Blue light induces leaf color change by modulating carotenoid metabolites in orange-head Chinese cabbage (Brassica rapa L.ssp.pekinensis)

ZHANG Rui-xing, ZHANG Ni-nan, WANG Ya-xiu, Khan ABID, MA Shuai, BAI Xue, ZENG Qi, PAN Qi-ming, LI Bao-hua#, ZHANG Lu-gang#

1 State Key Laboratory of Crop Stress Biology for Arid Areas, College of Horticulture, Northwest A&F University, Yangling 712100,P.R.China

2 Department of Horticulture, The University of Haripur, Haripur 22620, Pakistan

Abstract Carotenoids are involved in the formation of plant leaf color as well as photosystem photoprotection.This study showed that blue light significantly induced up-regulation of the total carotenoid content in the inner leaves of orange-head Chinese cabbage (OHCC).Furthermore, the transcriptomic analysis revealed that blue light treatment induced upregulation of genes in photosynthesis (BrHY5-2, BrCOP1 and BrDET1) and the methylerythritol 4-phosphate pathways(BrGGPS, BrDXS and BrHDR) upstream of the carotenoid metabolic pathway.Carotenoid metabolomic analysis revealed that the accumulation of several orange and red carotenoids (lycopene, zeaxanthin, β-carotene, lutein, and β-cryptoxanthin) after blue light treatment contributed to the deepening of the leaf coloration, suggesting that short-term blue light treatment could be used to boost nutritional quality.The light signal gene BrHY5-2 participated in the blue light-induced transcriptional regulation of carotenoid biosynthesis in OHCC.Overexpression of BrHY5-2 in Arabidopsis significantly increased the total carotenoid content and the sensitivity to blue light.The above findings revealed new insights about blue-light-induced carotenoid synthesis and accumulation in OHCC lines.They suggested a new engineering approach to increase the nutritional value of vegetables.

Keywords: orange-head Chinese cabbage (OHCC), carotenoid, nutrition, blue LED light, secondary metabolite,transcriptome

1.Introduction

Carotenoids are a large group of plant secondary metabolites present across the plant kingdom.Representative carotenoids generally absorb blue and green light and provide coloration to leaves ranging from yellow-orange to red.Carotenoids are also important nutrients, and the colors they provide are highly appealing to consumers (Rodriguez-Concepcionet al.2018).The sensory index of deep orange and red could boost vegetable product demand.Furthermore, previous research has shown that a higher level of carotenoids in deep orange cassava prevents vitamin A deficiency(VAD), resulting in greater nutritional and health benefits for consumers (Atunaet al.2021).The inner leaves of orange-head Chinese cabbage (OHCC) are rich in carotenoids, providing higher nutritional value than normal white-head Chinese cabbage (Watanabeet al.2011;Zhang Jet al.2015).

Light can induce carotenoid biosynthesis in plants, and notably, blue light has been demonstrated to increase the expression of carotenoid biosynthesis genes and stimulate the accumulation of carotenoids in apricots(Martyet al.2005).Blue light is also vital in pigment deposition regulation (Kopsell and Sams 2013), and shortterm blue light treatment could considerably boost the contents of β-carotene, violaxanthin, and total lutein cycle pigments in the stems of pre-harvest cauliflower seedlings compared with a mixed treatment of red and blue light(Kopsell and Sams 2013).Notably, lycopene is uniquely accumulated in OHCC, as it is not present in white head Chinese cabbage (Watanabeet al.2011).Lycopene and β-carotene have been shown to contribute significantly to the color of the inner orange leaves (ZhangJet al.2015).

Transcription regulation is the primary regulatory mechanism for carotene production, and the ELONGATED HYPOCOTYL5 (HY5) of bZIP transcription factor family could induce the accumulation of carotenoids in tomato fruits (Liuet al.2004).Through phy-interacting factors (PIF) and HY5 transcription factors, light signals can also directly target the methylerythritol 4-phosphate(MEP) pathway in the chloroplast (Chenge-Espinosaet al.2018).Furthermore, the light-responsive transcription factorsHY5andPIF1control the expression of carotenoid biosynthesis genes (Toledo-Ortizet al.2010).Transcription abundances of MEP pathway-related genes are induced by light treatment, and the main regulatory factors of light signal transduction, HY5 and PIFS, interact directly with the upstream elements of the flux-controlling genes 1-deoxy-D-xylulose 5-phosphate synthase (DXS1),1-deoxy-D-xylulose 5-phosphate reductoisomerase(DXR), and 4-hydroxy-3-methylbut-2-enyl-diphosphate reductase (HDR).Meanwhile, the MEP pathway can effectively increase the metabolic flux of carotenoids and isoprene (Cordobaet al.2009).However, the molecular pathways linking light signal perception and regulation of carotenoid gene expression are still poorly understood.

The mechanism of changes in photochemical substances, including carotenoids under varying light conditions has remained unknown, while research largely has been limited to broccoli (Kopsell and Sams 2013),citrus (Zhanget al.2012), and tomato (Xieet al.2019).Little is known about the transcriptional and metabolic changes in OHCC across different light conditions.

Although we have previously studied the genetic basis of theBr-orOHCC mutant and cloned it (Zhanget al.2013), the regulatory mechanisms of carotenoid accumulation in OHCC are still unknown.The inner leaves of Chinese cabbage change from yellow to orange after exposure to sunlight for 10 min (Suet al.2015).A recent study has shown that blue light treatment induces the accumulation of lycopene content, while the contents of carotenoids do not significantly change in heading leaves of OHCC ‘14za1’ after red, yellow, and green light treatments (Ma 2020).In our current study,the leaves of mature OHCC were treated with blue light and then examined.This study aimed to (i) identify the carotenoid metabolites that affect the color change of Chinese cabbage leaves under blue light, (ii) compare the differences in carotenoid metabolism responses of different Chinese cabbage genotypes under blue light,and (iii) explore the role of the transcription factor BrHY5-2 in the regulation of carotenoid metabolism.Our work provides unique insights into the molecular mechanisms of carotenoid accumulation in vegetables.Also, it suggests that blue LED light could be used to effectively transform leaf color of OHCC as a powerful tool for nutrition enrichment.

2.Materials and methods

2.1.Plant materials and growth conditions

Three OHCC inbred lines ‘14S837’, ‘15S1094’, and‘20S530’ were initially incubated in growth chambers(16/8 h, 25/20°C, day/night) and then were planted in an experimental farm of the Horticultural Center,Northwest A&F University, Yangling, Shaanxi Province,China (108°1′E, 34°18′N(xiāo)) in mid-August 2020, with row and plant spacing of 50 cm and 45 cm, respectively.Standard management procedures were followed during plant growth.The ripe leaf heads were harvested on December 13, 2020, immediately, and details of the subsequent blue light treatment are provided in the next subsection.Arabidopsisecotype Columbia-0 (WT)and transgenicArabidopsislines were raised in a light incubator (GDN-1000D-4, Ningbo Southeast Instrument Co., Ltd., Ningbo, China) in black plastic bowls(7 cm×7 cm) containing peat and vermiculite (3:2, v/v) with a 60% relative humidity and 16 light/8 h dark (22/18°C)photoperiod cycle during the seed germination stage.Each sample was analyzed in triplicate, and three biological replicates were tested.

2.2.Light treatments

The 7th leaf inward from the outermost leaf of the three OHCC lines (‘14S837’, ‘15S1094’, and ‘20S530’) was selected, and blue light (470 nm) treatments with LEDs were conducted, with a constant light intensity (120 μmol m–2s–1).First, control treatments were conducted to investigate the effects of blue light on the inner leaves for 0, 30, 60, 90, and 120 min, respectively.Based on the results of the pre-experiment, we found that the carotenoid content of inner leaves significantly accumulated after 60 min of blue light treatment compared to the other time treatments (Appendix A).Thus, the samples of ‘14S837’,‘15S1094’, and ‘20S530’ were coded as 14DCK, 15DCK,and 20DCK, respectively, as the control at 0 min.Meanwhile, the treated samples of ‘14S837’, ‘15S1094’,‘20S530’ were coded as 14DT, 15DT, and 20DT at 60 min,respectively.The DCK and DT stand for metabolome data processing, CK and T stand for transcriptome data processing, respectively.All experiments were performed with three independent biological replicates.

2.3.Chromaticity values of leaves

Leaf samples were subjected to chromaticity value measurement, and their L＊, a＊, b＊, C, and color contribution index (CCI) values were recorded using a CR-400 Chroma Meter Colorimeter (Konica Minolta, Tokyo, Japan)as reported by Zhouet al.(2020).CCI is defined as follows:

The OHCC chromaticity values indicated various colors, and positive L＊, a＊, b＊, and CCI values indicated orange-red, red-purple, yellow-orange, and red coloration,respectively.In contrast, negative L＊, a＊, b＊, and CCI values indicate the mean of blue, blue-green, yellow-blue,and blue-green coloration, respectively.

2.4.Pigment, soluble solid, total phenol, and flavonoid contents analysis

Total carotenoid contents were calculated from the absorbances at 665, 649, and 470 nm according to the method described by Lightenthaler (1987).Under dark conditions, 0.2 g of fresh leaves were thoroughly ground with 95% (v/v) ethanol with a small amount of CaCO3powder and treated in the dark at 4°C for 12 h.Then,the mixed solution was centrifuged at 5 000 r min–1for 15 min, and the supernatant was measured using an ultraviolet spectrophotometer UV-1800 (Shimadzu, Kyoto,Japan), as Liet al.(2019) reported.The flavonoids and total phenols contents were measured according to the method of Wilson (1985) with slight modification.Briefly,0.5 g samples were ground with 1% (v/v) HCl (Haohua Chemical Reagent Co., Ltd., Luoyang, China) and methanol (Merck, Darmstadt, Germany) in the dark, fully mixed, and extracted for 30 min, after which samples were each filtered into a 15 mL test tube.Then, OD325g–1and OD280g–1values were used to calculate the relative amounts of fla vonoids and total phenols, respectively.The soluble solids of the OHCC leaves were measured with an Atago PAL-1M portable refractometer (Atago Co.,Ltd., Tokyo, Japan) by the degree Brix.The refractometer was calibrated before each measurement with distilled water and wiped with mirror paper.

2.5.Detection of carotenoids by ultra-highperformance liquid chromatography–tandem mass spectrometry (UPLC–MS/MS) analysis

The chemicals, reagents sample preparation, extraction,and metabolite quantitative analysis for UPLC–MS/MS were provided by MetWare Science Biotechnology Co., Ltd.(Wuhan, China), and the metabolite data were analyzed according to the method of Liuet al.(2020).Leaf samples were quickly frozen in liquid nitrogen, and 50 mg samples were weighed.A pre-prepared mixed solution of acetone(Sinopharm Chemical Reagent Co., Ltd., Shanghai,China): ethanol (Merck, Darmstadt, Germany):n-hexane(CNW, Shanghai, China) was added to it, and the extract was vortexes at 20°C for 15 min.After drying samples with nitrogen, methanol:methyl tert-butyl ether (MTBE)(Aladdin Inc., Shanghai, China) was added to redissolve the samples, and finally, the mixed solution was slowly filtered through a 0.22 μm organic filters into a brown vial for further UPLC–MS/MS system analysis according to standard procedures (Zhouet al.2020).Mass spectral data were processed using the software package Analyst 1.6.3 (Sciex,Ontario, Canada).The total ion chromatograms (TIC) and TIC overlay of the QC samples are shown in Appendix B.In addition, the linear equations of the standard curves of various carotenes are listed in Appendix C, and the metabolite names and transition information are listed in Appendix D.The carotenoid substance contents identified in all samples are in Appendix E.

2.6.Transcriptome analysis

Leaves from five individual OHCCs were collected and combined into one biological replicate.Three biological replicates were used for transcriptome analysis.RNA extraction from leaf samples was carried out using the Trizol (Tiangen, Beijing, China) method, and the RNA purity was checked by electrophoresis with a 1% agarose gel (HydraGene, Xiamen, China).The integrity of RNA was examined using a Bioanalyzer 2100 System (Agilent Technologies, Santa Clara, CA, USA), and fastp v.0.19.3 was used to filter the original data, mainly to remove reads with adapters and keep high-quality bases (Q≥20).The reads were then used to create RNA-seq libraries.Libraries were sequenced using the Illumina HiSeq platform (Illumina,San Diego, CA, USA) (Liuet al.2020).The filtered reads were mapped to the reference genome ofB.rapa(version:3.0) (http://brassicadb.cn, accessed on April 5, 2022) (Zhang Jet al.2015), and HISAT v2.1.0 was used to construct the index.DESeq2 (version 1.20.0) was used to identify differentially expressed genes (DEGs) (Loveet al.2014),and the filter condition was |log2(fold change)|≥1.5, while the false discovery rate was (FDR)<0.05.DEGs were subjected to Gene Ontology (GO) enrichment analysis using the GO Seq R package (correctedP<0.05).All raw read data were uploaded to the National Center for Biotechnology Information (NCBI) Short Read Archive under BioProject accession number PRJNA858610.We constructed the coexpression networks using the weighted correlation network analysis (WGCNA, v1.47) package in R (Langfelder and Horvath 2008).

2.7.Generation of BrHY5-2 overexpressing Arabidopsis lines

The open reading frame ofBrHY5-2(438 bp) was cloned from ‘20S530’ Chinese cabbage leaves using specific primers (Appendix F) atXbal andKpnI restriction sites and ligated into the pVBG2307 overexpression vector to generate pVBG2307:BrHY5-2 overexpression plasmid.Agrobacteriumtumefaciensstrain GV3101 was used to obtain transgenic lines (Clough and Bent 1998).The seeds of T0generationArabidopsiswere screened using Murashige and Skoog (MS) (Phytotechlabs, Lenexa, KS,USA) medium containing 50 mg L–1kanamycin followed by PCR verification.The T3generation transgenic seeds were used to generate the data shown.Arabidopsismutant (SALK_096651) was obtained from AraShare(https://www.arashare.cn, accessed on April 15, 2022).TheArabidopsisoverexpression (OE) and mutant lines were identified by primer-specific PCR (Appendix F).BrHY5-2sequences were submitted to GenBank under accession number ‘20S530’ (OP104961).

2.8.Gene expression analysis

First-strand cDNA (1 μg per sample) was synthesized using PrimeScript? Kit (TaKaRa, Dalian, China) according to the manufacturer instructions, and all gene-specific primers used in this study were shown in Appendix G.By using NCBI Primer BLAST (https://www.ncbi.nlm.nih.gov/tools/primer-blast/, accessed on September 17, 2022)primer specificity was confirmed.Quantitative reverse transcription-PCR (qRT-PCR) was performed as described previously (Zhang Jet al.2015).The reaction mixture contained a total volume of 20 μL followed by 7.2 μL of ddH2O, 0.8 μL of gene-specific primers (10 μmol L–1),2 μL of cDNA template (50 ng μL–1), and 10 μL of SYBR qPCR Master Mix (Vazyme, Shanghai, China).The PCR procedure was as follows: initial denaturation at 90°C for 30 s, followed by 40 cycles of 94°C for 15 s,60°C for 30 s and 72°C for 30 s.The relative transcript levels were calculated using the 2-ΔΔCTmethod (Livak and Schmittgen 2001).BrGAPDH(GO0048316) andAtACT2(AT3G18780) were used as internal reference genes in Chinese cabbage andArabidopsisthaliana, respectively.All samples were determined simultaneously for three biological replicates.

2.9.Statistical analysis

All raw data and experiments are expressed as the means of three independent replicates.Data analyses were performed using the R statistical computing environment (www.r-project.org/, accessed on October 21, 2022) except for Duncan multiple range (DMR) tests,which were performed using SPSS 19.0 (version 19.0;IBM Corp., Armonk, NY, USA).The experimental data between various treatments were analyzed by one-way analysis of variance (ANOVA) and Pearson’s correlation tests at a threshold ofP≤0.05.The correlation network diagram was performed using the Metware Cloud, a free online platform for data analysis (https://cloud.metware.cn, accessed on October 25, 2022).

3.Results

3.1.The leaf color and carotenoid content of three representative OHCC lines after blue light treatment

The inner leaf color of ‘14S837’, ‘15S1094’, and ‘20S530’were yellow, as shown in Fig.1-A.In the detached leaf assay, the color of the seventh inner leaf of three OHCC lines changed from yellow to orange after blue light treatment for 60 min; the leaves of ‘14S837’ and ‘15S1094’changed to slightly orange, while the leaves of ‘20S530’changed to a deeper orange, as showed in Fig.1-B.The highest carotenoid content (1 014.07 μg g–1dry weight(DW)) was observed in line ‘20S530’, followed by line‘14S837’ (572.16 μg g–1DW) and line ‘15S1094’ (383.41 μg g–1DW), consistent with their leaf colors as shown in Fig.1-C.After 60 min of blue light treatment, the carotenoid contents of ‘20S530’, ‘14S837’ and ‘15S1094’ lines reached 1 221.36, 605.30, and 393.18 μg g–1DW, increases of 20.44, 5.79, and 2.55%, respectively (Fig.1-C).The total soluble solid, total phenol and flavonoid contents were also increased in the leaf heads of all the OHCC lines when treated with blue light (Fig.2-A).Unlike carotenoids,the basal soluble solid content of the ‘15S1094’ line was higher than that of the ‘20S530’ and ‘14S837’ lines, while this increase in the soluble solid content of the ‘20S530’,‘14S837’, and ‘15S1094’ lines was 7.84, 4.49, and 4.64%respectively, after blue light treatment.The total phenol content of the ‘20S530’, ‘14S837’, and ‘15S1094’ lines increased by 15.84, 12.12, and 7.14%, respectively, while their flavonoid content increased by 22.92, 12.89, and 7.82%, respectively (Fig.2-B and C).The color of ‘20S530’was the brightest after blue light treatment, followed by‘14S837’ and ‘15S1094’ (Fig.2-D).The leaf color of‘20S530’ turned more quickly to yellow-orange after blue light treatment than did that of ‘14S837’ and ‘15S1094’ (Fig.2-E, F and G).After 60 min of blue light treatment, the CCI values of all three lines became positive, indicating that the color changed from sightly green to orange (Fig.2-H).Moreover, the correlation coefficient between the total carotenoid content and CCI was 0.85 (Appendix A).This suggests that blue light could significantly induce color changes of the OHCC lines ‘20S530’ and ‘14S837’ from light yellow to orange, and ‘15S1094’ from yellowish to light orange.

3.2.Differential carotenoid metabolites (DCMs) accumulated in the head leaf of OHCC treated with blue light

Fig.2 Physicochemical indices of the orange-head Chinese cabbage (OHCC) lines under blue light treatment for 60 min.A, contents of soluble solids.B, contents of total flavonoid.C, contents of total phenolic.D–H, chromaticity values (L＊, a＊, b＊,C, and color contribution index (CCI)).Values presented are the mean±SD (n≥3).Different letters using Duncan’s test represent significance at P<0.05 levels.

To further understand the dynamic changes of carotenoid metabolite flux during the inner leaf color change in OHCC, the carotenoid components were extracted and analyzed by UPLC-MS/MS analysis from the head leaves of ‘14S837’, ‘15S1094’, and ‘20S530’ lines after blue light treatment (Fig.3).In the leaf heads of ‘14S837’,‘15S1094’, and ‘20S530’, 32, 31,and 31 carotenoid components were detected, respectively, including 3 lycopene metabolites (component 1), 22 β-carotene branch products(component 2) and 8–9 species of α-carotene branch products(component 3) (Appendix F).The maximum difference between the 15DCK and 20DCK was 2.64 times,while the content of component 1 was 959.1000 μg g–1fresh weight (FW),528.1300 μg g–1FW, and 365.8200 μg g–1FW, which accounted for 94.32,92.02, and 95.30% of the total carotenoid content, respectively.The carotenoid synthesis mechanism of the three OHCC lines was basically the same, and the carotenoid components that differed most among the three OHCC lines were phytoene and lycopene in component 1,corresponding to a difference between the three varieties of 5.9 and 1.6 times.

The effects of blue light differed among steps of carotenoid production in the three OHCCs (Appendix H).After blue light treatment in 20DT, components 1, 2, and 3 were increased by 13.58, 142.48, and 87.28% respectively.However, in 15DT, component 1 decreased by 4.11%, while components 2 and 3 increased by 146.18 and 83.26%,respectively; in 14DT, components 1 and 2 increased by 2.63 and 43.27%,respectively, while component 3 decreased by 0.69%.These results suggested that blue light treatment activates all three pathways of carotene synthesis in 20DT but mainly activates two downstream branch pathways (components 2 and 3) in 15DT and one pathway of the downstream β-carotene branch(component 2) in 14DT.Further analysis of the specific carotenoid composition analysis showed that phytoene and lycopene were significantly increased (by 5.85 and 24.63% respectively); similarly,the β-carotene, zeaxanthin, and antheraxanthin of the β-carotene branch were increased by 512.11, 708.29,and 722.63% respectively; meanwhile, the contents of β-cryptoxanthin and lutein of the α-carotene branch were increased by 287.27 and 537.01%, respectively, in 20DT.The significantly increased components of 14DT were β-carotene, zeaxanthin, and antheraxanthin in the β-carotene pathway branch, which were increased by 176.03, 334.46, and 449.14%, respectively, while the β-cryptoxanthin and lutein in the α-carotene branch were increased by 170.83 and 325.98%, respectively.In contrast to the above results, the upstream metabolites phytoene and phytofluene of the main carotene branch in 15DT leaves were reduced by 16.20 and 32.61%,respectively, after blue light treatment.Blue light also triggered the xanthophyll cycle, including increases in the contents of β-cryptoxanthin, lutein, β-carotene,zeaxanthin, and antheraxanthin in the leaf of all the OHCC lines (Fig.3-B).Because different carotenoid components have different colors (Appendix I), our findings showed that complex carotenoid components together contributed to the color change in the OHCC lines caused by blue light treatment.Notably, carotenoid esters, including zeaxanthin-dilaurate, zeaxanthindimyristate, and β-cryptoxanthin-myristate, were downregulated in ‘14S837’ and ‘20S530’ (Fig.3-B).

Fig.3 Heatmap of the relative differences in the carotenoid metabolites profiles among the three orange-head Chinese cabbage(OHCC) lines.A, the heatmap of all types of carotenoid components normalized in OHCC lines.The red row shows the higher carotenoid content; the green row indicates relatively a lower carotenoid content.B, heat maps of differential carotenoid metabolites(DCMs) in 14DCK vs.14DT, 15DCK vs.15DT, and 20DCK vs.20DT pair comparisons.14, 15 and 20, represent the samples of‘14S837’, ‘15S1094’, and ‘20S530’, respectively.DCK (0 min) and DT (60 min) stand for metabolome data processing.

3.3.Transcriptomic analysis of OHCC lines treated with blue light

Fig.4 Transcriptomic data analysis in differently processed samples.A, correlation heat map of the different treatment samples.B, PCA analysis of 18 sample scores diagram transcript profiles from different blue-treatment groups.C, Venn diagram showing differentially expressed genes (DEGs) of three pair samples.D, heatmap of 1 762 common DEGs between 14CK and 14T, 15CK and 15T, 20CK and 20T.14, 15 and 20, represent the samples of ‘14S837’, ‘15S1094’, and ‘20S530’, respectively.CK (0 min) and T (60 min) stand for transcriptome data processing.E, KEGG classification of 1 762 common DEGs in three treatment samples.

Transcriptomic analysis was used to further determine the molecular mechanisms by which blue light-induced leaf color changes in OHCC lines.A total of 120.38 Gb of high-quality sequencing data from 18 samples were obtained.The clean reads of each sample reached 6 Gb of mapped data, ranging from 87.74 to 89.38%, and the Q30 base percentage was 92% or higher (Appendix J).The data in Fig.4-A exhibited a significant correlation between the different treatments in OHCC lines.In the PCA plot of transcript profiles of 18 samples, PC1 and PC2 explained 29.07 and 20.23% of the variance in gene expression, respectively (Fig.4-B).Furthermore,the heatmap diagram of 1 762 common DEGs further confirmed the PCA results that 14CK and 20CK together and 14T and 20T together were clustered,respectively (Fig.4-C and D).There were 4 400 DEGs(2 476 and 1 924 up- and down-regulated, respectively)in the 14CKvs.14T comparison, 3 839 DEGs (2 377 and 1 462 up- and down-regulated, respectively) in the 15CKvs.15T comparison, and 4 351 DEGs (2 484 and 1 867 up- and down-regulated, respectively) in the 20CKvs.20T comparison (Appendix K).The KEGG enrichment analysis of the DEGs revealed that the major enriched pathways, such as secondary metabolic biosynthesis,photosynthesis, photosynthesis-antenna proteins, and carotenoid biosynthesis were affected by the blue light treatment (Fig.4-E).The KEGG enrichment analysis also showed that the photosynthetic pathways (photosystem I, photosystem II, photosynthetic electron transport, and F-type ATPase) were the most enriched pathways (P<0.05)(Appendices L and M).On the other hand, a total of 103(3.32%) DEGs were involved in the secondary metabolite biosynthetic process in the 14CKvs.14T comparison,followed by the 15CKvs.15T comparison, in which 88(3.29%) DEGs were found to be involved in the secondary metabolite biosynthetic process, and out of them, 33(1.23%) DEGs responded to the blue light.However, in the 20CKvs.20T comparison, a total of 113 (3.71%) DEGs were identified to be involved in the secondary metabolite biosynthetic process, and out of them, 71 (2.33%) were found to be involved in the pigment metabolism process(Appendix N).Overall, these findings revealed that most DEGs are involved in photosynthetic signal transduction and secondary metabolism, implying that blue light may activate the light signaling pathway and stimulate the accumulation of carotenoid metabolites.

3.4.WGCNA identified carotenoid-related DEGs

To explore the effects of dynamic regulation of gene networks on carotenoid synthesis in the head leaves of OHCC lines, we used weighted gene co-expression network analysis to identify carotenoid-related DEGs.These DEGs were clustered into 32 main branches, with each branch representing a module and marked in a different color in Fig.5-A.The DEGs in the carotenoid biosynthesis pathways are mainly in the lightsteelblue1,darkgrey, darkturquoise, steelblue, darkred, and bisque4 modules.A highly positive coefficient was observed for antheraxanthin (0.85), lutein (0.90), zeaxanthin (0.83),β-carotene (0.88), and β-cryptoxanthin (0.91) with the darkturquoise module (Fig.5-B).Notably, the darkgrey module had a strong negative correlation with carotenoid esters, including β-cryptoxanthin myristate (–0.73),zeaxanthin dilaurate (–0.64), zeaxanthin-laurate-myristate(–0.66), and zeaxanthin dimyristate (–0.63).Meanwhile,the darkgrey module had a highly positive coefficient with antheraxanthin (0.68), lutein (0.62), zeaxanthin (0.78),β-carotene (0.58), and β-cryptoxanthin (0.57) (Fig.5-B).According to KEGG enrichment analysis, genes in both the darkgrey and darkturquoise modules were mostly involved in the synthesis and transport of primary and secondary metabolites (Figs.5-C and 6-D), including the two most prominent pathways involved in photosynthesisrelated pathways in leaf color regulation (Appendix O).

3.5.Effects of blue light on the carotenoid metabolic flux pathway and light signal-related genes

Blue light can affect the expression of light signal-related genes, promoting the expression of light signal activators,thus suppressing the expression of transcription repressorsBrCRY1(BraA09g029430.3C),BrPIF1(BraA07g000720.3C), andBrPIF3-1(BraA06g006410.3C)(Appendices P and Q).In particular,BrHY5-2(BraA02g003870.3C) was also highly positively correlated with antheraxanthin (r=0.88,P=1.72E–06), lutein (r=0.82,P=3.7E–05), and zeaxanthin (r=0.91,P=2.03E–07)(Fig.5-E).The fold changes of gene expression ratios measured by RT-qPCR and RNA-seq data were positively correlated (R2=0.7346), supporting the accuracy of the transcriptome data (Fig.6).MEP pathway genes upstream of the carotenoid pathway were induced by blue light (Appendices P and R).The expression level ofBrCOP1(BraA05g011840.3C) was significantly positively correlated with the contents of antheraxanthin(r=0.91,P=1.15E–07), lutein (r=0.86,P=3.53E–06),zeaxanthin (r=0.93,P=7.9E–09) and β-carotene (r=0.81,P=4.62E–05).Similarly, the expression levels ofBrDET1(BraA03g027150.3C) were highly correlated with the contents of antheraxanthin (r=0.86,P=3.73E–06),lutein (r=0.80,P=6.46E–05) and zeaxanthin (r=0.887,P=9.62E–07).Blue light further caused significant transcriptional up-regulation of carotenoid synthesisrelated genes such asBrPSY1,BrCRTISO,BrLYCB, andBrLYCE, which led to the upregulation of related genes in the xanthophyll metabolism pathway (Fig.6).In summary,these results indicated thatBrHY5may promote the accumulation of carotenoids in the inner leaves of OHCC,and further contribute to the observed change of leaf color under blue light treatment.

3.6.Overexpression of BrHY5-2 promotes carotenoid metabolism in Arabidopsis under blue light treatment

Arabidopsis transgenic lines overexpressingBrHY5-2(OE2 and OE5) were used for blue light treatment experiments (Fig.7-A; Appendix S).The total carotenoid content in transgenic lines OE2 and OE5 was higher than that in Col-0, and theAthy5mutant accumulated a lower level of total carotenoid content, as expected.Notably, only the total carotenoid content of OE2 and OE5 were increased under blue light treatment(Fig.7-B).Meanwhile, the transcriptional abundances of the carotenoid-related genesAtLCYB,AtLCYE,AtZEP,AtCCD4,AtZDS,AtPDS, andAtPSYinArabidopsisOE5 were significantly higher than those in WTArabidopsis(Fig.7-C).After 60 min of blue light induction, the transcript abundances ofAtPSY,AtZDS,AtLCYE,andAtZEPin both OE2 and OE5Arabidopsislines were significantly higher than those in WTArabidopsis(Fig.7-D), and especiallyAtPSYandAtZEPwere upregulated, by approximately five-fold relative to WTArabidopsis.These results suggested thatBrHY5-2may be involved in carotenoid metabolism by inducingAtPSYandAtZEPexpression under blue light conditions inArabidopsisthaliana.

4.Discussion

Fig.6 The expression profile of light signal-related genes and carotenoid synthesis genes under the blue light treatment.Scatter diagrams indicate correlation analysis between RNA sequencing data and qRT-PCR data results.The RNA-seq result was obtained using log2(fold change) measurements of the 17 genes.Values represent the mean±SD from three biological replications.Different letters using Duncan’s test represent significance at P<0.05 levels.

Chinese cabbage is an important leafy vegetable, especially in Asia, where malnutrition, including vitamin A deficiency,threatens the health and lives of millions of people,including infants and children.Enhancing the nutritional value of vegetables proves to be one of the most effective and affordable ways to battle malnutrition worldwide(Rodriguez-Concepcionet al.2018).Here, we studied how the nutritional value of OHCC could be further enhanced by blue light treatment.The shifting of the leaf colors in these cultivars after blue light treatment not only provides more nutrients but also makes vegetables more attractive to consumers.The molecular mechanisms uncovered in this study should be valuable for both plant breeders as well as researchers studying postharvest processes to produce vegetables with high nutritional value (Fig.8).

4.1.The different carotenoids accumulated in the inner leaves of different OHCC lines

OHCC lines are valuable germplasm resources with unique health-promoting compounds (Watanabeet al.2011;Zhanget al.2013; Zhang Jet al.2015).The differences in the inner leaf color of OHCC lines may be owing to differences among genotypes (Parket al.2019).Similarly,the accumulation of secondary metabolites and individual primary metabolism are frequently affected by differences in gene regulation (Kirket al.2012; Parket al.2020).

The three OHCC lines in this study have different colors,with differential accumulation of carotenoid components that result in the darkest orange color occurring in ‘20S530’,followed by the ‘14S837’ and ‘15S1094’ OHCC lines(Fig.1).We discovered that the content of red, orangecolored γ-carotene, β-cryptoxanthin, and apocarotenoid in ‘20S530’ heads was higher than that in ‘14S837’ and‘15S1094’ heads, which might contribute to the orange color of ‘20S530’ (Appendix G).According to previous research, β-cryptoxanthin and apocarotenoid are orange or yellow-red pigments and increase color deepening(Sumiasihet al.2018; Zhouet al.2020).Moreover, lutein,antheraxanthin, and zeaxanthin were significantly upregulated (Fig.3), and the accumulation of other various carotenoids also deepens the color of leaves (Appendix I).Obviously, the accumulation of these carotenoid components contributed to the color formation of the inner leaves of OHCC lines.

4.2.Blue light-induced carotenoid accumulation in the inner leaves of OHCC lines

In our study, the color of the inner leaves of the OHCC lines changed from light yellow to dark orange after blue light treatment (Fig.2).The lycopene content reached 501 μg g–1FW in 20T, an increase of 24.63%.Notably,our results also demonstrated that blue light induced a rapid up-regulation of β-carotene, lutein, and zeaxanthin in the ‘14S837’, ‘15S1094’, and ‘20S530’ lines (Appendix H).Zhang Let al.(2015) reported that 100 μmol m–2s–1blue light increased the content of carotenoids in mandarin oranges, and this process was beneficial to the accumulation of lycopene and shunting of the β-carotenoid pathway (Estebanet al.2015), making it a potential source of increased levels of vitamin A and antioxidants(Rodriguez-Concepcionet al.2018).Notably, blue light treatment induces echinenone accumulation in OHCC lines(Appendix H).Despite its low content, echinenone has high market potential owing to its role as an edible orange pigment (Matsuuraet al.2012), and similar induction was partially shown inArabidopsismutantch1(Ramelet al.2013).Our findings also revealed that short-term blue light treatment induces the accumulation of numerous bioactive substances (carotenoids, phenolics, flavonoids,and soluble solids) in orange Chinese cabbage leaves(Fig.2).Kopsellet al.(2014) reported that blue light can also induce soluble solids, total phenols, and flavonoids to be up-regulated, causing significant changes in a number of physiological and biochemical processes in sprouted broccoli microgreens; additionally, the accumulation of these substances also helps to improve the palatability of vegetables to consumers (Klee and Giovannoni 2011),which enhances the health benefits of OHCC.Blue light treatment significantly reduced rubixanthin-laurate,zeaxanthin-dilaurate, zeaxanthin-laurate-myristate, and β-cryptoxanthin-palmitate in the ‘14S837’ and ‘20S530’lines (Fig.3-B; Appendix T).According to previous studies,the carotenoid esterification reaction helps in the isolation of carotenoids as well as their stability and accumulation in the chromatophore (Enfissiet al.2019; Lewiset al.2021).Our study indicated that blue light may consume esterified carotenoids, lessening the damage to plants subjected to short-term blue light treatment.

Fig.8 Effects of blue light on carotenoid components, light signal-related genes, and carotenoid metabolism-related genes in orangehead Chinese cabbage-head (OHCC) lines.This carotenoid metabolic flux pathway was constructed based on the KEGG-enriched pathways and published references (Appendices E and J).Enzymes are shown in bold fonts, while the key carotene and xanthophyll are shown in the red-orange yellow box.

4.3.Blue light-induced photosynthesis and the MEP and carotenoid metabolic pathways in OHCC leaves

Light is the primary stimulus inducing the transcription of genes involved in photosynthesis (Larkin and Ruckle 2008), while carotenoids are part of an important lightharvesting pigment-protein complex and also serve as a blue light filter (Jahns and Holzwarth 2012).Our KEGG enrichment analysis results show that a variety of photosynthesis-related genes, especially photosystem II PSB27 and PSB28, are rapidly up-regulated (Appendices L, M, and U), acting as a molecular switch to provide a role for drastic changes during carotenoid biosynthesis(Sch?ttler and Tóth 2014).Furthermore, photosystem activation has the ability to improve carbon pool energy storage, favoring the biosynthesis of carotenoid metabolites to adapt to excessive light stimulation (Jahns and Holzwarth 2012).

Most MEP pathway genes were co-expressed with carotenoid metabolic genes, resulting in an increased source pool of carotenoid metabolic flux in plants(Cordobaet al.2009).Our findings indicated that blue light increases the expression of MEP pathway structural genes and strongly increases the metabolic flow to the carotenoid metabolic pool (Fig.6; Appendix P).Overexpression ofDXScan indicate significantly increased downstream carotenoid content (Estévezet al.2000).However, theDXRgene was not induced by light in rubber tree (Seetang-Nunet al.2008), suggesting that there may be a more complex cascade network regulating MEP pools in plants.

When exposed to blue light, theBrPSY1andBrLYCBgenes were up-regulated 9.08 and 4.89 fold, respectively(Appendix R).Previously, blue light treatment of citrus induced the expression ofCitPSYandCitLCYB1(Zhanget al.2012).Recent results in pepper (Capsicumannuum)showed that blue LEDs significantly increased the expression ofPSY,Lcyb, andCrtZ, which encode genes involved in the carotenoid biosynthesis pathway (Polaet al.2019).These findings suggested that blue light treatment induces the transcription of light signal and MEP pathwayrelated genes and increases carotenoid metabolism.

4.4.The light signal transcription factor BrHY5-2 is involved in blue light-induced carotenoid accumulation in the inner leaves of OHCC lines

The light signal activator HY5 transcription factor can be induced by light (Toledo-Ortizet al.2014; Xiaoet al.2022),and HY5 acts as a positive regulator and forms a complex with PIF to regulate various metabolic fluxes (Toledo-Ortizet al.2014), especially affecting carotenoid metabolism to improve biological adaptability.We observed that blue light induces the up-regulation ofBrHY5-2transcription factors in Chinese cabbage (Fig.6), and the expression level ofBrHY5-2was significantly positively correlated with the antioxidant enzyme gene level (Appendices V and W), suggesting that this gene may be involved in the antioxidant enzyme system to regulate blue light stress in OHCCs.BrHY5transcript abundance was positively co-expressed with antheraxanthin, β-carotene,lutein, and zeaxanthin content (Fig.5-E).Meanwhile,overexpression ofBrHY5inArabidopsisinduces the significant accumulation of carotenoid content and carotenoid pathway genes in transgenicArabidopsis(Fig.7), indicatingthat BrHY5may regulate carotenoid metabolism.In general, environmental factors, especially blue light treatment, offer an effective and efficient artificial means for increasing the accumulation of carotenoids in response to people’s negative attitudes towards genetically modified (GM) foods (Rodriguez-Concepcionet al.2018).Overall, our results showed that blue light as an “inducer”increases the accumulation of various carotenoids in OHCC and improves its nutritional value and marketability,and the light signal geneBrHY5may function as a key transcriptional regulator in controlling the carotenoid metabolism of OHCC.

5.Conclusion

In this study, carotenoids of the leaves of three OHCC lines were analyzed by UHPLC-APCI-MS/MS.The highest total carotenoid content (1 014.07 μg g–1) was noted in line‘20S530’, followed by line ‘14S837’ (572.16 μg g–1) and line ‘15S1094’ (383.41 μg g–1).The total carotenoid, the total soluble solid, total phenol, and flavonoid contents were induced by blue light treatment in the inner leaves of three OHCC lines.In particular, the β-carotene,antheraxanthin, zeaxanthin, and lutein contents were significantly accumulated in three OHCC lines.The content of carotenoid esters, especially rubixanthin laurate and β-cryptoxanthin myristate, was significantly reduced in three OHCC lines.Furthermore, blue light induced the up-regulation of the light signal, photosynthesis, and MEP pathway genes, includingBrHY5-2, thus increasing the accumulation of downstream metabolites of carotenoids.Functional analysis confirmed that overexpression ofBrHY5-2increased carotenoid content inArabidopsis.Our study explored and uncovered the molecular basis of how short-term blue light treatment could induce significant leaf color change and carotenoid enrichment in orange-head Chinese cabbage and will facilitate the production of highquality Chinese cabbage in future cultivation efforts.

Acknowledgements

This work was supported by the National Key Research and Development Program of China (2017YFD0101802 and 2016YFD0101701), the Key Research and Development Program of Yangling Seed Innovative Center, China (Ylzy-sc-04) and the Key Research and Development Program of Shaanxi Province, China (2023-YBNY-078).

Declaration of competing interest

The authors declare that they have no conflict of interest.

Appendicesassociated with this paper are available on https://doi.org/10.1016/j.jia.2023.09.029