods exhibit a tendency to dehisce, releasing seeds at the maturity stage. Unfortunately, this natural process leads to significant yield loss. Usually, up to 15% of seeds are lost during machine harvesting due to pod shattering. In harsh climatic conditions, this yield loss can escalate to a staggering 50% (Cavalieri et al. 2016). To mitigate this challenge, the adoption of pod shatter-resistant varieties is of importance. These specialized varieties offer a convenient solution to reduce yield loss and enhance overall rapeseed production. The development and promotion of such shatter-resistant varieties could safeguard rapeseed production and contribute to sustainable agriculture.

The pod structure of rapeseed is similar to that of Arabidopsis. The seeds are enclosed by two valves, which are joined together through the replum at the margin of valves. At maturity, the pods dehisce along the valve margin, resulting in seed shattering. In Arabidopsis, SHATTERPROOF1 (SHP1), SHP2, INDESHISCENT (IND), and ALCATRAZ (ALC) are involved in valve margin development (Liljegren et al. 2000, 2004; Rajani and Sundaresan 2001). REPLUMLESS (RPL) and FRUITFULL (FUL) repress the expression of valve margin identity genes in the replum and valves, respectively (Gu et al. 1998; Ferrándiz et al. 2000; Roeder et al. 2003). NAC SECONDARY WALL THICKENING PROMOTING FACTOR1 (NST1) and NST3 are involved in secondary cell wall formation in the dehiscence zone and endocarp b (Mitsuda and Ohme-Takagi 2008). ARABIDOPSIS DEHISCENCE ZONE POLYGALACTURONASE1 (ADPG1), ADPG2, CELLULASE6 (CEL6), and MANNANASE7 (MAN7) are cell wall regulators and function in pod shattering (Ogawa et al. 2009; He et al. 2018). Phyto-hormones such as auxin, gibberellins, and cytokinin play important roles in pod shattering (Arnaud et al. 2010; van Gelderen et al. 2016; Marsch‐Martínez et al. 2012). Recent studies have shown that BnSHP1, BnIND, BnALC, and BnGA4 also regulate pod dehiscence in rapeseed (Zaman et al. 2021; Liu et al. 2020; Zhai et al. 2019; Braatz et al. 2018a, b; Stephenson et al. 2019), indicating that the regulatory network of pod shattering is conserved between Arabidopsis and rapeseed.

QTL mapping has become a powerful tool to detect genetic loci underlying complex traits in many crops (Xu et al. 2017). Taking advantage of next-generation sequencing and chip technology, high-throughput marker systems, especially high-density SNP markers have enabled the construction of high-density genetic maps (Brown et al. 2014; Trick et al. 2009). Based on high-density genetic maps, several QTL for pod shatter resistance have been identified. For example, Hu et al. (2012) identified a major QTL on chromosome A09 in an F₂ population. Raman et al. (2017) constructed a high-density genetic map in Brassica carinata and identified five QTL for shatter resistance in an F₂ population. Another study identified six QTL through genome-wide association analysis, two of which, qSRI.A06 and qSRI.A09, were verified in DH and IF₂ populations (Liu et al. 2016). Recently, qSRI.A09 was divided into qSRI.A09-1 and qSRI.A09-2, with BnSHP1.A9 underlines qSRI.A09-1 (Liu et al. 2020).

Although QTL for shatter resistance have been identified in Brassica crops, these QTL were originally derived from a limited number of lines exhibiting moderate level of shatter resistance. Consequently, numerous studies have indicated that the genetic variation for shatter resistance in rapeseed is limited (Wen et al. 2008; Morgan et al. 1998; Raman et al. 2014). Furthermore, the identified QTL have not effectively addressed the shatter problem. Therefore, it is imperative to detect and fine-map more stable QTL for shatter resistance to enhance rapeseed breeding. In the present study, using the combination of genetic mapping and comparative transcriptome analysis, a stable QTL for shatter resistance, qSRI.A06, was mapped and candidate gene underlying this QTL was predicted, providing useful information for both genetic mechanism elucidation and marker-based assistant selection for the improvement of pod shattering resistance in B. napus.

Materials and methods

Plant materials

Both Zhongshuang11 (ZS11 hereafter, susceptible line) and R11 (resistant line) are semi-winter-type inbred lines, which were developed by the Oil Crops Research Institute, Chinese Academy of Agricultural Sciences (OCRI-CAAS). The ZRDH population with 280 lines were derived from the cross of ZS11 and R11 (Wang et al. 2019).

Field Trials and Trait Measurement

The DH and parental lines were planted at four environments: Wuhan (114.33°E, 30.50°N), Hubei province in 2014–2015, 2015–2016 and 2016–2017, Huanggang (114.87°E, 30.44°N), Hubei in 2016–2017, hereafter named as WH2015, WH2016, WH2017 and HG2017. The meteorological character for the four growth cycles were presented in Supplementary Table 1. All lines and parents were grown in field following a randomized complete block design with three replications during October to May. Each line was grown in two rows of 40 plants, with 10 cm between plants and 33 cm between rows. Phenotypic data of all lines in four individual environments was used for QTL analysis.

At maturity, pods were harvested and subjected to random impact test (RIT) for the assessment of shatter resistance index (SRI) (Liu et al. 2016). The pods were dried at 45C for 8 h, and placed into 20-cm-diameter cylindrical containers with eight 14-mm steel balls (20 pods in each container). The containers were shaken at 220 r min⁻¹ and the number of broken pods were record five times at 2 min interval. The SRI was calculated as following: (1≤i≤5) where Χ_i is the number of cracked pods recorded at the i^th time (1 ≤ i5).

Data analysis

Basic descriptive statistics (i.e. mean value, standard deviation, minimum and maximum values), analysis of variance (ANOVA), and phenotypic t test were performed by IBM SPSS Statistics version19 software (IBM SPSS, USA). The broad-sense heritability was calculated as H² = σ2g/(σ2g + σ2ge/n + σ2e/rn), where σ2g is genetic variance, σ2ge is the interaction variance of genotype with environment, σ2e is environment variance, n is number of environments, and r is the number of replications.

QTL analysis

The SNP-based genetic map with 14,658 SNP markers was reported previously (Wang et al. 2019). QTL analysis was conducted using composite interval mapping (CIM) model (model 6) in WinQTLCart2.5 (Wang 2005). The number of control markers, windows size, and walk speed were 5, 10, 1 cM, respectively. The threshold of LOD for the detection of a QTL was determined by 1000 permutation test. QTL identified in multiple environments at similar location (1-LOD support interval) were assigned as the same QTL. 1-LOD support interval was assigned as confidence interval of each QTL.

Fine-Mapping

According to the whole-genome sequencing results of ZS11 and R11, SNPs between the two parents flanking the target QTL region were screened. The 250-bp sequences flanking a target SNP on either side were extracted from the Darmor.v10 genome and used for KASP marker primer design. Specific primers were labeled FAM (GAAGGTGACCAAGTTCATGCT) or HEX (GAAGGTCGGAGTCAACGGATT) at 5’ end, and target the SNP locus at 3’ end. KASP-TF Master Mix was ordered from LGC company (Teddington, UK). SNP assays followed the methods by Trick et al. (2012) with primer parameters (e.g., annealing temperature 60 °C, amplicon length 100–150 bp). Information of all markers is presented in Supplementary Table 2.

Marker-assisted backcrossing (MABC) was practiced, in which the qSRI.A06 allele from ZS11 was introgressed into R11. Briefly, ZS11 was used as donor to cross with R11 for generating F₁. A single ZS11 × R11 F₁ plant was backcrossed (BC) with the recurrent parent R11. From BC₁ to BC₂, foreground selection was conducted using two flanking markers of S188 and S194 for qSRI.A06. At BC₃F₁, plants that were heterozygous at the qSRI.A06 region but homozygous for R11 alleles at all other marker loci were self-pollinated. The resulting BC₃F₂ plants carrying homozygous qSRI.A06 allele from ZS11, qSRI.A06-ZS11, were near-isogenic with R11 at the target locus. Then, recombinants were identified in a large BC₃F₂ population with 1656 individuals using two flanking markers S188 and S194 and 5 new developing markers. Different haplotypes or homozygous recombinants in BC₃F₂and BC₃F₃ were subjected to SRI measurement and used for fine mapping of qSRI.A06.

RNA sequencing and data analysis

During pod developmental stage, the pods of ZS11 and R11 at 3 weeks after flowering were collected with three biological replicates. This stage coincides with the critical period of secondary wall thickening before pod dehiscence (Mitsuda and Ohme-Takagi 2008). Total RNA from the sampled pods was isolated using an RNAprep Pure Kit (TIANGEN, China). The RNA requirements (purity and integrity) and construction of the sequencing library subjected to Illumina NovaSeq 6000 platform (illumina, USA). Raw reads were initially processed and filtered using the NGSQC Toolkit (Patel and Jain 2012). The obtained clean reads were then mapped to the B. napus genome Darmor.v10 https://yanglab.hzau.edu.cn/BnIR/germplasm_info? id=Darmor.v10 with software Hisat2 v2.1.0 (Kim et al. 2015). To assemble and quantify reads, Stringtie v1.3.6 was used to count the fragment numbers and normalize the Fragments Per Kilobase of exon model per mllion mapped reads (FPKM) as an indicator of gene expression (Pertea et al. 2015). Differential analysis was conducted by DESeq2 (Love et al. 2014) and the thresholds, |log2(FoldChange)| ≥ 1 and padj ≤ 0.05, were used for the designation of differentially expressed genes (DEGs).

Candidate gene analysis

The genes and annotation information within the fine mapping region were extracted from Darmor.V10 genome (Yang et al. 2023). RNA was extracted from 3 weeks pod of ZS11 and R11 using RNA Easy Fast kit (Tiangen Biotech, Beijing, China). The RNA was then subject to reverse transcription using a HiScript III 1 st Strand cDNA Synthesis Kit (Vazyme, Nanjing, China). RT-PCR amplification was performed using 2 × Phanta Max Master Mix (Vazyme). Sequence clone was performed using Planta Super-Fidelity DNA Polymerase (Vazyme). Sequence of 2.5 kb upstream of the candidate gene was extracted and subjected to cis-element prediction by PLANT CARE (http://bioinformatics.psb.ugent.be/webtools/plantcare/html/) (Lescot et al. 2002) and plotted with GSDS2.0 (http://gsds.gao-lab.org/index.php) (Hu et al. 2015). Transcriptional information was referred to BnTIR (Liu et al. 2021) (http://yanglab.hzau.edu.cn/BnTIR).

Regional association and haplotype analysis

One hundred and twenty-five accessions of semi-winter rapeseed were re-sequenced on Illumina HiSeq 4000 platforms (Yu et al. 2024). Average sequencing depth of accessions is 10×, and the genome coverage reaches up to 90%. The 6,730,436 high-quality SNPs/Indels markers were used for GWAS and regional association analysis (Yu et al. 2024). TASSEL 5.0 Software (Bradbury et al. 2007) and Mixed Linear Model (MLM) were used for trait association with Q (population structure) and K (kinship matrix) corrections. The False Discovery Rate (FDR) value was set at FDR < 0.01 based on the Q-value software package (http://www.Bioconductor.org/packages/release/bioc/qvalue.html). The final determination of the P-value for significance level was p ≤ 1 × 10 ^–4.

To investigate haplotypes of the candidate gene for pod SRI, we performed haplotype analysis using SNPs/Indels around the candidate gene, and assigned the allele with low SRI as the dominant allele. We conducted significance test using T-test with p < 0.05 as the significance level and analyzed the dominant allele sites for SRI. R package ggplot2 was used to build violin plots.

Results

Phenotypic variation and heritability of SRI

Descriptive statistics for the SRI of the parental lines and ZR DH lines in four different environments are presented in Table 1; Fig. 1. The SRI in the DH population across four environments ranged from 0.00 to 0.71, indicating a continuous distribution. In contrast, the highest SRI values for ZS11 and R11 were 0.17 and 0.44, respectively. Notably, the SRI in the DH population exhibited transgressive segregation. ANOVA revealed significant effects of genotype, environment, and genotype × environment interaction on SRI (Table 2). The broad-sense heritability of SRI was estimated to be 0.84 (Table 2).

Table 1.

Descriptive statistics on SRI of ZR DH and parental lines in four environments

Environment	Year	Parents		DH population
		ZS11	R11	Range	Mean ± SD
		(mean ± SD)	(mean ± SD)
WH	2015	0.17 ± 0.07	0.44 ± 0.06	0.00–0.71	0.19 ± 0.12
	2016	0.05 ± 0.01	0.09 ± 0.05	0.00–0.68	0.14 ± 0.13
	2017	0.04 ± 0.02	0.10 ± 0.05	0.00–0.69	0.13 ± 0.12
HG	2017	0.03 ± 0.01	0.15 ± 0.04	0.00–0.28	0.05 ± 0.04

Fig. 1.

Distribution of pod shattering resistance index (SRI) in ZR DH population derived from ZS11 × R11. WH15, WH16, WH17, and HG17 represent four environments with different colors. WH15: Wuhan, 2014–2015 growing cycle; WH16, Wuhan, 2015–2016 growingcycle; WH17, Wuhan, 2016–2017 growingcycle; and HG17, Huanggang, 2016–2017 growingcycle

Table 2.

Genetic variance and effect analysis for SRI in the DH population

Source	df	Sum of square	Mean square	F value	P	h2
Genotype (G)	279	24.28	0.09	17.20	< 0.0001	0.84
Environment (E)	3	8.10	2.70	533.35	< 0.0001
G × E	837	13.04	0.02	3.08	< 0.0001
Error	1960	1.76	0.00

QTL mapping

A stable QTL for SRI were identified on chromosomes A06 across four environments with LOD value ranging from 2.77 to 6.89 (Table 3). This QTL explains 4.80% − 15.00% of the phenotypic variation with an additive effect of 0.02–0.06 (Table 3; Fig. 2). All the peak positions of qSRI.A06 were located within 59.2 ~ 59.6 cM, and the confidence interval covers 59.0 ~ 60.1 cM.

Table 3.

Stable QTL qSRI.A06 for pod shattering resistance at four environments

QTL	Chr	Environment	Peak position (cM)	Confidence interval (cM)	LOD	Additive effect	R2 (%)
qSRI.A06	A06	WH15	59.2	59.0-59.3	4.99	0.05	10.65
		WH16	59.2	59.0-59.3	6.89	0.06	15.00
		WH17	59.6	59.1–59.9	6.75	0.04	9.76
		HG17	59.6	59.3–60.1	2.77	0.02	4.80

Fig. 2.

Genetic mapping of qSRI.A06 in ZR DH population from four independent environments. Lod threadshold ≥ 2.5. WH2015, WH2016, WH2017, and HG2017 represent four environments with different colors. WH15: Wuhan, 2014–2015 growing cycle; WH16, Wuhan, 2015–2016 growing cycle; WH17, Wuhan, 2016–2017 growing cycle; and HG17, Huanggang, 2016–2017 growing cycle

Fine mapping of qSRI.A06

Based on QTL mapping, qSRI.A06 was mapped between two bin markers Block13367 and Block13379, spanning 59.0 cM to 59.3 cM, corresponding to a physical distance of 664 kb according to the Darmor.v10 genome (Fig. 3A). Seven KASP markers were developed to cover the whole region (Supplemental Table 2). Using these markers, 198 BC₃F₁ plants were categorized into heterozygotes (101 plants), R11 allele homozygotes (91 plants), and recombinants (6 plants) between markers S188 and S194. The SRI of heterozygotes (0.20) was significantly higher than that of R11 allele homozygotes (0.12), verifying the qSRI.A06 in the BC₃F₁ population (Fig. 3C). To narrow down the target region of qSRI.A06, a BC₃F₂ population with 1656 plants was developed and genotyped using the seven above mentioned KASP markers. There were six types of homozygous recombinants in BC₃F₂ and BC₃F₃ populations. The SRI of 4–06, 3–56, and 4–33 types was significantly lower than that of 1–91, 8–21, and 8–12 types (Fig. 3B). Finally, the SRI.A06 was confined to the interval between S190 and S1917 with the recombination rate = 0.38%, with a physical length of 90.8 kb.

Fig. 3.

Fine mapping of qSRI.A06. A qSRI.A06 in B. napus based on darmor v10 genome. (B) Graphical genotypes of homozygous recombinants of 6 types BC3F2 individuals. White blocks represent genomic segments from ZS11 and black blocks represent genomic segments from R11. Recombination breakpoints were placed in the middle between flanking markers; the SRI of 6 types was obtained from BC3F3 plants derived from 6 BC3F2 lines. (C) In BC3F1, SRI of heterozygotes was significantly higher than that of R11 allele homozygotes. A represent homozygote of R11 allele; H represent heterozygote. ** indicate statistical significance at P < 0.01

Comparative transcriptome analysis

Transcriptome analysis was performed using 3 week pods to identify differentially expressed genes in the target region. Six samples from parental line ZS11 and R11 were collected for RNA sequencing. On average, 29.76 million reads per replicate were generated, with a Q30 of 95.12%. On average, 90.17% of the reads were aligned to the B. napus genome, of which 87.31% were aligned to the exon regions of genes. In total, the expression of 30,038 genes was detected. Using log2FC ≥ 1 and FDR < 0.05 as screening criteria, a total of 9,215 differentially expressed genes were obtained (Fig. 4A). Among them, 3,639 genes were up-regulated and 5,576 genes were down-regulated in ZS11. All differentially expressed genes were hierarchical clustered as shown in Fig. 4C. KEGG enrichment analysis (Fig. 4B) was used to determine the pathways where these differentially expressed genes were located, using Q-value < 0.05 as the threshold for significant enrichment. The pathway enrichment analysis revealed that these differentially expressed genes (DEGs) were annotated to 130 synthetic pathways, derived from 17 secondary pathways, with 76% of these pathways being associated with metabolism. Among these, the plant hormone signal transduction pathway exhibited the highest number of differentially expressed genes, totaling 109. Additionally, 92 DEGs were associated with phenylpropanoid metabolism, a crucial pathway for plant lignin synthesis.

Fig. 4.

RNA-seq analysis between ZS11 and R11. (A) DEGs between ZS11 and R11 pods at 3 weeks after flowering. (B) Gene ontology (GO) analysis of DEGs. (C) Hierarchical clustering diagram of differentially expressed DEGs. (D) Differential expression of 11 genes within the targeted region

Candidate gene for qSRI.A06

Based on the micro-collinearity of the target region between B. napus and Arabidopsis, 11 predicted genes were identified within this region: BnaA06G27810D (AT5G25550), BnaA06G27820D (AT5G25570), BnaA06G27830D (AT5G25580/DDR5), BnaA06G27840D (AT5G25590), BnaA06G27850D (AT5G25754), BnaA06G27860D (AT5G25760/PEX4) BnaA06G27870D (AT5G25790/TCX7), BnaA06G27880D (AT5G25800/SDN5), BnaA06G27890D (AT5G11530/EMF1) BnaA06G27900D (AT5G25810/TINY) and BnaA06G27910D (AT5G25820) (Supplementary Table 3). By screening the DEGs within the target region of qSRI.A06, expression level of the 11 genes showed only one gene, BnaA06g27900D was significantly differentially expressed (log2FC = 1.02; FDR = 0.0016488) (Fig. 4D; Supplementary Table 4). The fragments per kilobase transcript per million reads (FPKM) values of BnaA06g27900D were 5.92 and 2.90 in ZS11 and R11, respectively.

BnaA06g27900D encodes an ethylene response transcription factor TINY. The expression level of BnaA06g27900D is significantly higher than that of the other three homologues in rapeseed silique (Supplementary Fig. 1). The RT-PCR data showing specific high expression of BnaA06g27900D in the pod abscission zone, with significantly higher levels compared to other tissues (Supplementary Fig. 2). The expression level of BnaA06g27900D in ZS11 was higher than that of R11 by RT-PCR analysis (Fig. 5A), which is consistent with the result of comparative transcriptome. No sequence alteration was found in the coding sequence between BnaA06g27900D-ZS11 and BnaA06g27900D-R11. However, mutations were detected in the promoter region of BnaA06g27900D between ZS11 and R11, resulting in variation in the cis-element profile.

Fig. 5.

Expression model and haplotype analysisi of BnaA06g27900D. (A) Semi-quantitative RT-PCR analysis of BnaA06g27900D. (B) Regional association plot around the gene BnaA06g27900D. (C) Haplotype analysis of Indel A06-41975887 at the promoter region of the candidate gene BnaA06g27900D. *** indicate statistical significance at P < 0.001

To verify the genomic variation of BnaA06g27900D is associated with SRI, 125 accessions of a GWAS population were used to investigate different haplotypes. Altogether, a total of 120 unique suggestive-SNPs/Indels were observed inclusively for all haplotype blocks from 41.973 Mb to 41.977 Mb including BnaA06g27900D gene and its promoter region (Fig. 5B). Subsequently, SNPs/Indels were mapped to BnaA06g27900D (Supplementary Table 5). The maximum number of genome wide suggestive significant SNPs/Indels (n = 4) from all the haplotype blocks were found in BnaA06g27900D. A regional association plot with all associated loci in BnaA06g27900D revealed that A06-41975887 showed the strongest association among surrounding SNPs/Indel (Fig. 5B). A06-41975887 showed suggestive association with SRI (p = 8.80E-06) with a TG allele variation in the promoter region, which is 1310 bp from the start code. Based on these findings, we prioritized that A06-41975887 is correlated to pod SRI. The haplotype result suggests that A06-41975887 is significantly associated with SRI p-value of 0.000136 (Fig. 5C).

Discussion

Developing pod shatter-resistant cultivars is of utmost significance in modern agriculture, especially as the reliance on efficient machine harvesting continues to grow. Pod shattering during maturity and harvest not only leads to yield losses but also increases production costs. In rapeseed, a major oil - producing crop, this problem can be particularly devastating. In the past decade, the scientific community has been actively searching for the genetic architecture related to pod shatter resistance. Although over 10 quantitative trait loci (QTLs) have been found, the number of stable QTLs reported is disappointingly low (Liu et al. 2016; Raman et al. 2014). This scarcity can be attributed to multiple factors. The genetic diversity among different rapeseed populations can cause variability in QTL detection, and environmental factors like temperature, humidity, and harvest time can also influence the expression of these QTLs, making it difficult to identify truly stable loci. Cloning QTLs with significant effects is a complex and challenging task, and to date, only one such QTL has been successfully cloned (Liu et al. 2020), emphasizing the need for more efforts in pod shatter resistance research.

In this study, the repeated detection of qSRI.A06 across four distinct environments and its validation in the BC2F1 population is a remarkable achievement. The fact that qSRI.A06 shares the same chromosomal location (58.3–59.6 cM region) as qSRI.A06b detected by Liu et al. (2016) in different populations strongly indicates its reliability and stability as a QTL for pod shatter resistance. The stability of qSRI.A06 across four environments may be attributed to its role in regulating fundamental cell wall development processes, which are less susceptible to short-term environmental fluctuations (e.g., temperature, humidity). Cavalieri et al. (2016) to support that pod shattering is predominantly genetically controlled, with environmental factors indirectly influencing secondary metabolism (e.g., lignin synthesis). This stability is fundamental for further genetic research and breeding efforts. Fine-mapping of qSRI.A06 is crucial for understanding the genetic basis of pod shatter resistance and provides breeders with a powerful tool for marker-assisted selection (MAS). By precisely identifying the genes underlying this QTL, breeders can more efficiently introduce the shatter resistant trait into new cultivars, thus enhancing the efficiency and effectiveness of breeding programs.

Traditional QTL mapping techniques are hampered by low marker density, which makes them time-consuming, labor-intensive, and ineffective in providing comprehensive and accurate information about QTLs related to target traits (Salvi et al. 2005; Chen et al. 2014; Fletcher et al. 2016). The limited marker density often results in large QTL intervals, making it difficult to precisely identify the causal genes. In contrast, the QTL analysis in our study utilized a high-density genetic map (Wang et al. 2019), allowing us to obtain relatively narrow QTL intervals for qSRI.A06. The development of advanced genome sequencing technologies has also made it easier to obtain genic markers. For qSRI.A06, we initially determined its physical interval to be approximately 664 Kb. Through the development of 5 markers between S188 and S194, we were able to narrow down the region. Notably, two markers, S190 and S1917, successfully delimited the region to a 93.5 Kb segment containing 11 candidate genes. However, the lack of recombination events between S190 and S1917 in the BC3F2 population prevented further narrowing of this region. In the future, it`s necessary with expanding the BC₄F₂ population of over 5,000 individuals to narrow down the target region. This situation highlights the importance of using larger populations and more diverse genetic backgrounds in future mapping studies to overcome such limitations.

Among the 11 candidate genes in the target region, BnaA06g27900D has emerged as a prime candidate for qSRI.A06. This gene encodes an TINY, which belongs to the ERF/AP2 transcription factor family. In Arabidopsis, overexpression of TINY leads to thickened cell walls with reduced lignin content in various cell types (Sakamoto et al. 2018). Lignin is a key component of the cell wall that contributes to the mechanical strength and tension of the pod wall. Changes in lignin content can directly affect pod shattering. For example, the nst1nst3 mutant in Arabidopsis (Mitsuda and Ohme-Takagi 2008) and the pdh1 mutant in soybean (Funatsuki et al. 2014) show increased shatter resistance due to reduced lignin in specific pod wall tissues. In the case of BnaA06g27900D, despite being a transcription factor with relatively low overall expression, its expression in the abscission zone of pod is significantly higher than that of its three homologues, as revealed by BnTIR. The discovery of numerous genic variations in the promoter region of BnaA06g27900D between R11 and ZS11, along with the RT-PCR results showing higher expression of BnaA06g27900D-ZS11 compared to BnaA06g27900D-R11, strongly suggest that BnaA06g27900D is the key gene underlying qSRI.A06.

Looking ahead, the future of rapeseed breeding for pod shatter resistance holds great promise. With the identification of qSRI.A06 and its candidate gene BnaA06g27900D, researchers can now focus on validating the gene’s function through advanced genetic transformation experiments. For instance, the use of CRISPR/Cas9 gene-editing technology can precisely manipulate the BnaA06g27900D gene in rapeseed to directly observe its impact on pod shatter resistance. This could lead to the development of new rapeseed varieties with enhanced shatter resistance. Moreover, exploring the interaction between BnaA06g27900D and other genes in the pod-shattering regulatory network is another crucial area for future research. By understanding these complex interactions, we can gain a more comprehensive understanding of the molecular mechanisms underlying pod shattering. This knowledge can then be used to develop more targeted breeding strategies.

Furthermore, the insights gained from this research on rapeseed may have broader applications in other crops that also suffer from pod shattering issues like legumes. The knowledge and techniques developed can be transferred and adapted to improve the resistance of these crops, ultimately contributing to more sustainable and efficient global agriculture. Overall, the findings of this study pave the way for significant advancements in rapeseed breeding and have the potential to revolutionize the way we approach pod shatter resistance in multiple crop

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

J.L. and T.H. planned and supervised the research. W.X.W. and W.C. conducted the experiment and wrote the first version of this manuscript. W.H., M.H., W.W. and H.T.C. collected the experimental data, D.S.M. and Q.H. reviewed the manuscript, and J.L. and T.H. supervised the experiment, modified and reviewed the manuscript. All authors have read and agreed to the version of the manuscript.

Funding

This work was supported by Strategic Cooperation Projects on Agricultural Science and Technology Innovation between Chongqing Municipal People’s Government and Chinese Academy of Agricultural Sciences, the National Natural Science Foundation of China (32001587 and 32101814), China Agriculture Research System of MOF and MARA and the Science and Technology Innovation Project of the Chinese Academy of Agricultural Sciences (CAAS-ASTIP-2021- OCRI).

Data availability

The RNA-seq data for ‘ZS11’ and ’R11’ have been submitted to the Genome Sequence Archive (GSA) in National Genomics Data Center, China National Center for Bioinformation, Chinese Academy of Sciences, under accession number CRA013668 that are publicly accessible at https://bigd.big.ac.cn/gsa.

Declarations

Conflict of interest

The authors declare no conflict of interest.