Genome-wide identification and expression profiling of the carotenoid cleavage dioxygenase (CCD) gene family in Brassica napus L

Carotenoid cleavage dioxygenase (CCD), a key enzyme in carotenoid metabolism, cleaves carotenoids to form apo-carotenoids, which play a major role in plant growth and stress responses. CCD genes had not previously been systematically characterized in Brassica napus (rapeseed), an important oil crop worldwide. In this study, we identified 30 BnCCD genes and classified them into nine subgroups based on a phylogenetic analysis. We identified the chromosomal locations, gene structures, and cis-promoter elements of each of these genes and performed a selection pressure analysis to identify residues under selection. Furthermore, we determined the subcellular localization, physicochemical properties, and conserved protein motifs of the encoded proteins. All the CCD proteins contained a retinal pigment epithelial membrane protein (RPE65) domain. qRT-PCR analysis of expression of 20 representative BnCCD genes in 16 tissues of the B. napus cultivar Zhong Shuang 11 (‘ZS11’) revealed that members of the BnCCD gene family possess a broad range of expression patterns. This work lays the foundation for functional studies of the BnCCD gene family.


Introduction
"Carotenoids" is the general term for the class of natural pigments widely present in animals, plants, and microorganisms. Carotenoids are lipid-soluble isoprene-like compounds that contain 40 carbon molecules and comprise more than 750 pigments with different structures [1]. These pigments have numerous important biological functions; for example, they are photoprotective and are indispensable components of photosynthesis [2,3]. They also can scavenge free radicals and have antioxidant properties [4]. Furthermore, carotenoids are components of cell membranes, potential anti-cancer agents, and can interact with proteins [5]. Carotenoids are exploited as coloring agents in flowers and fruits to attract pollinators and agents of seed dispersal [6][7][8]. In addition, carotenoids are precursors of important phytohormones, such as abscisic acid and strigolactones, which regulate plant development and plant-environment interactions [9][10][11].
The carotenoid biosynthesis pathway in plants has been fully elucidated; the catalytic oxidative cracking of carotenoids is a key process in this pathway. Multiple conjugated double bonds exist within the carotenoid center chain, which can be specifically cleaved by carotenoid cleavage dioxygenases (CCDs) to form a variety of apo-carotenoids, and some apo-carotenoids can be further degraded into small biologically active molecules [12]. In plants, the addition of two oxygen atoms to the cleavage product means that CCDs possess the characteristics of a dioxygenase [13]. Moreover, CCDs are also a class of non-heme oxygenases, whose catalytic activity requires Fe 2+ as a cofactor [14,15]. CCD proteins contain four highly conserved histidine residues bound to Fe 2+ and all contain a retinal pigment epithelial membrane protein (RPE65) domain that is characteristic of enzymes involved in carotenoid cleavage [16,17].
In plants, CCD proteins are encoded by an ancient gene family. CCD gene family consists of two subfamilies: carotenoid cleavage dioxygenases (CCDs) and 9-cis epoxycarotenoid dioxygenases (NCEDs) [12]. Through analysis of a novel ABA-deficient mutant of maize, the first protein found to specifically cleave carotenoids, viviparous14 (VP14), was identified by Schwartz et al [14]. Vallabhaneni et al. described the characteristics of the CCD gene family in the grass species maize (Zea mays), rice (Oryza sativa), and sorghum (Sorghum bicolor) [18]. During water stress and seed dormancy, TaNCED of Triticum aestivum might play a primary role in regulation of ABA content [19]. Besides, drought stress can induce expression of NCED3 and accumulation of endogenous ABA in Nicotiana tabacum [20]. Wei et al. found at least seven CCD genes in Solanum lycopersicum genome sequence and analyzed their expression patterns [21]. CCD genes have also been identified or functionally expressed in a variety of other plant species, such as Glycine max [22], Gossypium hirsutum, Solanum tuberosum [23], Cucurbita pepo [24], Saccharum officinarum [25], Crocus sativus, Osmanthus fragrans [26], Vitis vinifera [27], Mangifera indica [28] and Amygdalus persica [29].
In Arabidopsis thaliana, the CCD gene family consists of nine members: four CCD genes (AtCCD1, 4, 7, and 8) and five NCED genes (AtNCED2, 3, 5, 6 and 9) [30]. Carotenoid cleavage dioxygenase homologs in other plant species are named according to the system used for Arabidopsis CCD family members.
The CCD1 and CCD4 enzymes catabolize a variety of carotenoids and produce volatile apo-carotenoids, which are important for the biosynthesis of aromas and flavors of flowers and fruits, respectively [31,32]. The carotenoid content of mature seeds of AtCCD1 mutants was higher than that in the wild type, indicating a role for CCD1 in carotenoid catabolism [33]. CCD4 regulates carotenoid homeostasis and CmCCD4a is specifically expressed in white petals of Chrysanthemum morifolium [34]. RNAi-mediated inhibition of CmCCD4a expression in white-petaled plants results in the production of yellow flowers [35]. In line with this observation, loss or downregulation of CmCCD4a function led to an increase in carotenoid content in petals [12]. CCD7 and CCD8 function in strigolactone biosynthesis, regulate axillary bud growth, and inhibit branching [36,37]. Using RNAi to reduce Actinidia chinensis CCD8 expression increases in branch development and delays leaf senescence [38].
The reaction catalyzed by 9-cis-epoxycarotenoid dioxygenase (NCED) is a rate-limiting step in ABA biosynthesis, and thus influences plant tolerance to diverse abiotic stresses [39]. AtNCED5, AtNCED6 and AtNCED9 are the dominant contributors to developmentally regulated ABA synthesis in seeds and thereby regulate seed embryo maturation and dormancy [30]. Overexpression of OsNCED3 increased drought resistance in rice and caused an increased ABA level [40]. AtNCED2 and AtNCED3 transcripts are abundant in Arabidopsis roots and their encoded proteins function in abscisic acid biosynthesis and thereby regulate lateral root growth [30]. Furthermore, heterologous expression of Brassica napus NCED3 led to ABA accumulation and NO and ROS generation in transgenic Arabidopsis plants, thereby enhancing abiotic stress tolerance [41].
B. napus, an important oilseed crop globally, is an allopolyploid derived from a natural interspecific cross between Brassica rapa (turnip; 2n = 2x = 20) and Brassica oleracea (kohlrabi; 2n = 2x = 18). The important physiological functions of carotenoid cleavage products in plants, such as abscisic acid and strigolactones, have prompted studies of the lyases involved in carotenoid metabolism. Little is known about the BnCCD gene family. The availability of the B. napus genome sequence would enable the identification and analysis of members of this family [42].
In this study, we identified 30 BnCCD genes. In addition to analyzing their gene evolution and structure, chromosomal localization, conserved motifs, and cis-acting promoter elements, we determined their tissue-and organ-specific expression and examined the physicochemical properties of their encoded proteins. The results form a solid basis for further studies on the biological functions of the BnCCD gene family.

Plant materials
The B. napus cultivar Zhong Shuang 11 ('ZS11') was planted in Chongqing, China (29˚45' N, 106˚22' E). To analyze the expression patterns of BnCCD genes, four different tissues of 'ZS11' were harvested when plants were in full flower: mature leaves (Le), sepals (Se), flowers (F), and stems (St). The seeds (S) and silique pericarps (Sp) were harvested at different timepoints following the termination of flowering (25,30,35,40,45, and 50 days after flowering). Samples were immediately frozen in liquid nitrogen and stored at -80˚C for further use.

Brassica. oleracea
The coding sequences of AtCCD genes were downloaded from TAIR (https://www. arabidopsis.org/) and used as reference sequences. The BnCCD, BrCCD, and BoCCD genes were identified through the BLASTN analysis [43] of AtCCD genes against the Brassica Database (BRAD, http://brassicadb.org/brad/index.php) [44]. The local database was established using Geneious 4.8.5 software (http://www.geneious.com/; Biomatters, Auckland, New Zealand). The genome-wide alignment was verified by the MEGABLAST program [45], and the following screening standards were used: the consistency of aligned sequences with the reference sequence was �80% and gene sequences <700 bp were discarded. Because enzymes involved in carotenoid cleavage contain a retinal pigment epithelial membrane protein (RPE65) domain [16,17], gene sequences from the Pfam database [46] that did not contain this domain were excluded.

Multiple sequence alignment and phylogenetic analysis
CCD amino acid sequences were subjected to multiple sequence alignment using MUSCLE [47] with default parameters. The evolutionary relationships of B. napus CCD proteins with those of A. thaliana, B. oleracea, and B. rapa were analyzed using MEGA7 [45]. To identify the conserved blocks of all predicted sequences, the Gblocks program was used [48]. The substitution saturation was detected using DAMBE [49]. A phylogenetic tree was constructed using the neighbor-joining method implemented in MEGA7. The number of bootstrap replications was 1,000, and paired deletion was performed. The phylogenetic tree was visualized using

Protein properties, sequence analysis, and duplication time inference
The chromosomal location of BnCCD genes, which was queried from the Brassica napus genome browser, and the genes were mapped onto chromosomal linkage groups by Mapchart software [50]. The Gene Structure Display Server (GSD 2.0) (http://gsds.cbi.pku.edu.cn/index. php) was used to portray the exon-intron structures of the BnCCD genes. The ExPASy proteomics server database (http://expasy.org/) [51]was used to predict the relative molecular weight, theoretical isoelectric point, protein stability, and aliphatic amino acid content of BnCCD proteins. The conserved motifs were identified using the MEME Version 5.0.5 online tool (http:// meme-suite.org/tools/meme) [52] and the maximum motif retrieval value was set to 20; other parameters used the default settings. Annotations of the identified motifs were obtained from InterProScan (www.ebi.ac.uk/Tools/InterProScan/) [53].
To determine whether the CCD protein-coding sequences are under selective pressure, the ratios of synonymous substitution rate (ks) and non-synonymous substitution rate (ka) of homologous gene pairs were calculated using TBtools software [54]. The approximate date of duplication events was inferred by substituting Ks values into the formula T ¼ Ks= 2 � 1:5 � 10 À 8 À � � 10 À 6 million years ago (MYA) [55].
All BnCCD genes expression levels were quantified in terms of FPKM (fragments per kilobase of exon per million mapped fragments) using Cufflinks with default parameters [60], and then extracted from RNA-seq data according to their B. napus code, including expression data from thirteen different organs (roots, stems, leaves, buds, anthocauli, calyxes, petals, pistils, stamens, anthers, capillaments, seeds and silique pericarps) at different developmental stages in B.napus cultivar ZS11. The heatmap for BnCCD genes was constructed using TBtools software.

RNA extraction and quantitative real-time PCR
Total RNA was extracted using the RNeasy Extraction Kit (Invitrogen, Carlsbad, CA, USA) according to the manufacturer's instructions. Contaminating genomic DNA was removed with DNase I. Total RNA (1 μg) was used to synthesize cDNA by reverse transcriptase (TaKaRa). The gene-specific primer pairs used to analyze BnCCD genes expression by qRT-PCR were designed using Primer Premier 5 [61] and the ACTIN 7 gene was used as an endogenous reference gene. All the primers were listed in S1 Table. Quantitative RT-PCR was carried out using TB Green Premix Ex Taq on a Real-Time PCR Detection System (Bio-Rad, Hercules, CA, USA). The reaction mixture included 10 μL TB Green Premix Ex Taq, 0.5 μL forward primer, 0.5 μL reverse primer, 2 μL cDNA template, and 7 μL nuclease-free H 2 O in a total volume of 20 μL. The PCR cycling conditions were as follows: 95˚C for 30s, followed by 40 cycles of 95˚C for 5 s and 60˚C for 30 s. Following PCR amplification, the dissolution curve was analyzed to ensure the specificity of the amplified products. Three biological replicates and three technical replicates were performed for each reaction. The relative gene expression levels were calculated using the 2 −ΔΔCt method [61]. All the results were plotted as the mean ± standard error of mean (SEM) from three independent biological replicates using Graph Pad Prism 5.0 software (GraphPad Software Inc., San Diego, CA, USA).

Identification and characterization of BnCCD genes
Using the nine AtCCD coding sequences as a BLASTN query, we identified 30 B. napus CCD genes (BnCCDs), 16 originating from B. rapa (BrCCDs) and 14 from B. oleracea (BoCCDs)(S2 Table). All the genes contained sequences encoding the RPE65 domain. RPE65 belongs to a family of carotenoid oxygenases in plant, bacterial, and animal systems that typically oxidatively cleave conjugated double bonds in the polyene backbone of carotenoids [16,17]. In addition to light related functions, the function and mechanism of RPE65 in plants have not been elucidated. To determine the evolutionary relationships among 60 CCD proteins, we constructed a phylogenetic tree by the neighbor-joining method. Based on the topology of the phylogenetic tree, CCD proteins could be grouped into nine distinct subgroups (I, II, III, IV, V, VI, VII, VIII and IX), each of which contained one of the AtCCD proteins and the corresponding subgroup members of B. rapa, B. oleracea, and B. napus (Fig 1).
The physical and chemical properties of the BnCCD proteins are summarized in Table 1. The length of BnCCD proteins ranged from 255 amino acids (aa) (BnNCED3f) to 668 aa (BnCCD7a), with a mean length of 551 aa. The predicted molecular weights varied from 28.65 kDa (BnNCED3f) to 74.79 kDa (BnCCD7a) and the theoretical isoelectric point (PI) ranged from 4.75 (BnNCED3f) to 8.83 (BnNCED3e). Out of the 30 BnCCD proteins, 27 had PI values of less than 7, and the values of the remaining three BnCCD proteins were greater than 7. Based on the hydrophilicity index of amphoteric proteins between -0.5~+0.5 (a negative GRAVY value indicates hydrophilicity and a positive value indicates hydrophobicity), all BnCCD proteins are amphiphilic proteins.

Chromosomal localization and gene structure
By Mapchart, an analysis of the chromosomal distribution of the BnCCD loci showed that these genes were not evenly distributed across the chromosomes (Fig 2). The 30 BnCCD loci are distributed on 13 chromosomes in B. napus; the 14 BnCCDs are located on the A subgenome and the 16 BnCCDs are located on the C subgenome. Chromosomes A01 and C01 possess the most BnCCD loci, each containing four BnCCDs. Chromosomes A03, A04, A05, A07, A08, C06, A03-random, A07-random, and Ann-random each contain only one BnCCD locus, whereas chromosomes A09, C04, C07, C08, and Cnn-random each contain two BnCCD loci. BnNCED3e, BnNCED3f, and BnNCED3g (subgroup III) are located on chromosome C05.
The positions of exons and introns in members of a gene family might have played crucial roles during evolution [62]. Therefore, we next analyzed the gene structure (exon-intron organization) of the 30 BnCCDs (Fig 3). Genes with similar structures were closely related.

PLOS ONE
for annotations for the conserved motifs. Motifs 1-9 and 19 were associated with carotenoid oxygenase (IPR004294) and the remaining motifs had not been annotated. Prediction of secondary structure using SOPMA showed that the main structure of the BnCCD proteins was a random coil. In addition to BnNCED3e, the alpha helix was more than the extension chain. The alpha helix was equal to the extension chain in BnCCD7a. The secondary structure characteristics of other BnCCD proteins ranged from more to less: random curl, extended strand, alpha helix, beta turn. Analysis using the TMHMM Server v. 2.0 showed that no BnCCD proteins possessed a transmembrane domain.
Nineteen of the 30 BnCCD proteins were predicted to be localized to the chloroplast, five to the cytoplasm, and five to the mitochondrion. BnNCED6b was predicted to be localized to the peroxisome.

Ka and Ks calculation for orthologous CCD genes between Brassica napus and Arabidopsis thaliana
To determine whether the CCD protein-coding genes in B. napus and A. thaliana are under selective pressure, the Ka/Ks ratios for 30 pairs of orthologous genes were calculated using TBtools software ( Table 2). The Ka/Ks ratios of all the gene pairs were considerably lower than

The cis-acting elements predicted to be present in BnCCD promoters
The cis-acting elements present in the promoters are essential for transcriptional gene regulation. In total, 87 types of cis-acting elements were predicted to be present in 30 BnCCD promoters (S3 Table), using the online software PlantCARE (http://bioinformatics.psb.ugent.be/ webtools/plantcare/html/). In addition to CAAT and TATA boxes, each promoter contained more than 10 cis-acting elements. Among these, 26 different elements are associated with responses to light, with 206 occurrences across all promoters. The remaining elements are associated with hormonal and stress responses and tissue-specific expression. Hormoneresponsive cis-elements account for a large proportion of the total and include ABRE (abscisic acid response element), AuxRR-core and TGA-elements (auxin response), ERE (ethyleneresponse element), TCA-element (salicylic acid response), P-box and GARE-motif (gibberellin-response), and the CGTCA and TGACG motifs (MeJA-response). Notably, all BnCCDs contain ABRE, CGTCA, and TGACG motifs, suggesting that this gene family functions in hormone response pathways. Additional cis-elements related to stress response were identified; for example, LTR (low-temperature responsiveness) is present in all BnCCD promoters except for BnNCED3f and BnCCD8a. Additional elements include the drought-inducible MBS element, TC-rich repeats, a cis-acting element involved in defense and stress responsiveness, and the WUN-motif, a wound-responsive element.
Some cis-acting elements in BnCCD promoters might determine the tissue-specific expression pattern, such as the CAT-box, which is associated with meristematic expression; GCN4_motif, involved in endosperm expression; HD-Zip 1, involved in differentiation of the palisade mesophyll cells; and RY-element, which confers seed-specific expression.

RNA-seq analysis
We analyzed the expression of 30 BnCCDs at different developmental stages and in various tissues from the B. napus cultivar ZS11 and constructed an expression heat map (Fig 5). Whereas no BnNCED3f and BnNCED3g transcripts were detected in any tissue, the remaining 28 BnCCDs showed tissue-and development-specific expression levels. BnCCD1b and BnCCD1c were highly expressed in leaves, stems, buds, flowers, seeds, and silique pericarp, but expression was relatively low in roots. BnNCED9a and BnNCED9b had similar expression patterns and were highly expressed in seeds. Genes belonging to the same subgroup had a wide range of expression patterns: BnCCD4a and BnCCD4c were highly expressed in leaves, stems, and pericarps; however, BnCCD4b, which belongs to the BnCCD4 subgroup, had a low expression level in leaves, stems, and pericarps but was highly expressed in anthers, suggesting that it might function in stamen growth and development.

Expression profiling of BnCCDs in different organs
To gain insight into the biological functions of BnCCDs, we analyzed their expression patterns in flowers, leaves, sepals, stems, seeds, and silique pericarps, and for seeds and silique pericarps also at different developmental stages (25, 30, 35 (Fig 6). Except for BnCCD8b, the expression level of CCD genes was low in the stem. BnNCED5b and BnCCD8a transcripts were abundant in flowers, BnNCED5a and BnNCED5b transcripts were abundant in sepals, and BnNCED9a and BnNCED9b transcripts were abundant in seeds. However, the expression patterns of some genes within the same family differed: the transcript levels of BnCCD1b and BnCCD1c in leaves were higher than those in other tissues, whereas BnCCD1d was most highly expressed in sepals. In addition, BnNCED2a and BnNCED2c transcript levels were higher in seeds at 40-50 days after flowering (40S, 45S, and 50S) than at 25-35 days after flowering (25S, 30S and 35S). BnCCD4a and BnCCD4c transcript levels were abundant in silique pericarps collected at the 25 days after flowering (25Sp), 35Sp, and 45Sp stages, but were present at low levels at the 30Sp, 40Sp, 45Sp and 50Sp stages.

Discussion
Arabidopsis contains nine CCD genes. Brassicaceae, including Arabidopsis, underwent an ancient α, β and γ whole-genome duplication polyploidization events, and also experienced an

PLOS ONE
additional recent genome-wide tripling (WGT) event [63]. Because it is an allotetraploid formed by interspecific hybridization between B. rapa and B. oleracea, B. napus would be expected to contain six copies of each Arabidopsis thaliana gene (i.e., 54) [64]. Because of genome shrinkage and gene loss, the number of BnCCDs identified in this study was a little lower (i.e., 30), with only 2-6 copies of each Arabidopsis gene present. The B. rapa and B. oleracea genomes contain 16 BrCCD and 14 BoCCD genes, respectively, and the number of BnCCDs is the sum of these. However, only two members of the BnNCED9 subgroup and three of the BnCCD4 subgroup were identified in this study, whereas two members were found in the corresponding subgroups of both B. rapa and B. oleracea, indicating that gene loss might have occurred among subgroups BnCCD4 and BnNCED9.
The molecular characteristics of all BnCCD protein members were found to differ, but proteins in the same subgroup had similar molecular weights and isoelectric points. The position of exon-intron boundaries reflects the evolution of these genes [62]. Subcellular localization predictions suggested that BnCCD proteins are localized to the chloroplast, mitochondrion, cytoplasm, and peroxisome. BnCCD1b, BnCCD1c, BnNCED3b, BnNCED3f, and BnNCED3h were predicted to be cytoplasmic, suggesting that BnCCD1 and BnNCED3 might interact in the cytoplasm. It also showed that these genes may not participate in chlorophyll photosynthesis, which was consistent with the research by Zhang et al. [28]. BnNCED2 subgroup is located in the chloroplast. Studies have shown that histidine residues in NCED amino acids can bind Fe 2+ to make NCED proteins function [65], and there is a chloroplast transit peptide structure in them [66]. Wang et al. [67] found that NCED2 protein in Camellia sinensis is localized in chroloplast and has N-terminal chloroplast targeting signal peptide sequences, which further proves that CsNCED2 has the biological activity of cleaving epoxy carotenoids to generate ABA precursors in plastids. Besides, in both sugarcane and rice, CCD8 are proteins located in chloroplast [25,68]. Through online analysis we found that BnCCD8 is also localized in the chloroplast.
Seven cis-acting elements associated with various stress responses were predicted among the promoters of the 30 BnCCDs; LTR, MBS, TC-rich repeats, and nine cis-acting elements were associated with responses to abscisic acid, MeJA, salicylic acid, and auxin. CCD1, 4, and 8 and NCED2 and 9 are transcriptionally upregulated following treatment of B. rapa with the phytohormones ABA and SL [16]. Furthermore, Nicotiana tabacum plants that heterologously expressed NCED1 of Stylosanthes guianensis had increased tolerance to light, oxidative, drought, salt, and cold stress, indicating that CCD might regulate plant tolerance against these abiotic stresses [69]. The G-Box, GT1, and ACE motifs, which are transcriptionally responsive to light, are present in the BnCCD gene promoters. In addition, BnCCD promoters contain circadian response elements, RY-elements, and MSA-like and MBSI elements, indicating that most CCD genes function in plant growth and development and stress responses.
Based on RNA-seq data (S4 Table), we constructed an expression heatmap, which demonstrated that the 30 BnCCDs were differentially expressed in different B. napus tissues. Members of the same gene subgroup occasionally exhibited different expression patterns, such as BnCCD1 and BnCCD4, potentially reflecting new functionalization among the gene family during evolution. Genes of the CCD1 and CCD4 subgroups function in the formation of a variety of apo-carotenoids, which confer unique colors, tastes, and aromas [70][71][72]. A CCD1 lossof-function mutant showed a decreased level of β-ionone in tomato fruit (Solanum lycopersicum) [31] and petunia flowers (Petunia hybrida) [73]. Carotenoid homeostasis is regulated by CCD4 in different tissues, such as Arabidopsis seeds [74] and potato tubers [23]. BnCCD4b was mainly expressed in stamens, suggesting that it might function in stamen growth and development. Species such as Solanumly copersicum, Prunus persica, and Crocus, were also reported to preferentially express CCD4 in floral organs, indicating that the evolution of CCD4 genes might have been adaptive and have enhanced specific physiological traits unique to flowering plants [21,75,76]. CCD4 with normal function or lack of function will change the color of fruit and flower organs. Previous studies have suggested CCD4 gene can fade the yellow petals of Rhododendron japonicum and Eustoma grandiflorum [77,78], because its ability of cleaving carotenoids. Inactivation of CCD4 will change the color of B.napus and Chrysanthemum morifolium petals from white to yellow [79,80]. Controlling the expression of CCD gene family provides a way to change the color of plant petals.

Conclusion
In conclusion, we performed a comprehensive study of CCD gene family in B.napus. We identified 30 putative BnCCDs that were classed into nine subgroups (I-IX) on the basis of their phylogenetic relationships. The length of the BnCCD proteins ranged from 255 to 668 aa, and the Ks values for B. napus relative to A. thaliana ranged from 0.28852 to 0.65433. In addition, RNA-seq data and qRT-PCR analysis revealed that BnCCDs were differentially expressed in different tissues and organs, and had tissue/organ specificity and expression preference, suggesting that BnCCDs had clear function differentiation. Our results will help lay the foundation for the functional characterization of the CCD gene family and better understand the structural and functional relationships among these family members.
Supporting information S1 Table. qRT-PCR primers used to analyze BnCCDs expression.