Gene 555 (2015) 458–463
Contents lists available at ScienceDirect
Gene journal homepage: www.elsevier.com/locate/gene
Short communication
De novo characterization of the Lycium chinense Mill. leaf transcriptome and analysis of candidate genes involved in carotenoid biosynthesis Gang Wang a,b, Xilong Du a, Jing Ji a,b,⁎, Chunfeng Guan b, Zhaodi Li a, Tchouopou Lontchi Josine b a b
School of Chemical Engineering and Technology, Tianjin University, Tianjin 30072, People's Republic of China School of Environmental Science and Engineering, Tianjin University, Tianjin 300072, People's Republic of China
a r t i c l e
i n f o
Article history: Received 12 July 2014 Received in revised form 14 October 2014 Accepted 30 October 2014 Available online 1 November 2014 Keywords: De novo RNA-seq Lycium chinense Mill. Transcriptome Carotenoid biosynthesis
a b s t r a c t Lycium chinense Mill. (Chinese wolfberry), enriching in carotenoids, is an important Chinese herbal medicine. However, studies on the functional genomics research, especially the carotenoid biosynthesis and accumulation, are limited because of insufficiently available datasets. RNA-Seq was performed by the Illumina sequencing platform. Approximately 26 million clean reads were generated after filtering. Clean reads were assembled by SOAPdenovo and subsequently annotated. Among all 61,595 unigenes, 37,816 (61.39%), 25,266 (41.02%), and 17,598 (28.57%) unigenes were annotated in NCBI non-redundant protein, Swiss-Prot, and Kyoto Encyclopedia of Genes and Genomes (KEGG) database, respectively. A total of 16,073 and 11,394 unigenes were assigned to Gene Ontology and Cluster of Orthologous Group, respectively. Furthermore, the majority of genes encoding the enzymes in the carotenoid biosynthesis pathway were identified in the unigene datasets. We first found several genes related to L. chinense carotenoid biosynthesis. The expression levels and the biological functions of these genes involved in carotenoid biosynthesis in the leaf and the green ripening fruit were further confirmed by qPCR and high performance liquid chromatography (HPLC). In the present study, we first characterized the transcriptome of L. chinense leaf, which may provide useful data for functional genomics investigations in L. chinense in the future. And essential genes involved in the carotenoid biosynthesis pathway may contribute to elucidate the expression patterns in different stages of development and fruit ripening and the specific mechanisms of carotenoid biosynthesis/accumulation in L. chinense. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Lycium chinense (Mill.) of the family Solanaceae, which has been widely used as healthy food because of its biological and pharmacological activities, is an important Chinese herbal medicine (Luo et al., 2006). Fruits of L. chinense play essential role in reducing blood glucose and serum lipids, anti-aging, immunomodulation, antitumor, etc. (Gan et al., 2004). Moreover, L. chinense is widely distributed in northwest
Abbreviations: CHYB, β-carotene hydroxylase; COG, Cluster of Orthologous Group; CRTISO, carotene isomerase; DMAPP, dimethylallyl pyrophosphate; Gbp, gigabase pairs; GGPP, geranylgeranyl pyrophosphate; GGPS, geranylgeranyl diphosphate synthase; GO, Gene Ontology; IPP, isopentenyl pyrophosphate; KEGG, Encyclopedia of Genes and Genomes; LCY-B, lycopene β-cyclase; LCY-E, lycopene ε-cyclase; NR, non-redundant; NXS, neoxanthin synthase; PDS, phytoene desaturase; PSY, phytoene synthase; qRT-PCR, quantitative real-time PCR; RACE, Rapid amplification of cDNA ends; RT-PCR, Reverse Transcription-Polymerase Chain Reaction; VDE, violaxanthin de-epoxidase; ZDS, ζcarotene desaturase; ZEP, zeaxanthin epoxidase. ⁎ Corresponding author at: 92 Weijin Road, Nankai District, Tianjin, People's Republic of China. E-mail addresses: [email protected] (G. Wang), [email protected] (X. Du), [email protected] (J. Ji), [email protected] (C. Guan), [email protected] (Z. Li), [email protected] (T.L. Josine).
http://dx.doi.org/10.1016/j.gene.2014.10.058 0378-1119/© 2014 Elsevier B.V. All rights reserved.
areas of China and planted in other warm and subtropical countries, such as Japan, Korea, and certain European countries. Carotenoids are especially abundant as one of the most important pigments and nutritional components in Chinese wolfberry. It has demonstrated that carotenoids are essential for photosynthesis and plant photoprotection in the plants and human nutrition (Ji et al., 2009). Although progress in exploring the mechanisms involved in carotenoid biosynthesis and accumulation in plants has been recently made in tomato and watermelon fruit (Fraser et al., 2007; Grassi et al., 2013), limited data is available in Chinese wolfberry. Previous studies on L. chinense have mainly focused on its basic biological studies including extraction and separation of nutritional components (Tian and W.M., 2006), pharmacology and medical function (Li et al., 2007; Chang and So, 2008), genetic engineering and cultivation (Hu et al., 2006). However, there are few reports on functional genomics of the biosynthesis of nutritional components, especially the carotenoid biosynthesis and accumulation in L. chinense. Currently, only 1,681,129 nucleotide sequences are available in NCBI for Lycium and for L. chinense, respectively, which are not abundant to find functional genes for L. chinense. RNA-Seq is a cost-effective and reliable method for transcriptome analysis, providing a specific analysis of gene expression and regulation at different development stages. Several non-model organisms, such as
G. Wang et al. / Gene 555 (2015) 458–463
Ma Bamboo, chili pepper, Chinese fir, and blueberry, have been investigated by transcriptome sequencing (Huang et al., 2012; Li et al., 2012b; Liu et al., 2012, 2013), which provided a better standing of these plants. Although most of these studies on transcriptome characterization of non-model organisms have been naturally descriptive, they have provided valuable resource for further analysis and applications in the future. For example, these sequences may be used as reference sequences for studies on functional genomics (Ong et al., 2012), in developing molecular markers for marker-assisted selection breeding (Li et al., 2012a), and in the study of alternative splicing (Harr and Turner, 2010). In the present study, we characterize the gene profiles in L. chinense for functional genomics by the de novo assembly of leaf transcriptome due to the lack of relative genome sequencing. The expression levels and the biological functions of these genes involved in carotenoid biosynthesis in the leaf and the green ripening fruit were further confirmed by qRT-PCR and HPLC analyses. 2. Materials and methods 2.1. Plant materials and RNA isolation L. chinense used in this study was grown at the experimental garden of the Institute of Genetic Engineering, Tianjin University (Tianjin, China) in 2010. Four young leaves were collected and the tissues were stored in liquid nitrogen. Total RNA was extracted with a PureLink® RNA Mini Kit (Ambion®) according to the manufacturer's instructions. The concentration and quality of each RNA sample were determined by NanoDrop 2000™ micro-volume spectrophotometer (Thermo Scientific, Waltham, MA, USA). Equal amounts of total RNA from the four samples were pooled together to construct the cDNA library. Pooling is an efficient and cost-effective strategy which has been well-justified when the goal of primary research is to characterize gene expression profiles (Auer and Doerge, 2010; Xu et al., 2012). The RNA samples were then stored at −80 °C. The ripening fruits used in this study were collected at the green ripening stage when the fruit color was changing from green to yellow. Triplicate replicates were performed in the study. 2.2. Illumina sequencing and de novo assembly Transcriptome sequencing was performed using Illumina HiSeq™ 2000. Raw reads with 3′ adaptors and repeating reads were removed, while nucleotides with a quality score lower than 20 were trimmed from the end of raw reads. Then, de novo assembly of the clean reads was conducted with the de novo transcriptome assembler SOAPdenovo (http://soap.genomics.org.cn/soapdenovo.html) as previously described (Li et al., 2010), to generate non-redundant unigenes. 2.3. Sequence annotation Clean reads were aligned to NCBI non-redundant (NR) protein database, Swiss-Prot, KEGG, and Cluster of Orthologous Group (COG) databases using Blastx (E-value ≤ 1E− 05). And the best aligning results were used to determine the sequence direction of unigenes with the priority of NR, Swiss-Prot, KEGG and COG if alignment results of different databases conflicted with each other. Unigene sequences were aligned to these protein databases to retrieve proteins with the highest sequence similarity to the given unigenes as their protein functional annotations. The coding sequences of unigenes were determined by proteins with the highest ranks. The coding sequences were then translated into amino sequences with the standard codon table. When a unigene could not be aligned to the above databases, ESTScan (Iseli et al., 1999) was introduced to predict its sequence direction. For unigenes with sequence directions, their sequences were identified from 5′ end to 3′ end; for those without any direction, nucleotide sequence (5′–3′) and amino sequence of the coding regions were determined with assembly software.
459
Furthermore, Blast2GO (Conesa et al., 2005) was used to obtain Gene Ontology (GO) annotation of unigenes with NR database. The WEGO software (Ye et al., 2006) was then used to perform functional classification of GO term for all unigenes and to comprehend the distribution of gene functions of L. chinense from the macro level. The unigene sequences were aligned to the COG database to predict and classify possible functions. Pathway assignments were performed according to the KEGG database. 2.4. Analysis of L. chinense unigenes related to carotenoid biosynthesis The unigenes were analyzed based on the KEGG pathways and functional annotations. The corresponding unigenes involved in carotenoid biosynthesis were aligned with Solanum lycopersicum and other plant protein sequences from public databases using the local blast program in BioEdit software (E-value b 1E−05). If identical alignment sequences were showed, we concluded that the corresponding genes were expressed in L. chinense. According to the unigene sequences, we identified a set of genes essential for the carotenoid biosynthesis. Total RNA was extracted from the leaf and green ripening fruit and cDNA was generated using High Capacity cDNA Reverse Transcription Kit (Applied Biosystems®). Then qRT-PCR was performed to analyze the relative with expression of GGPS1 and GGPS2, PSY, PDS, ZDS, CRTISO, LYC-E, LYC-B, CHYB, VDE and NCED by TransStart Top Green qPCR SuperMix. Specific primers were listed in Table S1. 2.5. Carotenoid extraction and HPLC analysis Approximately 50 mg fresh leaf and fruit tissues were ground in liquid nitrogen, and the powder of frozen samples was suspended in 20 mL methanol with 10% potassium hydroxide. Then the solution was incubated at 60 °C for 20 min, and total carotenoids were extracted by 50% ether in petrol ether. The extracts were used for HPLC analysis. The solvent was evaporated under a stream of N2 at 37 °C and dissolved in 50 μL of acetone, and 20 μL of aliquot was injected immediately. The samples were separated on a nucleosil 100-3 C18, 250 × 4.6 mm (MN, Germany) column with column temperature at 32 °C. The mobile phase was composed of acetonitrile/methanol/2-propanol (85:10:5 volume ratio) with a flow rate of 1 mL/min Detection was done at 450 nm using a UV detector. All carotenoids were identified with authentic reference compounds and comparisons of their spectra, and standards were purchased from Sigma (Sigma-Aldrich, USA). 3. Results 3.1. Illumina sequencing and de novo assembly After stringent quality assessment and data filtering, a total of 25,911,114 clean reads with average length of 90 bp were obtained. The high-quality reads produced in this study have been deposited in the NCBI SRA database (accession number: SRP033262). All high-quality reads were assembled using SOAPdenovo. Summary of the transcriptome sequencing and de novo assembly of L. chinense unigenes were shown in Table 1. 471,054 contigs were generated with an average length of 143 bp and a N50 of 119 bp, ranging from 50 bp to 6299 bp. The majority of the contigs were shorter than 100 bp
Table 1 Summary of Illumina transcription sequencing and de novo assembly of L. chinense unigenes.
Reads Contig Scaffold Unigene
Number
Mean size
N50 size
Total nucleotides
25,911,114 471,054 101,870 61,595
90 143 363 511
90 119 509 627
2,332,000,260 67,591,029 36,974,687 31,477,312
460
G. Wang et al. / Gene 555 (2015) 458–463
(73.50%), and only 4289 contigs (0.91%) were longer than 1000 bp. The size distribution of these contigs was demonstrated in Fig. 1a. The contigs were further assembled into scaffolds using paired-end joining. As a result, 101,870 scaffolds were obtained with an average length of 363 bp and a N50 of 509 bp, including 6344 scaffolds longer than 1000 bp (Fig. 1b). A total of 78,739 scaffolds (77.29%) had no gaps, and paired-end reads were used for the gap filling to get unigenes. Totally, 61,595 unigenes were obtained with an average length of 511 bp and a combined length of 31.48 Mb. The length of the assembled unigenes ranged from 203 bp to 6299 bp and the length distribution of the unigenes was similar to that of the scaffolds (Fig. 1c). 3.2. Annotation and classification of L. chinense unigenes All the assembled unigenes were subjected to Blastx against public protein database for annotation using a cut-off E-value of 1E− 05.
37,816 (61.39%), 25,266 (41.02%), and 17,598 (28.57%) unigenes were annotated in NR, Swiss-Prot, and KEGG database, respectively. A total of 38,016 (61.72%) unigenes were successfully annotated, whereas 23,579 (38.28%) unigenes had no alignments in the NR, Swiss-Prot, or KEGG databases. The detailed annotations were addressed in Table S2. 3.3. GO classification The unigenes annotated in the NR database were further analyzed with GO term using Blast2GO (Conesa et al., 2005). 16,073 (26.09%) unigenes were assigned to GO classes (Table S3), which were classified into three main GO categories (biological process, cellular component, and molecular function) and 44 subcategories (Fig. 2). Among these unigenes, 28,701 were assigned to at least one GO term in biological process category, 19,560 in cellular component category, while 34,170 in molecular function category. Within the biological process category, 25.96% of the unigenes were assigned to metabolic process, followed by cellular process (23.37%), response to stimulus (8.96%), and biological regulation (5.9%) (Fig. 2). In the cellular component category, cell (31.94%) and cell part (31.94%) were the dominant groups, followed by organelle (23.20%) and organelle part (5.51%) (Fig. 2). In the molecular function category, binding (46.24%) was the most dominant group, followed by catalytic activity (41.81%), transporter activity (4.90%), and molecular transducer activity (3.06%) (Fig. 2). 3.4. COG classification All of the assembled unigenes were subjected to the COG database to further evaluate the functional prediction and classification. Out of 37,816 NR hits, 11,394 unigenes were assigned to COG classifications. And a total of 18,803 COG functional annotations were obtained, which were classified into 25 functional categories (Fig. 3). The category of general function prediction suggested the largest group (3152, 16.76%), followed by transcription (1590, 8.46%), posttranslational modification, protein turnover and chaperones (1588, 8.45%), replication, recombination and repair (1565, 8.32%). However, only 4 and 1 unigenes were associated with extracellular and nuclear structures, respectively. In addition, 759 unigenes were assigned to the unknown function category. 3.5. KEGG pathway analysis The unigenes were compared to the KEGG database for better understanding of the complicated biological behaviors and biological pathways in L. chinense (Kanehisa et al., 2008). Here, 17,598 (28.57%) unigenes were identified and assigned to 119 KEGG pathways. The pathways with the most genes annotated were metabolic pathways (3992 unigenes, 22.68%), biosynthesis of secondary metabolites (2128, 12.09%), plant–pathogen interaction (1442, 8.19%), etc. And there are 118 (0.67%) unigenes in the carotenoid biosynthesis pathway (Table S4). 3.6. Analysis of the genes involved in carotenoid biosynthesis pathway
Fig. 1. Length and number distribution of assembled contigs, scaffolds and unigenes. (a) Length and number distribution of L. chinense contigs. (b) Length and number distribution of L. chinense scaffolds. (c) Length and number distribution of L. chinense unigenes.
Based on the transcriptome sequencing and annotation analysis, the majority of the genes encoding the enzymes in the carotenoid biosynthesis pathway were identified in the unigene dataset (Fig. 4a). LcGGPS1 and LcGGPS2, LcPSY, LcPDS, LcZDS, LcCRTISO, LcLYC-E, LcLYC-B, LcCHYB LcVDE and LcNCED were isolated by Rapid amplification of cDNA ends (RACE) and Reverse Transcription-Polymerase Chain Reaction (RTPCR). When the Sanger-sequenced, full-length cDNA were aligned against the unigene dataset, 27 unigenes were mapped to different regions of these genes respectively. It is noteworthy that most of the unigenes showed high similarity with the full-length genes (Table 2). These data further demonstrated that the unigenes obtained by RNA-
G. Wang et al. / Gene 555 (2015) 458–463
461
Fig. 2. GO classifications of the L. chinense unigenes. A total of 16,073 unigenes were assigned to at least one GO term and were grouped into three main GO categories and 44 subcategories. Right Y-axis represents number of genes in a category. Left Y-axis indicates percentage of a specific category of genes in each main category.
seq were efficiently assembled and the quality of sequencing data may contribute to metabolic pathway research in L. chinense. qRT-PCR analysis was further used to compare the expression levels of the genes involved in carotenoid biosynthesis in leaf and ripening fruit. It has been demonstrated that the expression levels of the majority of the genes related to carotenoid biosynthesis in the ripening fruit were higher than that in the leaf. LcGGPS1, LcCHYB and LcVDE were highly expressed in the leaf, while LcCHYB and LcNCED were highly expressed in the ripening fruit (Fig. 4b). Intriguingly, the expression of LcGGPS1 was significantly higher than LcGGPS2 in the leaf, while the
expression of LcGGPS2 was much higher than LcGGPS1 in the ripening fruit. Thus, LcGGPS1 may be essential in the leaf, while LcGGPS2 play an important role in the ripening fruit. In addition, the expression level of LcLYC-B was much higher than those of LcLYC-E in both the leaf and the ripening fruit. And LcCHYB was expressed at very high level both in the leaf and in the ripening fruit, which may contribute to the carotenoid biosynthesis and accumulation in L. chinense. Furthermore, LcNCED was expressed at a high level in the ripening fruit compared to that in the leaf, and it may play an important role in the stage of fruit ripening.
Fig. 3. COG function classification of the L. chinense unigenes. In total, 18,803 of the 37,816 sequences with NR hits were grouped into 25 COG classifications. Y-axis indicates number of unigenes in a specific functional cluster.
462
G. Wang et al. / Gene 555 (2015) 458–463
Fig. 4. Carotenoid metabolism pathway (a) and relative expression levels of related genes in the leaf and the ripening fruit (b). The numbers in the brackets following the gene name in subpanel a indicate the number of unigenes mapped to that gene. IPP, isopentenyl pyrophosphate; DMAPP, Dimethylallyl pyrophosphate; GGPP, geranylgeranyl pyrophosphate; GGPS, geranylgeranyl diphosphate synthase; PSY, phytoene synthase; PDS, phytoene desaturase; ZDS, ζ-carotene desaturase; CRTISO, carotene isomerase; LCY-E, lycopene ε-cyclase; LCY-B, lycopene β-cyclase; CHYB, β-carotene hydroxylase; ZEP, zeaxanthin epoxidase; VDE, violaxanthin de-epoxidase; NXS, neoxanthin synthase.
3.7. Analysis of carotenoid contents in leaf and green ripening fruit of L. chinense HPLC analysis was used to determine the carotenoid composition. It demonstrated that the total contents of carotenoids in leaf were much higher than those in the green ripening fruits. The β-carotene content (5.75 ± 0.61 μg/g) was found to be the highest in the leaf and the highest β-carotene content (5.47 ± 0.20 μg/g) was observed in the green ripening fruit. And the amount of lutein and zeaxanthin in the leaf (4.03 ± 0.46 μg/g), was more than three times higher compared to that in green ripening fruit. The content of neoxanthin in the leaf (0.80 ± 0.08 μg/g) was also observed more than three times higher Table 2 Sequence analysis of the cloned genes involved in carotenoid biosynthesis. Gene
Length
LcGGPS1 LcGGPS2 LcPSY LcPDS LcZDS LcCRTISO LcLCY-B LcLCY-E LcCHYB LcVDE LcNCED
1203 1246 1341 2021 2042 1815 1506 1847 915 1413 1827
bp bp bp bp bp bp bp bp bp bp bp
Characteristics of correspondent unigenes Coverage
ORF
Identities
RPKM
100.0% 100.0% 100.0% 97.5% 100.0% 99.9% 100.0% 94.4% 100.0% 100.0% 99.4%
Complete Complete Complete Part Complete Complete Complete Part Complete Complete Part
1201/1203 (99.8%) 1246/1246 (100.0%) 1320/1341 (98.4%) 1955/1970 (99.2%) 2025/2032 (99.7%) 1813/1813 (100%) 1496/1506 (99.3%) 1739/1744 (99.7%) 915/915 (100.0%) 1412/1413 (99.9%) 1811/1817 (99.7%)
157.17 27.91 50.80 31.89 81.40 14.96 62.03 31.40 128.70 152.26 114.24
compared to that in green ripening fruit (0.26 ± 0.01 μg/g) (Table 3). Thus, the accumulation of carotenoids was dependent on the regulation of different carotenogenic enzymes.
4. Discussion RNA-Seq has become an important tool for comprehensive studies of gene expression, detection of novel transcripts, and discovery of genes or expressed SNPs associated with valuable traits. In this study, we generated de novo transcriptome sequences and characterized the leaf transcriptome of L. chinense, and subsequent annotation of gene functions. Chinese wolfberries contain high levels of carotenoids, such as βcarotene, lutein and zeaxanthin, which are essential nutrients for human health. In the present study, almost all of genes encoding related enzymes in the carotenoid biosynthesis pathway were found in the transcriptome dataset. qRT-PCR analysis showed significantly higher expression levels of LcGGPS2, LcPSY, LcLCY-B, LcCHYB, and LcVDE genes in the ripening fruit, thereby possibly explaining the high β-carotene, lutein and zeaxanthin contents in Chinese wolfberry. As a key gene in ABA biosynthesis and closely related to carotenoid accumulation in plants, LcNCED was one of the most highly expressed carotenoidrelated genes in the ripening fruits, and probably plays an important role in fruit ripening. In higher plant, the condensation of three IPP and one DMAPP units to form GGPP is a crucial step in the carotenoid biosynthesis pathway, which is catalyzed by GGPS. Two isoforms of the GGPS gene, LcGGPS1 and LcGGPS2, were identified and cloned in L. chinense. LcGGPS1 was
G. Wang et al. / Gene 555 (2015) 458–463
463
Table 3 The carotenoid compositions of leaf and the green ripening fruit. Samples
Total carotenoids (μg/g)
Neoxanthin (μg/g)
Vioxanthin (μg/g)
Lutein + zeaxanthin (μg/g)
β-Carotene (μg/g)
Leaf Ripening fruit
10.85 ± 1.17 7.52 ± 0.25
0.80 ± 0.08 0.26 ± 0.01
0.26 ± 0.036 0.50 ± 0.01
4.03 ± 0.46 1.29 ± 0.04
5.75 ± 0.61 5.47 ± 0.20
highly expressed in leaf, while LcGGPS2 was detected at a rather low expression level in leaf but highly expressed in the ripening fruit. Thus, we presumed that LcGGPS2 may encode a fruit-specific isoform and have important roles in ripening fruit at specific development stages, while LcGGPS1 dominated in the leaf. The expression patterns of LcGGPS2 and LcGGPS1 were probably similar to the PSY2 and PSY1 in tomato, respectively (Fraser et al., 2007). Also we noticed that the expression level of LcLYC-B was much higher than LcLYC-E in both the leaf and the ripening fruit. Cyclization of lycopene marks a branch point involved in the pathway of the plant carotenoid biosynthesis to either α- or βcarotene. Lycopene β-cyclase (LCY-B) catalyzes lycopene to produce β-carotene, whereas conversion to α-carotene is catalyzed by lycopene ε-cyclase (LCY-E). LCY-B is one of the crucial enzymes for carotenoid biosynthesis and believed to predominate in the formation of vegetative carotenoids (Fraser et al., 2007). The higher expression level of LYC-B in Chinese wolfberry may permanently maintain higher metabolic flux toward β-carotene while the lower expression level of LYC-E may maintain lower metabolic flux to α-carotene, thereby leading to the high βcarotene and low α-carotene contents in the fruits of Chinese wolfberry. The expression pattern of LcCHYB with the highest expression level in both the leaf and the ripening fruit may reveal its important role in the carotenoid biosynthesis and accumulation in L. chinense. Based on HPLC analysis, the lutein and zeaxanthin contents in the green ripening fruits were much less than in the leaf, which may be resulted from the expression level of related genes. In the leaf, relative expressions of almost all upstream genes of lutein and zeaxanthin in the carotenoid biosynthetic pathway were much lower than that in the green ripening fruit. However, the accumulation of lutein and zeaxanthin was dependent on the regulation of different carotenogenic enzymes. In the ripening fruit, the downstream genes of lutein and zeaxanthin may regulate their expression. The expression of LcVDE was much lower, while the expression of LcNCED was higher in the ripening fruit (Fig. 4). The high expression levels of LcNCED may be responsible for lutein and zeaxanthin accumulation at the early stages of ripening and lead to the low expression of neoxanthin (0.26 ± 0.01 μg/g in the fruit compared to 0.80 ± 0.08 μg/g in the leaf). In summary, the identification of these key genes in the carotenoid biosynthesis pathway might contribute to further elucidate the expression patterns of carotenoid pathway genes at different stages of development and fruit ripening and the mechanisms of carotenoid biosynthesis/accumulation in L. chinense. This will be of great significance to cultivate elite Chinese wolfberry varieties with high carotenoid contents for improved nutritional properties. As to our knowledge, this study is the first application of RNA-Seq in investigating the leaf transcriptome of L. chinense. A large number of transcriptomic sequences may provide sufficient data to discover almost all of the known genes in major metabolic pathways. The functional annotations and further analysis of these genes are useful in exploring the biosynthesis mechanisms of nutritional components in L. chinense. Consequently, these datasets may provide an important public information platform for further research in gene expression and functional genomics in L. chinense. Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.gene.2014.10.058. Acknowledgments This work was supported by the National Natural Science Foundation of China (grant nos. 31271419 and 31271793) and National Science
and Technology Key Project of China on GMO Cultivation for New Varieties (No. 2014ZX08003-002B). This work was also supported by High Performance Computing Center of Tianjin University, China. The authors thank the Beijing Genome Institute at Shenzhen, China for the technical support.
References Auer, P.L., Doerge, R.W., 2010. Statistical design and analysis of RNA sequencing data. Genetics 185, 405–416. Chang, R.C., So, K.F., 2008. Use of anti-aging herbal medicine, Lycium barbarum, against aging-associated diseases. What do we know so far? Cell. Mol. Neurobiol. 28, 643–652. Conesa, A., Götz, S., García-Gómez, J.M., Terol, J., Talón, M., Robles, M., 2005. Blast2GO: a universal tool for annotation, visualization and analysis in functional genomics research. Bioinformatics 21, 3674–3676. Fraser, P.D., Enfissi, E.M., Halket, J.M., Truesdale, M.R., Yu, D., Gerrish, C., Bramley, P.M., 2007. Manipulation of phytoene levels in tomato fruit: effects on isoprenoids, plastids, and intermediary metabolism. Plant Cell 19, 3194–3211. Gan, L., Zhang, S.H., Yang, X.L., Xu, H.B., 2004. Immunomodulation and antitumor activity by a polysaccharide–protein complex from Lycium barbarum. Int. Immunopharmacol. 4, 563–569. Grassi, S., Piro, G., Lee, J.M., Zheng, Y., Fei, Z., Dalessandro, G., Giovannoni, J.J., Lenucci, M.S., 2013. Comparative genomics reveals candidate carotenoid pathway regulators of ripening watermelon fruit. BMC Genomics 14, 781. Harr, B., Turner, L.M., 2010. Genome‐wide analysis of alternative splicing evolution among Mus subspecies. Mol. Ecol. 19, 228–239. Hu, Z., Wu, Y.-R., Li, W., Gao, H.-H., 2006. Factors affecting Agrobacterium tumefaciensmediated genetic transformation of Lycium barbarum L. In Vitro Cell. Dev. Biol. Plant 42, 461–466. Huang, H.H., Xu, L.L., Tong, Z.K., Lin, E.P., Liu, Q.P., Cheng, L.J., Zhu, M.Y., 2012. De novo characterization of the Chinese fir (Cunninghamia lanceolata) transcriptome and analysis of candidate genes involved in cellulose and lignin biosynthesis. BMC Genomics 13, 648. Iseli, C., Jongeneel, C.V., Bucher, P., 1999. ESTScan: a program for detecting, evaluating, and reconstructing potential coding regions in EST sequences. Proc. Int. Conf. Intell. Syst. Mol. Biol. 138–148. Ji, J., Wang, G., Wang, J., Wang, P., 2009. Functional analysis of multiple carotenogenic genes from Lycium barbarum and Gentiana lutea L. for their effects on beta-carotene production in transgenic tobacco. Biotechnol. Lett. 31, 305–312. Kanehisa, M., Araki, M., Goto, S., Hattori, M., Hirakawa, M., Itoh, M., Katayama, T., Kawashima, S., Okuda, S., Tokimatsu, T., Yamanishi, Y., 2008. KEGG for linking genomes to life and the environment. Nucleic Acids Res. 36, D480–D484. Li, X.M., Li, X.L., Zhou, A.G., 2007. Evaluation of antioxidant activity of the polysaccharides extracted from Lycium barbarum fruits in vitro. Eur. Polym. J. 43, 488–497. Li, R., Zhu, H., Ruan, J., Qian, W., Fang, X., Shi, Z., Li, Y., Li, S., Shan, G., Kristiansen, K., 2010. De novo assembly of human genomes with massively parallel short read sequencing. Genome Res. 20, 265–272. Li, D., Deng, Z., Qin, B., Liu, X., Men, Z., 2012a. De novo assembly and characterization of bark transcriptome using Illumina sequencing and development of EST-SSR markers in rubber tree (Hevea brasiliensis Muell. Arg.). BMC Genomics 13, 192. Li, X., Sun, H., Pei, J., Dong, Y., Wang, F., Chen, H., Sun, Y., Wang, N., Li, H., Li, Y., 2012b. De novo sequencing and comparative analysis of the blueberry transcriptome to discover putative genes related to antioxidants. Gene 511, 54–61. Liu, M., Qiao, G., Jiang, J., Yang, H., Xie, L., Xie, J., Zhuo, R., 2012. Transcriptome sequencing and de novo analysis for Ma bamboo (Dendrocalamus latiflorus Munro) using the Illumina platform. PLoS One 7, e46766. Liu, S., Li, W., Wu, Y., Chen, C., Lei, J., 2013. De novo transcriptome assembly in chili pepper (Capsicum frutescens) to identify genes involved in the biosynthesis of capsaicinoids. PLoS One 8, e48156. Luo, Q., Li, Z., Huang, X., Yan, J., Zhang, S., Cai, Y.-Z., 2006. Lycium barbarum polysaccharides: protective effects against heat-induced damage of rat testes and H2O2induced DNA damage in mouse testicular cells and beneficial effect on sexual behavior and reproductive function of hemicastrated rats. Life Sci. 79, 613–621. Ong, W.D., Voo, L.-Y.C., Kumar, V.S., 2012. De novo assembly, characterization and functional annotation of pineapple fruit transcriptome through massively parallel sequencing. PLoS One 7, e46937. Tian, M., Wang, M., 2006. Studies on extraction, isolation and composition of Lycium barbarum polysaccharides. China J. Chin. Mater. Med. 31, 1603–1607. Xu, D.-L., Long, H., Liang, J.-J., Zhang, J., 2012. De novo assembly and characterization of the root transcriptome of Aegilops variabilis during an interaction with the cereal cyst nematode. BMC Genomics 13, 133. Ye, J., Fang, L., Zheng, H., Zhang, Y., Chen, J., Zhang, Z., Wang, J., Li, S., Li, R., Bolund, L., 2006. WEGO: a web tool for plotting GO annotations. Nucleic Acids Res. 34, W293–W297.
Contents lists available at ScienceDirect
Gene journal homepage: www.elsevier.com/locate/gene
Short communication
De novo characterization of the Lycium chinense Mill. leaf transcriptome and analysis of candidate genes involved in carotenoid biosynthesis Gang Wang a,b, Xilong Du a, Jing Ji a,b,⁎, Chunfeng Guan b, Zhaodi Li a, Tchouopou Lontchi Josine b a b
School of Chemical Engineering and Technology, Tianjin University, Tianjin 30072, People's Republic of China School of Environmental Science and Engineering, Tianjin University, Tianjin 300072, People's Republic of China
a r t i c l e
i n f o
Article history: Received 12 July 2014 Received in revised form 14 October 2014 Accepted 30 October 2014 Available online 1 November 2014 Keywords: De novo RNA-seq Lycium chinense Mill. Transcriptome Carotenoid biosynthesis
a b s t r a c t Lycium chinense Mill. (Chinese wolfberry), enriching in carotenoids, is an important Chinese herbal medicine. However, studies on the functional genomics research, especially the carotenoid biosynthesis and accumulation, are limited because of insufficiently available datasets. RNA-Seq was performed by the Illumina sequencing platform. Approximately 26 million clean reads were generated after filtering. Clean reads were assembled by SOAPdenovo and subsequently annotated. Among all 61,595 unigenes, 37,816 (61.39%), 25,266 (41.02%), and 17,598 (28.57%) unigenes were annotated in NCBI non-redundant protein, Swiss-Prot, and Kyoto Encyclopedia of Genes and Genomes (KEGG) database, respectively. A total of 16,073 and 11,394 unigenes were assigned to Gene Ontology and Cluster of Orthologous Group, respectively. Furthermore, the majority of genes encoding the enzymes in the carotenoid biosynthesis pathway were identified in the unigene datasets. We first found several genes related to L. chinense carotenoid biosynthesis. The expression levels and the biological functions of these genes involved in carotenoid biosynthesis in the leaf and the green ripening fruit were further confirmed by qPCR and high performance liquid chromatography (HPLC). In the present study, we first characterized the transcriptome of L. chinense leaf, which may provide useful data for functional genomics investigations in L. chinense in the future. And essential genes involved in the carotenoid biosynthesis pathway may contribute to elucidate the expression patterns in different stages of development and fruit ripening and the specific mechanisms of carotenoid biosynthesis/accumulation in L. chinense. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Lycium chinense (Mill.) of the family Solanaceae, which has been widely used as healthy food because of its biological and pharmacological activities, is an important Chinese herbal medicine (Luo et al., 2006). Fruits of L. chinense play essential role in reducing blood glucose and serum lipids, anti-aging, immunomodulation, antitumor, etc. (Gan et al., 2004). Moreover, L. chinense is widely distributed in northwest
Abbreviations: CHYB, β-carotene hydroxylase; COG, Cluster of Orthologous Group; CRTISO, carotene isomerase; DMAPP, dimethylallyl pyrophosphate; Gbp, gigabase pairs; GGPP, geranylgeranyl pyrophosphate; GGPS, geranylgeranyl diphosphate synthase; GO, Gene Ontology; IPP, isopentenyl pyrophosphate; KEGG, Encyclopedia of Genes and Genomes; LCY-B, lycopene β-cyclase; LCY-E, lycopene ε-cyclase; NR, non-redundant; NXS, neoxanthin synthase; PDS, phytoene desaturase; PSY, phytoene synthase; qRT-PCR, quantitative real-time PCR; RACE, Rapid amplification of cDNA ends; RT-PCR, Reverse Transcription-Polymerase Chain Reaction; VDE, violaxanthin de-epoxidase; ZDS, ζcarotene desaturase; ZEP, zeaxanthin epoxidase. ⁎ Corresponding author at: 92 Weijin Road, Nankai District, Tianjin, People's Republic of China. E-mail addresses: [email protected] (G. Wang), [email protected] (X. Du), [email protected] (J. Ji), [email protected] (C. Guan), [email protected] (Z. Li), [email protected] (T.L. Josine).
http://dx.doi.org/10.1016/j.gene.2014.10.058 0378-1119/© 2014 Elsevier B.V. All rights reserved.
areas of China and planted in other warm and subtropical countries, such as Japan, Korea, and certain European countries. Carotenoids are especially abundant as one of the most important pigments and nutritional components in Chinese wolfberry. It has demonstrated that carotenoids are essential for photosynthesis and plant photoprotection in the plants and human nutrition (Ji et al., 2009). Although progress in exploring the mechanisms involved in carotenoid biosynthesis and accumulation in plants has been recently made in tomato and watermelon fruit (Fraser et al., 2007; Grassi et al., 2013), limited data is available in Chinese wolfberry. Previous studies on L. chinense have mainly focused on its basic biological studies including extraction and separation of nutritional components (Tian and W.M., 2006), pharmacology and medical function (Li et al., 2007; Chang and So, 2008), genetic engineering and cultivation (Hu et al., 2006). However, there are few reports on functional genomics of the biosynthesis of nutritional components, especially the carotenoid biosynthesis and accumulation in L. chinense. Currently, only 1,681,129 nucleotide sequences are available in NCBI for Lycium and for L. chinense, respectively, which are not abundant to find functional genes for L. chinense. RNA-Seq is a cost-effective and reliable method for transcriptome analysis, providing a specific analysis of gene expression and regulation at different development stages. Several non-model organisms, such as
G. Wang et al. / Gene 555 (2015) 458–463
Ma Bamboo, chili pepper, Chinese fir, and blueberry, have been investigated by transcriptome sequencing (Huang et al., 2012; Li et al., 2012b; Liu et al., 2012, 2013), which provided a better standing of these plants. Although most of these studies on transcriptome characterization of non-model organisms have been naturally descriptive, they have provided valuable resource for further analysis and applications in the future. For example, these sequences may be used as reference sequences for studies on functional genomics (Ong et al., 2012), in developing molecular markers for marker-assisted selection breeding (Li et al., 2012a), and in the study of alternative splicing (Harr and Turner, 2010). In the present study, we characterize the gene profiles in L. chinense for functional genomics by the de novo assembly of leaf transcriptome due to the lack of relative genome sequencing. The expression levels and the biological functions of these genes involved in carotenoid biosynthesis in the leaf and the green ripening fruit were further confirmed by qRT-PCR and HPLC analyses. 2. Materials and methods 2.1. Plant materials and RNA isolation L. chinense used in this study was grown at the experimental garden of the Institute of Genetic Engineering, Tianjin University (Tianjin, China) in 2010. Four young leaves were collected and the tissues were stored in liquid nitrogen. Total RNA was extracted with a PureLink® RNA Mini Kit (Ambion®) according to the manufacturer's instructions. The concentration and quality of each RNA sample were determined by NanoDrop 2000™ micro-volume spectrophotometer (Thermo Scientific, Waltham, MA, USA). Equal amounts of total RNA from the four samples were pooled together to construct the cDNA library. Pooling is an efficient and cost-effective strategy which has been well-justified when the goal of primary research is to characterize gene expression profiles (Auer and Doerge, 2010; Xu et al., 2012). The RNA samples were then stored at −80 °C. The ripening fruits used in this study were collected at the green ripening stage when the fruit color was changing from green to yellow. Triplicate replicates were performed in the study. 2.2. Illumina sequencing and de novo assembly Transcriptome sequencing was performed using Illumina HiSeq™ 2000. Raw reads with 3′ adaptors and repeating reads were removed, while nucleotides with a quality score lower than 20 were trimmed from the end of raw reads. Then, de novo assembly of the clean reads was conducted with the de novo transcriptome assembler SOAPdenovo (http://soap.genomics.org.cn/soapdenovo.html) as previously described (Li et al., 2010), to generate non-redundant unigenes. 2.3. Sequence annotation Clean reads were aligned to NCBI non-redundant (NR) protein database, Swiss-Prot, KEGG, and Cluster of Orthologous Group (COG) databases using Blastx (E-value ≤ 1E− 05). And the best aligning results were used to determine the sequence direction of unigenes with the priority of NR, Swiss-Prot, KEGG and COG if alignment results of different databases conflicted with each other. Unigene sequences were aligned to these protein databases to retrieve proteins with the highest sequence similarity to the given unigenes as their protein functional annotations. The coding sequences of unigenes were determined by proteins with the highest ranks. The coding sequences were then translated into amino sequences with the standard codon table. When a unigene could not be aligned to the above databases, ESTScan (Iseli et al., 1999) was introduced to predict its sequence direction. For unigenes with sequence directions, their sequences were identified from 5′ end to 3′ end; for those without any direction, nucleotide sequence (5′–3′) and amino sequence of the coding regions were determined with assembly software.
459
Furthermore, Blast2GO (Conesa et al., 2005) was used to obtain Gene Ontology (GO) annotation of unigenes with NR database. The WEGO software (Ye et al., 2006) was then used to perform functional classification of GO term for all unigenes and to comprehend the distribution of gene functions of L. chinense from the macro level. The unigene sequences were aligned to the COG database to predict and classify possible functions. Pathway assignments were performed according to the KEGG database. 2.4. Analysis of L. chinense unigenes related to carotenoid biosynthesis The unigenes were analyzed based on the KEGG pathways and functional annotations. The corresponding unigenes involved in carotenoid biosynthesis were aligned with Solanum lycopersicum and other plant protein sequences from public databases using the local blast program in BioEdit software (E-value b 1E−05). If identical alignment sequences were showed, we concluded that the corresponding genes were expressed in L. chinense. According to the unigene sequences, we identified a set of genes essential for the carotenoid biosynthesis. Total RNA was extracted from the leaf and green ripening fruit and cDNA was generated using High Capacity cDNA Reverse Transcription Kit (Applied Biosystems®). Then qRT-PCR was performed to analyze the relative with expression of GGPS1 and GGPS2, PSY, PDS, ZDS, CRTISO, LYC-E, LYC-B, CHYB, VDE and NCED by TransStart Top Green qPCR SuperMix. Specific primers were listed in Table S1. 2.5. Carotenoid extraction and HPLC analysis Approximately 50 mg fresh leaf and fruit tissues were ground in liquid nitrogen, and the powder of frozen samples was suspended in 20 mL methanol with 10% potassium hydroxide. Then the solution was incubated at 60 °C for 20 min, and total carotenoids were extracted by 50% ether in petrol ether. The extracts were used for HPLC analysis. The solvent was evaporated under a stream of N2 at 37 °C and dissolved in 50 μL of acetone, and 20 μL of aliquot was injected immediately. The samples were separated on a nucleosil 100-3 C18, 250 × 4.6 mm (MN, Germany) column with column temperature at 32 °C. The mobile phase was composed of acetonitrile/methanol/2-propanol (85:10:5 volume ratio) with a flow rate of 1 mL/min Detection was done at 450 nm using a UV detector. All carotenoids were identified with authentic reference compounds and comparisons of their spectra, and standards were purchased from Sigma (Sigma-Aldrich, USA). 3. Results 3.1. Illumina sequencing and de novo assembly After stringent quality assessment and data filtering, a total of 25,911,114 clean reads with average length of 90 bp were obtained. The high-quality reads produced in this study have been deposited in the NCBI SRA database (accession number: SRP033262). All high-quality reads were assembled using SOAPdenovo. Summary of the transcriptome sequencing and de novo assembly of L. chinense unigenes were shown in Table 1. 471,054 contigs were generated with an average length of 143 bp and a N50 of 119 bp, ranging from 50 bp to 6299 bp. The majority of the contigs were shorter than 100 bp
Table 1 Summary of Illumina transcription sequencing and de novo assembly of L. chinense unigenes.
Reads Contig Scaffold Unigene
Number
Mean size
N50 size
Total nucleotides
25,911,114 471,054 101,870 61,595
90 143 363 511
90 119 509 627
2,332,000,260 67,591,029 36,974,687 31,477,312
460
G. Wang et al. / Gene 555 (2015) 458–463
(73.50%), and only 4289 contigs (0.91%) were longer than 1000 bp. The size distribution of these contigs was demonstrated in Fig. 1a. The contigs were further assembled into scaffolds using paired-end joining. As a result, 101,870 scaffolds were obtained with an average length of 363 bp and a N50 of 509 bp, including 6344 scaffolds longer than 1000 bp (Fig. 1b). A total of 78,739 scaffolds (77.29%) had no gaps, and paired-end reads were used for the gap filling to get unigenes. Totally, 61,595 unigenes were obtained with an average length of 511 bp and a combined length of 31.48 Mb. The length of the assembled unigenes ranged from 203 bp to 6299 bp and the length distribution of the unigenes was similar to that of the scaffolds (Fig. 1c). 3.2. Annotation and classification of L. chinense unigenes All the assembled unigenes were subjected to Blastx against public protein database for annotation using a cut-off E-value of 1E− 05.
37,816 (61.39%), 25,266 (41.02%), and 17,598 (28.57%) unigenes were annotated in NR, Swiss-Prot, and KEGG database, respectively. A total of 38,016 (61.72%) unigenes were successfully annotated, whereas 23,579 (38.28%) unigenes had no alignments in the NR, Swiss-Prot, or KEGG databases. The detailed annotations were addressed in Table S2. 3.3. GO classification The unigenes annotated in the NR database were further analyzed with GO term using Blast2GO (Conesa et al., 2005). 16,073 (26.09%) unigenes were assigned to GO classes (Table S3), which were classified into three main GO categories (biological process, cellular component, and molecular function) and 44 subcategories (Fig. 2). Among these unigenes, 28,701 were assigned to at least one GO term in biological process category, 19,560 in cellular component category, while 34,170 in molecular function category. Within the biological process category, 25.96% of the unigenes were assigned to metabolic process, followed by cellular process (23.37%), response to stimulus (8.96%), and biological regulation (5.9%) (Fig. 2). In the cellular component category, cell (31.94%) and cell part (31.94%) were the dominant groups, followed by organelle (23.20%) and organelle part (5.51%) (Fig. 2). In the molecular function category, binding (46.24%) was the most dominant group, followed by catalytic activity (41.81%), transporter activity (4.90%), and molecular transducer activity (3.06%) (Fig. 2). 3.4. COG classification All of the assembled unigenes were subjected to the COG database to further evaluate the functional prediction and classification. Out of 37,816 NR hits, 11,394 unigenes were assigned to COG classifications. And a total of 18,803 COG functional annotations were obtained, which were classified into 25 functional categories (Fig. 3). The category of general function prediction suggested the largest group (3152, 16.76%), followed by transcription (1590, 8.46%), posttranslational modification, protein turnover and chaperones (1588, 8.45%), replication, recombination and repair (1565, 8.32%). However, only 4 and 1 unigenes were associated with extracellular and nuclear structures, respectively. In addition, 759 unigenes were assigned to the unknown function category. 3.5. KEGG pathway analysis The unigenes were compared to the KEGG database for better understanding of the complicated biological behaviors and biological pathways in L. chinense (Kanehisa et al., 2008). Here, 17,598 (28.57%) unigenes were identified and assigned to 119 KEGG pathways. The pathways with the most genes annotated were metabolic pathways (3992 unigenes, 22.68%), biosynthesis of secondary metabolites (2128, 12.09%), plant–pathogen interaction (1442, 8.19%), etc. And there are 118 (0.67%) unigenes in the carotenoid biosynthesis pathway (Table S4). 3.6. Analysis of the genes involved in carotenoid biosynthesis pathway
Fig. 1. Length and number distribution of assembled contigs, scaffolds and unigenes. (a) Length and number distribution of L. chinense contigs. (b) Length and number distribution of L. chinense scaffolds. (c) Length and number distribution of L. chinense unigenes.
Based on the transcriptome sequencing and annotation analysis, the majority of the genes encoding the enzymes in the carotenoid biosynthesis pathway were identified in the unigene dataset (Fig. 4a). LcGGPS1 and LcGGPS2, LcPSY, LcPDS, LcZDS, LcCRTISO, LcLYC-E, LcLYC-B, LcCHYB LcVDE and LcNCED were isolated by Rapid amplification of cDNA ends (RACE) and Reverse Transcription-Polymerase Chain Reaction (RTPCR). When the Sanger-sequenced, full-length cDNA were aligned against the unigene dataset, 27 unigenes were mapped to different regions of these genes respectively. It is noteworthy that most of the unigenes showed high similarity with the full-length genes (Table 2). These data further demonstrated that the unigenes obtained by RNA-
G. Wang et al. / Gene 555 (2015) 458–463
461
Fig. 2. GO classifications of the L. chinense unigenes. A total of 16,073 unigenes were assigned to at least one GO term and were grouped into three main GO categories and 44 subcategories. Right Y-axis represents number of genes in a category. Left Y-axis indicates percentage of a specific category of genes in each main category.
seq were efficiently assembled and the quality of sequencing data may contribute to metabolic pathway research in L. chinense. qRT-PCR analysis was further used to compare the expression levels of the genes involved in carotenoid biosynthesis in leaf and ripening fruit. It has been demonstrated that the expression levels of the majority of the genes related to carotenoid biosynthesis in the ripening fruit were higher than that in the leaf. LcGGPS1, LcCHYB and LcVDE were highly expressed in the leaf, while LcCHYB and LcNCED were highly expressed in the ripening fruit (Fig. 4b). Intriguingly, the expression of LcGGPS1 was significantly higher than LcGGPS2 in the leaf, while the
expression of LcGGPS2 was much higher than LcGGPS1 in the ripening fruit. Thus, LcGGPS1 may be essential in the leaf, while LcGGPS2 play an important role in the ripening fruit. In addition, the expression level of LcLYC-B was much higher than those of LcLYC-E in both the leaf and the ripening fruit. And LcCHYB was expressed at very high level both in the leaf and in the ripening fruit, which may contribute to the carotenoid biosynthesis and accumulation in L. chinense. Furthermore, LcNCED was expressed at a high level in the ripening fruit compared to that in the leaf, and it may play an important role in the stage of fruit ripening.
Fig. 3. COG function classification of the L. chinense unigenes. In total, 18,803 of the 37,816 sequences with NR hits were grouped into 25 COG classifications. Y-axis indicates number of unigenes in a specific functional cluster.
462
G. Wang et al. / Gene 555 (2015) 458–463
Fig. 4. Carotenoid metabolism pathway (a) and relative expression levels of related genes in the leaf and the ripening fruit (b). The numbers in the brackets following the gene name in subpanel a indicate the number of unigenes mapped to that gene. IPP, isopentenyl pyrophosphate; DMAPP, Dimethylallyl pyrophosphate; GGPP, geranylgeranyl pyrophosphate; GGPS, geranylgeranyl diphosphate synthase; PSY, phytoene synthase; PDS, phytoene desaturase; ZDS, ζ-carotene desaturase; CRTISO, carotene isomerase; LCY-E, lycopene ε-cyclase; LCY-B, lycopene β-cyclase; CHYB, β-carotene hydroxylase; ZEP, zeaxanthin epoxidase; VDE, violaxanthin de-epoxidase; NXS, neoxanthin synthase.
3.7. Analysis of carotenoid contents in leaf and green ripening fruit of L. chinense HPLC analysis was used to determine the carotenoid composition. It demonstrated that the total contents of carotenoids in leaf were much higher than those in the green ripening fruits. The β-carotene content (5.75 ± 0.61 μg/g) was found to be the highest in the leaf and the highest β-carotene content (5.47 ± 0.20 μg/g) was observed in the green ripening fruit. And the amount of lutein and zeaxanthin in the leaf (4.03 ± 0.46 μg/g), was more than three times higher compared to that in green ripening fruit. The content of neoxanthin in the leaf (0.80 ± 0.08 μg/g) was also observed more than three times higher Table 2 Sequence analysis of the cloned genes involved in carotenoid biosynthesis. Gene
Length
LcGGPS1 LcGGPS2 LcPSY LcPDS LcZDS LcCRTISO LcLCY-B LcLCY-E LcCHYB LcVDE LcNCED
1203 1246 1341 2021 2042 1815 1506 1847 915 1413 1827
bp bp bp bp bp bp bp bp bp bp bp
Characteristics of correspondent unigenes Coverage
ORF
Identities
RPKM
100.0% 100.0% 100.0% 97.5% 100.0% 99.9% 100.0% 94.4% 100.0% 100.0% 99.4%
Complete Complete Complete Part Complete Complete Complete Part Complete Complete Part
1201/1203 (99.8%) 1246/1246 (100.0%) 1320/1341 (98.4%) 1955/1970 (99.2%) 2025/2032 (99.7%) 1813/1813 (100%) 1496/1506 (99.3%) 1739/1744 (99.7%) 915/915 (100.0%) 1412/1413 (99.9%) 1811/1817 (99.7%)
157.17 27.91 50.80 31.89 81.40 14.96 62.03 31.40 128.70 152.26 114.24
compared to that in green ripening fruit (0.26 ± 0.01 μg/g) (Table 3). Thus, the accumulation of carotenoids was dependent on the regulation of different carotenogenic enzymes.
4. Discussion RNA-Seq has become an important tool for comprehensive studies of gene expression, detection of novel transcripts, and discovery of genes or expressed SNPs associated with valuable traits. In this study, we generated de novo transcriptome sequences and characterized the leaf transcriptome of L. chinense, and subsequent annotation of gene functions. Chinese wolfberries contain high levels of carotenoids, such as βcarotene, lutein and zeaxanthin, which are essential nutrients for human health. In the present study, almost all of genes encoding related enzymes in the carotenoid biosynthesis pathway were found in the transcriptome dataset. qRT-PCR analysis showed significantly higher expression levels of LcGGPS2, LcPSY, LcLCY-B, LcCHYB, and LcVDE genes in the ripening fruit, thereby possibly explaining the high β-carotene, lutein and zeaxanthin contents in Chinese wolfberry. As a key gene in ABA biosynthesis and closely related to carotenoid accumulation in plants, LcNCED was one of the most highly expressed carotenoidrelated genes in the ripening fruits, and probably plays an important role in fruit ripening. In higher plant, the condensation of three IPP and one DMAPP units to form GGPP is a crucial step in the carotenoid biosynthesis pathway, which is catalyzed by GGPS. Two isoforms of the GGPS gene, LcGGPS1 and LcGGPS2, were identified and cloned in L. chinense. LcGGPS1 was
G. Wang et al. / Gene 555 (2015) 458–463
463
Table 3 The carotenoid compositions of leaf and the green ripening fruit. Samples
Total carotenoids (μg/g)
Neoxanthin (μg/g)
Vioxanthin (μg/g)
Lutein + zeaxanthin (μg/g)
β-Carotene (μg/g)
Leaf Ripening fruit
10.85 ± 1.17 7.52 ± 0.25
0.80 ± 0.08 0.26 ± 0.01
0.26 ± 0.036 0.50 ± 0.01
4.03 ± 0.46 1.29 ± 0.04
5.75 ± 0.61 5.47 ± 0.20
highly expressed in leaf, while LcGGPS2 was detected at a rather low expression level in leaf but highly expressed in the ripening fruit. Thus, we presumed that LcGGPS2 may encode a fruit-specific isoform and have important roles in ripening fruit at specific development stages, while LcGGPS1 dominated in the leaf. The expression patterns of LcGGPS2 and LcGGPS1 were probably similar to the PSY2 and PSY1 in tomato, respectively (Fraser et al., 2007). Also we noticed that the expression level of LcLYC-B was much higher than LcLYC-E in both the leaf and the ripening fruit. Cyclization of lycopene marks a branch point involved in the pathway of the plant carotenoid biosynthesis to either α- or βcarotene. Lycopene β-cyclase (LCY-B) catalyzes lycopene to produce β-carotene, whereas conversion to α-carotene is catalyzed by lycopene ε-cyclase (LCY-E). LCY-B is one of the crucial enzymes for carotenoid biosynthesis and believed to predominate in the formation of vegetative carotenoids (Fraser et al., 2007). The higher expression level of LYC-B in Chinese wolfberry may permanently maintain higher metabolic flux toward β-carotene while the lower expression level of LYC-E may maintain lower metabolic flux to α-carotene, thereby leading to the high βcarotene and low α-carotene contents in the fruits of Chinese wolfberry. The expression pattern of LcCHYB with the highest expression level in both the leaf and the ripening fruit may reveal its important role in the carotenoid biosynthesis and accumulation in L. chinense. Based on HPLC analysis, the lutein and zeaxanthin contents in the green ripening fruits were much less than in the leaf, which may be resulted from the expression level of related genes. In the leaf, relative expressions of almost all upstream genes of lutein and zeaxanthin in the carotenoid biosynthetic pathway were much lower than that in the green ripening fruit. However, the accumulation of lutein and zeaxanthin was dependent on the regulation of different carotenogenic enzymes. In the ripening fruit, the downstream genes of lutein and zeaxanthin may regulate their expression. The expression of LcVDE was much lower, while the expression of LcNCED was higher in the ripening fruit (Fig. 4). The high expression levels of LcNCED may be responsible for lutein and zeaxanthin accumulation at the early stages of ripening and lead to the low expression of neoxanthin (0.26 ± 0.01 μg/g in the fruit compared to 0.80 ± 0.08 μg/g in the leaf). In summary, the identification of these key genes in the carotenoid biosynthesis pathway might contribute to further elucidate the expression patterns of carotenoid pathway genes at different stages of development and fruit ripening and the mechanisms of carotenoid biosynthesis/accumulation in L. chinense. This will be of great significance to cultivate elite Chinese wolfberry varieties with high carotenoid contents for improved nutritional properties. As to our knowledge, this study is the first application of RNA-Seq in investigating the leaf transcriptome of L. chinense. A large number of transcriptomic sequences may provide sufficient data to discover almost all of the known genes in major metabolic pathways. The functional annotations and further analysis of these genes are useful in exploring the biosynthesis mechanisms of nutritional components in L. chinense. Consequently, these datasets may provide an important public information platform for further research in gene expression and functional genomics in L. chinense. Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.gene.2014.10.058. Acknowledgments This work was supported by the National Natural Science Foundation of China (grant nos. 31271419 and 31271793) and National Science
and Technology Key Project of China on GMO Cultivation for New Varieties (No. 2014ZX08003-002B). This work was also supported by High Performance Computing Center of Tianjin University, China. The authors thank the Beijing Genome Institute at Shenzhen, China for the technical support.
References Auer, P.L., Doerge, R.W., 2010. Statistical design and analysis of RNA sequencing data. Genetics 185, 405–416. Chang, R.C., So, K.F., 2008. Use of anti-aging herbal medicine, Lycium barbarum, against aging-associated diseases. What do we know so far? Cell. Mol. Neurobiol. 28, 643–652. Conesa, A., Götz, S., García-Gómez, J.M., Terol, J., Talón, M., Robles, M., 2005. Blast2GO: a universal tool for annotation, visualization and analysis in functional genomics research. Bioinformatics 21, 3674–3676. Fraser, P.D., Enfissi, E.M., Halket, J.M., Truesdale, M.R., Yu, D., Gerrish, C., Bramley, P.M., 2007. Manipulation of phytoene levels in tomato fruit: effects on isoprenoids, plastids, and intermediary metabolism. Plant Cell 19, 3194–3211. Gan, L., Zhang, S.H., Yang, X.L., Xu, H.B., 2004. Immunomodulation and antitumor activity by a polysaccharide–protein complex from Lycium barbarum. Int. Immunopharmacol. 4, 563–569. Grassi, S., Piro, G., Lee, J.M., Zheng, Y., Fei, Z., Dalessandro, G., Giovannoni, J.J., Lenucci, M.S., 2013. Comparative genomics reveals candidate carotenoid pathway regulators of ripening watermelon fruit. BMC Genomics 14, 781. Harr, B., Turner, L.M., 2010. Genome‐wide analysis of alternative splicing evolution among Mus subspecies. Mol. Ecol. 19, 228–239. Hu, Z., Wu, Y.-R., Li, W., Gao, H.-H., 2006. Factors affecting Agrobacterium tumefaciensmediated genetic transformation of Lycium barbarum L. In Vitro Cell. Dev. Biol. Plant 42, 461–466. Huang, H.H., Xu, L.L., Tong, Z.K., Lin, E.P., Liu, Q.P., Cheng, L.J., Zhu, M.Y., 2012. De novo characterization of the Chinese fir (Cunninghamia lanceolata) transcriptome and analysis of candidate genes involved in cellulose and lignin biosynthesis. BMC Genomics 13, 648. Iseli, C., Jongeneel, C.V., Bucher, P., 1999. ESTScan: a program for detecting, evaluating, and reconstructing potential coding regions in EST sequences. Proc. Int. Conf. Intell. Syst. Mol. Biol. 138–148. Ji, J., Wang, G., Wang, J., Wang, P., 2009. Functional analysis of multiple carotenogenic genes from Lycium barbarum and Gentiana lutea L. for their effects on beta-carotene production in transgenic tobacco. Biotechnol. Lett. 31, 305–312. Kanehisa, M., Araki, M., Goto, S., Hattori, M., Hirakawa, M., Itoh, M., Katayama, T., Kawashima, S., Okuda, S., Tokimatsu, T., Yamanishi, Y., 2008. KEGG for linking genomes to life and the environment. Nucleic Acids Res. 36, D480–D484. Li, X.M., Li, X.L., Zhou, A.G., 2007. Evaluation of antioxidant activity of the polysaccharides extracted from Lycium barbarum fruits in vitro. Eur. Polym. J. 43, 488–497. Li, R., Zhu, H., Ruan, J., Qian, W., Fang, X., Shi, Z., Li, Y., Li, S., Shan, G., Kristiansen, K., 2010. De novo assembly of human genomes with massively parallel short read sequencing. Genome Res. 20, 265–272. Li, D., Deng, Z., Qin, B., Liu, X., Men, Z., 2012a. De novo assembly and characterization of bark transcriptome using Illumina sequencing and development of EST-SSR markers in rubber tree (Hevea brasiliensis Muell. Arg.). BMC Genomics 13, 192. Li, X., Sun, H., Pei, J., Dong, Y., Wang, F., Chen, H., Sun, Y., Wang, N., Li, H., Li, Y., 2012b. De novo sequencing and comparative analysis of the blueberry transcriptome to discover putative genes related to antioxidants. Gene 511, 54–61. Liu, M., Qiao, G., Jiang, J., Yang, H., Xie, L., Xie, J., Zhuo, R., 2012. Transcriptome sequencing and de novo analysis for Ma bamboo (Dendrocalamus latiflorus Munro) using the Illumina platform. PLoS One 7, e46766. Liu, S., Li, W., Wu, Y., Chen, C., Lei, J., 2013. De novo transcriptome assembly in chili pepper (Capsicum frutescens) to identify genes involved in the biosynthesis of capsaicinoids. PLoS One 8, e48156. Luo, Q., Li, Z., Huang, X., Yan, J., Zhang, S., Cai, Y.-Z., 2006. Lycium barbarum polysaccharides: protective effects against heat-induced damage of rat testes and H2O2induced DNA damage in mouse testicular cells and beneficial effect on sexual behavior and reproductive function of hemicastrated rats. Life Sci. 79, 613–621. Ong, W.D., Voo, L.-Y.C., Kumar, V.S., 2012. De novo assembly, characterization and functional annotation of pineapple fruit transcriptome through massively parallel sequencing. PLoS One 7, e46937. Tian, M., Wang, M., 2006. Studies on extraction, isolation and composition of Lycium barbarum polysaccharides. China J. Chin. Mater. Med. 31, 1603–1607. Xu, D.-L., Long, H., Liang, J.-J., Zhang, J., 2012. De novo assembly and characterization of the root transcriptome of Aegilops variabilis during an interaction with the cereal cyst nematode. BMC Genomics 13, 133. Ye, J., Fang, L., Zheng, H., Zhang, Y., Chen, J., Zhang, Z., Wang, J., Li, S., Li, R., Bolund, L., 2006. WEGO: a web tool for plotting GO annotations. Nucleic Acids Res. 34, W293–W297.
