Fine Mapping of BrTCP1 as a Key Regulator of Branching in Flowering Chinese Cabbage (Brassica rapa subsp. chinensis)

Abstract

Branching is a critical agronomic trait in flowering Chinese cabbage (Brassica rapa subsp. chinensis), influencing plant architecture and yield. In this study, there was a highly significant difference between CX010 (single primary rosette branches) and BCT18 (multiple primary rosette branches). Phenotypic analysis revealed significant differences in primary rosette branch numbers, with BCT18 showing up to 15 branches and CX010 displaying only one main stem branch. Genetic analysis indicated that branching was controlled by quantitative trait loci (QTL) with a normal distribution of branch numbers. Using bulked segregant analysis coupled with sequencing (BSA-seq), we identified a candidate interval of approximately 2.96 Mb on chromosome A07 linked to branching. Fine mapping narrowed this to a 172 kb region containing 29 genes, with BraA07g032600.3C (BrTCP1) as the most likely candidate. cDNA cloning of the BrTCP1 gene revealed several variations in BCT18 compared to CX010, including a 6 bp insertion, 10 SNPs, and two single-nucleotide deletions. Expression analysis indicated that BrTCP1 was highly expressed in the rosette stems of CX010 compared to BCT18, consistent with its role as a branching suppressor. The heterologous mutants in Arabidopsis confirmed the conserved role of BrTCP1 in branch inhibition. These findings reveal that BrTCP1 might be a key regulator of branching in flowering Chinese cabbage, providing insights into the molecular mechanisms underlying this trait and offering a framework for genetic improvement in Brassica crops.

1. Introduction

Flowering Chinese cabbage is an important and popular vegetable because of its high nutrient content and good flavor in Asia. As the number of primary rosette branches directly impacts yield, branching was a critical agronomic trait in flowering Chinese cabbage. This vegetable is characterized by its edible flower stalks [1]. The flower stalk of flowering Chinese cabbage developed from the apical meristem of lateral buds. Its branching architecture included the main branch, which arose from the apical meristem; rosette branches, which developed from the rosette axis; and axillary branches, which originated from the axillary meristems of the stem and leaves [2]. Manipulating plant architecture, particularly branching patterns, is a promising approach to enhance production, and identifying the key gene controlling branching is essential for developing improved cultivars.

Branching development is a complex process influenced by a multitude of factors, including environmental conditions, genetic determinants, and phytohormones [3]. The regulation of branch development is stage-dependent, with distinct gene expression profiles and regulatory mechanisms observed at different developmental phases [4]. Genetic factors play a major role in axillary bud initiation, whereas the later stages of growth and development involve a synergistic interaction between phytohormones and genes [5]. In Arabidopsis thaliana, las loss-of-function mutants exhibited a novel phenotypic characteristic: the inability to form lateral branches during the vegetative phase of development [6]. Isolated from a low-tillering mutant, MOC1, the first cloned gene affecting rice tillering, showed contrasting phenotypes: loss-of-function (moc1) eliminated tillering buds, while MOC1 overexpression promoted tiller development from these buds [7]. The moc3 mutant exhibits the formation of lateral bud-like structures; however, a developmental block in the axillary meristems precludes tiller outgrowth [8]. The TB1 gene, a member of the TCP protein family, plays a role in the developmental processes of axillary meristems. In maize, TB1 negatively regulates axillary meristem formation; its expression diminishes as the plant progresses through its developmental stages [9]. In Arabidopsis, homologs of the TB1 genes, both AtBRC1 and AtBRC2, exert inhibitory effects on lateral branch development, with AtBRC1 demonstrating a more prominent regulatory role. AtBRC1 gene expression is observed in axillary buds and found to be negatively correlated with bud growth [10]. In tomato, homologs of the AtBRC1 gene are implicated in the inhibitory regulation of stem branching [11]. In apple, MdWUSCHEL2 interacts with the co-repressor MdTPR9, which is activated by cytokinin, and regulates branching by suppressing MdTCP12 activity [12]. However, there are fewer studies on the identification of the key genes of the rosette branching in flowering Chinese cabbage.

Shoot branching is not only genetically controlled but also strongly modulated by multiple phytohormones, particularly auxin, strigolactones (SLs), and gibberellins (GAs), which interact with key transcription factors such as members of the TCP family. These hormonal signals coordinate axillary bud initiation, dormancy, and outgrowth through both local and systemic regulation. Auxin, synthesized at the shoot apex, is transported basipetally and indirectly inhibits axillary bud outgrowth. It does not suppress bud activity directly but acts by promoting the biosynthesis of strigolactones and repressing cytokinin production in the stem, thereby creating an inhibitory environment for bud activation [13]. Strigolactones, in turn, directly inhibit bud outgrowth and enhance the expression of BRANCHED1 (BRC1), a TCP transcription factor known to suppress lateral branching [10]. The interaction between SL signaling and TCP activity is particularly critical; in Arabidopsis, BRC1 integrates signals from SLs, auxin, and sugars to determine the fate of axillary buds [14]. Disruption in SL biosynthesis or signaling leads to reduced BRC1 expression and increased branching phenotypes, emphasizing their interdependence. GAs and SLs synergistically regulate rice tillering through the SLR1-OsMADS23-D14 module [15].

Bulked segregant analysis (BSA) has been widely used to quickly identify markers linked to candidate genes or QTL regions. This method involves observing the genotype and selecting two highly divergent DNA samples from a segregating population that exhibit extreme traits of interest [16]. BSA-seq was employed to identify nine QTLs associated with heading date, plant height, and panicle length in a large F₂ population derived from two rice landraces. Among these, qPH8 represented a major novel QTL for plant height [17]. Using BSA and the Brassica 60K SNP array with 7785 BC₁F₃ individuals, two major QTLs (qDBA09 and qDBC06) for branching were mapped to a 270 kb region on chromosome A09, and BnaA09.ELP6 was identified as a candidate gene [18]. In chrysanthemum research, BSA-seq technology, combined with SNP/InDel-index and exponential decay (ED) algorithms, was employed to map candidate gene regions associated with the silver lotus flower phenotype. Additionally, competitive allele-specific PCR (KASP) markers based on the identified candidate SNPs were developed, which can effectively distinguish between the silver lotus flower and non-silver lotus flower phenotypes [19]. This technology enhanced the efficiency of plant QTL mapping and candidate gene localization.

In this study, CX010 (single primary rosette branch) and BCT18 (multiple primary rosette branches) were used as experimental materials to statistically analyze the number of primary rosette branches. The candidate gene was mapped using BSA-seq analysis, and cloning and sequencing were performed to identify mutation characteristics. The validation was carried out by characterizing the mutant traits of the homologous gene in Arabidopsis. These findings offer a valuable genetic resource for the development of improved cultivars with enhanced branching traits.

2. Materials and Methods

2.1. Plant Materials and Mapping Populations

‘CX010’ is a doubled haploid (DH) line derived from Guangdong flowering Chinese cabbage with one main shoot. The multi-branching mutant ‘BCT18’ was obtained by microspore culture. The plants were grown on the experimental farm of Shenyang Agricultural University, Liaoning, China, in spring 2021. For genetic analysis, wild-type CX010 (P₁) was crossed with the BCT18 (P₂) to generate the F₁ and F₂ populations. And F₂ segregation mapping population (840 individuals) was used for BSA-seq and traditional QTL mapping. Phenotypic identification began at the seedling stage (three biological repetitions), and sampling was performed at the most obvious stage of branching phenotype (60 days after germination). The criteria for investigation were primary rosette branches within 5 cm of the stem base. Performing analysis of variance on phenotypic data was conducted using SPSS 19.0 software (IBM SPSS, Chicago, IL, USA).

2.2. DNA Isolation and Pool Construction

For BSA-seq, an F₂ population of the cross BCT18 × CX010 was selected to construct DNA pools. The more-branching pool (MB) consisted of 50 F₂ individuals with 10–15 primary rosette branches, while the less-branching pool (LB) comprised 50 individuals with only 1 branch. Total DNA was extracted from the two parents and two extreme pools from fresh leaves using a DNA Secure Plant Kit (Tiangen, Beijing, China). The DNA of mixed pool samples that passed the quality inspection was subjected to BSA sequencing experiments. Qualified DNA samples were fragmented into 400 bp fragments using a fragmentation kit for library construction. The sequencing platform is a NovaSeq 6000 sequencer (Illumina, San Diego, CA, USA), and the sequencing mode is PE150. The clean data were obtained to perform quality control on offline data, remove low-quality sequences, and sequence splice sequences. Next, the clean data were compared with the reference genome (http://brassicadb.cn/#/Download/) (accessed on 7 June 2021), and SNP and InDel detection and annotation were performed based on the comparison results [20]. Next, the SNP index and the differences in the offspring mixed pool were calculated. The intersection regions of the top 1% SNP-index values and the top 1% ED values were, respectively, screened, and the candidate interval was finally mapped [21]. A sliding window of 1 Mb with a step size of 10 kb was used for SNP-index analysis. Finally, candidate genes were selected within the regions according to the functional annotations on the Chinese cabbage database and homologous Arabidopsis.

2.3. Genotyping and Linkage Map Construction

The interval obtained by preliminary mapping using the BSA-seq method is too large, which is not conducive to accurately predict candidate genes. In order to narrow down the QTL region of the gene controlling branching, SNP markers and Indel markers were uniformly designed within the candidate interval. KASP genotyping technology was used to analyze the genotype of the 190 plants and parents [22]. The genotype that is the same as the female parent is recorded as “0”, the genotype that is the same as the male parent is recorded as “2”, the heterozygous genotype is recorded as “1”, and the missing genotype is recorded as “−1”. QTL IciMapping (v4.2) was then used to analyze the KASP genotyping data, combining with the phenotypic data for precise localization. Interval mapping (IM) was first used to detect potential QTL, and a significant QTL was defined based on LOD > 3.0. If the confidence interval of a site was higher than the LOD threshold, we considered the site to exist. Finally, the SNPs and Indels in the QTL interval were used to identify candidate genes by positioning their physical positions on the cabbage genome.

2.4. Cloning and Analysis of Gene Sequences

Based on searching for candidate gene sequences, the primers were used to amplify full-length coding sequences and promoter sequences of candidate genes using cDNA from CX010 and BCT18. The amplification products were separated by agarose gel electrophoresis; then, using a gel extraction kit (Omega, Bridgewater, NJ, USA), the isolated gene fragments. The recovered products were ligated with pGEM-T Easy vector (Promega, Madison, WI, USA) using T4 ligase. The recombinants were transformed into TOP10 competent cells (CWBIO) for blue and white shift screening [23]. The recombinant plasmids were sequenced by Sangon Biotech (Shanghai, China). The sequences were aligned by DNAMAN (Lynnon Biosoft; San Diego, CA, USA).

2.5. Determination of Expression Level by RT-qPCR

The different organs (roots, stems, leaves, flowers, and siliques) and shortened stems from different stages of BCT18 and CX010 were extracted and used as templates, in three biological repetitions. Total RNA was extracted using the PLANTpure universal plant total RNA rapid extraction kit (Aidlab, Beijing, China). FastKing cDNA first-strand synthesis kit (Tiangen, China) was used to reverse RNA into complementary DNA (Complementary DNA, cDNA). A 20 µL reaction system was prepared using cDNA as the template and UltraSYBR mixture (Kangwei Century, Beijing, China) as the fluorescent detection dye. The specific ingredients were as follows: 8.8 µL of distilled deionized water (ddH2O), 0.8 µL of diluted cDNA, 10 µL of 2.5 × Real MasterMix/20 × SYBR solution, and 0.4 µL of mixed primer solution. QuantStudio™ 6 Flex (Applied Biosystems, Waltham, MA, USA) instrument was used to perform qRT-PCR reactions. Actin was used as an internal reference. Each sample was repeated three times, and the relative expression was calculated using the 2^−ΔΔct calculation method.

2.6. Bioinformatics Analysis

Candidate gene information was retrieved from the BRAD database (http://brassicadb.cn/#/, accessed on 27 July 2024) and searched for homologs using NCBI (https://blast.ncbi.nlm.nih.gov/Blast.cgi, accessed on 27 July 2024). The protein secondary structure was predicted using the online website SOPMA (https://npsa-pbil.ibcp.fr/cgi-bin/npsa_automat.pl?page=npsa_sopma.html, accessed on 27 July 2024). The three-dimensional structure of the proteins was synthesized using the online website SWISS-MODEL (https://swissmodel.expasy.org/interactive/, accessed on 27 July 2024). The transmembrane domains were studied using TMHMM-2.0 (https://services.healthtech.dtu.dk/services/TMHMM-2.0/, accessed on 27 July 2024).

2.7. Identification of Homologous Mutants in Arabidopsis

The multi-branched phenotype mutant AT1G67260 was purchased from the proposed Southern mustard TAIR database. The plants were grown on the experimental farm of Shenyang Agricultural University, Liaoning, China, in spring 2021. Observing the branching phenotype during the bolting period.

3. Results

3.1. Phenotypic Observation and Inherent Characteristics of Branching in Flowering Chinese Cabbage

Phenotypic analysis showed considerable differences in branch number between the single-branch line CX010 (Figure 1a) and the multi-branch line BCT18 (Figure 1b). The primary rosette branches on the rosette stem reached a maximum value of about 15 branches during the reproductive growth stage of BCT18 (Figure 1c). In contrast, only one visible main stem branch was observed during the reproductive growth stage of CX010 (Figure 1c). After counting primary rosette branches of 840 F₂ plants, it was found that the number of primary rosette branches on the rosette stem showed obvious separation and showed a normal distribution (Figure 1d). This means that branching is controlled by quantitative trait loci and does not follow the laws of Mendelian inheritance.

3.2. Preliminary Localization of Branching Candidate Intervals by BSA-Seq

BSA-seq was performed to identify candidate intervals based on the genetic characteristics. DNA from BCT18, CX010, and F₂ individuals with extreme phenotypes of the two pools was sequenced. After removing sequencing adapters and filtering out unqualified sequences to obtain valid data, BWA was used to compare the reads to the Chinese cabbage (v3.0) version reference genome. 97,789,644, 114,718,700, 149,017,888, and 149,097,810 effective reads were obtained from the four mixed pools, respectively, which could be compared to the reference genome could reaching more than 98.73%. A total of 3,410,337 SNP mutations and 946,750 indel mutations were obtained. The average coverage depths of the four mixing pools were 36.25%, 41.80%, 54.59%, and 54.21%, respectively. According to the sliding window calculation, the intersection area of the top 1% of the SNP-index value (Figure 2a) and the top 1% of the ED value (Figure 2b) was screened, respectively. The final interval was located in the 22,160,000–25,120,000 bp region of the A07 chromosome, approximately 2.96 Mb.

3.3. Fine Mapping Candidate Gene

We utilized comprehensive genetic maps in conjunction with phenotypic data to conduct QTL mapping for the purpose of fine mapping. More SNP and InDel markers were excavated in the candidate interval, including 18 pairs of SNP molecular markers with good polymorphism and high genotyping quality. A total of 840 F₂ individuals were used in KASP, which were analyzed using QTL IciMapping software v42 (accessed on 27 July 2024). A QTL associated with primary rosette branches was identified between markers 21,248 and 21,251 (Figure 3). The genetic distance between the two polymorphic base markers 21,248 and 21,251 starts from 23,041,502 bps, ending at 23,646,811 bps on chromosome A07, respectively (Figure 3). The peak LOD score of the QTL between markers 21,248 and 21,251 was 4.07, exceeding the threshold of 3.0, indicating a statistically significant association with branching traits. To further narrow down the candidate region, we selected a total of 10 pairs of polymorphic Indel molecular markers with high genotyping quality. After F₂ genotyping through KASP in conjunction with QTL IciMapping software analysis, it was found that the 21,407 and 21,316 markers were found to be most significantly associated with branching (Figure 3, Table S1). The physical locations are 23,386,598 bp and 23,563,210 bp on chromosome A07, respectively. The distance between markers is approximately 172 kb, and it contains 29 genes (Figure 3 and Table S2).

3.4. Identification of Candidate Gene

Within the candidate region, BraA07g032600.3C encodes a protein with a TCP (TB1, CYC, PCF) domain, which may be involved in DNA binding and protein–protein interactions. The OsTB1 gene in rice suppressed the formation of lateral branches [24]. The homologous gene of TB1, BRC1, in Arabidopsis also suppresses the formation of lateral branches [25]. They all encode transcription factor TCP family proteins, which contain a TCP domain. Based on this, BraA07g032600.3C was the most likely candidate gene for branching in flowering Chinese cabbage, and it was named BrTCP1.

3.5. Cloning and Expression Pattern Analysis of BrTCP1 Gene

The full length of the BrTCP1 gene is 1122 bp, consisting of two exons and one intron. cDNA cloning and sequence analysis of the BrTCP1 gene revealed several variations in BCT18 compared to CX010, including a 6 bp insertion, 10 SNPs, and two single-nucleotide deletions (Figure 4).

The expression level of the BrTCP1 gene at the root, rosette, leaf, and flower bud was quantitatively analyzed in both CX010 and BCT18 using qRT-PCR to determine its physiological function. The qRT-PCR results show that the expression in the rosette of CX010 was significantly higher than BCT18, and the expression level in root, leaf, and flower bud was significantly lower than BCT18 (Figure 5a). Based on the higher expression level of BrTCP1 in the stems of CX010 compared to BCT18, it was predicted that BrTCP1 inhibited the number of lateral branches in flowering Chinese cabbage.

The expression levels of BrTCP1 in the rosette stem at different stages were also analyzed, and the results are shown in Figure 5b. Only in the second stage, the expression level of BrTCP1 in BCT18 was significantly higher than that in CX010, while in other stages, CX010 was higher than BCT18.

3.6. Structural Analysis of BrTCP1 Protein

The BrTCP1 protein in CX010 had 341 amino acid residues, a molecular weight of 38.79 kDa, and an isoelectric point of 6.80. The BrTCP1 protein in BCT18 possessed 343 amino acid residues, a molecular weight of 38.97 kDa, and an isoelectric point of 7.62. The BrTCP1 protein contained a TCP domain (83rd–225th) (Figure 6a). At the same time, the insertion of two amino acids (168th) within the TCP conserved domain might also affect the structure of the BrTCP1 protein. In addition, comparing the secondary structure of the BrTCP1 protein found that extended strands (CX010: 43; BCT18: 39) were significantly increased, especially at the mutation site (Figure 6b,c). Then, we online predicted the tertiary protein structure of the BrTCP1 protein before and after amino acid insertion, and found that these two amino acid insertions caused the tertiary structural changes in the BrTCP1 protein (Figure 6d,e). Therefore, we speculated that the insertion of 6 bp nucleoside acid might lead to the loss of the original function of the protein.

3.7. The Arabidopsis Homologous TCP1 Mutant Exhibited a Multi-Branching Phenotype

The function of the candidate gene BrTCP1 was validated by introducing the target gene into an Arabidopsis mutant. The mutant AT1G67260 was purchased from the Arabidopsis TAIR database, and compared to the wild type (Figure 7a), the mutant exhibited increased branching phenotypes (Figure 7b). The above results indicated that BrTCP1 was the target gene responsible for the increased branching in BCT18.

4. Discussion

Multiple branching significantly increased the number of branches in flowering Chinese cabbage, thereby optimizing plant architecture and increasing yield per plant. We identified BraA07g032600.3C (BrTCP1), a TCP domain transcription factor, as a key regulator of branching in flowering Chinese cabbage. The mutation of BrTCP1 likely weakened its inhibitory effect on axillary bud outgrowth, resulting in an increased branching number. The branching gene BrTCP1 was successfully cloned in flowering Chinese cabbage, providing insights into the molecular breeding of multiple branching.

Branching ability was a critical trait in plant breeding, directly influencing various aspects of plant growth and development, yield, and stress resistance [26]. From a production perspective, moderate branching ability significantly increased individual plant yield by enhancing the number of effective inflorescences. Under high-density planting systems, branching varieties achieved higher yield through efficient spatial utilization. However, excessive branching restricted main stem development, resulting in thinner stalks and reduced key commercial traits, such as the thickness and weight of the main stalk. It also impaired field ventilation and light penetration, increasing the risk of pest and disease infestations. The PAT1 gene was identified in leafy Brassica juncea, where it regulated branching and flowering by interacting with BRC1 and COL13 [27]. A branching trait control gene, qDBA09, was identified in Brassica napus and mapped to a 270 kb genomic region on chromosome A09 [18]. A shoot branching gene, Bra007056, was identified in non-heading Chinese cabbage, and its expression exhibited a positive correlation with branch development [28]. Two tandem GA2ox1 homologs on chromosome A07 (BraA07g041560.3C and BraA07g041570.3C) were identified, revealing their key roles in regulating primary rosette branching in flowering Chinese cabbage through the gibberellin pathway [2]. BraA07g034950.3C regulated primary rosette branching in flowering Chinese cabbage through the IAA pathway and was localized within two QTL intervals on chromosome A07 [29]. This study identified a novel branching-related gene, BrTCP1, in flowering Chinese cabbage, which contains a TCP domain associated with branching function, and its function was validated through heterologous transformation in Arabidopsis.

In this study, we identified a gene, BrTCP1, containing a TCP domain, and found that its mutation significantly increased branching in flowering Chinese cabbage. This finding is consistent with the known role of the TCP gene family in regulating plant branching. TCP transcription factors, unique to plants, played diverse roles in plant development, encompassing branching, leaf shape, and floral organogenesis [30,31]. TCP transcription factors influenced plant branching architecture through the control of cell proliferation and differentiation. For example, TCP1 and TCP12 in Arabidopsis regulated branching by inhibiting axillary bud development [32]. BRC1a, a TCP gene in potato, affected branching number through the modulation of axillary bud growth [33]. Our findings suggested that BrTCP1 might regulate branching in flowering Chinese cabbage through a similar mechanism. The mutation of BrTCP1 likely weakened its inhibitory effect on axillary bud outgrowth, resulting in an increased branching number. Moreover, TCP genes exhibited interactions with other genes involved in branching. For example, in Arabidopsis, TCP genes and BRC1 cooperated to control branching [32]. Future studies could further investigate whether BrTCP1 exhibits similar interactions with BRC1 or other branching-related genes to comprehensively understand its role in regulating branching in flowering Chinese cabbage.

Here, we successfully identified the key gene BrTCP1 regulating branching traits in flowering Chinese cabbage using BSA-seq combined with SNP-index and ED analysis. BSA-seq is an efficient and cost-effective method capable of rapidly locating genomic regions associated with target traits [16]. By constructing extreme phenotype pools (high-branching and low-branching pools), we obtained genome-wide SNP information through high-throughput sequencing technology and screened candidate regions significantly associated with branching traits using SNP-index and ED analysis. The SNP-index method, which compares allele frequency differences between the two extreme pools, effectively identified loci related to the trait [21], while ED analysis further improved localization accuracy by calculating the genetic distance between the two pools [34]. To validate the BSA-seq results, we performed genotyping of SNPs within the candidate regions using the KASP method. KASP is a high-throughput, low-cost, and highly accurate genotyping technology, particularly suitable for large-scale SNP validation [35]. Through KASP analysis, we successfully confirmed the association between the BrTCP1 gene and branching traits in flowering Chinese cabbage and identified its key mutation site. This result not only validated the reliability of the BSA-seq method but also laid the foundation for subsequent functional validation and molecular mechanism studies. The gene localization approach employed in this study provided an efficient technical route for the genetic dissection of complex traits. The strategy of combining BSA-seq with KASP validation is not only applicable to the study of branching traits in flowering Chinese cabbage but can also be widely applied to the localization of important agronomic traits in other crops.

Recent studies have demonstrated that shoot branching is not solely regulated by genetic factors but also tightly controlled by a network of phytohormones, including auxin, SLs, and GAs. TCP transcription factors, particularly the TB1/BRC1 subgroup, serve as integrators of hormonal signals to modulate axillary bud outgrowth. In Arabidopsis, BRC1 expression is promoted by SLs and repressed by cytokinins, acting downstream of auxin-mediated SL biosynthesis to suppress bud activation [16]. Similarly, in rice, the SLR1-OsMADS23-D14 module mediates the crosstalk between SL and GA signaling to regulate tillering through OsTB1 [19]. In our study, BrTCP1, a homolog of TB1/BRC1, was identified as a key repressor of branching in flowering Chinese cabbage. Given the conserved role of TCP proteins, it is likely that BrTCP1 mediates its inhibitory effect on rosette branching through interactions with hormone signaling pathways. Notably, a previous study in flowering Chinese cabbage reported that exogenous GA₃ application inhibits rosette branching by altering auxin transport and suppressing cytokinin levels [5]. These findings suggest that BrTCP1 may function downstream of or in coordination with these hormone signals to repress axillary bud outgrowth. Further investigation into the expression dynamics of BrTCP1 under hormonal treatments and its regulatory network would provide deeper insight into its role in hormone-mediated branching control.

The identification of BrTCP1 as a branching regulator has practical implications for breeding programs. Manipulation of BrTCP1 alleles through marker-assisted selection or gene editing could enable the development of cultivars with optimized branching patterns tailored to different cultivation systems. For instance, low-branching lines may be preferred in high-density planting systems to reduce shading and improve airflow, whereas moderately branching lines could be selected for low-input or open-field conditions to increase yield potential. Exploring natural allelic variation in BrTCP1 in diverse germplasm collections may further accelerate breeding for ideal plant architecture.

5. Conclusions

This study identifies BrTCP1 as the key gene of primary rosette branching in flowering Chinese cabbage. Fine-mapping delimited the QTL to a 172 kb interval on chromosome A07, pinpointing BrTCP1. Sequence variations (6 bp insertion, 10 SNPs, two deletions) in BCT18 and higher BrTCP1 expression in low-branching CX010 suggest its role in suppressing branching. These findings provide molecular insights for Brassica crop breeding.