CmBr confers fruit bitterness under CPPU treatment in melon

Mingyan Wang; Naiyu Jiang; Yuanchao Xu; Xinxiu Chen; Cui Wang; Chuangjiang Wang; Shiqi Wang; Kuipeng Xu; Sen Chai; Qing Yu; Zhonghua Zhang; Huimin Zhang

doi:10.1111/pbi.14399

. 2024 May 30;22(10):2724–2737. doi: 10.1111/pbi.14399

CmBr confers fruit bitterness under CPPU treatment in melon

Mingyan Wang ^1,^{^†}, Naiyu Jiang ^1,^{^†}, Yuanchao Xu ^2,^{^†}, Xinxiu Chen ^1,^{^†}, Cui Wang ¹, Chuangjiang Wang ¹, Shiqi Wang ¹, Kuipeng Xu ¹, Sen Chai ¹, Qing Yu ¹, Zhonghua Zhang ^1,^✉, Huimin Zhang ^1,^✉

PMCID: PMC11536449 PMID: 38816932

Summary

Many biotic or abiotic factors such as CPPU (N‐(2‐chloro‐pyridin‐4‐yl)‐N′‐phenylurea), a growth regulator of numerous crops, can induce bitterness in cucurbits. In melon, cucurbitacin B is the major compound leading to bitterness. However, the molecular mechanism underlying CuB biosynthesis in response to different conditions remains unclear. Here, we identified a set of genes involved in CPPU‐induced CuB biosynthesis in melon fruit and proposed CmBr gene as the major regulator. Using CRISPR/Cas9 gene editing, we confirmed CmBr's role in regulating CuB biosynthesis under CPPU treatment. We further discovered a CPPU‐induced MYB‐related transcription factor, CmRSM1, which specifically binds to the Myb motif within the CmBr promoter and activates its expression. Moreover, we developed an introgression line by introducing the mutated Cmbr gene into an elite variety and eliminated CPPU‐induced bitterness, demonstrating its potential application in breeding. This study offers a valuable tool for breeding high‐quality non‐bitter melon varieties and provides new insights into the regulation of secondary metabolites under environmental stresses.

Keywords: CPPU, melon, bitterness, cucurbitacin, CmBr, RSM1

Introduction

Plant secondary metabolites are produced in various plants as a natural defence system against herbivores attacks and environmental stresses (Li et al., 2023; Tao et al., 2022; Thakur et al., 2019; Yamagishi, 2022; Yang et al., 2012). In the Cucurbitaceae family, cucurbitacin, a group of bitterness triterpenoid compounds, serves either as protection against herbivores and microbes or as Chinese herbal medicine with anti‐cancer effects (Luo et al., 2019; Wink, 2016). Cucurbitacin B is the major bitter triterpenoid compound in melon, while CuC and CuE are mainly found in cucumber and watermelon respectively (Horie et al., 2007). In our previous study in cucumber, we demonstrated a pathway involving nine genes, including an oxidosqualene cyclase (OSC), seven cytochromes P450 (CYPs) and an acyltransferase (ACT), responsible for CuC biosynthesis. Similarly, the CuB biosynthesis pathway in melon has been reported, involving an OSC (CmBi), six CYPs (Cm160, Cm170, Cm180, Cm710, Cm890 and Cm490) and an ACT (CmACT). Remarkably, four biosynthetic steps have been experimentally validated in the CuB biosynthesis pathway. In the initial committed step, CmBi catalyses the 2,3‐oxidosqualene, resulting in the formation of cucurbitadienol. Subsequently, Cm890 catalyses the next two steps to produce 11‐carbonyl‐20β‐hydroxy‐cucurbitadienol. Cm180 is responsible for introducing the C2‐hydroxyl to produce 11‐carbonyl‐2β,20β‐dihydroxy‐cucurbitadienol. In the last step, CmACT acetylates cucurbitacin D (CuD) at the C25‐hydroxyl position, generating the CuB product (Zhou et al., 2016). The genetic basis underlying the transcriptional regulation of cucurbitacin biosynthesis has been extensively studied. The CuC biosynthetic genes are regulated by two tissue‐specific basic helix–loop–helix (bHLH) transcription factors (TFs) named CsBl and CsBt in leaves and fruits respectively (Shang et al., 2014). In melon, it is also reported a bHLH cluster, containing two bHLH transcription factors CmBr and CmBt, is responsible for CuB biosynthesis which is located on chromosome 9. CmBr was found to be predominantly expressed in roots, suggesting its role in the regulation of CuB biosynthesis in roots (Zhou et al., 2016). In contrast, CmBt exhibited specific expression in fruit and is considered to regulate the biosynthesis of CuB in fruit (Zhao et al., 2019; Zhou et al., 2016). Previous studies have proposed that CmBt responds to CPPU, activating the expression of three cucurbitacin B biosynthetic genes (CmBi, Cm710 and CmACT), and subsequently promotes CuB accumulation (Luo et al., 2020). Recent studies have identified another two QTLs on chromosomes 2 and 5 regulating melon bitterness (Shang et al., 2020). The mechanisms of these genes regulating cucurbitacin biosynthesis remain largely unknown.

Widely consumed as vegetables and fruits, cucurbits have been domesticated from their wild progenitors that had extremely bitter fruit (Shang et al., 2014; Yuan et al., 2023; Zhao et al., 2019). Nonetheless, various environmental factors such as unsuitable temperature, excessive nitrogen, grafting and the use of plant growth regulators can also contribute to fruit bitterness in cultivars (Kano and Goto, 2003; Luo et al., 2020; Shang et al., 2014; Zhang et al., 2019). N‐(2‐chloro‐pyridin‐4‐yl)‐N′‐phenylurea (CPPU) is a synthetic cytokine‐like plant regulator that can promote fruit setting and enlargement (Matsuo et al., 2012). CPPU is initially applied in crops such as grapes (Nickell, 1986), pears (Banno, 1986), kiwifruit (Iwahori et al., 1988), apples (Ogata et al., 1989), watermelons (Hayata et al., 1995) and melon (Hayata et al., 2000). In melon, CPPU application become a prominent technique for promoting fruit setting, increasing yield and reducing the labour costs associated with artificial pollination (Cheng et al., 2021; Hayata et al., 2000; Huang et al., 2017; Li et al., 2021; Liu et al., 2023). However, the use of CPPU often leads to bitter‐tasting fruit, significantly impacting melon fruit quality and thereby restricting CPPU's regular application (Luo et al., 2020).

Previous studies have proposed that CmBt responds to CPPU, activating the expression of three cucurbitacin B biosynthetic genes (CmBi, Cm710 and CmACT), and subsequently promotes CuB accumulation (Luo et al., 2020). Although multiple genes have been reported involving in CuB biosynthetic pathway in melon, the mechanisms of CuB biosynthesis in response to CPPU and the breeding strategies for producing non‐bitter fruit in response to CPPU in production are still unclear. In this study, we investigated the dynamics of CuB and the profiles and transcriptome at four stages (3, 10, 20 and 49) day after anthesis (DAA) of melon fruits in response to CPPU. Through a combination analysis of CuB metabolite and transcriptome data, we constructed a regulatory network controlling CuB biosynthesis. We validated the role of CmBr in regulating CuB biosynthesis in response to CPPU by CRISPR‐Cas9‐mediated modification of CmBr, leading to reduced CuB content in melon fruits. CmBr functions by directly activating the expression of eight biosynthetic genes, including CmBi, Cm160, Cm170, Cm180, Cm710, Cm890, Cm490 and CmACT. We further identified a CPPU‐induced MYB‐related transcription factor CmRSM1, which directly binds to the Myb motif within the CmBr promoter, activating its expression and regulating CuB biosynthesis. Importantly, by employing marker‐assisted background selection, we generated introgression lines (ILs) harbouring nonfunctional CmBr alleles with a similar background from the donor parent of bitter‐tasting fruit. The IL‐Cmbr produced non‐bitter fruit even when subjected to CPPU treatment. This study sheds light on the regulation of plant secondary metabolites in response to environmental conditions and provides a breeding strategy for non‐bitter fruit production using CPPU.

Results

The regulatory network provides key transcription factors regulating CuB biosynthesis in response to CPPU

To investigate the impact of CPPU on cucurbitacin B accumulation, a high concentration (20 mg/L) of CPPU was sprayed onto ovaries at the day of anthesis, as previously reported (Luo et al., 2020). The dynamics of CuB content in both CPPU‐treated fruit (CPPU) and hand‐pollination control fruit (CK) at 3, 10, 20 and 49 days after anthesis (DAA) of agrestis variety were analysed (Figure 1a). The results showed that the CuB contents were consistently higher in CPPU‐treated fruit than control fruit throughout the corresponding developmental stages. Notably, CuB content significantly increased at 3 DAA, peaked at 10 DAA and gradually decreased during fruit development and maturation in both CPPU‐treated and control fruit (Figure 1b). At 49 DAA, when the fruit reaches maturity, the CPPU‐treated fruit exhibited a distinct bitter taste, while the control fruit remained non‐bitter. These findings indicate that CPPU induces CuB accumulation during the early stages of fruit development, ultimately resulting in the presence of a bitter taste in mature fruit.

Identification of *CmBr* as a key transcription factor regulating the CuB biosynthetic pathway in response to CPPU. (a) The whole melon fruits during four developmental and ripening periods (3, 10, 20 and 49 DAA) with CPPU treatment group (CPPU) and hand‐pollination group (CK). DAA, day after anthesis; CPPU, 20 mg/L CPPU treatment fruits; CK, hand‐pollination fruits; Scale bar = 2 cm. (b) CuB content of melon fruits during four developmental and ripening periods of CK and CPPU groups. Data are given as mean ± SEM (n = 3 biological replicates). **P < 0.01 (multiple t‐tests). (c) CuB biosynthetic pathway with the expression pattern of eight biosynthetic genes in response to CPPU during four developmental and ripening of melon fruit. *CmBi*, cyclase gene; *Cm160*, *Cm170*, *Cm180*, *Cm710*, *Cm890* and *Cm490*, cytochrome oxidase P450 genes; *CmACT*, acetyltransferase gene. Expression data were Z‐score standardized to −2 to 2.5 per gene. (d) A co‐expression transcriptional regulatory network in response to CPPU based on correlations between structural genes and transcription factors (TFs). Pearson correlation coefficient values (PCC) were calculated for each pair of genes. Pink circles represent structural genes involved in CuB metabolism during fruit development and ripening. Diamonds with different colours represent different families of transcription factors (PCC >0.8) in the same module. Strong correlations (PCC ≥0.9) are indicated by blue lines in the first circle, while moderate correlations (PCC from 0.8 to 0.9) are shown by grey lines in the second circle. (e) The relative fold change of bHLH transcription factors (PCC >0.8) in response to CPPU at 3 DAA. Data are given as mean ± SEM (n = 3 biological replicates). *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001 (one‐tailed t‐test).

To establish the regulatory network controlling CuB biosynthesis in melon fruit in response to CPPU, we conducted transcriptome analyses of CPPU‐treated fruit and control fruit at four development and ripening stages. A total of 12 376 genes were subjected to weighted gene co‐expression network analysis (WGCNA), showing 11 co‐expression modules in different expression patterns (Figure S1a). Interestingly, the previously reported CuB biosynthetic pathway (Zhou et al., 2016) was observed in the turquoise module. These genes encompass the cyclase gene CmBi, cytochrome oxidase P450 genes Cm160, Cm170, Cm180, Cm710, Cm890, Cm490 and the acetyltransferase gene CmACT. The expression levels of these biosynthetic genes were consistently higher in CPPU‐treated fruit compared to the control fruit throughout all stages of melon fruit development and ripening (Figure 1c). Remarkably, there was a significant increase especially at the 3 DAA stage in CPPU‐treated fruit (Figure 1c). This suggests that the 3 DAA is the crucial stage in the CPPU‐induced CuB biosynthesis.

To investigate potential transcription factors involved in regulating CuB metabolism in response to CPPU, we identified 49 transcription factors, including bHLH (10), MYB (4) and C2H2 (3) etc., that showed strong correlations (Pearson correlation coefficient values, PCC >0.8) with the eight structural genes of CuB in the turquoise module via regulatory network analysis (Figures 1d and S1b). It is well known that bHLH family transcription factors play essential roles in regulating the cucurbitacin biosynthesis in cucurbit crops (Shang et al., 2014; Zhou et al., 2016). Therefore, we focused on bHLH factors that exhibited strong correlations with the eight structural genes of CuB (PCC >0.9). At the 3 DAA stage, CmBr showed a remarkable 22‐fold up‐regulation in CPPU‐treated fruit compared to the control fruit, surpassing the up‐regulation observed for other bHLH factors (Figure 1e). Furthermore, we examined the expression pattern of CmBr in different tissues and found high expression levels in young fruit (DAF 0 d, 2 d and 4 d), as well as in the root, while it exhibited low expression levels in the stem and leaf (Figure S2) (Yano et al., 2018). These findings collectively suggest that CmBr may play an important role in the early‐stage CuB accumulation during fruit development in response to CPPU.

CmBr regulates the CuB biosynthesis in fruit induced by CPPU

To explore the role of CmBr in CPPU‐induced CuB biosynthesis, we generated loss‐of‐function mutants of Cmbr. We have established an efficient genetic transformation system using the P147 accession and generated T₀ transgenic plants expressing sgRNA^CmBr in the P147 background using CRISPR/Cas9 gene editing system. However, P147 produces non‐bitter fruits even when exposed to high concentrations (20 mg/L) of CPPU treatment (Table S1), which cannot be used for bitterness investigation. To investigate the role of CmBr in fruit bitterness, we performed crosses between P147‐CmBr ^CR T₀ transgenic plants and ivf05, an elite cultivar known for developing fruit bitterness when exposed to high concentrations (20 mg/L) of CPPU (Table S1). Using an InDel marker, we screened homozygous Cmbr mutant (Cmbr‐1) with an 8‐bp deletion at the target site in the resulting F₁ plants carrying the Cas9 transgenic fragment (Figures 2a and S3a). The 8‐bp deletion caused a frameshift mutation and premature translation termination (Figure 2b). In parallel, we conducted crosses between P147 and ivf05, and obtained wild‐type F₁ plants (WT) that served as control plants (Figures 2a and S3a).

CRISPR‐Cas9 engineered *CmBr* mutations result in reduced CuB content in response to CPPU. (a) Schematic representation of *Cmbr* mutant of F₁ progeny crossing ivf05 and *Cmbr*‐edited line in P147 background (*Cmbr*‐1), wild type of F₁ progeny crossing ivf05 and P147 (WT) as the control. (b) Position of the target site in *CmBr* and Sanger sequencing of the CRISPR‐Cas9 edited sites in *Cmbr*‐1, WT as the control. SgRNA‐targeted sequences are highlighted in red and a protospacer adjacent motif (PAM) is shown in blue. Deletion and insertion are represented by dashed line and in grey background. (c) Melon fruits of WT and *Cmbr*‐1 mutant in response to CPPU at 3 DAA. Bar = 2 cm. (d) The CuB contents of melon fruits of WT and *Cmbr*‐1 mutant in response to CPPU at 3 DAA. Data are given as mean ± SEM (n ≥ 18). Different letters above the bars indicate a significant difference (P < 0.05) obtained by the one‐way ANOVA test.

We analysed CuB content of young fruits (3 DAA) treated with CPPU in Cmbr mutants (Cmbr‐1) and wild‐type plants (WT) using liquid chromatography mass spectrometry (LC–MS) (Figure 2c). In WT plants, CPPU‐treated young fruits exhibited significantly higher CuB content compared to hand‐pollinated control fruits. However, in Cmbr‐1 mutants, CPPU‐treated fruits did not show a CuB content increase compared to control fruits (Figure 2d). Additionally, CuB content in CPPU‐treated fruits of Cmbr‐1 mutants decreased by 86.4% relative to CPPU‐treated fruits in WT plants (Figure 2d). Considering that CPPU increases CuB content and induces bitterness in mature fruits, we further assessed the occurrence of bitter mature fruits treated with CPPU. The proportion of bitter maturity fruits induced by CPPU was 8.96% in the Cmbr‐1 mutants, which was significantly lower than the 30% observed in the WT plants (Figures S3b and S3c). These results suggested that CmBr plays an essential role in regulating CuB biosynthesis in melon fruit in response to CPPU.

IL‐Cmbr exhibits non‐bitter fruit with high quality induced by CPPU during melon production

To further verify the role of CmBr in CPPU‐induced CuB biosynthesis, we generated another type of mutant of Cmbr by constructing introgression lines (ILs) and evaluated their fruit bitterness when treated with CPPU. We conducted three generations of backcrossing between P147‐CmBr ^CR T₀ transgenic plants and the recurrent parent ivf05 (Figure S4a). From the Cas9‐free plants of BC₃ population, the V‐12 individual was selected using the whole‐genome array MELON2K, showing the highest background similarity of 81.8% with ivf05. This individual carried at least four chromosome introgressions (chr02, chr05, chr09 and chr11), wherein the known bitterness genes are located (Figure S4b). Subsequently, we generated the BC₃S₂ population and obtained another type of mutant ivf05^IL(Cmbr‐2) containing homozygous Cmbr allele (CmBr has an in‐frame 46 bp deletion and is a complete loss‐of‐function allele), and ivf05^IL(CmBr) individuals with homozygous CmBr allele as control (Figures 3a and S5). Furthermore, no mutations were detected at potential off‐target positions in ivf05^IL(Cmbr‐2) mutant, as determined by Sanger sequencing (Figure S6).

We further tested the fruit bitterness in ivf05^IL(Cmbr‐2) mutant treated with CPPU by conducting field experiments. Measurement of CuB content in young fruit at 3 DAA in response to CPPU revealed a significant increase in ivf05^IL(CmBr) lines, while no significant change was observed in ivf05^IL(Cmbr‐2) lines (Figure 3b,c). Particularly, CuB content in ivf05^IL(CmBr) reached approximately 24 μg/g, while it was nearly undetectable in ivf05^IL(Cmbr‐2) mutant lines, resulting in a non‐bitter taste (Figure 3c).

To further evaluate the utility of the CmBr gene in melon breeding for producing non‐bitter fruit when treated with CPPU, we also assessed the presence of bitter‐tasting ripe fruits at 35 DAA after an application of a high CPPU concentration through field experiments (Figure 3d,e). The results revealed that in the ivf05^IL(CmBr) lines, 51.65% of the fruits displayed bitterness induced by CPPU. In contrast, in ivf05^IL(Cmbr‐2) mutant lines, the percentage of bitter fruit significantly decreased to 4.17% (Figure 3e). Thus, the nonfunctional allele of CmBr effectively eliminates fruit bitterness when CPPU is used in melon production.

Considering that melon fruits are primarily grown for commercial purposes, we further evaluated the commercial viability of the ivf05^IL(Cmbr‐2) lines (Figure S7). We analysed important fruit traits, including fruit weight, fruit length, fruit width and soluble solids content (SSC). There were no significant differences in fruit weight between the ivf05^IL(CmBr) lines and the ivf05^IL(Cmbr‐2) lines (Figure 3f). Interestingly, CPPU‐induced fruits exhibited slightly higher fruit weight than hand‐pollinated fruits (Figure 3f), confirming CPPU's role in enhancing fruit yield. Moreover, evaluation of the quality trait of soluble solid content (SSC) revealed no differences in flesh or placenta between the ivf05^IL(CmBr) lines and the ivf05^IL(Cmbr‐2) lines (Figure 3g). Consequently, the nonfunctional allele of CmBr has no apparent impact on fruit weight and quality when CPPU is employed in melon production.

CmBr activates the expression of CuB biosynthetic pathway genes in response to CPPU

To illustrate the mechanism of CmBr regulates CuB biosynthesis in melon fruits, we performed transcriptome analyses of melon fruits treated with CPPU at 3 DAA in both the ivf05^IL(CmBr) lines and ivf05^IL(Cmbr‐2) lines (Figure 3b). Moreover, 788 and 893 differentially expressed genes (DEGs) were obtained between CPPU‐treated and control fruits in ivf05^IL(CmBr) and ivf05^IL(Cmbr‐2) respectively (Figure 4a). We focused on genes exclusively upregulated by CPPU in the ivf05^IL(CmBr) lines but not in the ivf05^IL(Cmbr‐2) lines. These genes potentially respond to CmBr regulation and remain inactive when CmBr is nonfunctional. We subsequently analysed these genes using Gene Ontology enrichment analysis and the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses, which revealed significant enrichment in the ‘Biosynthesis of secondary metabolites’ and ‘Diterpenoid biosynthesis’ pathways (Figure 4b,c). Moreover, eight CuB biosynthetic genes (the cyclase gene CmBi, the cytochrome oxidase P450 genes Cm160, Cm170, Cm180, Cm710, Cm890 and Cm490, and the acetyltransferase gene CmACT) consistently exhibited higher expression induced by CPPU in ivf05^IL(CmBr) lines, while remaining unresponsive in ivf05^IL(Cmbr‐2) lines (Figure 4d). These results indicate that CmBr activates the expression of eight biosynthetic genes (CmBi, Cm160, Cm170, Cm180, Cm710, Cm890, Cm490 and CmACT), and thereby positively regulating CuB accumulation in melon fruits in response to CPPU.

Analysis of differentially expressed genes (DEGs) between ivf05^IL(*CmBr*) and ivf05^IL(*Cmbr*‐2) mutant in response to CPPU. (a) Venn diagram showing the DEGs of melon fruits between ivf05^IL(*CmBr*) and ivf05^IL(*Cmbr*‐2) mutant induced by CPPU respectively. Red number represents DEGs in ivf05^IL(*CmBr*) induced by CPPU but not in ivf05^IL(*Cmbr*‐2) mutant. (b, c) Gene Ontology‐biological process (GO‐BP) enrichment analysis (b) and the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis (c) of DEGs in ivf05^IL(*CmBr*) and ivf05^IL(*Cmbr*‐2) mutant induced by CPPU respectively. (d) Expression levels of CuB biosynthetic pathway genes of melon fruits in ivf05^IL(*CmBr*) and ivf05^IL(*Cmbr*‐2) mutant in response to CPPU. Data are given as mean ± SEM (n = 3 biological replicates). **P < 0.01; ***P < 0.001 (one‐tailed t‐test).

CPPU induces MYB factor CmRSM1 to regulate the transcription of CmBr

To investigate how CPPU induces the expression of CmBr, Analysis of the 2 kb promoter region of CmBr using the PlantCare program revealed four MYB‐related cis‐elements, including two Myb motif ((T/C)AACTG), one MRE motif (AACCTAA), one MYB motif (CAACCA), which are known as potential MYB protein‐binding cis‐elements (Mu et al., 2009) (Figure S8a). MYB transcription factors play important roles in terpenoid metabolism (Li et al., 2022). In the turquoise module derived from regulatory network analysis, we found the MYB factor CmRSM1 that exhibited strong correlations with CmBr (PCC >0.9) (Figure S8b). CmRSM1 belongs to the MYB‐related subfamily and contains a single SANT/MYB DNA‐binding domain which consists of four members in Arabidopsis: RADIALIS‐LIKE SANT/MYB 1 (RSM1), RSM2, RSM3 and RSM4 (Figure S8c). We further examined the expression pattern of CmRSM1 in different tissues and observed high expression levels in young fruit (DAF 0 d, 2 d, 4 d), while displaying low expression levels in the root, stem and leaf (Figure S8d) (Yano et al., 2018).

To further confirm the function of CmRSM1, we overexpressed CmRSM1 in ‘ivf05’ fruit at 10 DAA by using a transient infiltration expression system (Figure 5a). The infiltrated fruits with CmRSM1 (CmRSM1‐INF) exhibited higher CuB content (Figure 5b). We also confirmed the function of CmRSM1 using a hairy root transgenic system in melon. CmRSM1‐overexpressing (CmRSM1‐OE) transgenic hairy roots also showed increased CuB content (Figure S9a,b). Furthermore, qRT‐PCR analysis revealed that overexpression of CmRSM1 activated the expression of the target gene CmBr and the biosynthetic gene CmBi (Figures 5c and S9c). Yeast one‐hybrid (Y1H) assay demonstrated that CmRSM1 can bind to the promoter of CmBr (Figure 5d). The tobacco transient reporter (luciferase) activation system demonstrated that CmRSM1 activated the expression of CmBr (Figure 5e,f). Electrophoretic mobility shift assay (EMSA) further revealed that CmRSM1 specifically bound to the P1 probe (Myb motif) of CmBr promoter (Figure 5g), while it did not bind to the P2 probe (MYB, MBS motifs) and P3 probe (MRE motif) of CmBr promoter (Figure S10). These findings suggest that CmRSM1 specifically binds to the Myb motif and activates the expression of CmBr, playing a crucial role in CPPU‐induced CuB biosynthesis in melon fruit.

CPPU‐induced *CmRSM1* promotes *CmBr* expression to regulate CuB biosynthesis by binding to Myb motif in the *CmBr* promoter. (a) Schematic diagram of transient infiltration expression system of melon fruit. (b) Relative CuB content of melon fruits detected 8 days post‐infiltration. *CmRSM1*‐INF, sample infiltrated with *CmRSM1*; INF, sample infiltrated with empty vector. Data are given as mean ± SEM (n = 3 biological replicates). *P < 0.05 (one‐tailed t‐test). (c) Expression level of *CmRSM1*, *CmBr* and *CmBi* determined 8 days post‐infiltration. Data are given as mean ± SEM (n = 3 biological replicates). *P < 0.05 (one‐tailed t‐test). (d) Yeast one‐hybrid (Y1H) assay showed that *CmRSM1* binds the promoter of *CmBr*. Yeast cells were co‐transformed with *CmRSM1* and *CmBr* promoter. SD/−Leu, Simple Dropout medium lacking leucine. AbA, Aureobasidin A with different concentrations of 100, 300 and 500 mM. Spots from left to right in each row indicate the gradient dilution (1, 10⁻¹, 10⁻² and 10⁻³) of yeast clones. (e) Schematic diagram of the constructs using dual‐luciferase (LUC) analysis in a tobacco transient expression system. Effector constructs contained coding sequence of *CmRSM1*. Reporter constructs contained 2 kb of the promoters of *CmBr* upstream of the translation initiation sites, fused to the LUC reporter gene. Renilla luciferase (REN) was used as the internal control. (f) The activation effects of *CmRSM1* on the *CmBr* promoter. Effects of *CmRSM1* on the CmBr promoter are presented as a ratio of LUC to REN. Data are given as mean ± SEM (n = 4 biological replicates). *P < 0.05 (one‐tailed t‐test). (g) EMSA showing specific binding of CmRSM1‐GST to the P1 probe (Myb motif) within the *CmBr* promoter. P1, P2 and P3 probes were marked by red lines representing Myb, MYB, MBS and MRE motif in the promoter of *CmBr* respectively. GST, GST‐tag; Competitor probe (unlabelled probe); +/−, presence or absence of protein or probe; 5×, 10×, 20× represents the rates of competitor probes.

Discussion

Cucurbitacin plays an important role in mediating plant and biotic or abiotic factors interaction. CPPU, a synthetic cytokine‐like plant regulator, can lead to fruit bitterness, adversely affecting fruit quality. However, the genetic mechanisms underlying cucurbitacin B biosynthesis in melon fruit and the breeding strategies to obtain non‐bitter fruit in response to CPPU remain poorly understood. Here, we applied a comprehensive analysis of CuB dynamics and transcriptome profiles at four stages (3, 10, 20 and 49 DAA), generating a regulatory network controlling CuB accumulation in CPPU‐induced fruits (Figure 1a–d). These networks provide valuable insights into the genetic basis of metabolic pathways and guide future breeding strategies for improving fruit quality (Li et al., 2020; Wang et al., 2022; Yang et al., 2023).

According to our datasets, CmBr displayed a significant 22‐fold up‐regulation in CPPU‐treated fruit at the 3 DAA stage, demonstrating higher responsiveness to CPPU than other bHLH factors (Figure 1e). In a previous study, CmBt was identified as a regulator of CuB biosynthesis in melon fruits (Zhou et al., 2016). It was induced by CPPU to promote CuB accumulation by activating its biosynthetic genes (CmBi, CmACT, Cm710 and Cm890) through transient expression assays (Luo et al., 2020). However, our study revealed that CmBr exhibited an approximately 22‐fold up‐regulation in CPPU‐treated fruits, which was significantly higher than CmBt (Figure 1e). To explore the role of CmBr in CPPU‐induced CuB biosynthesis, we demonstrated that the CuB content in the Cmbr‐1 mutant exhibited a 70.3% decrease compared to the WT plants in hand‐pollinated melon fruits, suggesting the involvement of CmBr in regulating CuB levels. Upon CPPU treatment, the Cmbr‐1 mutant exhibited an 86.4% reduction in CuB content relative to the WT plants (Figure 2c,d). These results indicate that CmBr plays a crucial role as a transcription factor in regulating CuB biosynthesis in melon fruit in response to CPPU. Since both CmBr and CmBt regulate CPPU‐induced CuB biosynthesis and have been found to directly bind to CuB biosynthesis genes (Zhou et al., 2016), our investigation revealed that in ivf05^IL(Cmbr‐2) mutant, CmBt did not show significant changes (Figure S11). This suggests that CmBr may not transcriptionally regulate CmBt expression. CmBr and CmBt encode bHLH family transcription factors, and it is widely recognized that bHLH family transcription factors commonly function by forming heterodimeric complexes (Lei et al., 2020). Therefore, it is worth further investigating whether CmBr interacts with CmBt.

Previous studies have examined CmBr was expressed in roots, regulating CuB biosynthesis. In our study, we further examined the expression pattern of CmBr in various tissues, including young fruit (DAF 0 d, 2 d and 4 d), root, low stem, high stem, old leaf and young leaf. We observed high expression levels of CmBr in young fruit and root, whereas stem and leaf exhibited lower expression levels (Figure S2). Consistent with these findings, our analysis of CuB content of ILs in roots revealed higher CuB content in ivf05^IL(CmBr) compared to the ivf05^IL(Cmbr‐2) mutant (Figure S7e). These results are consistent with the previously reported role of CmBr in CuB biosynthesis in roots (Zhou et al., 2016).

To reveal the mechanism of CmBr regulate CuB biosynthetic pathway genes of melon fruit, transcriptome analyses of melon fruit at 3 DAA after CPPU treatment in both the ivf05^IL(CmBr) lines and ivf05^IL(Cmbr‐2) lines indicated that eight genes consistently remained unresponsive in the ivf05^IL(Cmbr‐2) mutant lines (Figure 4d). Previous studies have also reported that CmBr has the capacity to directly activate the transcription of co‐expressed genes responsible for CuB biosynthesis in roots (Zhou et al., 2016). These results suggest that CmBr can bind to the E‐box of the promoters of eight biosynthetic genes (CmBi, Cm160, Cm170, Cm180, Cm710, Cm890, Cm490 and CmACT), activating their expression and thereby promoting CuB accumulation in melon fruits.

MYB transcription factors also play important roles in terpenoid metabolism (Li et al., 2022). For instance, AtMYB21 and AtMYB24 regulate JA‐signalling mediated terpenoid biosynthesis by interacting with AtMYC2 and controlling AtTPS14 and AtTPS21 promoters (Song et al., 2011; Yang et al., 2020). FhMYB21L1 and FhMYB21L2 collaborate to negatively modulate FhTPS1 expression in Freesia hybrida, regulating monoterpene linalool biosynthesis (Yang et al., 2020). CsMYB147 (AtMYB21 homologues) and CsMYB68 respond to MeJA treatment, which is likely involved in volatile terpenoid synthesis in tea plants (Li et al., 2022). SlMYB75 influences volatile aroma compound accumulation in tomato fruits through modulating LOXC, AADC2 and TPS genes (Jian et al., 2019). It also interacts with SlMYB52 and SlTHM1 to control SlTPS12, SlTPS31 and SlTPS35, which catalyse sesquiterpene synthesis (Gong et al., 2021). VvMYB5b overexpression in tomatoes decreases phenylpropanoid while increasing terpenoid metabolism (Mahjoub et al., 2009). DoMYB26, DoMYB29 and DoMYB31 directly regulate sesquiterpene synthase gene DoECS, responsible for (E)‐β‐caryophyllene sesquiterpenoid biosynthesis in Dendrobium officinale (Lv et al., 2022). SmMYB9b and SmMYB98 promote tanshinone production in Salvia miltiorrhiza hairy roots by activating MEP pathway genes like SmGGPPS (Hao et al., 2020; Zhang et al., 2017). MYB‐related subfamily RADIALIS‐LIKE SANT/MYB (RSM) contains a single SANT/MYB DNA‐binding domain, consists of four members in Arabidopsis: RADIALIS‐LIKE SANT/MYB 1 (RSM1), RSM2, RSM3 and RSM4 (Hamaguchi et al., 2008). Arabidopsis MATERNAL EFFECT EMBRYO ARREST 3 (MEE3) /RSM1 represses the floral transition by activating transcription of Flowering Locus C (FLC) (Li et al., 2015). The expression of MEE3 is induced by the cytokinin BA treatment and upregulated in a strong freezing tolerant mutant esk1 (Lee et al., 2007; Xin et al., 2007). RSM1 also interacts with HY5/HYH to bind to ABI5 promoter and regulate seed germination and seedling development in response to ABA and salinity in Arabidopsis (Yang et al., 2018). Despite these diverse functions, MYB‐related family transcription factors RSM have not been previously associated with terpenoid regulation. In our study, we identified that CmRSM1, a homologue of Arabidopsis RSM1, is induced by the synthetic cytokine‐like CPPU, which is consistent with previous research indicating that the expression of MEE3 is induced by cytokinin BA treatment (Lee et al., 2007; Xin et al., 2007). Biochemical and biological evidence demonstrated that CmRSM1 directly binds to the Myb motif within the promoter of CmBr, activating its expression, and thereby regulating CuB biosynthesis under CPPU treatment (Figure 5a–g). These findings indicate that CmRSM1, a member of the MYB‐related family, may directly activate CmBr and be involved in terpenoid metabolism.

With the knowledge that CmBr plays an essential role in modulating CuB biosynthesis in melon fruits under CPPU stress (Figure 2d), there is potential to engineer non‐bitter melon crops exposed to CPPU‐induced stress. In our study, we constructed ILs by introducing the Cmbr mutant allele from the donor parent transformed T₀ plants (P147‐CmBr ^CR) into the ivf05, which can produce bitter fruit treated with high concentration of CPPU (Figure S4a). Since bitterness genes have been found on four chromosomes (chr02, chr05, chr09 and chr11), we employed marker‐assisted background selection and identified the BC3‐V‐12 individual, which exhibited the highest background similarity of 81.8% with ivf05, encompassing at least four chromosome introgressions (Figure S4b). In our field experiment, we conducted an assessment of fruit bitterness in ILs of ivf05^IL(CmBr) lines and ivf05^IL(Cmbr‐2) mutant lines. The results showed that the percentage of bitter fruit reached 51.65% in the ivf05^IL(CmBr) lines when subjected to a high concentration of CPPU, but significantly decreased to 4.17% in the ivf05^IL(Cmbr‐2) mutant lines (Figure 3e). These findings indicate that the nonfunctional CmBr allele effectively eliminates fruit bitterness when CPPU is employed in melon production. Furthermore, we observed no significant differences in fruit weight, soluble solid content (SSC) of flesh or placenta between the ivf05^IL(CmBr) lines and the ivf05^IL(Cmbr‐2) lines (Figure 3f,g). It is noteworthy that there was a slight increase in the fruit shape index (Figure S7b–d). This phenomenon may be attributed to the influence of CmBr on fruit development genes, as well as the remaining background from the donor parent. Consequently, the nonfunctional CmBr allele does not have a noticeable impact on fruit weight, length, width and quality when CPPU is applied in melon production. In conclusion, we have effectively demonstrated the potential to improve melon crops by reducing CPPU‐induced fruit bitterness through CmBr modification. This strategy may have practical utility in breeding high‐quality, non‐bitter melon varieties and addressing the issue of CPPU‐induced fruit bitterness in melon production.

Materials and methods

Plant materials and growth conditions

Seeds of the ‘38’, ‘ivf05’ and ‘P147’ inbred lines were sourced from Huaisong Wang at the Institute of Vegetables and Flowers, Chinese Academy of Agricultural Sciences and Guangwei Zhao at the Zhengzhou Fruit Research Institute, Chinese Academy of Agricultural Sciences respectively. The seedlings were cultivated in a greenhouse located at the Jimo Agricultural High‐tech Industrial Development Zone, Qingdao, Shangdong, China, from 2020 to 2023.

CPPU treatment and sample preparation

CPPU (Yuanye, S18139) was applied at a concentration of 20 mg/L, which was sprayed onto the ovaries of the biosexual flowers of ‘38’ wild agrestis inbred lines on the day of anthesis at the 15th to 20th internode position of lateral branches (Luo et al., 2020). Hand‐pollinated biosexual flowers served as the control. The CPPU‐treated fruits and control fruits were collected at different developmental and ripening stages of 3, 10, 20 and 49 days after anthesis (DAA). Pericarp samples (without skin and seeds) were cut into small pieces, flash‐frozen in liquid nitrogen and stored at −80 °C for subsequent CuB measurement and transcriptomic analysis. All experiments were conducted with three biological replicates, each consisting of three fruits.

UPLC‐MS quantification of cucurbitacin B

The sample material was ground into powder using a grinder in liquid nitrogen. Weighing 100 mg of the finely ground powder, it was then mixed with 1 mL of methanol. The mixture was vortexed and incubated in an ultrasonic bath for 15 min, followed by centrifugation at 12000 rpm for 10 min. The supernatant was collected, diluted 1:100 in methanol and filtered through a 0.22 μm membrane before injection. Chromatography was conducted using a Waters ACQUITY UPLC I‐Class‐Xevo TQ‐S Micro system, equipped with a BEH C18 column (2.1 mm × 50 mm i.d., 1.7 μm). The mobile phase consisted of methanol (solvent A) and water with 0.1% formic acid (0.1% formic acid as mobile phase modifier, v/v, solvent B). The flow rate was set to 0.2 mL/min, and the injection volume was 1 μL. The UPLC was coupled with an electrospray ionization and a hybrid triple quadrupole (QQQ) mass spectrometer (model: Xevo TQ‐S Micro, Waters). Mass acquisition was performed in positive ionization and multiple reaction monitoring modes. The ion pairs used for the quantitative analysis of CuB were 576.0/499.0. The sheath gas was set to 350 °C (650 L/h), and the capillary voltage was fixed at 3.4 kV. The CuB content was quantified using an authentic standard.

Transcriptome data analyses

The high‐quality RNA was separately extracted from CPPU‐treated fruits and control fruits (hand‐pollination) at four developmental and ripening stages (3, 10, 20 and 49 days), with three biological replicates. Libraries were constructed according to the protocol for the Illumina HiSeq4000 platform (Annoroad Gene Technology). All clean RNA‐seq reads from each sample were mapped onto the reference genome sequences (Castanera et al., 2020), using Hisat2 (version 2.1.0) with default parameters (Kim et al., 2015). The alignment results were converted to BAM files using SAMtools (1.11). The generated BAM format alignments, together with the gene GFF annotation file, were then fed to StringTie (v1.3.4d) software to compute FPKM (the fragments per kilobase of exon model per million reads mapped) values of genes (Pertea et al., 2015).

Construction of co‐expression transcriptional regulatory network analysis of CuB biosynthetic genes and transcription factor

Co‐expression and co‐regulation analyses were performed on samples collected from eight different time points and treatments using the R package WGCNA (v1.7.1). All eight biosynthetic genes of CuB were found within the turquoise module. The Pearson's correlation algorithm was employed to construct a regulatory network between TF‐related genes and biosynthetic genes of CuB in the turquoise module. The R software was used to calculate mutual information for assessing the expression level similarities between TFs and biosynthetic genes of CuB. The associations among TFs and biosynthetic genes of CuB were visualized using Cytoscape software (v3.9.1; https://cytoscape.org/).

Reverse transcription quantitative PCR analyses

CPPU‐treated fruits and control fruits at 3 DAA were collected for RT‐qPCR analysis. Total RNA extraction was isolated using the Quick RNA isolation Kit (Huayueyang, Cat. No. 0416GK). First‐strand cDNA synthesis was performed using 1 μg of total RNA with a GoScript™ Reverse Transcription Mix (Promega, Cat. No. A2790). The quantitative PCR assays were conducted on a CFX96™ Real‐Time System (BIO‐RAD) using SYBR Premix (Roche, Cat. No. 04707516001), following the manufacturer's instructions. Three independent biological replicates were performed, with each replicate consisting of a mixture of three to six fruits. The relative gene expression was determined using the comparative 2^−∆∆Ct method proposed by Schmittgen and Livak (2008), with CmActin2 (MELO3C008032.2.1) serving as the internal reference gene. Primer design was conducted using Primer Premier 5 software (Premier Biosoft, Palo Alto, CA, USA), and the primer sequences are provided in Table S2.

CRISPR/Cas9 vector construction

The single guide RNA (sgRNA) for CmBr was designed using the CRISPR‐GE (Genome Editing) website (http://skl.scau.edu.cn/targetdesign/). The sgRNA primer was annealed and then assembled into the pBSE402 vector through the Golden Gate cloning method (Xin et al., 2022). The resulting clones pBSE401‐sgRNA‐CmBr were confirmed through Sanger sequencing, and transformed into Agrobacterium tumefaciens strain EHA105 for melon transformation. The primers are shown in Table S2.

Melon transformation

Melon transformation was performed using the previously described method (Hu et al., 2017; Liu et al., 2023; Xin et al., 2022). Agrobacterium tumefaciens strain EHA105 carrying the pBSE401‐sgRNA‐CmBr vectors was cultured in liquid LB medium supplemented with 50 mg/L kanamycin and 25 mg/L rifampicin at 28 °C until reaching an optical density (OD600) of 0.4–0.8. The cell culture was then collected by centrifugation, and the resulting pellet was diluted to an OD600 of 0.2 in an inoculation medium (4.43 g/L Murashige and Skoog (MS), 0.5 g/L 2‐Morpholinoethanesulphonic acid monohydrate, 2 mg/L 6‐Benzylaminopurine (6‐BA), 80 mg/L Acetosyringone (AS), 1 mg/L ABA, 30 g/L sucrose). The seeds of the cultivated melo inbred line P147 were peeled and sterilized, then germinated in darkness at 28 °C for 1–2 days. Subsequently, the cotyledons were prepared by removing the hypocotyl and cutting them in half transversely. The proximal part of the cotyledon, including the U‐shaped end, was utilized as the explant. The explants were immersed in an Agrobacterium suspension and sonicated using an ultrasonic cleaning instrument (SB‐5200DTD). Following that, the explants were placed into a 20 mL syringe containing 15 mL of the Agrobacterium suspension. Vacuum infiltration was then performed by gently pulling up on the plunger to ensure comprehensive penetration of the suspension into the explants.

After infection, the explants were co‐cultured with Agrobacterium on three layers of moist filter papers placed on solid medium B (4.43 g/L MS, 3.3 g/L Gelzan, 1 mg/L ABA, 0.5 g/L MES, 80 mg/L AS, 2 mg/L 6‐BA, 150 mg/L dithiothreitol (DTT) and 30 g/L sucrose) for 72 h in the dark. Furthermore, the explants were rinsed with sterile water and transferred to solid medium C (4.43 g/L MS, 1 mg/L ABA, 3.3 g/L Gelzan, 200 mg/L Timentin, 30 g/L sucrose, 0.5 mg/L 6‐BA, 2 mg/L Basta) for further development. Shoots started to emerge within 3–4 weeks, and the transformed buds were selected by screening for GFP (Green Fluorescent Protein) fluorescence. The shoots were then transferred to rooting medium (4.43 g/L MS, 15 g/L sucrose, 9.4 g/L Agar, 1 mL/L (v/v) Plant Preservative Mixture (PPM), 200 mg/L Timentin) to induce root formation. Successfully rooted shoots were subsequently transplanted into soil for further growth.

Detection of targeted mutation analysis

Genomic DNA was extracted from transformed plants using cetyltrimethylammonium bromide (CTAB) method. The PCR fragments were amplified from genomic DNA using KOD‐Plus‐Neo (TOYOBO) and analysed on a 1.5% agarose gel. The gene‐specific primers are provided in Table S2. Potential off‐target sites were identified using the CRISPR website (http://crispor.tefor.net/). The gene information of potential off‐targets is listed in Table S3. The primers used are listed in Table S2.

Construction and whole‐genome background selection of ILs of Cmbr

The melon ILs ivf05^IL(CmBr) and ivf05^IL(Cmbr‐2) were developed from two varieties ivf05 and P147(CmBr ^CR). Ivf05, a cultivated agrestic variety, exhibiting bitterness induced by 20 mg/L of CPPU, was used as the recipient due to its functional CmBr allele. While P147(CmBr ^CR), a Cmbr mutant containing Cas9, was used as the donor. A cross was performed between ivf05 and P147(CmBr ^CR), following a recurrent backcross procedure consisting of three generations of backcrosses and two generations of selfing. In the backcrossing generations (BC₁ to BC₃), selected individuals heterozygous for CmBr were backcrossed to ivf05.

In BC₃, we screened Cas9‐free individuals by using PCR or fluorescence microscopy to identify the absence of the Cas9 or GFP linkaged with Cas9 (Hu et al., 2017). Then we performed background selection on the recombinants that were heterozygous for CmBr using whole‐genome array MELON2K. We selected background with four chromosome substitutions consisting of chr02, chr05, chr09 and chr11 where the known bitterness genes are located. Subsequently, we developed the BC₃S₂ population by selecting an individual from BC₃ with the highest proportion of genome recovery from the sequence of the recurrent parent Within the BC₃S₂ population, we successfully obtained ivf05^IL(Cmbr‐2) individuals with a homozygous Cmbr allele and ivf05^IL(CmBr) individuals with a homozygous CmBr allele.

Identification, multiple alignments and phylogenetic analysis of RSM genes

The information and sequences of A. thaliana RSMs were retrieved from the TAIR database (https://www.arabidopsis.org). Melon genome sequence data were obtained from the Melonomics database (www.melonomics.net). A BLASTP search of A. thaliana RSMs was performed against the melon database. Melon proteins with more than 50% identity were further predicted from InterPro (https://www.ebi.ac.uk/interpro/) and used as queries to search the SANT/Myb domains from HMMER (version 3.1b2) software (http://hmmer.janelia.org) and the SMART database (http://smart.embl‐heidelberg.de).

Multiple sequence alignments of identified RSM domains were carried out by muscle software with default parameters (http://www.drive5.com/muscle/). The conserved motifs of RSM proteins were determined by MEME (http://meme‐suite.org/tools/meme) and the NCBI database (https://www.ncbi.nlm.nih.gov/cdd). Based on the result of multiple sequence alignment, a neighbour‐joining tree was constructed using MEGA X (Kumar et al., 2018) using a bootstrap test with 1000 replicates based on ‘Jones–Taylor–Thornton(JTT)’ model and 80% partial deletion for gap treatment.

Transient agrobacterium infiltration expression system in melon fruit

Transient expression system in melon fruit was revised from the previously described method (Gao et al., 2023; Shang et al., 2014; Zhou et al., 2016). The full‐length CDS of CmRSM1 was cloned into the pCAMBIA1305.4 vector using the In‐Fusion HD Cloning Kit (Clontech, Cat. No. 639650), with a green fluorescent protein (GFP) serving as the reporter gene. The primers are shown in Table S2. CmRSM1‐1305.4 vector was transformed into Agrobacterium tumefaciens strain EHA105. After cultivation, the cells were harvested by centrifugation and resuspended in 10 mM MES buffer containing 10 mM MgCl₂ and 200 μM acetosyringone. For fruit infiltration, approximately 100 μL of the EHA105 suspension (OD₆₀₀ = 0.6) was infiltrated into the middle part of ‘ivf05’ melon fruit at 10 DAA. The infiltrated fruit sections were collected 8 days post‐infiltration for further experiments.

Agrobacterium rhizogenes‐mediated hairy root transgenic system in melon

The hairy root transgenic system in melon was performed following previously described methods (Xu et al., 2022; Zhong et al., 2022). CmRSM1‐1305.4 vector was transformed into Agrobacterium rhizogenes Ar.Qual. Melon seeds were subjected to a 30‐min soaking in ddH₂O at 50–55 °C, followed by removal of their seed coats. The naked seeds were then sterilized by a 30‐s treatment with 75% ethanol and a 15‐min treatment with 0.3% sodium hypochlorite solution. After sterilization, the seeds were germinated on Murashige and Skoog (MS30) medium at 28 °C in darkness for 2 days and then transferred to a light‐controlled environment at 25 °C for 1 week until the cotyledons fully expanded. The cotyledons were cut at the base and tip and incubated with Ar. Qual (OD₆₀₀ = 0.2) for 20 min. Subsequently, the explants were co‐cultured on MS30 medium in darkness for 2 days at 23 °C. The explants were transferred to a solid MS30 medium supplemented with 200 mg/L Timentin and grown at 25 °C under light for 2 weeks. Positive root tissues were identified based on GFP fluorescence. To obtain sufficient roots for biological replication, the positive roots were transferred to a liquid Gamborg B5 medium (Phytotech, Cat. No. G398) supplemented with 200 mg/L Timentin and cultivated at 25 °C in the dark on a shaker at 80 rpm for 1 week. These roots were collected for subsequent experiments.

Yeast one‐hybrid (Y1H) assay

The yeast one‐hybrid (Y1H) assays were conducted following the previously described method. The full‐length coding sequence (CDS) of CmRSM1 was cloned into the pGADT7 pray vector, while the 2 kb sequence upstream of the translation initiation start site of the CmBr gene was cloned into the pAbAi bait vector. The pAbAi‐bait vectors were linearized and transformed into the yeast strain Y1H Gold using the lithium chloride‐polyethylene glycol (LiCl‐PEG) method, according to the manufacturer's manual. Positive clones were screened for optimal AbA (Aureobasidin A) concentration on SD/‐Ura medium. The bait yeast strains containing the AD prey were then grown on SD/−Leu/AbA (optimal AbA concentration) plates for 3 days at 30 °C. The primers used in this study are listed in Table S2.

Dual‐luciferase reporter assay

The full‐length CDS of CmRSM1 was incorporated into the pGreenII‐62‐SK plasmid, generating the CmRSM1‐pGreenII‐62‐SK effector. The 2 kb sequence upstream of the translation initiation start site of the CmBr gene was cloned into the pGreenII 0800‐LUC vector, generating pCmBr‐LUC double reporter vector. The CmRSM1‐pGreenII‐62‐SK effector and pCmBr‐LUC reporter vector were transformed into Agrobacterium tumefaciens strain GV3101(pSoup‐p19). Co‐infiltration of the reporter and effector was carried out in N. benthamiana leaves at a volume ratio of 9:1, with the empty pGreenII‐62‐SK vector used as a control. Samples were collected within 60 h after injection to assess luciferase activity using the Dual‐Luciferase Reporter Assay System (Promega, Cat. No. E1910), using a GloMax 20/20 Luminometer (Promega) according to the manufacturer's instructions. The LUC/REN ratio was used to determine relative gene expression levels, and four independent transformations were performed for each sample. The primers used in this experiment are listed in Table S2.

Electrophoretic mobility shift assays (EMSA)

The full‐length CDS of CmRSM1 was cloned into the pGEX4T‐2 vector. CmRSM1‐ pGEX4T‐2 vector was transformed into the Rosetta (DE3) E. coli strain. The CmRSM1 fusion protein expression was induced by adding 1 μM IPTG and shaking overnight at 16 °C. After induction, the protein was purified using Glutathione Sepharose 4B (GE, Cat. No. 17–0756‐01) and eluted with 20 μM Glutathione Elution Buffer.

Three probes, namely P1, P2 and P3, were designed to target the 2 kb promoter region of CmBr. P1 contained Myb motif (TAACTG), P2 contained MYB motif (CAACCA) and MBS (CAACTG), P3 contained MRE motif (AACCTAA). The primers are shown in Table S2. The probes were labelled using the Biotin 3′‐End DNA Labeling Kit (Thermo, Cat. No. 89818), and unlabelled probes were used as competitors. EMSA assays were performed following the instructions of the LightShift Chemiluminescent EMSA Kit (Thermo, Cat. No. 20148), with minor adjustments. Binding reactions consisted of 30 μg of fusion protein and 50fmol of Biotin‐labelled probe for each reaction. The binding buffer used contained 25 mM HEPES (pH 7.5), 50 mM KCl, 2.5 mM MgCl₂, 0.1% NP‐40, 1 mM ZnSO₄ and 5% glycerol. Biotin‐labelled DNA was detected by Chemiluminescent Nucleic Acid Detection Module (Thermo, Cat. No. 89880). Competitor unlabelled probes were used at 5, 10 or 20 times the concentration of the labelled probes.

Conflict of interest

The authors declare no conflict of interests.

Author contributions

H.M.Z and Z.H.Z supervised the project. M.Y.W, N.Y.J, C.W, C.J.W and S.Q.W performed experiments. S.C. provided support for the experiments. Y.C.X, X.X.C, K.P.X and Q.Y conducted the data analysis. H.M.Z prepared the manuscript. All authors read and approved the final manuscript.

Supporting information

Data S1 Transcription factors correlated with eight CuB biosynthetic genes (Pearson correlation coefficient values >0.8) in turquoise module.

Data S2 Transcription factors correlated with CmBr (Pearson correlation coefficient values >0.6) in turquoise module.

Data S3 Differentially expressed genes (DEGs) between CPPU‐treated and control fruits that upregulated by CPPU in the ivf05^IL(CmBr) lines but not in the ivf05^IL(Cmbr‐2) lines.

PBI-22-2724-s001.xlsx^{(144.2KB, xlsx)}

Figure S1 Transcriptomic analysis of CPPU‐treated fruit and hand‐pollination control fruit during developmental and ripening of melon.

Figure S2 The expression level of CmBr and CmBt in different tissues.

Figure S3 The whole plant lines and maturity fruit bitterness of Cmbr‐1 mutant.

Figure S4 IL population construction in ivf05 background and whole genome selection.

Figure S5 The absence of Cas9 in the ivf05^IL(Cmbr‐2) line.

Figure S6 Analysis of potential off‐target effects of CmBr in the ivf05^IL(Cmbr‐2) lines by Sanger sequencing.

Figure S7 Fruit quality traits and CuB in leaves and root of ivf05^IL(CmBr) and ivf05^IL(Cmbr‐2).

Figure S8 CPPU‐induced CmRSM1 co‐expressed with CmBr and its expression patterns.

Figure S9 CmRSM1 participates in regulating CuB biosynthesis.

Figure S10 EMSA shows CmRSM1 binding to the P2 and P3 motifs of the CmBr promoter.

Figure S11 The expression level of CmBt in CmBr ILs of ivf05^IL(CmBr) and ivf05^IL(Cmbr‐2) lines.

PBI-22-2724-s003.docx^{(5.1MB, docx)}

Table S1 The bitter phenotype of P147 and ivf05 in response to high concentration of 20 mg/L CPPU in development and ripening stages.

Table S2 Sequences of primers used in this study.

PBI-22-2724-s002.docx^{(20.9KB, docx)}

Table S3 The gene information of potential off‐target of the sgRNA of CmBr.

PBI-22-2724-s004.docx^{(10.8KB, docx)}

Acknowledgements

We thank Dr. Huaisong Wang (Institute of Vegetables and Flowers, CAAS, Beijing) and Dr. Guangwei Zhao (Zhengzhou Fruit Research Institute, CAAS, Zhengzhou) for providing ‘38’, ‘P147’ and ‘ivf05’ inbred lines. We thank Dr. Jinjing Sun (Institute of Vegetables and Flowers, CAAS, Beijing) for providing the experimental guidance for EMSA assays. This study was supported by the National Natural Science Foundation of China (32372693 and 32130093).

Contributor Information

Zhonghua Zhang, Email: zhangzhonghua@qau.edu.cn.

Huimin Zhang, Email: zhanghm@qau.edu.cn.

Data availability statement

The RNA sequencing data generated in this study have been deposited in the NCBI under the accession number PRJNA1021806.

References

Banno, K. (1986) Effect of KT‐30 spray on the growth and quality of pear fruit. J. Jpn. Soc. Hortic. Sci. 55, 521–522. [Google Scholar]
Castanera, R. , Ruggieri, V. , Pujol, M. , Garcia‐Mas, J. and Casacuberta, J.M. (2020) An improved melon reference genome with single‐molecule sequencing uncovers a recent burst of transposable elements with potential impact on genes. Front. Plant Sci. 10, 484967. [DOI] [PMC free article] [PubMed] [Google Scholar]
Cheng, J.T. , Chen, H.W. , Ding, X.C. , Shen, T. , Peng, Z.W. , Kong, Q.S. , Huang, Y. et al. (2021) Transcriptome analysis of the influence of CPPU application for fruit setting on melon volatile content. J. Integr. Agric. 20, 3199–3208. [Google Scholar]
Gao, G. , Yang, F. , Wang, C. , Duan, X.Y. , Li, M. , Ma, Y. , Wang, F. et al. (2023) The transcription factor CmERFI‐2 represses CmMYB44 expression to increase sucrose levels in oriental melon fruit. Plant Physiol. 192, 1378–1395. [DOI] [PMC free article] [PubMed] [Google Scholar]
Gong, Z.H. , Luo, Y.Q. , Zhang, W.F. , Jian, W. , Zhang, L. , Gao, X.L. , Hu, X.W. et al. (2021) A SIMYB75‐centred transcriptional cascade regulates trichome formation and sesquiterpene accumulation in tomato. J. Exp. Bot. 72, 3806–3820. [DOI] [PubMed] [Google Scholar]
Hamaguchi, A. , Yamashino, T. , Koizumi, N. , Kiba, T. , Kojima, M. , Sakakibara, H. and Mizuno, T. (2008) A small subfamily of Arabidopsis RADIALIS‐LIKE SANT/MYB genes: a link to HOOKLESS1‐mediated signal transduction during early morphogenesis. Biosci. Biotechnol. Biochem. 72, 2687–2696. [DOI] [PubMed] [Google Scholar]
Hao, X.L. , Pu, Z.Q. , Cao, G. , You, D.W. , Zhou, Y. , Deng, C.P. , Shi, M. et al. (2020) Tanshinone and salvianolic acid biosynthesis are regulated by SmMYB98 in Salvia miltiorrhiza hairy roots. J. Adv. Res. 23, 1–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
Hayata, Y. , Niimi, Y. and Iwasaki, N. (1995) Synthetic Cytokinin‐1‐(2=chloro=4=pyridyl)‐3‐phenylurea (CPPU)‐promotes fruit set and induces parthenocarpy in watermelon. J. Am. Soc. Hort. Sci. 120, 997–1000. [Google Scholar]
Hayata, Y. , Niimi, Y. , Inoue, K. and Kondo, S. (2000) CPPU and BA, with and without pollination, affect set, growth, and quality of muskmelon fruit. HortScience 35, 868–870. [Google Scholar]
Horie, H. , Ito, H. , Ippoushi, K. , Azuma, K. , Sakata, Y. and Igarashi, I. (2007) Cucurbitacin C—Bitter principle in cucumber plants. JARQ 41, 65–68. [Google Scholar]
Hu, B.W. , Li, D.W. , Liu, X. , Qi, J.J. , Gao, D.L. , Zhao, S.Q. , Huang, S.W. et al. (2017) Engineering Non‐transgenic Gynoecious Cucumber Using an Improved Transformation Protocol and Optimized CRISPR/Cas9 System. Mol. Plant 10, 1575–1578. [DOI] [PubMed] [Google Scholar]
Huang, Y. , Li, W.H. , Zhao, L. , Shen, T. , Sun, J.Y. , Chen, H.W. , Kong, Q.S. et al. (2017) Melon fruit sugar and amino acid contents are affected by fruit setting method under protected cultivation. Sci. Hortic. 214, 288–294. [Google Scholar]
Iwahori, S. , Tominaga, S. and Yamasaki, T. (1988) Stimulation of fruit growth of kiwifruit, Actinidia chinensis Planch., by N‐(2‐chloro‐4‐pyridyl)‐N'‐phenylurea, a diphenylurea‐derivative cytokinin. Sci. Hortic. 35, 109–115. [Google Scholar]
Jian, W. , Cao, H. , Yuan, S. , Liu, Y. , Lu, J. , Lu, W. , Li, N. et al. (2019) SlMYB75, an MYB‐type transcription factor, promotes anthocyanin accumulation and enhances volatile aroma production in tomato fruits. Hortic. Res. 6, 22. [DOI] [PMC free article] [PubMed] [Google Scholar]
Kano, Y. and Goto, H. (2003) Relationship between the occurrence of bitter fruit in cucumber (Cucumis sativus L.) and the contents of total nitrogen, amino acid nitrogen, protein and HMG‐CoA reductase activity. Sci. Hortic. 98, 1–8. [Google Scholar]
Kim, D. , Landmead, B. and Salzberg, S.L. (2015) HISAT: a fast spliced aligner with low memory requirements. Nat. Methods 12, 357–360. [DOI] [PMC free article] [PubMed] [Google Scholar]
Kumar, S. , Stecher, G. , Li, M. , Knyaz, C. and Tamura, K. (2018) MEGA X: Molecular Evolutionary Genetics Analysis across Computing Platforms. Mol. Biol. Evol. 35, 1547–1549. [DOI] [PMC free article] [PubMed] [Google Scholar]
Lei, R.H. , Li, Y. , Cai, Y.R. , Li, C.Y. , Pu, M.N. , Lu, C.K. , Yang, Y.J. et al. (2020) bHLH121 Functions as a Direct Link that Facilitates the Activation of FIT by bHLH IVc Transcription Factors for Maintaining Fe Homeostasis in Arabidopsis . Mol. Plant 13, 634–649. [DOI] [PubMed] [Google Scholar]
Li, C.N. , Zhou, Y.Y. and Fan, L.M. (2015) A novel repressor of floral transition, MEE3, an abiotic stress regulated protein, functions as an activator of FLC by binding to its promoter in Arabidopsis . Environ. Exp. Bot. 113, 1–10. [Google Scholar]
Li, Y. , Chen, Y. , Zhou, L. , You, S.J. , Deng, H. , Chen, Y. , Alseekh, S. et al. (2020) MicroTom Metabolic Network: Rewiring Tomato Metabolic Regulatory Network throughout the Growth Cycle. Mol. Plant 13, 1203–1218. [DOI] [PubMed] [Google Scholar]
Li, J.F. , Lin, T. , Ren, D.D. , Wang, T. , Tang, Y. , Wang, Y.W. , Xu, L. et al. (2021) Transcriptomic and Metabolomic Studies Reveal Mechanisms of Effects of CPPU‐Mediated Fruit‐Setting on Attenuating Volatile Attributes of Melon Fruit. Agronomy‐Basel 11, 1007. [Google Scholar]
Li, P. , Xia, E. , Fu, J. , Xu, Y. , Zhao, X. , Tong, W. , Tang, Q. et al. (2022) Diverse roles of MYB transcription factors in regulating secondary metabolite biosynthesis, shoot development, and stress responses in tea plants (Camellia sinensis). Plant J. 110, 1144–1165. [DOI] [PubMed] [Google Scholar]
Li, Y. , Zhang, J. , Wang, S. , Zhang, H. , Liu, Y. and Yang, M. (2023) Integrative transcriptomic and metabolomic analyses reveal the flavonoid biosynthesis of Pyrus hopeiensis flowers under cold stress. Hortic. Plant J. 9, 395–413. [Google Scholar]
Liu, Y. , Li, Y. , Guo, H.X. , Lv, B.S. , Feng, J. , Wang, H.H. , Zhang, Z.H. et al. (2023) Gibberellin biosynthesis is required for CPPU‐induced parthenocarpy in melon. Hortic. Res. 10, uhad084. [DOI] [PMC free article] [PubMed] [Google Scholar]
Luo, H. , Vong, C.T. , Chen, H.B. , Gao, Y. , Lyu, P. , Qiu, L. , Zhao, M.M. et al. (2019) Naturally occurring anti‐cancer compounds: Shining from Chinese herbal medicine. Chinas Med. 14, 48. [DOI] [PMC free article] [PubMed] [Google Scholar]
Luo, F. , Li, Q. , Yu, L. , Wang, C. and Qi, H. (2020) High concentrations of CPPU promotes cucurbitacin B accumulation in melon (Cucumis melo var. makuwa Makino) fruit by inducing transcription factor CmBt. Plant Physiol. Biochem. 154, 770–781. [DOI] [PubMed] [Google Scholar]
Lv, M. , Sun, X. , Li, D.C. , Wei, G. , Liu, L. , Chen, F. , Cai, Y.P. et al. (2022) Terpenoid Biosynthesis in Dendrobium Officinale: Identification of (E)‐Beta‐Caryophyllene Synthase and the Regulatory MYB Genes. Ind. Crop Prod. 182, 114875. [Google Scholar]
Mahjoub, A. , Hernould, M. , Joubes, J. , Decendit, A. , Mars, M. , Barrieu, F. , Hamdi, S. et al. (2009) Overexpression of a grapevine R2R3‐MYB factor in tomato affects vegetative development, flower morphology and flavonoid and terpenoid metabolism. Plant Physiol. Biochem. 47, 551–561. [DOI] [PubMed] [Google Scholar]
Matsuo, S. , Kikuchi, K. , Fukuda, M. , Honda, I. and Imanishi, S. (2012) Roles and regulation of cytokinins in tomato fruit development. J. Exp. Bot. 63, 5569–5579. [DOI] [PMC free article] [PubMed] [Google Scholar]
Mu, R. , Cao, Y. , Liu, Y. , Lei, G. , Zou, H. , Liao, Y. , Wang, H. et al. (2009) An R2R3‐type transcription factor gene AtMYB59 regulates root growth and cell cycle progression in Arabidopsis . Cell Res. 19, 1291–1304. [DOI] [PubMed] [Google Scholar]
Nickell, L.G. (1986) The Effects of N‐(2‐Chloro‐4‐Pyridyl)‐N'‐Phenylurea and the 3‐Chlorobenzyl Ester of Dicamba on the Growth and Sugar Content of Grapes, pp. 805–806. Leuven, Belgium: International Society for Horticultural Science (ISHS). [Google Scholar]
Ogata, R. , Saito, T. and Oshima, K. (1989) Effect of N‐Phenyl‐N'‐(4‐Pyridyl)Urea (4‐PU) on Fruit Size: Apple Japanese Pear, Grapevine and Kiwi Fruit, pp. 395–398. Leuven, Belgium: International Society for Horticultural Science (ISHS). [Google Scholar]
Pertea, M. , Pertea, G.M. , Antonescu, C.M. , Chang, T.C. , Mendell, J.T. and Salzberg, S.L. (2015) StringTie enables improved reconstruction of a transcriptome from RNA‐seq reads. Nat. Biotechnol. 33, 290–295. [DOI] [PMC free article] [PubMed] [Google Scholar]
Shang, Y. , Ma, Y. , Zhou, Y. , Zhang, H. , Duan, L. , Chen, H. , Zeng, J. et al. (2014) Biosynthesis, regulation, and domestication of bitterness in cucumber. Science 346, 1084–1088. [DOI] [PubMed] [Google Scholar]
Shang, J.L. , Kong, S.N. , Li, N. , Wang, J.M. , Zhou, D. , Li, N.A. and Ma, S.W. (2020) Genetic mapping and localization of major QTL for bitterness in melon (Cucumis melo L.). Sci. Hortic. 266, 109286. [Google Scholar]
Song, S. , Qi, T. , Huang, H. , Ren, Q. , Wu, D. , Chang, C. , Peng, W. et al. (2011) The Jasmonate‐ZIM domain proteins interact with the R2R3‐MYB transcription factors MYB21 and MYB24 to affect Jasmonate‐regulated stamen development in Arabidopsis . Plant Cell 23, 1000–1013. [DOI] [PMC free article] [PubMed] [Google Scholar]
Tao, H. , Miao, H.Y. , Chen, L.L. , Wang, M.Y. , Xia, C.C. , Zeng, W. , Sun, B. et al. (2022) WRKY33‐mediated indolic glucosinolate metabolic pathway confers resistance against Alternaria brassicicola in Arabidopsis and Brassica crops. J. Integr. Plant Biol. 64, 1007–1019. [DOI] [PubMed] [Google Scholar]
Thakur, M. , Bhattacharya, S. , Khosla, P.K. and Puri, S. (2019) Improving production of plant secondary metabolites through biotic and abiotic elicitation. J. Appl. Res. Med. Aromat. Plants 12, 1–12. [Google Scholar]
Wang, R.C. , Shu, P. , Zhang, C. , Zhang, J.L. , Chen, Y. , Zhang, Y.X. , Du, K. et al. (2022) Integrative analyses of metabolome and genome‐wide transcriptome reveal the regulatory network governing flavor formation in kiwifruit (Actinidia chinensis). New Phytol. 233, 373–389. [DOI] [PubMed] [Google Scholar]
Wink, M. (2016) Evolution of Secondary Plant Metabolism. eLS. Chichester: John Wiley & Sons, Ltd. 10.1002/9780470015902.a0001922.pub3 [DOI] [Google Scholar]
Xin, T.X. , Tian, H.J. , Ma, Y.L. , Wang, S.H. , Yang, L. , Li, X.T. , Zhang, M.Z. et al. (2022) Targeted creation of new mutants with compact plant architecture using CRISPR/Cas9 genome editing by an optimized genetic transformation procedure in cucurbit plants. Hortic. Res. 9, uhab086. [DOI] [PMC free article] [PubMed] [Google Scholar]
Xu, Y. , Zhang, H. , Zhong, Y. , Jiang, N. , Zhong, X. , Zhang, Q. , Chai, S. et al. (2022) Comparative genomics analysis of bHLH genes in cucurbits identifies a novel gene regulating cucurbitacin biosynthesis. Hortic. Res. 9, uhac038. [DOI] [PMC free article] [PubMed] [Google Scholar]
Yamagishi, M. (2022) High temperature enhances anthocyanin coloration in Asiatic hybrid lily flowers via upregulation of the MYB12 positive regulator. Hortic. Plant J. 8, 769–776. [Google Scholar]
Yang, C. , Fang, X. , Wu, X. , Mao, Y. , Wang, L. and Chen, X. (2012) Transcriptional Regulation of Plant Secondary Metabolism. J. Integr. Plant Biol. 54, 703–712. [DOI] [PubMed] [Google Scholar]
Yang, B.C. , Song, Z.H. , Li, C.N. , Jiang, J.H. , Zhou, Y.Y. , Wang, R.P. , Wang, Q. et al. (2018) RSM1, an Arabidopsis MYB protein, interacts with HY5/HYH to modulate seed germination and seedling development in response to abscisic acid and salinity. PLoS Genet. 14, e1007839. [DOI] [PMC free article] [PubMed] [Google Scholar]
Yang, Z.Z. , Li, Y.Q. , Gao, F.Z. , Jin, W. , Li, S.Y. , Kimani, S. , Yang, S. et al. (2020) MYB21 interacts with MYC2 to control the expression of terpene synthase genes in flowers of Freesia hybrida and Arabidopsis thaliana . J. Exp. Bot. 71, 4140–4158. [DOI] [PubMed] [Google Scholar]
Yang, W. , Zhang, S. , Hou, Q. , Gao, J. , Wang, H. , Chen, X. , Liao, X. et al. (2023) Transcriptomic and metabolomic analysis provides insights into lignin biosynthesis and accumulation in different lodging‐resistance of hybrid wheat. J. Integr. Agric. 23, 1105–1117. [Google Scholar]
Yano, R. , Nonaka, S. and Ezura, H. (2018) Melonet‐DB, a Grand RNA‐Seq Gene Expression Atlas in Melon (Cucumis melo L.). Plant Cell Physiol. 59, e4. [DOI] [PubMed] [Google Scholar]
Yuan, P.L. , Xu, C.P. , He, N. , Lu, X.Q. , Zhang, X.P. , Shang, J.L. , Zhu, H.J. et al. (2023) Watermelon domestication was shaped by stepwise selection and regulation of the metabolome. Sci. China Life Sci. 66, 579–594. [DOI] [PubMed] [Google Scholar]
Zhang, J.X. , Zhou, L.B. , Zheng, X.Y. , Zhang, J.J. , Yang, L. , Tan, R.H. and Zhao, S.J. (2017) Overexpression of SmMYB9b enhances tanshinone concentration in Salvia miltiorrhiza hairy roots. Plant Cell Rep. 36, 1297–1309. [DOI] [PubMed] [Google Scholar]
Zhang, S. , Nie, L. , Zhao, W. , Cui, Q. , Wang, J. , Duan, Y. and Ge, C. (2019) Metabolomic analysis of the occurrence of bitter fruits on grafted oriental melon plants. PLoS One 14, e0223707. [DOI] [PMC free article] [PubMed] [Google Scholar]
Zhao, G. , Lian, Q. , Zhang, Z. , Fu, Q. , He, Y. , Ma, S. , Ruggieri, V. et al. (2019) A comprehensive genome variation map of melon identifies multiple domestication events and loci influencing agronomic traits. Nat. Genet. 51, 1607–1615. [DOI] [PubMed] [Google Scholar]
Zhong, Y. , Xun, W. , Wang, X. , Tian, S. , Zhang, Y. , Li, D. , Zhou, Y. et al. (2022) Root‐secreted bitter triterpene modulates the rhizosphere microbiota to improve plant fitness. Nat. Plants 8, 887–896. [DOI] [PubMed] [Google Scholar]
Zhou, Y. , Ma, Y. , Zeng, J. , Duan, L. , Xue, X. , Wang, H. , Lin, T. et al. (2016) Convergence and divergence of bitterness biosynthesis and regulation in Cucurbitaceae. Nat. Plants 2, 1–8. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials