Introduction

BIAs are low molecular weight secondary metabolites that are produced in higher plants often in response to environmental abiotic and biotic stresses, and have been pharmaceutically useful. They are derived from L-tyrosine, and diversified to generate 2500 known structures (Stranska et al. 2013). Opium poppy, P. somniferum L., one of the ancient, herbal medicines, is the only commercial source of many precious BIAs such as noscapine, morphine (Choe et al. 2011), codeine, papaverine, and sanguinarine which have been proven to have anticancer, analgesic, cough suppressant, muscle drelaxants, and antimicrobial activities, respectively (Holková et al. 2010; Dang et al. 2012). Despite the apparent vital roles in various physiological programs, specially, regulation of secondary metabolites, MYB and WRKY TFs in P. somniferum L. are ill-explored.

MYB proteins are among the largest TF superfamilies in eukaryotes (Riechmann et al. 2000). They function in diverse biological processes including regulation of primary/secondary metabolisms, biotic/abiotic stress responses, hormone syntheses, signal transduction and control of cell cycle as well as its development (Petroni et al. 2008; Gomez-Gomez et al. 2012; Chen et al. 2014). COLORED1 (C1) was the first identified gene encoding a MYB domain protein found in plants. It is involved in anthocyanin synthesis in the aleurone layer of Zea mays kernels (Paz-Ares et al. 1987). Functionally, MYBs are Sequence specific DNA-binding TFs that regulate gene expressions at the mRNA level through interaction with specific motifs in promoter regions of their target genes (Gantet and Memelink 2002; Rabara et al. 2013). Members of the MYB family are characterized by possession of a MYB domain located near the amino terminus, and are composed of one to three imperfect tandem repeats. The MYB repeat comprises a region of 50 to 53 amino acids containing three regularly spaced tryptophan/phenylalanine residues. Typically, each repeat is capable of forming three α-helices. The two that are located at the C-terminus adopt a variation of the HTH conformation that binds to the DNA major groove at the specific recognition site. The three tryptophan residues together form a hydrophobic core that participates in protein–DNA interactions (Jia et al. 2004). According to the number of repeats in the MYB domain, MYB proteins can be divided into four subfamilies: 4R-MYB with four adjacent repeats, 3R-MYB (R1R2R3-MYB) with three adjacent repeats, R2R3-MYB with two adjacent repeats, and MYB-related proteins with one or two separated repeats (Dubos et al. 2010).

WRKY proteins also represent a superfamily of plant-specific TFs (Rushton et al. 2010). They play crucial roles in a variety of biological processes including wounding, responses to biotic/abiotic stresses, and consequently secondary metabolisms (Li et al. 2012; Liu et al. 2013; Qin et al. 2013). Several WRKYs are involved in the regulation of plant growth and developmental processes such as leaf senescence, seed and trichome developments, embryogenesis, and germination (Rushton et al. 2010; Li et al. 2014). The first WRKY protein, SPF1, which takes part in the regulation of sucrose induced genes, was isolated from Ipomoea batatasin in 1994 (Ishiguro and Nakamura 1994; Xiong et al. 2013). WRKY TFs recognize the conserved consensus motif TGAC(C/T) of the W-box which has been identified in the promoter sequence of many genes such as stress-inducible ones, in addition to WRKY genes themselves, and regulate expression of target genes (Ando et al. 2014; Li and Luan 2014). They contain a 60 residue domain with the conserved WRKYGQK core sequence in their N-terminus, which is followed by either a C2H2 (Cx4–5Cx22–23HxH), or C2HC (Cx7Cx23HxC) zinc finger motif (Liu et al. 2013; Qin et al. 2013). Based on the number of WRKY domains and type of zinc-finger, WRKY proteins are classified into three distinct groups. The first group contains two WRKY domains with a C2H2 zinc-finger motif, the second group contains one WRKY domain with a C2H2 zinc-finger motif, and the third group contains one WRKY domain with a C2HC zinc-finger motif (Eulgem et al. 2000; Liu et al. 2013).

Herein, we report the isolation and characterization of MYB and WRKY TFs, and study their expression profiles in various tissues of P. somniferum L. and P. bracteatum L. Furthermore, to investigate whether the members of MYB and WRKY family are present in promoters of the genes involved in BIAs biosynthesis, promoter analysis of 10-gene clustered involved in noscapine biosynthesis in P. somniferum L. are performed.

Materials and methods

EST assembly and identification of MYB and WRKY genes

EST sequences for P. somniferum L. were downloaded from several RNAseq projects (http://www.ncbi.nlm.nih.gov/sra/). Then, 468 ESTs and 943 ESTs that had high similarity respectively to the MYB and WRKY families were selected using the Offline BLAST software, and were subsequently used to build up two consensus sequences. After that, the EST sequences were assembled via the “align-then-assemble” approach utilizing the Codon Code Aligner v. 5.0.1. Program, and then two consensus sequences for each group of ESTs were created. Next, ORF of each chosen consensus sequence was found via the ORF finder (http://www.ncbi.nlm.nih.gov/gorf/gorf.html). Finally, according to the bioinformatics analysis, gene-specific PCR primer pairs were designed for PCR-amplification of MYB and WRKY genes, separately, based on the complete ORF cDNA sequences (Table 1). The primers were confirmed by the Oligo Analyzer v.3.1 (http://eu.idtdna.com/calc/analyzer).

Table 1 PCR primers
Full size table

Sequence alignment and phylogenetic construction

After a BLASTp search on the NCBI database, the MYB and WRKY protein sequences from different species that had more than 50 % identity with the coding region of the consensus sequences were selected. For determining the relationship between the identified MYB and WRKY TFs and the downloaded proteins from the BLASTp search, multiple alignments were run utilizing the web-based ClustalW2 program (http://www.ebi.ac.uk/Tools/msa/clustalw2/). The maximum likelihood method in MEGA 6.06 was used to conduct the phylogenetic trees (Tamura et al. 2013), and 1000 iterations were used for calculating bootstrap values.

RNA extraction, and cDNA synthesis

Total RNA of poppy’s capsule, leaf, root and stem tissues was extracted using a commercial RNA extraction kit (Top Plant and Fungi RNA Purification Kit, TOPAZ GENE RESEARCH., cat No.: TGK2004, Iran). The quantity and quality of the extracted RNA was determined via a NanoDrop spectrophotometer (BioTek, EPOCH, serial: 121004C, USA), and confirmed by agarose gel electrophoresis. RNase-free DNaseI (TIANGEN Biotech(Beijing)Co.,Ltd., cat No.: RT411, China) was used to eliminate genomic DNA content. TIANScript M-MLV (TIANGEN Biotech(Beijing)Co.,Ltd., cat No.: ER104, China) was used to synthesize the first-strand cDNA.

Gene isolation, and nested-PCR

The expected lengths of amplification fragments are shown (Fig. 1). Nested primers were designed based on different positions on consensus sequences to amplify fragments which overlap with each other to confirm the characterized ORFs. The following program was used for nested-PCR: 10 pmol.μl−1 forward primer, and 10 pmol.μl−1 reverse primer, 10 mmol.L−1 dNTPs, 50–100 ng.μl−1 cDNA, 5 U.μl−1 Taq DNA Polymerase (TOPAZ GENE RESEARCH., Cat No.: TGT3001.1, Iran), 10 X Taq Buffer and 10 X MgCl2. The total volume for each reaction was 20 μl. The following thermal cycler was used for amplification: 3 min at 94 °C; 35 cycles of 30 s at 94 °C, 45 s at 47–52 °C, 1 min at 72 °C; followed by 10 min at 72 °C. The PCR products of forward.ORF and reverse.ORF (for MYB primers), and forward.1 and reverse.ORF (for WRKY primers) were cleaned using a commercial gel recovery kit (Top Gel Recovery Kit, TOPAZ GENE RESEARCH., cat No.: TGK1006, Iran). The cleaned PCR products were cloned into pTG19- T vector (Vivantis, Malaysia) in accordance with the manufacturer’s instructions, and transformed into E. coli strain DH5α. Plasmid DNA templates were purified using plasmid purification kit (Top Plasmid DNA Purification Kit, TOPAZ GENE RESEARCH., cat No.: TGK1008, Iran), and subjected to sequencing. The cDNA sequences of both MYB, and WRKY genes were determined using the ClustalW2 alignment analysis with their corresponding consensus sequences obtained from the aforementioned ESTs assembly.

Quantitative real-time PCR analysis

qRT-PCR was conducted with genes' specific primers designed based on the non-conserved region of gene sequences to ensure specific amplification (Table 1). The amplification fragments of both genes were 168 bp. qRT-PCR was performed using a BioRad system with the fluorescent dye SYBR®Green Master Mix 2X (Ampliqon, cat No.: A325402, Denmark) in accordance with the manufacturer’s recommendation. The following program was used to perform qRT-PCR: 1 μL of cDNA as template, 1 μL (10 pmol. μl−1) of each primer, 10 μL SYBR®Green Master Mix 2X in a total volume of 20 μL. qRT-PCR was run at 95 °C (15 min), 35 cycles at 95 °C (30 s), 51 °C (30), 72 °C (20 s), followed by gradient: 60–75 °C by 5 °C (5 s). Three technical replicates were performed per sample. For quantifying transcription levels, the reference gene E.F.Iα (GeneBank Accession No.: JN399225.1) was used as an internal control. Cycle thresholds (C(t)s) were analyzed using Livak method (2 -∆∆Ct) and relative expression levels were calculated using the Microsoft Excel software.

Promoter analysis and TFBS identification

The promoter sequences of PSSDR1, PSCXE1, CYP82X1, CP82X2, PSAT1, PSMT2, CYP82Y1, PSMT3, CYP719A21 and PSMT1 genes (accession numbers of JQ659007, JQ659006, JQ659002, JQ659004, JQ659008, JQ659000, JQ659005, JQ659001, JQ659003 and JQ658999, respectively) involved in noscapine biosynthesis pathway in poppy were analyzed. Since poppy is a non-model plant, its complete genome sequence is not available, therefore four BACs with accession numbers of JQ659009, JQ659010, JQ659011 and JQ659012 contained these 10 genes were obtained from the NCBI database. These BACs were merged to build up a BAC containing all 10 genes. Afterwards, pairwise sequence alignment between individual genes and the BAC was performed to locate each gene. Then, orientation, introns, exons, and start and end codons of these 10 genes were obtained from NCBI database. Subsequently, considering orientation of each gene, 1500 bp upstream of the TSS was isolated by version 7.0.4.1 of the BioEdit software (Hall 1999) as promoter for analysis. After promoter regions were isolated, cis-regulatory elements were found in silico using the MotifSampler program (Thijs et al. 2002). In order to identify overrepresented motifs in the cluster, the POBO program was employed. The web-based POBO program, which compares the frequency of a given motif in a set of promoter sequences with a random set of Arabidopsis promoter sequences, confirmed whether general consensus promoter motifs are statistically overrepresented in set of promoter sequences or not. (Kankainen and Holm 2004) (ekhidna.biocenter.helsinki.fi/poxo/pobo). After overrepresented motifs were identified, putative trans-regulatory elements which interact to the motifs were found computationally using the TRANSFAC (Wingender et al. 1996), PLACE (Higo et al. 1999) and Soft Berry (NSITE) (Solovyev et al. 2010) databases. To determine the number and type of TFBSs, the whole promoter sequence of each gene was also screened using TRANSFAC database.

Results

Identification, and characterization of the two TFs

Nested-PCR was performed using designed primers to amplify fragments containing the coding region of MYB and WRKY proteins. The full-length MYB and WRKY were cloned from P. somniferum L. using the nested-PCR method. As a result of nested-PCR, the lengths of amplified fragments were exactly the same as expectation from bioinformatics analysis (Fig. 2). The cloned cDNAs of MYB and WRKY were 1513 and 1330 bp, which contained ORFs of 1029 and 981 bp encoding 342 and 326 amino acids, with nuclear localization scores larger than 0.3, and 0.4 examined via the Predict NLS website, (Rost et al. 2004), respectively. These values of nuclear localization signal indicate their transcriptional regulatory nature (submitted to GenBank as accession numbers; MYB: KP411870 WRKY: KP203854). Multiple alignments between isolated MYB, and WRKY proteins and other MYB and WRKY proteins by the ClustalW2 v.2.1 showed that the MYB protein contains two imperfect tandem repeats, and therefore belongs to the R2R3-MYB subfamily, and the WRKY protein possesses one WRKY domain with a C2H2 zinc-finger motif, hence belongs to the group II WRKY family (Fig. 3). Phylogenetic analyses suggest that the MYB is more closely related to the D.myb-like.Zea, E.MYB.Zea, C.myb-related.Setaria and F.SsMYB7. Saccharum proteins, whereas the WRKY relates to the PsWRKY58 protein (Fig. 4).

Fig. 1
figure 1

a. Positions of nested-PCR primers, and the expected lengths of PCR products containing MYB gene, b. and WRKY gene. Arrows and lines denote primers and PCR products of each primer pairs, respectively. The graph is built using the Microsoft Power Point software

Full size image
Fig. 2
figure 2

Nested-PCR analysis of P. somniferum L. using a. a: forward.ORF and reverse.ORF, b: forward.1 and reverse.ORF, c: forward.2 and reverse.ORF, d: forward.ORF and reverse.1, e: forward.1 and reverse.1, f: forward.2 and reverse.1 of MYB gene. The target fragments containing MYB gene are 1029, 1111, 1210, 1229, 1292, 1410 bp, respectively. Positive control (C+): Arabidopsis thaliana At5g43350 primers and b. a: forward.1and reverse.2, b: forward.ORF and reverse.2, c: forward.1 and reverse.1, d: forward.ORF and reverse.1, e: forward.1 and reverse.ORF and f: forward.ORF and reverse.ORF primers of WRKY gene. The target fragments containing WRKY gene are 1335, 1249, 1189, 1133, 1067 and 906 bp, respectively. Positive control (C+): Arabidopsis thaliana At5g43350 primers. M: Ladder 100 bp

Full size image
Fig. 3
figure 3

a The protein sequence of MYB TF was aligned with related MYB proteins from other plants. Proteins accession numbers are as follows: MYB-related308-Beta vulgaris: XP_010676421; MYB3-Amboralla trichopoda: XP_011629059; MYB-Zea mays: AFW57370; MYB-like-Zea mays: NP_001151578; MYB-related-Setaria italic: XP_004972853; ScMYB7-Saccharum hybrid: CCU64159; MYB86-like-Brassica rapa: XP_009112787; MYB103-Arabidopsis thaliana: NP_17657; Papaver-MYB: KP411870 and MYB2-Pinus taeda: ABD60283, b. The protein sequence of WRKY TF was aligned with related WRKY proteins from other plants. Proteins accession numbers are as follows: Populus euphratica WRKY40: XP_011020606.1; Populus trichocarpa WRKY40: EEE78831.2; Jatropha curcas hypothetical protein: KDP41972.1; Theobroma cacao WRKY40: EOY26053.1; Morus notabilis WRKY40: EXB50423.1; Coffea canephora unnamed protein: CDO98227.1; Citrus clementina hypothetical protein: ESR40862.1; Nelumbo nucifera WRKY40: XP_010275778.1; Elaeis guineensis WRKY40: XP_010929158.1; P. somniferum WRKY58: AFU81787.1 and P. somniferum WRKY: KP203840. The ClustalW2 program with default parameters was used for multiple alignment

Full size image
Fig. 4
figure 4

a. Phylogenetic analysis of identified P. somniferum L. MYB protein, and other MYB proteins from various species using the Mega 6.06 software. Proteins accession numbers are as follows: MYB-related308-Beta vulgaris: XP_010676421; MYB3-Amboralla trichopoda: XP_011629059; MYB-Zea mays: AFW57370; MYB-like-Zea mays: NP_001151578; MYB-related-Setaria italic: XP_004972853; ScMYB7-Saccharum hybrid: CCU64159; MYB86-like-Brassica rapa: XP_009112787; MYB103-Arabidopsis thaliana: NP_17657; Papaver-MYB: KP411870 and MYB2-Pinus taeda: ABD60283, b. Phylogenetic analysis of identified P. somniferum L. WRKY protein, and other WRKY proteins from various species using the Mega 6.06 software. Proteins accession numbers are as follows: Populus euphratica WRKY40: XP_011020606.1; Populus trichocarpa WRKY40: EEE78831.2; Jatropha curcas hypothetical protein: KDP41972.1; Theobroma cacao WRKY40: EOY26053.1; Morus notabilis WRKY40: EXB50423.1; Coffea canephora unnamed protein: CDO98227.1; Citrus clementina hypothetical protein: ESR40862.1; Nelumbo nucifera WRKY40: XP_010275778.1; Elaeis guineensis WRKY40: XP_010929158.1; P. somniferum WRKY58: AFU81787.1 and P. somniferum WRKY: KP203840

Full size image
Fig. 5
figure 5

a. Relative expression patterns of MYB and WRKY in leaf, stem and capsule tissues of P. somniferum L. in comparison with root tissue. b. Relative expression patterns of MYB in root, stem and leaf tissues of P. bracteatum L. in comparison with capsule tissue, and WRKY in root, leaf and capsule in comparison with stem tissue. c. Relative expression patterns of MYB and WRKY in leaf, stem, capsule and root tissues of P. somniferum L. in comparison with the same tissues of P. bracteatum L

Full size image
Fig. 6
figure 6

Frequency of occurrence of the three motifs in artificial clusters generated in POBO for A. thaliana background promoters compared to the promoters of genes involved in noscapin biosynthesis. wTGnTCACC and TwTGsCAAA are motis that bind to MYB and WRKY TFs, respectively

Full size image
Fig. 7
figure 7

Distributions of identified common TFs presented in promoter sequences of noscapine gene cluster by TRANSFAC database (Table 5). The graph is built using the Microsoft Power Point software

Full size image

Transcriptome expression analyses of MYB and WRKY genes

Distribution of proteins can show their function in tissues (Shui et al. 2008). To determine whether MYB and WRKY transcripts are expressed in all tissues, mRNA levels of MYB, WRKY and E.F.1α were determined in various tissues of P. somniferum L. and P. bracteatum L. via the qRT-PCR method. Expression analysis of P. somniferum L. showed that transcription levels of MYB in leaves, capsules and stems were respectively 123, 24 and 4 times more than its transcription level in roots, and transcripts of WRKY in leaves, stems and capsules were respectively 406, 91 and 3 times more than its transcripts in roots. Therefore, the highest expression levels of both TFs were detected in leaves of P. somniferum L., and the lowest expression levels were detected in roots of P. bracteatum L. (Fig. 5a). The highest transcripts of MYB and WRKY TFs observed in the leaf organ have also been reported for various species like transcripts of IbMYB1 in Ipomoea batatas (Kim et al. 2013). SpWRKY2 which was isolated from Solanum pimpinellifolium also shows the highest expression level in leaves (Li et al. 2014). Unlike, P. somniferum L. the highest transcription levels of MYB and WRKY were detected in the roots of P. bracteatum L. The transcripts of MYB in roots, stems and leaves were respectively 24, 12 and 4 times more than its expression level in stems, and transcripts of WRKY in roots, leaves and capsules were respectively 94, 7 and 3 times more than its expression level in stems (Fig. 5b). Transcripts of GmMYB12B2 which was isolated from Glycine max, also shows the highest expression level in roots (Li et al. 2013). In some species such a Brassica oleracea, Nicotiana tabacum, Arabidopsis thaliana and Oryza sativa also the highest expression levels of most WRKY TFs were observed in root tissues (Hu et al. 2012; Liu et al. 2013; Nuruzzaman et al. 2014; Yao et al. 2014).

Comparison of transcript levels in both species, showed that expression levels of MYB in leaves, capsules and stems of P. somniferum L. were respectively 141, 135 and 1.9 times more than their expression in same tissues of P. bracteatum L., however, the expression level of MYB in roots of P. bracteatum L. was 4 times more than P. somniferum L. Transcripts of WRKY in stems, leaves and capsules of P. somniferum L. were respectively 323, 209 and 4 times more than transcripts in same tissues of P. bracteatum L., but the expression level of WRKY in the roots of P. bracteatum L. was 26 times more than P. somniferum L. (Fig. 5c).

Promoter analysis

Transcriptomic analysis of a high noscapine–producing poppy, bacterial artificial chromosome sequencing and F2 mapping analysis showed that 10-gene clustered involved in noscapine biosynthesis are co-expressed. Genes in the same cluster might be co-expressed but not necessarily co-regulated. Co-regulated genes have similar regulatory motifs. TFs are one of the major components in the complex molecular network of gene expression regulation. Complete genome sequences and expression profiling technology have made the study of transcriptional regulation networks easier (Winzer et al. 2012). 28 motifs that were commonly shared by 100 % of genes were derived as result of promoter analysis by the MotifSampler (Table 2), and checked with the POBO program. There was a significant (t-test: p-value >0.0001) enrichment in the frequency of the all 28 consensus motifs identified by MotifSampler in 1.5 kb upstream promoter sequences of the selected genes compared to the A. thaliana background (Table 3). Graphs of some of motifs are presented in Fig. 6. There is a significant difference between background and cluster of genes in two motifs. Number of occurrences of the two motifs in cluster of genes is larger than background, which suggests that these motifs are tightly related to noscapine biosynthesis. As a result of TFBSs identification obtained using the TRANSFAC, PLACE and SoftBerry databases, multiple unique TFBSs of many known TFs in the promoter regions of these 10 co-expressed genes were found (Table 4). WRKY and MYB TFs are two of the main identified TF families.

Table 2 Consensus common motifs identified in promoter region of 10 gene cluster involved in noscapine biosynthesis from P. somniferum L. using MotifSampler program
Full size table
Table 3 Frequency of occurrence of overrepresented promoter motifs of 10 gene cluster involved in noscapine biosynthesis from P. somniferum L. using POBO program
Full size table
Table 4 Putative trans/cis-acting regulatory elements presented in identified consensus motif patterns of promoters of noscapine biosynthesis pathway genes from P. somniferum L. by TRANSFAC, PLACE and SoftBerry databases
Full size table

As a result of TFBSs identification of promoters of noscapine gene cluster, 18 TFs that were commonly shared by 50 to 100 % of promoters were found using the TRANSFAC database (Table 5). According to Table 5, members of GT-1 TF family were found widely in all promoters. Occurrence of this family of TFs has widely been reported in different plant families such as Solanaceae. These TFs are similar to the SBF-1 TF. Unlike most TFs, these groups are interacting to the inhibitory cis elements in promoter sequences (Argüello-Astorga and Herrera-Estrella 1996). It has been reported that they play negative roles in gene expression regulation of most transgenic plants. GT-1, GT-1a and GT-1b TFs, which are presented in Table 5, are members of this family. PBF TFs were also occurred in all 10 genes. They are activator TFs that have been found in potato (Desveaux et al. 2000). In addition to these groups, some of TFs such as WRKY, MYB and BZIP were also found in most promoters. There are 2 WRKYs, and no MYB listed in the plant TFs database for P. somniferum L. (Eulgem et al. 2000). All promoters also carry a number of putative cis-acting regulatory elements involved in response to various external factors (Table 5). Results of two studies on promoter regulatory network of noscapine gene cluster including motifs and TFBSs identifications were similar. For instance, GT-1, MYB, WRKY, and PBF TFs were observed widely in both studies. Distributions of TFs on promoter sequences of 10 genes, identified by the TRANSFAC database were obtained, and the schematic view of them was shown in Fig. 7. The position of same TFBSs differs from one promoter to another, and there is not a regular pattern for their distribution. In fact, there is not a definite position for a specific TF on different promoters. For instance, SEF TF in PSSDR1 promoter is located on position 82, whereas in the CYP82Y1 promoter, it is located on position 1280.

Table 5 Identification of TFs which occurred in 50 to 100 % of promoters of noscapine gene cluster in P. somniferum L. using TRANSFAC database
Full size table

Discussion

Two novel MYB and WRKY TFs were identified and cloned from P. somniferum L. using integrated bioinformatics and experimental data. P. somniferum L. was chosen for this study due to its valuable medicinal properties. Transcript levels of TFs were determined by means of qRT-PCR in capsules, leaves, roots, and stems of P. somniferum L. and P. bracteatum L. The two TFs were detected in all the studied tissues of both species. Transcripts of the two TFs in leaves of P. somniferum L. were much higher than other tissues, so in addition to their role in secondary metabolite biosynthesis, they might be involved in leaf developmental processes such as trichome development. The role of MYB TFs with a complex of other TFs including WRKY TFs, in trichome development of Arabidopsis thaliana has been reported (Du et al. 2009; Rushton et al. 2010). The highest expression level of both TFs in P. bracteatum L. was observed in roots. Unlike P. somniferum L., which is an annual plant, P. bracteatum L. is a perennial plant with stable roots. The highest expression level of TFs in roots of P. bracteatum L. might be related to strong roots of P. bracteatum L. and their high metabolisms. There has been no report on MYB TFs in P. somniferum L. yet, and the identified MYB in this work was the first one submitted to the NCBI database, whereas a WRKY TF named WRKY58 was isolated from P. somniferum L. by Mishra et al. (2013). WRKY58 showed the highest transcript levels in capsules, followed by roots, stems, and buds. Multiple alignments and phylogenetic analysis showed that both WRKYs belong to the group II of WRKY family, suggested by their structural similarity, they might have close functions. WRKY58 is responsible for response to wounding, and also binds to TYDC promoter to regulate BIAs biosynthesis pathway (Mishra et al. 2013). As a result, identified WRKY might also be able to bind to TYDC promoter, so its overexpression might result in increasing of BIAs accumulation.

Noscapin gene cluster was selected for promoter analysis, since it is the only pathway of secondary metabolites in P. somniferum L., which its promoters are identified and available. Noscapine mostly accumulates in capsule and leaf tissues (Winzer et al. 2012; Mishra et al. 2013; Nuruzzaman et al. 2014). Results of promoter analysis showed that not only these genes are neighboring and co-expressed (Winzer et al. 2012), but also they can be co-regulated, and share common regulatory TFs such as MYB and WRKY protein families. Chen and Facchini have shown that the 10-gene clustered which is present in noscapine-accumulating opium poppy, is absent from the genome of noscapine-free chemotype. The presence or absence of transcripts corresponding to clustered genes is also observed. Tough, more information is needed, existing evidence suggests coordinated regulation (Chen and Facchini 2014). Suppression of a gene involved in noscapine biosynthesis pathway (CYP82Y1) resulted in reduction of noscapin, and changes in expression level of noscapine producing genes. Expressions of genes which are out of the gene cluster, and encode enzymes involved in noscapine biosynthesis were not affected by the silencing of CYP82Y1, while expression of clustered genes was affected. Co-suppression of only clustered genes is also indicates coordinated regulation (Dang and Facchini 2014). Overexpression of CjWRKY raised expression level of genes involved in BAIs biosynthesis which indicates although TFs are not the only regulators of co-expression in gene clusters, undoubtedly they are key regulatory elements (Agarwal et al. 2015). Most of the investigated promoters contain effective stress responsive elements, which states in addition to their similar expression pattern in noscapine biosynthesis, they might also show similar pattern in expression of stress response genes.

Despite of the apparently crucial role of WRKY TFs in noscapine biosynthesis pathway in capsules which was observed as a result of promoter analysis, identified WRKY showed a low expression level in capsule tissue in comparison with leaf tissue, so it seems that the transcript level of WRKY in capsules should have been higher. Furthermore, transcripts of MYB in capsules were relatively higher in comparison with transcripts of WRKY, while their negative regulatory role in BIAs biosynthesis in P. somniferum L. has been reported. In fact, in P. somniferum L. there is a competition between BIAs and MIAs production. STR is the main enzyme in MIAs biosynthesis, and MYB TFs have positive role in its regulation. If expression of MYB is decreased, expression of STR will also be decreased. As a result, accumulation of MIAs will also be mitigated, and accumulation of BIAs, which they are more valuable alkaloids, will be increased (Zhou et al. 2010). Thereupon, overexpression of WRKY using capsule tissue specific promoters, and suppression of MYB might results in more BIAs and less MIAs accumulation in capsules. In a study conducted by Yang et al. (2015), the presence of two to four W-boxes in promoter region of several genes involved in artemisinin biosynthesis in Artemisia annua was reported. This study also suggests that AaWRKY is capable of enhancing expression of these genes (Yang et al. 2015). In another study on Artemisia annua, overexpression of AaWRKY1 led to the increasing of artemisinin accumulation (1.8-fold)) Han et al. 2014).

It seems that genetic manipulation of TFs is a complicated, but more effective approach for controlling metabolites pathways compare to specific enzyme gene itself. Metabolic pathways in plants are very complicated and consist of different enzymatic steps (Gantet and Memelink 2002). Therefore, one of the main and practical applications of TFs is genetic engineering of these genes as valuable tools for plant manipulations. This will be only achieved by deep understanding of TFs genes, and regulatory networks of promoters (Grotewold et al. 1998).

Promoter is the central processor of transcription regulation. Understanding of how promoter networks are organized will shed light onto the control mechanism of specific genes. According to the promoter analysis method, co-regulatory genes are categorized based on their promoter organization status. This is a powerful approach for description of co-regulated genes properties using their promoter structure. After a model or framework is developed, it may be used to search for possible co-regulatory genes. In this order, consideration of the difference between co-regulation and co-expression is crucial (Werener et al. 2003). Promoter structure Studies of co-regulate genes, and identification and isolation of common regulatory TFs are efficient strategies for breeding quantitative traits such as production of secondary metabolites by engineering specific TFs, which is sometimes more effective than complicated enzymatic steps (Gantet and Memelink 2002; Yuan and Grotewold 2015).