Identification and fine mapping of a locus controlling multi-main-stem trait in Brassica napus

Abstract

Background

The main stem is a crucial component determining individual plant yield in rapeseed (Brassica napus). However, the genetic and developmental basis underlying the multi-main-stem trait remains largely unclear.

Results

In this study, we identified a multi-main-stem mutant, mms1, which exhibited a significantly increased silique number per plant and abnormal shoot apical meristem (SAM) development. Genetic analysis demonstrated that the multi-main-stem trait was controlled by a recessive gene. Using bulked segregant analysis combined with a Brassica napus 50 K SNP array and map-based cloning, the locus was mapped to a 340-kb interval on chromosome A09 of the ZS11 reference genome and was designated BnaA09.MMS1. Candidate gene analysis revealed that BnaA09G0254500ZS, which harbors sequence variations in both the promoter and coding regions and shows significantly increased expression in the mutant, was the most likely candidate gene. In addition, phytohormone analysis revealed reduced auxin accumulation in mutant SAMs, together with transcriptomic changes in genes associated with the CLAVATA3 (CLV3)-WUSCHEL (WUS) feedback loop.

Conclusions

These findings provide an important foundation for elucidating the genetic basis of the multi-main-stem trait and offer a valuable genetic resource for rapeseed improvement.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s12870-026-09152-y.

Introduction

Rapeseed (Brassica napus) is an important oilseed crop worldwide with substantial economic value in edible oil production, feed and vegetable supply, and bioenergy production [1]. In response to rising demand and scarce land resources, improving yield per unit area remains a major objective in rapeseed breeding. Because planting density is constricted during the seedling stage, enhancing seed yield per plant through optimization of plant architecture represents an effective strategy for yield improvement [2]. Seed yield per plant is mainly determined by total silique number per plant, seed number per silique and thousand-seed weight, among which total silique number per plant is most highly correlated with seed yield [3]. As an important component of total silique number per plant, silique number on the main inflorescence also plays a significant role in yield improvement [4].

The aerial architecture of plants originates from SAM development, including leaves, primary stem, lateral branches, and inflorescences [5]. Therefore, the maintenance of SAM homeostasis is essential for the proper establishment and stability of plant architecture. SAM homeostasis is a dynamic balance between stem cell self-renewal and differentiation, which ensures the sustained organ formation [6]. Disruption of this balance can produce diverse plant architectures [7–10]. The maintenance of the SAM homeostasis relies on a precise genetic regulatory network, in which the CLV-WUS negative feedback loop represents the core regulatory module [6]. As a homeodomain transcription factor expressed in the organizing center (OC) of SAM, WUS activates stem cell fate through intracellular protein movement in a non-cell-autonomous manner and positively activates CLV3 expression in the central zone (CZ) of SAM through its transcriptional regulatory activity [11]. In turn, CLV3, a secreted signaling peptide, restricts both the expression region and expression level of WUS through the signal transduction of receptor complex involving CLV1/CLV2/CORYNE, thereby maintaining a stable stem cell population within the SAM [8].

Phytohormones are key regulators of SAM development, among which auxin plays a vital role in the maintain of stem cell activity and organogenesis [12]. Auxin distribution and signaling response are spatially regulated in the SAM. In peripheral zone (PZ) of SAM, auxin accumulates at specific locations and promotes cell differentiation and organ initiation through PIN-FORMED 1 (PIN1)-mediated polar transport and AUXIN RESISTANT1/LIKE AUX1 (AUX1/LAX)-mediated cellular influx [13]. In CZ, auxin is present at high concentration, but only a minimal auxin response is observed using a DR5 auxin-inducible reporter [12]. This phenomenon is mediated by WUS, which regulates histone acetylation of auxin pathway-related genes, thereby conferring resistance to auxin-mediated differentiation and maintaining apical stem cell identity [14]. However, the low levels of auxin signaling in central zone is still required to maintain the stem cell fate. Previous studies have revealed that auxin signaling in stem cells is mediated by the transcription factor AUXIN RESPONSE FACTOR5/MONOPTEROS (ARF5/MP) [15]. ARF5/MP directly represses the transcription of DORNROSCHEN/ENHANCER OF SHOOT REGENERATION 1 (DRN/ESR1), which expresses in stem cells and positively regulates CLV3 expression [16].

In recent years, a novel agronomic trait, multi-main-stem (MMS) or multi-inflorescence (MI), has been identified in rapeseed. Plants exhibiting this trait usually develop two or more main stems or main inflorescences in a single plant and may exhibit greater yield potential. Previous studies on such materials have primarily focused on phenotypic characterization and transcriptomic analysis, in which differentially expressed genes were inferred to be associated with MMS or MI formation [17, 18]. In addition, a few studies have performed preliminary genetic mapping. Zhao et al. identified 43 QTLs associated with MMS using QTL mapping analysis and Gene-Fishing technique [19]. Zhang et al. and Lu et al. identified the same major QTL controlling MI and further narrowed it down to a 27.18-kb region on chromosome A05 by integrating QTL mapping, microarray analysis, and whole-genome sequencing [20, 21]. These studies provide initial insights into the potential genetic basis of multi-main-stem trait, whereas indicate that it is a complex trait controlled by multiple genes and regulatory pathways. However, detailed gene identification and functional validation of this trait remain largely unknown, and the contribution of MMS to rapeseed yield has not yet been explored, highlighting the need for further fine-mapping and candidate gene identification to uncover the underlying genetic mechanism of this important agronomic phenotype.

In this study, we characterized a multi-main-stem mutant mms1, which exhibits multiple main stems and main inflorescences. Cytological observations and phytohormone measurement were conducted to investigate the mechanism underlying the multi-main-stem phenotype, while field evaluation of agronomic traits was performed to assess its breeding potential. Using bulked segregant analysis (BSA) combined with a Brassica napus 50 K SNP array and map‑based cloning, we fine-mapped the gene controlling the multi-main-stem trait to a 340-kb region on chromosome A09, and named it as BnaA09.mms1. Our findings not only provide new insights into SAM development but also offer a valuable germplasm resources for improving plant architecture in B.napus.

Materials and methods

Plant materials and trait measurements

The mms1 mutant, which exhibits multiple main stems per plant, is a natural mutant identified from a local rapeseed germplasm resource and was provided by Hunan Academy of Agricultural Sciences. The conventional inbred line ZS10 exhibited only one main stem and was provided by the National Rapeseed Engineering Technology Research Center. The mms1 mutant was used as the female parent and crossed with ZS10 to generate the F₁ population. The F₁ plants were self-pollinated to generate the F₂ population and were backcrossed with mms1 to obtain the BC₁ population. The F₃ population was obtained through self-pollination of F₂ plants. Agronomic trait evaluation was conducted using parental lines and F₁ plants. Genetic analysis was conducted in the F₂ and BC₁ generations. During genetic analysis and phenotypic classification, plants with only one main stem were classified as “single-main-stem”, whereas plants with two or more main stems were classified as “multiple-main-stem”. All materials were planted in Wuhan, China from October to May, and in Zhangye, China from May to September.

Paraffin sections

Paraffin sections were used to observe SAM morphology. SAMs from mms1 and ZS10 seedlings at 18 days after germination (18 DAG) were collected and fixed in FAA solution (50% ethanol, 5% acetic acid, and 3.7% formaldehyde). Samples were vacuum-infiltrated for 30 min and fixed overnight at 4℃. Fixed tissues were dehydrated through a graded ethanol series, cleared in xylene, and infiltrated with paraffin. After embedding, the tissues were sectioned into 8 μm slices using a rotary microtome (Leica RM2016). Sections were mounted on slides, deparaffinized in xylene, rehydrated through a descending ethanol series, and stained with 0.05% toluidine blue, then dehydrated, cleared in xylene, and mounted with neutral balsam. Sections were observed and imaged using the Zeiss Axio Scope A1 microscope. Image J was used to measure SAMs size.

Observation of scanning electron microscopy (SEM) and stereo microscopy

The SAMs of mms1 and ZS10 during vegetable growth stage were collected and fixed in 2.5% glutaraldehyde solution overnight at 4℃. The samples were dehydrated through a graded ethanol series (30%, 50%, 70%, 85%, 95%, and 100%, 15 min each), vacuum dried and sputter-coated with gold. The prepared specimens were observed and imaged using a scanning electron microscope (SU3800, Hitachi, Japan). The inflorescence meristems of mms1 and ZS10 during reproductive growth stage were observed using a stereo microscopy in (Olympus Corporation, Tokyo, Japan) in flower bud differentiation stage.

Phytohormone measurements

The endogenous phytohormone levels in the SAMs of AA and aa individuals from the F₃ population were measured at 18 DAG, including auxin, cytokinin, gibberellin, salicylic acid and brassinosteroid. Three biological replicates were prepared for each genotype, with each replicate weighing at least 100 mg. Phytohormones were extracted using methanol/acetonitrile/water solution (2:2:1, v/v/v), followed by vortexing, homogenization, sonication, and centrifugation, and the supernatants were collected for analysis. Analyses were performed using an ultra-high-performance liquid chromatography system (UHPLC; LC-30 A, Shimadzu) coupled to a QTRAP mass spectrometer (7500+, AB SCIEX). The analytes were separated on a Waters UPLC HSS T3 column (2.1 mm × 100 mm, 1.8 μm) using water containing 0.1% formic acid and acetonitrile as the mobile phases. Mass spectrometry was conducted in positive and negative ion modes using multiple reaction monitoring (MRM). Sciex OS software was used to extract peaks from MRM raw data and quantify the phytohormone levels based on the ratio of analyte peak area to internal standard peak area. Phytohormone concentrations were quantified using authentic standards and internal standards, and the results were expressed as actual concentrations. Student’s t test was applied to evaluate the significance of differences between AA and aa.

Exogenous auxin treatment

Surface-sterilized seeds of mms1 were germinated on 1/2 MS medium, After germination, seedlings were transferred to 1/2 MS medium supplemented with different concentrations of α-naphthaleneacetic acid (NAA; 0, 0.1, 0.5, 1.0, and 2.0 µM). Seedlings were grown in a growth chamber under long-day conditions (16 h light/8 h dark) at 22 °C. After 7 days of treatment, a portion of the seedlings was collected for SAM observation by paraffin sectioning. The remaining seedlings were transplanted into soil and sprayed daily with the corresponding NAA solutions for an additional 7 days before SAM observation. Sections were observed and imaged using the Zeiss Axio Scope A1 microscope. Image J was used to measure SAM size.

Bulked segregant analysis with Brassica 50 K array

Thirty multi-main-stem plants and thirty single-main-stem plants from F₂ population are selected, and the genomic DNA was extracted using a DNAsecure Plant Kit (DP320; TIANGEN, Beijing, China). Three DNA pools for the multi-main-stem plants and three DNA pools for the single-main-stem plants were constructed. Together with two parental pools, a total of eight DNA pools were genotyped using the Brassica 50 K SNP assay [22]. The concentration of each pool was adjusted to 100 ng/µL. The array data were analyzed as previously described [23]. Specifically, Polymorphism analysis was performed in Microsoft Office Excel using the IF function. The screening criteria for SNPs were as follows: (I) polymorphic between the two parents, (II) monomorphic among the three multi-main-stem pools, with the genotype consistent with the mms1 parent, (III) monomorphic among three single-main-stem pools, and (IV) polymorphic between multi-main-stem pools and single-main-stem pools. The sequences of differential SNPs were aligned to the B. napus ZS11 genome (https://yanglab.hzau.edu.cn/) using BLAST to determine their chromosomal positions. Based on the distribution of the differential SNPs, the chromosomal interval harboring the target gene was determined.

Marker development

Based on the results of Brassica 50 K SNP assay and the resequencing data of two parents, insertion and deletion variants within the target interval were identified and subsequently used to design indel markers. PCR amplification was performed on the eight DNA bulks using the designed markers, and PCR products were examined by electrophoresis on 6% polyacrylamide gels. Markers that showed clear polymorphism were further validated in a small population to assess their reliability. Markers linked to the phenotype were then used for large-population genotyping and recombinant screening.

Fine mapping of the BnaA09.MMS1

Based on the interval initially defined by the Brassica 50 K SNP array, an F₂ population consisting of 1,930 individuals and an F₃ population consisting of 5,736 individuals were used for fine mapping. The two flanking indel markers were first used to genotype all individuals to identify recombinants. Recombinant individuals were then further genotyped using additional internal markers with increased density. The target interval of BnaA09.MMS1 was progressively narrowed according to the number and distribution of the recombinants.

Analysis of candidate genes

The gene information within the candidate interval were extracted from the BnIR website (https://yanglab.hzau.edu.cn/) and aligned against the TAIR website (http://www.arabidopsis.org/) to identify the corresponding Arabidopsis homologs. Based on the annotations of homologs, the functions of B. napus genes were predicted, allowing preliminary identification of potential candidate genes. The specific primers were designed to amplify the promoter region, full-length genomic sequence of the candidate genes using genomic DNA from individual F₃ plants with the AA (single-main-stem phenotype) and aa (multi-main-stem phenotype) genotype. The PCR products were sequenced and analyzed to further identify the candidate gene.

RNA extraction and quantitative real-time PCR (qRT-PCR) analysis

SAMs were collected from AA and aa individuals in the F₃ population at 18 DAG, with three biological replicates per genotype, each biological replicate consisted of a pool of 18–20 SAMs. Individuals were genotyped using two molecular markers tightly linked to BnaA09.MMS1, ID4-15 and ID4-22. Total RNA was extracted using the RNAprep Pure Plant Kit (TINGEN, China). First-strand cDNA was synthesized from 1.5 µg of total RNA using the One-Step gDNA Removal and cDNA Synthesis SuperMix (TransGen Biotech, China). Gene-specific primers for the candidate genes were designed and the qPCR reactions were performed in triplicate in a CFX384 instrument (Bio-Rad Ltd, USA) using ChamQ Universal SYBR qPCR Master Mix (Vazyme, China). Primer specificity was evaluated by preliminary PCR amplification followed by agarose gel electrophoresis, and each primer pair produced a single band of the expected size. The amplified products were further confirmed by Sanger sequencing. Relative gene expression levels were calculated using the 2^−ΔΔCt method, with BnaACTIN7 as the internal control. Statistical analysis was conducted with Prism software.

RNA-seq

SAMs from AA and aa individuals of the F₃ population at 18 DAG were collected for RNA sequencing analysis. Each sample consisted of three independent biological replicates, and each biological replicate consisted of a pool of 18–20 SAMs. Total RNA was extracted using the RNAprep Pure Plant Kit (TINGEN, China). RNA quality was assessed using an Agilent 2100 Bioanalyzer. Transcriptome sequencing was performed on an Illumina platform by Novogene (Beijing, China). After quality control, including adaptor removal and filtering of low-quality reads, clean reads were obtained and aligned to the B. napus ZS11 reference genome (https://yanglab.hzau.edu.cn/) using HISAT2. Gene expression levels were calculated as Fragments Per Kilobase of exon model per Million mapped reads (FPKM). Differential expression analysis was conducted using DESeq2 based on clean read counts. Genes with |log₂(Fold Change)|≥0.585 and padj ≤ 0.05 were identified as differentially expressed genes (DEGs). Gene Ontology (GO) annotation was performed using PANNZER [24]. GO enrichment analysis of DEGs was carried out using TBtools with padj ≤ 0.05 as the significance threshold [25], and ggplot2 was used for visualization [26]. TBtools was also used to perform a kyoto encyclopedia of genes and genomes (KEGG) enrichment pathway analysis. All heatmaps in this study were generated based on FPKM values with Z-score normalization for visualization.

Statistical analysis

Statistical analyses were performed using Microsoft Office Excel, and graphs were generated using GraphPad Prism software (www.graphpad.com). Data were expressed as mean ± standard deviation (SD). Statistically significant differences between groups were evaluated using Student’s t-test.

Results

Phenotypic characterization and genetic analysis

The mms1 mutant exhibited a multi-main-stem phenotype compared with normal plants. At the flowering stage, mms1 exhibited 2–11 main stems emerging from the basal region, whereas normal plants generally developed a single stem (Figs. 1A, B). Statistical analysis of the total silique number per plant at maturity revealed that mms1 produced significantly more siliques compared to ZS10 (Figs. 1C; S1). To elucidate the cytological basis underlying this phenotype, paraffin sections were performed using SAMs at 18 DAG. The results showed that ZS10 produced a single SAM with normal morphology, whereas mms1 had already differentiated multiple SAMs, and some SAMs exhibited fasciation or abnormal morphology (Figs. 2A-C). The SAM width and area of mms1 were also significantly larger than that of ZS10, whereas the SAM height did not show a significant difference (Fig. 2D, E). These observations were further confirmed by scanning electron microscopy at the same stage (Figs. 2E-G). During the floral meristem differentiation stage, scanning electron microscopy observations showed that the multiple SAMs in mms1 further differentiated into multiple inflorescence meristems (Figs. 2H-J), consistent with the results obtained from stereo microscopy at the same developmental stage (Fig. S2).

Fig. 1.

Phenotypic characterization of the mms1 mutant. A Flowering-stage phenotypes of ZS10, F₁ hybrid of ZS10 × mms1 and mms1, scale bar = 10 cm. B Quantification of main stem number in ZS10, F₁ hybrid of ZS10 × mms1 and mms1 (n = 10). C Statistical analysis of total silique number per plant at maturity in ZS10, F₁ hybrid of ZS10 × mms1 and mms1 (n = 10). Data are presented as mean ± SD. The significance of the data was tested using Student’ s t- test: ***, P < 0.001

Fig. 2.

Cytological analysis of the shoot apical meristem (SAM) and inflorescence meristem in ZS10 and mms1. A-C Paraffin sections of the SAM at 18 days after germination (DAG), scale bar = 100 µm. D Statistical analysis of SAM width and height at 18 DAG (n = 18). E Statistical analysis of SAM area at 18 DAG (n = 18). F-H Scanning electron microscopy (SEM) images of the SAM at 18 DAG. I-K SEM images of the inflorescence meristem at the floral transition stage, bar = 200µm. Data are presented as mean ± SD. The significance of the data was tested using Student’ s t-test: ***, P < 0.001; ns, not significant (P > 0.05)

To investigate the genetic pattern of the multi-main-stem trait, mms1 was crossed with the male parent ZS10. All F₁ individuals showed the single-main-stem phenotype. The F₂ and BC₁ populations were used to conduct a chi-square test. Among the 207 individuals in the F₂ population, there are 162 normal individuals and 45 multi-main-stem individuals, which matched the expected 3:1 segregation ratio (χ²=1.17 < χ²_0.05=3.8, P > 0.05). Among the 103 individuals in the BC₁ population, there are 60 normal individuals and 43 multi-main-stem individuals, which matched the expected 1:1 segregation ratio (χ²=2.85 < χ²_0.05=3.8, P > 0.05). The results indicated that the multi-main-stem phenotype of mms1 is controlled by a single recessive gene.

Phytohormones content analysis

Phytohormones play important roles in SAM maintenance and development. Therefore, to investigate whether phytohormone levels are altered in the multi-main-stem mutant, phytohormone accumulation in the SAMs of the F₃ population at 18 DAG were quantified. The results showed that the content of indole-3-acetic acid (IAA), the major form of auxin, was significantly reduced in the aa plants (Fig. 3A), whereas the content of other phytohormones, including cytokinin, brassinosteroid, salicylic acid, and gibberellin, did not differ significantly between the two genotypes (Fig. S9). To determine whether exogenous auxin could restore the abnormal SAM phenotype, mutant seedlings were treated with different concentrations of NAA and SAM morphology was examined by paraffin sectioning (Fig. 3B). After 7 days of continuous treatment following germination, treatment with 0.1 µM NAA did not significantly affect SAM morphology, whereas treatment with 0.5, 1.0, and 2.0 µM NAA significantly reduced SAM area and width to levels comparable to WT (Fig. 3C, D). However, SAM height did not show significant changes among treatments (Fig. 3E). In contrast, no significant differences in SAM morphology were observed among NAA-treated and untreated mms1 plants after 14 days of continuous treatment (Fig. S4).

Fig. 3.

Phytohormone content analysis at 18 DAG and SAM phenotype of mms1 seedlings after 7 days of NAA treatment on medium. A Indole-3-acetic acid (IAA) contents in the SAM at 18 DAG in AA and aa. AA and aa indicate normal plants and multi-main-stem plants from the F₃ population, respectively. B SAM morphology of mms1 plants treated with different concentrations of NAA (0, 0.1, 0.5, 1.0, and 2.0 µM) for 7 days after germination. Scale bar = 50 µm. Seedlings were grown on medium supplemented with different concentrations of NAA for 7 days after germination. C-F Quantification of SAM area (C), width (D) and height (E) under different NAA concentrations. Data are presented as mean ± SD (n = 6). The dashed line represents the corresponding ZS10 level for comparison. Data are presented as mean ± SD. The significance of the data was tested using Student’ s t-test: *, P < 0.05; ***, P < 0.001; ns, not significant (P > 0.05)

Genetic mapping based on the 50 K SNP array

To preliminarily map the gene controlling the multi-main-stem trait, 30 multi-main-stem and 30 single-main-stem individuals were selected from the F₂ population and used to construct three DNA pools for the multi-main-stem plants and three DNA pools for the single-main-stem plants, which were genotyped using the Brassica 50 K SNP array. A total of 169 polymorphic SNPs were identified between the two extreme bulks, among which 118 were located on chromosome A09, while the remaining SNPs were evenly distributed across the other chromosomes (Fig. 4A; Table S1). Further analysis of 118 polymorphic SNPs on chromosome A09 revealed that 85 of them were predominantly concentrated within the 11.20–20 Mb interval of the Darmor-bzh reference genome, which corresponds to the 14.56–22.90 Mb region in the ZS11 reference genome (Fig. 4B). These results showed that the candidate locus controlling the multi-main-stem trait was mapped to an 8.34 Mb interval on chromosome A09 of the ZS11 reference genome, and the gene was designated BnaA09.mms1.

Fig. 4.

Fine-mapping of BnaA09.MMS1 in B. napus. A Chromosomal distribution of SNPs retained for analysis based on the B. napus Darmor-bzh reference genome. B Distribution of polymorphic SNPs on chromosome A09. C Preliminary mapping of BnaA09.MMS1 using 1,930 individuals from F₂ population. Numbers below the markers represent the recombinants between the markers and BnaA09.MMS1. D Fine mapping of BnaA09.MMS1 using 5,736 individuals from F₃ population. Numbers below the markers represent the recombinants between the markers and BnaA09.MMS1

Fine mapping of BnaA09.MMS1

To determine the precise chromosomal location of BnaA09.MMS1, two flanking indel markers (ID1 and ID20) were first developed based on resequencing data and the SNP array to flank the target interval and used to screen an F₂ population consisting of 1,930 individuals. A total of 41 recombinant plants were identified, among which only 12 individuals exhibited the multi-main-stem phenotype. Seven additional indel markers (ID7, ID13, ID19, ID26, ID3-3, ID3-8, and ID3-9) (Table S2) were subsequently developed and applied to genotype these 12 recombinant individuals in combination with phenotypic evaluation. As a result, the candidate interval was further narrowed to the region between markers ID-13 and ID3-3, corresponding to a 2.24 Mb segment in the ZS11 genome (Fig. 4C).

To refine the candidate interval, a larger F₃ population comprising 5,736 individuals was constructed. Recombinants were firstly screened using the markers ID-13 and ID3-8. A total of 59 recombinant plants were identified, of which 12 exhibited the multi-main-stem phenotype. Five additional indel markers (ID4-3, ID4-11, ID4-15, ID4-17, ID4-22) (Table S2) were then developed and applied to genotype these 12 recombinant individuals, ultimately localizing BnaA09.MMS1 to the interval between markers ID4-15 and ID4-22, corresponding to a 340-kb segment (18.62–18.96 Mb) in the ZS11 genome (Fig. 4D).

Identification of candidate genes

The 340-kb candidate interval of BnaA09.MMS1 was analyzed using the B. napus ZS11 genome as a reference, revealing 39 predicted genes within this region. Homologs of these 39 genes in Arabidopsis were identified by performing BLAST searches against the TAIR database, whereas no homologs or no function annotations were detected for seven genes (Table S3). Based on functional annotation, the genes unrelated to the observed phenotype were further excluded from the remaining 32 genes, leaving 6 annotated genes for detailed analysis.

BnaA09G0252700ZS belongs to the protein phosphatase 2 C (PP2C) family, whose members are related to meristem maintenance and developmental signaling regulation [27]. BnaA09G0253100ZS encodes a F-box/RNI-like/FBD-like domains-containing protein, and BnaA09G0256000ZS belongs to the RING/U-box superfamily, both of which are involved in ubiquitin-mediated protein degradation in phytohormone signaling [28]. BnaA09G0253900ZS encodes a transmembrane protein, and some members of this family are known to be involved in hormone transport and responses in SAM [29]. BnaA09G0254500ZS belongs to the UDP-glycosyltransferase superfamily. Proteins in this family usually catalyze the transfer of a glycosyl moiety to hormones and other molecules, thereby modulating their biological activity [30]. BnaA09G0255800ZS encodes mitogen-activated protein kinase kinase kinase kinase 6 (MAP4K6), and members of the MAP4K family have been reported to participate in multiple developmental processes, including cell proliferation and cell expansion [31]. The 6 genes described above may be associated with multi-main-stem phenotype. Together with 7 genes that have not been previously studied (BnaA09G0254200ZS, BnaA09G0254300ZS, BnaA09G0254400ZS, BnaA09G0254700ZS, BnaA09G0254800ZS, BnaA09G0255300ZS, BnaA09G0255700ZS), they were considered as candidates for BnaA09.MMS1 and subjected to further analysis.

To further identify the candidate gene for the multi-main-stem trait, qRT-PCR was performed to examine their expression levels in the SAMs of AA and aa individuals from F₃ population. The results showed that BnaA09G0252700ZS, BnaA09G0254800ZS, BnaA09G0255300ZS, BnaA09G0256000ZS, and BnaA09G0255800ZS exhibited no significant differences in expression levels (Fig. S5A-E), while the expression of BnaA09G0253100ZS, BnaA09G0253900ZS, BnaA09G0254200ZS, BnaA09G0254400ZS, BnaA09G0254700ZS, and BnaA09G0255700ZS were extremely low and could not be detected in the SAM. Therefore, these genes are unlikely to represent the candidate gene for BnaA09.MMS1. In contrast, BnaA09G0254300ZS was not expressed in aa but showed low expression in AA (Fig. 5A), and BnaA09G0254500ZS was significantly up-regulated in aa (Fig. 5B). Furthermore, the expression level of BnaA09G0254500ZS was examined in the individual plants with three genotypes. The results showed that BnaA09G0254500ZS expression was significantly higher in aa plants than in either AA or Aa plants, whereas no significant difference was observed between AA and Aa plants (Fig. S5F). Accordingly, the promoter regions and full-length genomic sequences of these two candidate genes were amplified and sequenced. The genomic sequence of BnaA09G0254300ZS could not be amplified in aa. Synteny analysis further revealed its absence in several conventional rapeseed cultivars, including Westar, Gangan, and No2127 (Figs. S6A-C), and this result was further validated by PCR amplification and sequencing (Figs. S6D, E). Therefore, it is unlikely to be the candidate gene for BnaA09.MMS1. BnaA09G0254500ZS contained a 1,598 bp insertion located 45 bp upstream of the start codon, as well as 14 synonymous and 6 non-synonymous mutations within the coding sequence (Figs. 5C; S7; S8A). Sequence analysis using the Genetic Information Research Institute database (www.girinst.org) indicated that the 1,598 insertion is a hAT transposon. Based on expression analysis and sequence variation analysis, BnaA09G0254500ZS was considered as the most likely candidate gene for BnaA09.MMS1.

Fig. 5.

Candidate gene analysis of BnaA09.MMS1 in B. napus. A Relative expression levels of the candidate gene BnaA09G0254500ZS and its homologs in the SAM of AA and aa plants at 18 DAG. B Relative expression levels of the candidate gene BnaA09G0254500ZS and its homologs in the SAM of AA and aa plants at 18 DAG. A09 represents the candidate gene BnaA09G0254500ZS, A05 represents its homologous copy BnaA05G0321800ZS, and C05-2 represents the homologous copy BnaC05G0343800ZS. C Gene structure, SNP variations, and fragment replacement analysis of the candidate gene BnaA09G0254500ZS between the AA and aa plants. Data are presented as mean ± SD. The significance of the data was tested using Student’ s t-test: **, P < 0.01; ***, P < 0.001; ns, no significance

To evaluate whether the amino acid substitutions may affect protein function, we analyzed the non-synonymous substitutions identified in the coding sequence of BnaA09G0254500ZS. Domain prediction analysis revealed that several amino acid substitutions were located within the predicted GT1 glycosyltransferase domain (Fig. S8B). However, none of these substitutions were found within the highly conserved Plant Secondary Product Glycosyltransferase (PSPG) motif (Fig. S8B), which is recognized as a critical region for donor binding (Li et al., 2007). To further assess the potential structural impact of these substitutions, we performed protein structure modeling of both the AA and aa proteins. The predicted structures showed nearly complete overlap, with no obvious alterations in the overall folding (Figs. S8C-E). In addition, to determine whether the hAT transposon insertion of BnaA09G0254500ZS promoter co-segregates with the multi-main-stem phenotype, a three-primer marker was developed based on the 1,598-bp hAT transposon insertion identified in the promoter region of BnaA09G0254500ZS (Fig. S9A). Genotyping results showed that this insertion-based marker co-segregated with the multi-main-stem phenotype (Fig. S9B). Furthermore, the co-segregation pattern was independently validated using the previously developed linked marker ID-26, and the genotyping results obtained from the two markers were consistent (Fig. S9C). Therefore, the increased expression level of BnaA09G0254500ZS in aa plants is likely associated with the 1,598-bp hAT transposon insertion in its promoter region.

To determine whether the observed sequence and expression changes were specific to BnaA09G0254500ZS, its homologous genes were further analyzed for sequence variation and expression patterns. Homologous genes were searched using the Orthologous Groups Search tool on the BnPIR website, and five homologous genes were identified: BnaA05G0321800ZS on chromosome A05, BnaC05G0336300ZS and BnaC05G0343800ZS on chromosome C05, BnaC08G0105500ZS on C08. The promoter and full-length genomic sequences of these genes were analyzed using resequencing data, and no sequencing variations were detected between AA and aa (Figs. S10A-D). qRT-PCR was performed to determine the expression levels of these genes in the SAM. The results showed that the expression of BnaC05G0336300ZS (C05-1) and BnaC08G0105500ZS (C08) was undetectable in the SAM, while BnaA05G0321800ZS (A05) and BnaC05G0343800ZS (C05-2) exhibited extremely low expression with no significant difference between AA and aa plants (Fig. 5B). The expression patterns of the five homologs were further analyzed in other tissues, including roots, leaves, and buds. The results showed that A09 copy showed relatively high expression and exhibited significant difference between AA and aa plants in these tissues (Figs. S10E-G). In contrast, the other copies showed extremely low or nearly undetectable expression levels, similar to their expression patterns in the SAM. Collectively, these results indicated that among five homologous genes, BnaA09G0254500ZS exerts the primary function, which further supports that it is the most likely candidate gene for multi-main-stem phenotype.

Transcriptome analysis

To further explore how the BnaA09.MMS1 influences SAM development, RNA sequencing was performed on SAMs collected at 18 DAG. To validate the reliability of the RNA-seq data, several differentially expressed genes identified from the transcriptome analysis were selected for qRT-PCR verification. The qRT-PCR results were consistent with the RNA-seq data, confirming the reliability of the transcriptomic analysis (Fig. S11). A total of 347 upregulated and 293 downregulated differentially expressed genes (DEGs) were identified between AA and aa (Fig. 6A). Subsequently, GO enrichment analysis and KEGG enrichment analysis were conducted. Among the top 30 enriched GO terms, many terms were associated with meristem-related processes, including stem cell population maintenance, maintenance of cell number, meristem maintenance, and vegetative to reproductive phase transition of meristem (Fig. 6B). KEGG enrichment analysis showed that the DEGs were significantly enriched in glutathione metabolism, tryptophan metabolism, and amino acid metabolism pathways (Fig. S12A). Tryptophan is an important precursor for IAA biosynthesis, and it has been known that disturbances of tryptophan metabolism affect auxin metabolism and accumulation [32]. Accordingly, DEGs associated with SAM development and tryptophan metabolism were subjected to further analysis. Specifically, BnaA04.CLV3 and BnaC04.CLV3 were significantly downregulated in aa, whereas BnaC09.WUS, BnaA09.HAT2, BnaA02.TFL1, and BnaC09.TFL1 were significantly upregulated in aa (Fig. 6C; Table S5). Several genes associated with tryptophan metabolism also showed altered expression in aa, including the downregulation of BnaC03.CYP79B2 and BnaC07.CYP79B2, as well as the upregulation of BnaC07.IGMT1 and BnaA08.IGMT2 (Fig. S12B; Table S5). These results collectively indicate that BnaA09.MMS1 may influence the homeostasis of tryptophan metabolism and the CLV3-WUS feedback loop, thereby affecting SAM development.

Fig. 6.

Differentially expressed genes (DEGs) and enriched GO terms in the SAM at 18 DAG between AA and aa. A Volcano plot of DEGs in the SAM between AA and aa. Red dots represent up-regulated genes (log₂ fold change > 0.585, p-adj < 0.05), blue dots represent down-regulated genes (log₂ fold change < -0.585, p-adj < 0.05), and gray dots represent genes with no significant difference. B Top 30 GO terms enriched among DEGs. Dot size represents the number of enriched genes, and color indicates the -log₁₀ (FDR) value. C The heatmap showing the expression patterns of the representative SAM development-related genes in AA and aa based on RNA-seq data

Discussion

Potential applications of the multi-main-stem phenotype

Rapeseed is not only an important oilseed crop worldwide but also a widely consumed vegetable in China. Rapeseed shoots play a key role in the winter-spring vegetable supply system of the Yangtze River basin, thereby enhancing the economic value of rapeseed cultivation [33]. However, most rapeseed cultivars currently used for vegetable production are single-stem types, for which yield from a single harvest is limited. Therefore, improving both seed yield and vegetable yield is equally important in rapeseed breeding. In this study, we identified a multi‑main-stem rapeseed mutant that produces multiple main stems and main inflorescences, allowing the simultaneous growth of multiple edible stems (Figs. 1A, B). At maturity, the total silique number per plant was significantly higher than the wild type (Fig. 1C), suggesting its potential to increase shoot yield and individual plant yield, although its detailed agronomic performance still requires further systematic evaluation. Furthermore, the molecular markers developed in this study can be applied in marker‑assisted selection to accelerate the breeding of multi‑main-stem rapeseed varieties.

Multi-main-stem phenotype originates from abnormal SAM development

Although both excessive elongated lateral branches and increased number of main stems can lead to multi-main-stem phenotype, they arise from distinct developmental origins. Branching is typically initiated from axillary meristems formed in leaf axils, and its outgrowth depends on the release of apical dominance [34]. Previous studies have shown that loss of apical dominance can trigger abnormal activation of axillary meristems, resulting in excessive branching that resembles the multi-main-stem phenotype [35–37]. In contrast, true main stems are directly derived from the SAM, and the increase of main stem number generally originates during embryogenesis or early seedling development. Several studies have also indicated that developmental defects of the SAM at early seedling stages can exhibit multi‑main-stem phenotypes [17, 19]. Our paraffin sectioning results revealed that the SAM of mms1 seedlings was significantly wider and larger than that of ZS10 at 7 DAG (Fig. 3B). At 18 DAG, both paraffin sectioning and scanning electron microscopy further demonstrated that mms1 developed multiple SAM structures, rather than abnormal outgrowth of axillary meristems (Figs. 2; S2). This suggests that its multi‑main-stem phenotype is more likely attributable to abnormal SAM development than to abnormal activation of axillary meristems, providing important clues for further understanding this trait from the perspective of SAM development.

Altered auxin homeostasis may disrupt SAM maintenance

The maintenance of SAM homeostasis requires the precise control of auxin. In this study, the IAA content in the SAM of aa plants was significantly lower than that in AA plants (Fig. 3A). Continuous NAA treatment for 7 days after germination partially rescued the SAM phenotype of mms1, and SAM morphology became comparable to that of ZS10 under 0.5, 1.0, and 2.0 µM NAA treatments (Figs. 3B-E). These results suggest that reduced auxin levels contribute to the abnormal SAM development observed in mms1. Interestingly, the rescue effect was not maintained after an additional 7 days of treatment following transplantation to soil (Fig. S4). This may reflect developmental differences in auxin responsiveness, as younger seedlings are generally more sensitive to hormonal perturbations than mature plants. Alternatively, changes in treatment conditions after transplantation, such as reduced hormone uptake efficiency associated with foliar spraying, may have limited the effectiveness of prolonged NAA treatment. Further optimization of treatment conditions will be required to evaluate the long-term effects of auxin supplementation on SAM development.

A hAT transposon in BnaA09G0254500ZS promoter may be associated with altered gene expression

Glycosyltransferases have been reported to regulate the metabolism of phytohormones and other small molecules through glycosylation, thereby regulating diverse developmental processes throughout the plant life cycle [38–41]. BnaA09G0254500ZS encodes a glycosyltransferase, however, its biological function remains unknown. Relative expression analysis revealed that BnaA09G0254500ZS was significantly upregulated in the SAMs of aa plants at 18 DAG (Fig. 5B). Interestingly, the expression level of BnaA09G0254500ZS in heterozygous Aa plants was not significantly different from that in AA plants, whereas both genotypes exhibited substantially lower expression than aa plants (Fig. S5F). Consistent with this expression pattern, Aa plants displayed a normal single-main-stem phenotype, suggesting that the elevated expression level of BnaA09G0254500ZS may need to exceed a certain threshold before phenotypic effects become apparent. Several sequence variations were identified in BnaA09G0254500ZS between AA and aa plants, including non-synonymous substitutions in the coding region and a 1,598-bp hAT transposon insertion in the promoter region (Fig. 5C). Both domain prediction and protein structural modeling indicated no obvious impact on conserved functional motifs or overall protein conformation (Fig. S8), suggesting that coding variation is unlikely to be the primary causal factor. In contrast, the increased expression level implies that the promoter insertion may affect promoter activity. A three-primer marker designed based on this insertion demonstrated complete co-segregation with the multi-main-stem phenotype (Fig. S9), supporting a tight genetic association between the insertion and the trait. However, the precise regulatory mechanism by which this transposon insertion affects gene expression remains to be further investigated.

Conclusion

In this study, we identified a multi-main-stem mutant in rapeseed. Cytological observations indicated that the multi-main-stem trait was associated with abnormal development of SAM. In addition, significantly reduced IAA levels were detected in the SAMs of mutant plants. Using the Brassica napus 50 K SNP array and map-based cloning, the locus controlling the multi-main-stem trait was fine-mapped to a 340-kb genomic interval. Sequence and expression analyses identified BnaA09G0254500ZS as the most likely candidate gene. Transcriptome analysis suggested that the multi-main-stem trait may be associated with abnormal expression of genes related to stem cell maintenance and tryptophan metabolism. Our research provides new insights and germplasm resources for improving rapeseed plant architecture.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

SAM

Shoot apical meristem

OC

Organizing center

CZ

Central zone

PZ

Peripheral zone

MMS

Multi-main-stem

MI

Multi-inflorescence

DAG

Days after germination

SEM

Scanning electron microscopy

UHPLC

Ultra-high-performance liquid chromatography system

MRM

Multiple reaction monitoring

IAA

Indole-3-acetic acid

tZ

Trans-zeatin

tZR

Trans-zeatin riboside

iP

Isopentenyladenine

iPR

Isopentenyladenosine

cZ

Cis-zeatin

DHZR

Dihydrozeatin riboside

BL

Brassinolide

SA

Salicylic acid

GA₁

Gibberellin A₁

GA₃

Gibberellin A₃

GA₄

Gibberellin A₄

GA₈

Gibberellin A₈

NAA

α-naphthaleneacetic acid

BSA

Nulked segregant analysis

qRT-PCR

Quantitative real-time PCR

FPKM

Fragments Per Kilobase of exon model per Million mapped reads

DEGs

Differentially expressed genes

GO

Gene ontology

KEGG

Kyoto encyclopedia of genes and genomes

Author contributions

X.L. performed the experiments and wrote the original manuscript. S.Z. contributed to data analysis. S.Y. and Y.M. contributed to field sampling. J.W., B.Y., and C.M. contributed to manuscript revision. B.L., L.G., and J.T. supervised the project and provided critical revisions to the manuscript. All authors read and approved the final manuscript.

Funding

This work was funded by the National Natural Science Foundation of China (U25A20672, 32401939).

Data availability

All relevant data can be found within the manuscript and its supplementary materials. The raw transcriptome sequencing data used in this study have been deposited in the Genome Sequence Archive [42] in National Genomics Data Center [43] under the accession number CRA039879.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.