Long non-coding RNA RsLNC2361 positively regulate anthocyanin accumulation in radish Authors: Haiyi Liu, Xiaobo Luo, Linjun Wu, Guangqian Yang, Xianping Jiang, Dan Li, Maojun Mao, Yun Pei, Xiuhong Xu, Wanping Zhang Published: 2026 ABSTRACT AbstractBackgroundRadish (Raphanus sativus L.) is a crucial vegetable crop in the Brassicaceae family, which has a high nutritional, medicinal, and dietary therapeutic values. Radish varieties rich in anthocyanins are favored by consumers because of their bright color and high nutritional value. However, the reports on the role of lncRNA in regulating anthocyanin accumulation in radishes are relatively rare.ResultsBased on our previous LncRNA sequencing findings, the RsLNC2361 was differentially expressed in red and white skin radish. The results revealed that overexpression of RsLNC2361 can promote anthocyanin accumulation in radish leaves, whereas silencing of RsLNC2361 can decrease the anthocyanin content in radish flesh root using VIGS technology, indicating that RsLNC2361 has a positive regulatory effect on anthocyanin accumulation in radish. Overexpression of RsLNC2361 increased the expresssion anthocyanin structural genes, whereas silencing of RsLNC2361 had opposite effect. A total of 5,032 DEGs and 273 differental metabolites were identified in RsLNC2361-pTY and pTY (control) by transcriptome and metabolome sequencing, respectively. Combined metabolomic and transcriptomic analyses showed KEGG pathway associated with flavonoid biosynthesis and phenylpropanoid biosynthesis was commonly enriched. The results found that the miRNA858 was predicted to target RsLNC2361, with expectation values of 4.0. Therefore, we speculate that lncRNA participates in the accumulation of anthocyanins in radishes through the regulatory mechanism of the competitive endogenous RNA (ceRNA) module.ConclusionsThis study provides a new perspective on the regulation of anthocyanin biosynthesis by non-coding RNA and offers a theoretical basis for the breeding of high-anthocyanin radish.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12870-026-08909-9.Keywords: Anthocyanins, Radish, RsLNC2361, Transcriptome, Metabolome Introduction Anthocyanins are water-soluble pigments found in plants, which are primarily stored in plant cell vacuoles to perform their functions and play a critical role in the pigmentation of plant organs [1]. For instance, anthocyanins markedly influence purple-red Chrysanthemum × morifolium petals, deep-purple Solanum melongena fruits, and pink Lilium brownii var. viridulum flowers [2, 3]. The red leaves of Euphorbia pulcherrima and red blackberries were substantially correlated with anthocyanin accumulation [4]. Furthermore, the NsMYB1 gene promotes anthocyanin accumulation in the black fruit of Nitraria sibirica [5]. Transient overexpression of DbMYB2 markedly promotes anthocyanin accumulation in the leaves of Nicotiana benthamiana [6]. These results indicated that color enhancement is closely associated with an increase in anthocyanin levels. The biosynthetic pathways of anthocyanins have been clearly analyzed in plant. Anthocyanin metabolism primarily occurs in a branch of the phenylalanine pathway, the flavonoid biosynthetic pathway [7]. In this synthetic pathway, key enzymes include early biosynthetic genes (EBGs), such as phenylalanine ammonia-lyase (PAL), cinnamate 4-hydroxylase (C4H), chalcone synthase (CHS), chalcone isomerase (CHI), flavanone-3-hydroxylase (F3H), and flavone 3’-hydroxylase (F3’H). The late biosynthetic genes (LBGs) include flavonoid 3’,5’-hydroxylase (F3’5’H), dihydroflavonol-4-reductase (DFR), anthocyanidin synthase (ANS), and UDP-glucose: anthocyanidin 3-O-glucosyltransferase (UFGT) [8]. Additionally, multiple transcription factors regulate anthocyanin synthesis. The MYB-bHLH-WD40 ternary protein complex (MBW) comprises the MYB, basic Helix-Loop-Helix (bHLH), and WD repeat (WD40) transcription factors. This complex binds to the corresponding cis-acting elements in the promoters of the structural genes, thereby promoting or inhibiting the expression of downstream structural genes [9]. This complex is the most common transcriptional activation pathway in the anthocyanin biosynthesis pathway [10]. Furthermore, some R2R3-MYB transcription factors can directly regulate the expression of structural genes independently of other transcription factors [11]. Overexpression of the Solanum lycopersicum MYB genes SLANT1 and SLANT2 can result in a large increase in anthocyanin content in tomato fruits and the upregulation of structural genes involved in anthocyanin synthesis [12, 13]. Radish (Raphanus sativus L., 2n = 2x = 18) is a crucial vegetable crop in the genus Raphanus of the family Cruciferae and has high nutritional, medicinal, and therapeutic values. Radish germplasm resources are rich and their cultivation has a long history. After long-term natural and artificial selection, different types and varieties with different skin and flesh colors have formed. Among these, radish varieties rich in anthocyanins are favored by consumers because of their bright color and high nutritional value [2]. Most red radish varieties contain geranium pigment, whereas purple radish is primarily composed of cornflower pigments. RsMYB1 is a positive regulator in the anthocyanin synthesis pathway in red radish [14]. Overexpression of the radish RsMYB1 gene in Nicotiana benthamiana and Arabidopsis thaliana markedly increased the anthocyanin content of plants, as well as the expression levels of endogenous anthocyanin synthesis structural genes [14, 15]. RsMYB1 and RsTT8 co-expression induces anthocyanin accumulation and upregulates the expression of genes involve in the anthocyanin biosynthesis pathway in radish [16]. Furthermore, when RsMYB1a was overexpressed alone or co-expressed with RsbHLH4, a large amount of anthocyanins accumulated in radish cotyledons and leaves, and RsMYB1a interacted with RsbHLH4 to activate RsbHLH4 expression [17]. Red radish RsGST1 functions similarly to AtTT19, a gene responsible for anthocyanin transport in Arabidopsis. It can restore anthocyanin accumulation in the AtTT19 mutant of Arabidopsis. Furthermore, RsGST1 and RsMYB1a work synergistically to enhance anthocyanin accumulation in radish [18]. Long non-coding RNAs (lncRNAs) are non-coding RNAs longer than 200 nt, with a unique secondary structure that does not encode proteins [19]. Additionally, lncRNA can affect plant growth and development by regulating the expression of specific genes and auxin synthesis. When seed development in Arabidopsis ends the dormancy period, the antisense lncRNA DOG1 is transcribed near the promoter of the germination delay gene DOG1, which strongly inhibits DOG1 expression via cis-regulation and accelerates seed germination [20]. In Solanum lycopersicum, the expression of genes associated with ethylene and carotenoid synthesis was downregulated when knocking out of lncRNA1459, leading to the inhibition of ethylene production and lycopene synthesis [21]. After silencing the cold stress-related lncRNA XH123, Gossypium was found to be more sensitive to low temperatures, with chloroplast damage and increased levels of reactive oxygen species [22]. Overexpression of LINC15957 markedly increases anthocyanin accumulation in radish leaves and the expression levels of genes linked to anthocyanin biosynthesis, silencing of LINC15957 substantially decreases anthocyanin accumulation in fleshy roots and the expression levels of genes related to anthocyanin biosynthesis [23]. It was reported that MdLNC610, a positive regulatory factor, participates in the regulation of anthocyanin synthesis induced by strong light, thereby promoting MdACO1 expression and ethylene biosynthesis [24]. Previous study indicate that overexpression and silencing of MdLNC499 in apple fruits can promote and inhibit anthocyanin accumulation, respectively [25]. However, the specific functions of lncRNAs in radish still need to be verified, and the mechanism by which it regulate anthocyanin accumulation in radish remains elusive. Based on previous lncRNA sequencing in high- and low-anthocyanin-content varieties, the long non-coding RNA RsLNC2361 was differential expressed in two varieties, which was selected to subject further research. In the present study, the function of RsLNC2361 was verified by transient overexpression and virus-induced gene silencing (VIGS). The expression of anthocyanin structural genes were verified in overexpressing and silencing of RsLNC2361 using quantitative reverse transcription polymerase chain reaction (RT-qPCR). Differentially expressed genes (DEGs) and metabolites were explored by conducting transcriptome and targeted metabolome sequencing on the silenced plants and control group. This finding provides insights into how lncRNAs regulate anthocyanin accumulation and provide valuable gene resources for further research on radish lncRNA. Methods Plant materials Full seeds of red-skinned and red-fleshed ZiDan radish were selected and sown in cavity trays (5 × 10). The seedlings were raised in an artificial climate chamber (day temperature 25 °C, 16 h; night temperature 16 °C, 8 h; humidity 75%). Four true leaves were transplanted into plastic pots (outer diameter 29.2 cm, height 23.5 cm), with 15 pots of each material. Determination of anthocyanin content The pH difference method was performed according to the protocol reported by Mao et al. [26]. Samples (1 g) were finely ground in liquid nitrogen, and subsequently fully mixed with pre-chilled (4 °C) 0.05% hydrochloric acid-methanol solution to fill the tube. The mixture was incubated in the dark at 4 °C. For each sample, two test tubes were prepared (each containing 1 ml of extract), to which 4 ml of 0.4 mol/L KCl-HCl buffer (pH 1.0) or 0.4 mol/L citric acid-disodium hydrogen phosphate buffer (pH 5.0) was added. After thorough mixing, the tubes were allowed to stand at room temperature for 20 min, and the absorbance was measured at 510 and 700 nm using a UV spectrophotometer. All samples were analyzed in three biological repetitions. Histochemical b-glucuronidase (GUS) staining The function of RsLNC2361 in Nicotiana benthamiana leaves was verified using a GUS protein degradation assay. The complete sequence of RsLNC2361 was separately inserted into the pCAMBIA1301-GUS vector to generate the 35S:RsLNC2361-GUS and 35S:RsLNC2361 vectors. Tissues were stained with GUS as previously described [27]. A minimum of three experiments were performed and the pCAMBIA1301-GUS vector was used as a control. The primers were provided in Supplementary Table S1. Transient overexpression experiment of radish leaves and virus-induced gene silencing For transient overexpression, the full-length sequence of RsLNC2361 was introduced into the pGFC1008 plasmid, and its expression was driven by the 35S promoter. The recombinant plasmid and empty vector were transformed into the ZiDan radish leaves using Agrobacterium rhizocarcinae. The seedlings were subsequently recovered in darkness (28 °C) for 1–2 d, and were then transferred to a growth chamber for treatment under normal light culture. Leaf samples showing the phenotypes were collected. After taking photos and samples, they were immediately quick-frozen in liquid nitrogen and stored at − 80 °C until further use. To confirm the functions of RsLNC2361, a TYMV-based VIGS system was used [28, 29]. A palindromic DNA fragment of 80 nt corresponding to RsLNC2361 genes was designed and synthesized, and 15-nt homologous sequences corresponding to the vector were introduced to 3′ and 5′ ends. The pTY VIGS vector plasmid was cleaved with SnaB-I to construct the pTY-RsLNC2361 vector. The plasmid carrying the target gene was expanded and injected into radish root flesh using a syringe. Plants injected with empty pTY vectors served as controls. The fleshy root of ZiDan was used to injecte and then cultivated in an artificial climate chamber at 25 °C/22°C and subjected to 16-h/8-h light/dark cycles. The phenotype was assessed three weeks later. Supplementary Table S1 listed the primer sequences. Transcriptome sequencing and data analysis Total RNA was extracted from the fleshy root of RsLNC2361-pTY plants and control radish with three biological repetitions using an RNAprep Pure Plant Kit (Tiangen Biotech, Beijing). For each sample cDNA libraries were constructed using mRNA enrichment technology, followed by sequencing on an Illumina HiSeq-PE150 platform (Novogene Co., Ltd. (Beijing). Reference genome and gene model annotation files were downloaded directly from the genome website (https://39.100.233.196:82/download_genome/Brassica_Genome_data/Rapsa_Xiang_V1.0/?C=N;O=A). The reference genome index was constructed using Hisat2 v2.0.5, and clean paired-end reads were aligned using the software. Differential expression analysis was conducted on the two processed sample groups (with three biological replicates) using the DESeq2 R package (version 1.20.0). The transcripts with |log2(FoldChange)| >= 1 & padj < = 0.05 were defined as differential expression gene (DEG). The clusterProfiler R package was used to perform Gene Ontology (GO) enrichment analysis of DEG, with the gene length bias corrected. The GO terms were considered significantly enriched when the adjusted p-value was < 0.05. The same package was used to assess the statistical enrichment of the DEGs in the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway. Metabolome sequencing and data analysis Radish fleshy root tissue sample of RsLNC2361-pTY plants and control radish with three biological repetitions (100 mg) was taken to which 500 µL of 80% methanol water solution containing the internal standard was added. After vortexing in an ice bath and centrifugation, methanol was diluted to 53% and subsequently centrifuged for injection. Furthermore, 100 µL of the liquid sample was taken to which 400 µL of the above-mentioned methanol solution was added. The subsequent steps were the same as those used for the tissue samples. Cell/bacterial samples were added with 300 µL of 80% methanol aqueous solution containing internal standards, and subsequently subjected to rapid freezing in liquid nitrogen, ultrasonic treatment, etc., followed by centrifugation and freeze-drying, and finally dissolved in 10% methanol. After freeze-drying 1 mL of the culture medium supernatant, 100 µL of 80% methanol water solution containing the internal standard was added, and the steps for tissue samples were followed. Next, 1 g of the soil sample was added to 1000 µL of 80% methanol water solution containing the internal standard, which was then centrifuged two times, followed by freeze-drying and dissolution. The chromatographic flow rate was 0.4 mL/min. The gradient elution procedure was as follows: 98% A for 2 min, 0% A for 11–13 min, and 98% A for 13–15 min. In the positive and negative ion modes of the mass spectrometer, the Curtain Gas was 35 psi, the collision gas was Medium, and the temperature was 550 °C. The ion source gases 1 and 2 were both 60, with voltages of 5500 V and − 4500 V respectively. Metabolites were identified based on the NovoDB database established by Novogene using the multireaction monitoring mode. The mass spectrometry files were processed using SCIEX OS V1.4. Spectral peaks were screened according to standards, such as a minimum peak height of 500 and a signal-to-noise ratio of 5. The metabolites were annotated using the KEGG and HMDB databases for statistical analysis. Multivariate statistical analysis was conducted using metaX. Charts were plotted using the R package. The differential metabolites were screened with VIP > 1.0, FC > 1.5 or FC < 0.667 and P-value < 0.05. Correlation analysis between metabolome and transcriptome Orthogonal partial least squares (O2PLS) was used to conduct correlation analysis between metabolome and transcriptome and identify highly correlated metabolites and DEGs. The top 10 metabolites from the metabolic profiling enrichment results and the top 10 genes from the transcriptomics results were selected using the “mixOmics” package in R language, and draw a metabolic-transcription correlation network diagram. The correlation analysis of differential genes and differential metabolites is based on the Pearson statistical method. Expression analysis using RT-qPCR To determine the expression levels of RsLNC2361 and the genes associated with anthocyanin synthesis, total RNA was extracted using a plant RNA extraction kit (Omega, China) and reverse-transcribed to complementary DNA (cDNA) using a reverse transcription kit (GenStar, China). RT-qPCR was performed on a Roche LightCycler 96 System, and RsActin was selected as the reference gene. The relative expression levels were calculated using the 2−ΔΔCt method. The primers listed in Supplementary Table S1. Each experiment included 3 biological replicates with 3 technical replicates. Statistical analysis Anthocyanin determination and RT-qPCR were performed in replicates. Data are presented as mean ± standard deviation (SD). Student’s t-test was applied to compare different samples, with a difference considered statistically significant at P < 0.05 and highly significant at P < 0.01. Results Analysis of the coding potential of RsLNC2361 Based on previous lncRNA sequencing results, RsLNC2361 was differentially expressed between red and white skin radish [23]. We determined whether RsLNC2361 is a non-coding RNA using the ORF Finder online tool (https://www.ncbi.nlm.nih.gov/orffinder/) to verify the open reading frame. The three longest coding peptides were 67, 54, and 39 amino acids long (Fig. 1A). Furthermore, the coding potential of RsLNC2361 was calculated using the CPC2 online tool (https://cpc2.gaolab.org/), and the results predicted that RsLNC2361 is a non-coding RNA (Fig. 1B). The coding potential was further confirmed by GUS histochemical staining. The constructed expression vector 35S:RsLNC2361-GUS did not exhibit GUS activity (Fig. 1C). Thus, we concluded that RsLNC2361 is a non-coding RNA. Given that subcellular localization is critical for lncRNA function, we used lncLocator2.0 (https://www.csbio.sjtu.edu.cn/bioinf/lncLocator2/) to predict the subcellular localization of RsLNC2361 [30]. The prediction results revealed that RsLNC2361 was distributed in the cytoplasm and nucleus (Fig. 1D). Analysis of the coding potential of RsLNC2361. A Open Reading Frame (ORF) analysis of RsLNC2361 revealed three potential frames (marked by red boxes). Among these, the three ORFs were computationally predicted to encode polypeptides with lengths of 67, 54, and 39 aa, respectively. B The coding potential of RsLNC2361 was evaluated using the CPC2 program, which generated corresponding coding potential scores. RsActin was employed as a positive control in this analysis to validate the reliability of the results. C Recombinant vectors pCAMBIA1301-RsLNC2361-GUS were constructed and subsequently transformed into tobacco leaf tissues. β-Glucuronidase (GUS) staining assay was performed to visually assess the potential coding capacity of RsLNC2361, with n = 3 biological replicates for the assay. D The subcellular localization of RsLNC2361 was predicted through the online bioinformatics tool lncLocator2.0 RsLNC2361 promoted anthocyanin accumulation in radish We constructed an RsLNC2361-pGFC1008 overexpression vector containing the 35S promoter to explore the role of RsLNC2361 in anthocyanin biosynthesis. Two true leaves of ZiDan radish were selected for transient overexpression experiment with RsLNC2361 and pGFC1008 (control group). Anthocyanin accumulation could be clearly observed at the injection site of leaves in RsLNC2361 overexpressed lines (Fig. 2A). Additionally, PCR amplification indicated that RsLNC2361 was successfully transferred to the radish leaves (Fig. 2B). Compared to the control group, the transient overexpression of RsLNC2361 in radish leaves markedly increased the anthocyanin content and gene expression level of RsLNC2361 (Fig. 2C and D). The expression levels of genes related to the anthocyanin synthesis pathway was conducted to detect the relative expression level of genes by RT-qPCR. The anthocyanin synthase (RsANS), UDP-glucoflavone 3-O-glucosyltransferase (RsUFGT), Flavonoid-3-O-glucosyl transferase (Rs3GT), TRANSPARENT TESTA 8 (RsTT8), Chalcone Synthase (RsCHS), flavanone 3b hydroxylase (RsF3H), dihydroflavonol 4-reductase (RsDFR), and cinnamic acid 4-hydroxylase (RsC4H), were significantly increased when RsLNC2361 was overexpressed (Fig. 2E). RsLNC2361 promotes the accumulation of anthocyanins in radish. A Empty control (pFGC1008) and two independent transient overexpression lines of RsLNC2361 (OE-RsLNC2361-1 and OE-RsLNC2361-2). B The presence of transgenes in the RsLNC2361 transient overexpression lines was confirmed by PCR amplification. C The anthocyanin content of transiently overexpressing RsLNC2361 in radish leaves. D The expression levels of RsLNC2361 with transient overexpression in radish leaves. E The expression levels of genes related to anthocyanin biosynthesis in transient overexpressing RsLNC2361 in radish leaves. Data are presented as mean ± standard deviation (SD) with n = 3 biological replicates, and statistical significance was determined by Student’s t-test. Asterisks indicate significantly different values (*P < 0.05 and **P < 0.01) Silencing of RsLNC2361 decreased anthocyanin accumulation in radish fleshy root The recombinant plasmid RsLNC2361-pTY and empty plasmid pTY were injected into the fleshy roots of purple radish. Four weeks after transformation, the red color of plants infected with RsLNC2361-pTY weakened at the injection site, whereas no color change was observed in the pTY vector (Fig. 3A). Compared to the control (pTY vector), the anthocyanin content of RsLNC2361-pTY plants was markedly reduced (Fig. 3B). Additionally, the expression level of RsLNC2361 in fleshy root infected with RsLNC2361-pTY was substantially lower than that in the control group (Fig. 3C). Compared to the control group, the expression levels of anthocyanin-related structural genes (RsANS, Rs3GT, RsTT8, RsF3H, RsUFGT, RsANS, RsF3H RsDFR) in radish fleshy root infected with RsLNC2361-pTY were considerably reduced (Fig. 3D). These results revealed that RsLNC2361 positively regulates anthocyanin synthesis in radish. Silencing of RsLNC2361 can reduce the accumulation of anthocyanins in radish fleshy root. A The silencing of RsLNC2361 in radish fleshy root by VIGS, with the empty vector pTY as the negative control. B The anthocyanin content in silencing of RsLNC2361 in radish fleshy root. C The expression levels of RsLNC2361 with silencing of RsLNC2361 in radish fleshy root. D The expression levels of genes related to anthocyanin biosynthesis in silencing of RsLNC2361 in radish fleshy root. Data are presented as mean ± standard deviation (SD) with n = 3 biological replicates, and statistical significance was determined by Student’s t-test. Asterisks indicate significantly different values (*P < 0.05 and **P < 0.01) Transcriptome profiles of RsLNC2361-pTY and pTY To identify the genes regulated by RsLNC2361 in radish anthocyanin accumulation, two libraries (RsLNC2361-pTY and pTY controls) with three biological replicates were constructed for transcriptome sequencing. The raw data obtained by the RsLNC2361-pTY and control groups were 7.08 Gb and 7.03 Gb, respectively. After filtering, clean data 6.91 Gb and 6.9 Gb were obtained. In the RsLNC2361-pTY group, Q20 was ≥ 99.37% and Q30 was ≥ 97.33%. In the control group, Q20 was ≥ 99.38% and Q30 was ≥ 97.32%. The average GC content of all the samples was 46.56%. Approximately 83.5% of the clean reads mapped to 38,447,575 genes in the radish reference genome (Supplementary Table S2). A total of 22,470 genes were commonly identified between the RsLNC2361-pTY and pTY groups (Fig. S1). A totol of 5,032 DEGs were identified between RsLNC2361-pTY and pTY, including 2,161 upregulated and 2,871 downregulated genes (Fig. 4A; Supplementary Table S3). Herein, we identified 12 genes associated with anthocyanin biosynthesis. Notably, compared to the RsLNC2361-pTY group, RsCHS, RsF3H, RsTT8, RsUFGT and RsDFR in the silenced RsLNC2361 plants were considerably downregulated (Fig. 4B). GO enrichment analysis found that the the most highly represented GO terms were heme binding, tetrapyrrole binding, hydrolase activity, and hydrolase activity, hydrolyzing O-glycosyl compounds (Fig. 4C). KEGG pathway enrichment analysis showed that the DEGs were markedly enriched in the biosynthesis of flavonoids, phenylpropanes, stilbenoids, diarylheptanoids, and gingerol (Fig. 4D). Transcriptome profiles of RsLNC2361-pTY and pTY. A Volcano map of DEGs in pTY and RsLNC2361-pTY, with DEGs screened by DESeq2 based on n = 3 biological replicates. B The expression levels of 12 genes involved in anthocyanin biosynthesis in radish leaves. Different colors indicate difference in gene expression levels, ranging from high (red) to low (blue), with n = 3 biological replicates. C Gene ontology (GO) enrichment analysis is used to enrich DEGs, with GO functional enrichment using padj less than 0.05 as the threshold for significant enrichment. D Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis of the DEGs, with statistical significance determined by clusterProfiler R package with adjusted P < 0.05 Metabolome analysis of RsLNC2361-pTY and pTY Metabolomic analysis was conducted using the two samples with three biological replicates. The principal component analysis (PCA) graph clearly showed the significant differences in metabolites between the RsLNC2361-pTY and control groups (PTY). The proportions of the two main components, PC1 and PC2, in the fleshy root were 43.43% and 21.96% respectively (Fig. 5A). The larger the PC1 value, the more significant was the difference between the treatment and control groups. The PLS-DA results revealed that the Q2Y value was 0.78, indicating the reliability of the metabolomic data. The results of the liquid chromatography-tandem mass spectrometry (LC-MS/MS) technique and volcano plot showed that 1,358 metabolites in the RsLNC2361-pTY and PTY were identified, among which 72 metabolites were significantly upregulated, including organic oxides, benzene compounds, amino acids, and their derivatives (Supplementary Table S4). In total, 201 metabolites were substantially downregulated, including flavonoids, phenylpropanoids, polyketones, and methylated flavonoids (Fig. 5B). Five main categories was organic acids and their derivatives (22.09%), phenylpropyl and polyketone compounds (20.62%), and lipids and lipid-like molecules (14.65%), organoheterocyclic compounds (12.0%), organic oxygen compounds (11.34%) (Fig. S2). We performed logarithmic transformation on the fold-change values of the differential metabolites to the base of 2 and selected the top 20 metabolites (Fig. 5C). Further analysis of KEGG pathway enrichment indicated that differential metabolites revealed significant accumulation in pathways such as flavonoid biosynthesis, zeatin biosynthesis, phenylpropanoid biosynthesis, and flavone and flavonol biosynthesis (Fig. 5D). Metabolome analysis of RsLNC2361-pTY and pTY. A Principal Component Analysis (PCA) in pTY and RsLNC2361-pTY was performed using n = 3 biological replicates, with different colored scatter points representing samples from different experimental groups and ellipses representing 95% confidence intervals. B Volcanic maps of differential metabolites in pTY and RsLNC2361-pTY, with n = 3 biological replicates for metabolome detection and differential metabolites screened by VIP > 1 and P < 0.05. C Matchstick chart of differential metabolites between pTY and RsLNC2361-pTY, with blue represents down-regulation and red represents up-regulation. The length of the rod represents the size of log2(Fold Change). The size of the dot represents the size of the VIP value, with n = 3 biological replicates. D Enrichment analysis of differential metabolites conducted by the KEGG analysis, with statistical significance determined by P < 0.05 based on n = 3 biological replicates Correlation analysis between the metabolome and transcriptome We conducted a correlation analysis of the transcriptome and metabolome to further reveal the regulatory mechanism of RsLNC2361 in radish anthocyanin accumulation. The results revealed that 2,871 down-regulated differential genes, 2,161 up-regulated differential genes, 201 downregulated metabolites, and 72 upregulated metabolites were present (Fig. 6A). Two-way orthogonal partial least squares (O2PLS) can be used for bidirectional modeling and prediction in two-omics. The degree of association between the two omics datasets and the primary metabolites and genes responsible for this association were determined. The analysis results presented the DEGs and metabolites associated with the association of the two omics and displayed the top 10 genes and metabolites in terms of contribution (Fig. 6B). These are isoniipecotic acid, L-pipecolic acid, p-hydroxycinnamic acid, 3-phenylbutyric acid, succinic Acid, novel. 887, Rsa10028171, Rsa10002061, Rsa10031689, and Rsa10033817 (Fig. 6B). Correlation analysis was conducted on the DEGs and metabolites to screen the top 50 differentially expressed metabolites (ranked by p-value from smallest to largest) and the top 50 DEGs (ranked by p-value from smallest to largest), obtaining a correlation heat map showed the correlation between the expression levels of DEGs and metabolites. The red color indicated the stronger the positive correlation, and the bluer color indicated the stronger the negative correlation (Fig. 6C). KEGG association analysis revealed that the biosynthesis of phenylpropanoid biosynthesis, biosynthesis of cofactors, tryptophan metabolism alpha-Linolenic acid metabolism, and flavonoid biosynthesis were substantially enriched (Fig. 6D). Correlation analysis between metabolome and transcriptome. A Analysis of transcriptome and metabolome expression regulation based on n = 3 biological replicates for both transcriptome and metabolome sequencing. B O2PLS data integration analysis was conducted to present the top 10 genes and metabolites in terms of contribution, with data from n = 3 biological replicates. C Correlation analysis of differential genes and differential metabolites. The redder the color, the stronger the positive correlation; the bluer the color, the stronger the negative correlation. The flatter the ellipse, the higher the absolute value of the correlation. A correlation statistical test with a P value < 0.05 will be marked with an asterisk (*), with all data based on n = 3 biological replicates. D KEGG enrichment analysis was conducted based on the same comparison pair of metabolisms and genes, with statistical significance determined by P < 0.05 and n = 3 biological replicates Validation of DEGs via qRT-PCR To verify the reliability of the RNA sequencing results, we randomly selected 7 differentially expressed genes for RT-qPCR analysis. These differentially expressed genes of Rsa10013670 (RsHY5), Rsa10004324 (RsMYB308), Rsa10033817, Rsa10005513, Rsa10022856, Rsa10022857, and Rsa10018160 showed relatively low expression in pTY-RsLNC2361 plants when compared with the control pTY. Trends in the RNA sequencing data were highly consistent with those in the RT-qPCR data (Fig. 7). These results verified the reliability of the RNA sequencing results. The gene expression level of seven DEGs in RsLNC2361-pTY and pTY by qRT-PCR. Data are presented as mean ± standard deviation (SD) with n = 3 biological replicates , and statistical significance was determined by Student’s t-test. Asterisks indicate significantly different values (*P < 0.05 and **P < 0.01) Discussion In recent years, chronic diseases such as cancer, diabetes, and cardiovascular and cerebrovascular diseases have gradually become major threats to human health and safety. Anthocyanins have several functions that are beneficial to human health, including antioxidant, anti-inflammatory, and antitumor activities [31]. Radish varieties with rich anthocyanins possess good appearance and high nutritional value. Based on our previous LncRNA sequencing, the RsLNC2361 was differentially expressed in white and red skin radish. In this study, we confirmed that RsLNC2361 was positively regulate anthocyanins in radish. Furthermore, transcriptome and metabolome association analyses was conducted to explore differentially expressed genes and metabolites, providing a theoretical basis for the molecular mechanism by which LncRNA regulates anthocyanin accumulation. As a key component of non-coding RNAs, lncRNAs regulates gene expression at the transcriptional, post-transcriptional, and epigenetic levels and play a critical role in animals and plants [32]. The regulatory mechanism of the competitive endogenous RNA (ceRNA) module in plants, which is involved in various biological processes, has been widely reported [33]. In apples, MLNC3.2 and MLNC4.6 act as endogenous target mimics (eTMs) of miR156a by direct complementary interactions with miR156a, thereby inhibiting miR156a-mediated cleavage of SPL2-like protein and SPL33, and participating in SPL-mediated anthocyanin accumulation [34]. Silencing of lncRNA81246 can enhanced the resistance of tea plants to pathogens, whereas tea plants transiently overexpressing lncRNA81246 showed reduced pathogen resistance. Moreover, lncRNA81246 can efficiently bind to miR164d and weaken its silencing effect on the CsNAC1 gene [35]. CsmiR395 target CsSULTR2;1 positively regulated sulfate metabolism and enhanced disease resistance in tea plant [33]. In this study, we transient overexpression of RsLNC2361 in radish leaves markedly increased anthocyanin accumulation, whereas silencing of RsLNC2361 resulted in a substantial reduction in anthocyanin content in the roots of red-fleshed radishes. To determine which miRNA were targets the RsLNC2361, all miRNA genes from red and white skin radish were selected to predicted potential miRNA by the psRNATarget microRNA analysis website tool (https://www.zhaolab.org/psRNATarget/analysis). The results found that the miRNA858 was target RsLNC2361, with expectation values of 4.0. MiR858 target MsMYB62-like positively regulate anthocyanin content under the low-nitrogen condition in Malus spectabilis [36]. The mdm-miR858 targets the MdMYB9 and MdMYBPA1 to regulates anthocyanin accumulation in apple [37]. Therefore, we speculate that miRNA858 targets the RsLNC2361, thereby participating in the molecular mechanism of regulating anthocyanins accumulation in radish. The rapid development of high-throughput sequencing technology in recent years, especially the combined analysis technology of omics has provided a powerful tool for exploring complex intracellular metabolic processes [38]. The metabolomic and transcriptomic analysis found that the strong red color of red pericarp longan originates from the cyanidin derivative accumulation in its pericarp, and the F3’H and F3’5’H genes might play a crucial role in selecting which anthocyanin components to synthesize [39]. The specific mechanisms of AME degradation in the phase I and phase II reaction pathways (such as hydroxylation and dealkylation) were revealed via transcriptomic and metabolomic association analyses [40]. In Bletilla striata, metabolomics and transcriptomics association analyses have confirmed that BsPP2C17 and BsPP2C27, as positive regulatory factors of ABA signaling, are involved in regulating the germination process [41]. Transcriptome analysis showed that 263 transcription factors are involved in anthocyanin biosynthesis in pear peels. Additionally, previous study indicated that PpPIF8 gene overexpressed in pear peels and callus promoted anthocyanin biosynthesis and significantly upregulated the expression levels of related genes [42]. Transcriptome analysis was conducted on six legume varieties with differently colored seed coats, and 21 candidate genes associated with the color of legume seed coats were identified [43]. Transcriptomic research has shown that genes in the flavonoid biosynthesis pathway of Chinese cabbage interact with those related to defense and stress mechanisms [44]. Through a comprehensive analysis of the transcriptome and metabolome data of natural red and white rose varieties, nine anthocyanin glycosides were identified [45]. A combined transcriptomic and metabolomic analysis was conducted on five tea tree varieties with varying petal colors, and 187 flavonoids were identified [46]. Transcriptional and metabolic analyses were conducted on the immature pods of the two cowpea varieties, a total of 171 flavonoid metabolites and 4,142 DEGs were detected [47]. Similarly, in our transcriptome data analysis, 5,032 DEGs were identified, including anthocyanin-related structural genes (RsUFGT, RsF3H, RsDFR, RsTT8, and RsCHS) and RsMYB genes. Through metabolomic data analysis, 273 differential metabolites were identified, of which 72 were upregulated and 201 were downregulated. Moreover, the flavonoid and phenylpropane metabolic pathways were markedly enriched in the anthocyanins synthesis of radish. The expression levels of these DEGs and metabolites significantly differed with variations in anthocyanin content. Through association analysis, the top 10 differential metabolites and differential genes were screened, showing pairwise correlations between the top 50 differential metabolites and the top 50 differential genes. Additionally, flavonoid biosynthesis was significantly enriched in the anthocyanin anabolic pathway. These results indicated that RsLNC2361 is involve in regulating anthocyanin accumulation. Conclusions In this study, overexpression of RsLNC2361 promoted anthocyanin accumulation in radish leaves, whereas silencing RsLNC2361 reduced anthocyanin accumulationn in radish flesh root, indicating RsLNC2361 positively regulated anthocyanin accumulation in radishes. Overexpression of RsLNC2361 increased the expresssion anthocyanin-related structural genes, whereas silencing of RsLNC2361 inhibited its expression. Combined metabolomic and transcriptomic analyses indicated flavonoid biosynthesis and phenylpropanoid biosynthesis were markedly enriched. The miRNA858 was predicted to target RsLNC2361. These results suggest that RsLNC2361 plays a crucial role in radish anthocyanins accumulation. Supplementary Information Supplementary Material 1. Supplementary Material 2. Acknowledgements Not applicable. Authors’ contributions ZW participated in the conceptualization and design of this study. LH performed experiments, data collection, and analysis and wrote the first draft of the manuscript. LX provided experimental guidance and helped revise the article. YG and JX combined and optimized the picture. LD and MM helped to modify the language. XX, PY, WL assisted in the experiment. All authors reviewed the manuscript. Funding This research was supported by the National Natural Science Foundation of China (Project No. : 32360755), Guizhou Modern Agriculuture Research System (GZMARS)-Plateau characteristic vegetable industry, Key core technology research project for mountainous agriculture in Guizhou Province (GZNYGJHX-2023013). Data availability All sequencing raw data have been submitted to the National Center for Biotechnology Information (Bioproject accession: PRJNA1377310). Declarations Ethics approval and consent to participate Not applicable. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Supplementary Materials Supplementary Material 1. Supplementary Material 2. Data Availability Statement All sequencing raw data have been submitted to the National Center for Biotechnology Information (Bioproject accession: PRJNA1377310).