Cloning and Characterization of a Flavonol Synthase Gene from Scutellaria baicalensis

Flavonols are the most abundant of all the flavonoids and play pivotal roles in a variety of plants. We isolated a cDNA clone encoding flavonol synthase from Scutellaria baicalensis (SbFLS). The SbFLS cDNA is 1011 bp long, encodes 336 amino acid residues, and belongs to a family of 2-oxoglutarate-dependent dioxygenases. The overall structure of SbFLS is very similar to that of Arabidopsis thaliana anthocyanidin synthase (AtANS), with a β jelly-roll fold surrounded by tens of short and long α-helices. SbFLS was constitutively expressed in the roots, stems, leaves, and flowers, with particularly high expression in the roots and flowers. SbFLS transcript levels in the roots were 376-, 70-, and 2.5-fold higher than in the leaves, stems, and flowers. The myricetin content was significantly higher than that of kaempferol and quercetin. Therefore, we suggest that SbFLS mediates flavonol formation in the different organs of S. baicalensis. Our study may contribute to the knowledge of the role of FLS in S. baicalensis.


Introduction
Scutellaria baicalensis Georgi (Lamiaceae) is one of the most popular herbs in several oriental countries, where it is used to treat inflammation, respiratory tract infections, diarrhea, dysentery, jaundice/liver disorders, hypertension, hemorrhaging, and insomnia [1]. Scutellaria is a mild relaxant that affects the neural and muscular-skeletal systems [2]. S. baicalensis root has abundant flavones, a class of plant flavonoids [3].
In this study, a full-length cDNA encoding FLS was isolated from S. baicalensis by using next-generation sequencing platforms (NGS) ( analyzed by real-time PCR and high-performance liquid chromatography (HPLC). The cloning and characterization of SbFLS may provide a foundation to elucidate the mechanism of flavonol synthesis in S. baicalensis.

Plant Material and Growth
Conditions. The seeds of S. baicalensis were sown in May 2012, and then seedlings were transferred into pots filled with perlite-mixed soil. Seedlings were grown under 16 h light and 8 h dark conditions in the greenhouse (25 ∘ C and 50% humidity) at Chungnam National University (Daejeon, Korea). We used at least nine pots for biological repeats. Each organ (flowers, stems, leaves, and roots) was collected for two weeks after the first day of flowering. All samples were frozen in liquid nitrogen upon collection and stored at −80 ∘ C prior to RNA isolation and HPLC. forward, 5 -ATGGAGGTTGGGAGAGTG-3 ; reverse, 5 -TCACTGGGGAAGCTTATTAAGCTTAC-3 . The TM Calculator program (http://bioinfo.ut.ee/primer3-0.4.0/) was used to compute the PCR annealing temperatures. The PCR program consisted of denaturation at 94 ∘ C for 1 min, annealing at 45 ∘ C for 1 min, and an extension at 72 ∘ C for 1 min. Thirty cycles were preceded by denaturation step at 94 ∘ C for 3 min, followed by a final extension at 72 ∘ C for 10 min. The amplified product was purified and cloned into the Tblunt vector (SolGent, Daejeon, Korea) and sequenced. The full-length FLS sequence was aligned using BioEdit Sequence Alignment Editor, version 5.0.9 [17]. For putative conserved domains, we used a GenBank conserved domain database search and function analysis service in NCBI.

HPLC Analysis.
Flavonol content was determined as described in the studies by Li et al. [15] and Kovács et al. [21], with a modification. Flavonol standards, kaempferol, quercetin, and myricetin were purchased from Sigma-Aldrich. The harvested organs (flowers, stems, leaves, and roots) were freeze-dried for 48 h and ground into a fine powder using a mortar and pestle. Powdered samples (100 mg) were extracted by sonication in 80% (v/v) methanol at room temperature for 60 min. During extraction, samples were vortexed every 15 min. After extraction, the extracts were centrifuged at 14,000 rpm for 10 min and the supernatant was filtered with a 0.45 m Acrodisc syringe filter (Pall Corp.; Port Washington, NY) for HPLC. HPLC was performed with a C18 column (250 ×4.6 mm, 5 m; RStech; Daejeon, Korea). The mobile phase was a gradient prepared from mixtures of methanol and 0.1% acetic acid. The flow rate was maintained at 0.7 mL/min. An injection volume of 20 L and wavelength of 270 nm were used for detection. The compounds in the sample were determined using a standard curve. Determinations were performed after 3 separate extractions of each sample and each extract was analyzed in triplicate ( = 3).  with the published sequence from Arabidopsis [22]. G. biloba FLS has a 1023 bp ORF encoding a 340 amino acid protein, MW of 38.7 kDa, and pI of 5.75 [16]. The deduced SbFLS shared 77, 74, 72, 72, 71, and 70% identity with FLS proteins from Antirrhinum majus (ABB53382), Nicotiana tabacum (ABE28017), Camellia nitidissima (ADZ28516), Gentiana triflora (BAK09226), Solanum tuberosum (ACN81826), and Vitis vinifera (BAE75809), respectively (Supplementary Figure S1 available online at http://dx.doi.org/10.1155/2014/ 980740). Owens et al. [23] reported that the genome of Arabidopsis thaliana contains 5 sequences with high similarity to AtFLS1, a previously characterized flavonol synthase gene; 4 AtFLS isoforms are located on chromosome 5. A GenBank conserved domain database search and function analysis revealed that SbFLS has putative conserved domains belonging to the 20G-FeII Oxy super family and the highly conserved N-terminal region of proteins with 2-oxoglutarate/ Fe(II)-dependent dioxygenase activity. This is consistent with the results reported by Xu et al. [16]. Moreover, SbFLS had conserved sequence motifs, including the HXD motif [15] for ligating ferrous iron and the RXS motif for binding 2oxoglutarate (2OG) [16]. To determine the relationship between the putative SbFLS protein and other plant FLSs, we performed phylogenetic analysis (Figure 2). The SbFLS phylogeny was clustered into 2 distinct groups and showed the closest relationship with A. majus. The subcellular targeting of SbFLS was predicted by PSORT to be nuclear and cytosolic. AtFLS1 localizes to the cytoplasm and nucleus [22], whereas maize FLS localizes in the ER and perinuclear region [24]. SOPMA [20] indicated that SbFLS contains 121 (36%) alphahelices, 61 (18.6%) extended strands, 22 (6.6%) beta turns, and 132 (39.3%) random coils, respectively. synthase (AtANS; PDB 1GP6) [25]. The overall structure of SbFLS and AtANS is very similar, with a jelly-roll fold surrounded by tens of short and long -helices (Figure 3(a)). The iron metal may be ligated with a bidentate group of cosubstrate 2OG as well as 3 side-chain residues, that is, H222, D224, and H278, which are conserved in AtANS and SbFLS (Figures 3(a) and 3(b)) [15,25]. Hydrogen bonds may form between the other side of 2OG and 3 conserved residues, that is, Y207, R288, and S290 (Figure 3(b)) [15,25,26]. A substrate quercetin (QUE) might be located near an iron metal ion, in which 5 invariant residues H133, F135, K203, F294, and E296 may contribute to binding stability by forming a -stacking arrangement (especially F294) with the QUE ring and by hydrogen bonding. All interactions as shown here were less than 3.5Å distant. Kinetic analyses of AtANS revealed that most of the highly conserved residues play key roles in substrate/cosubstrate binding and enzymatic catalysis through metal coordination [27]. in mature leaves, whereas the lowest level was observed in the roots. Expression of AtFLS isoforms follows different patterns in different tissues [23]. The transcription of AtFLS1 was highest in floral buds, flowers, and siliques, whereas the roots and shoots of young seedlings and the lea-ves of later vegetative stages exhibited lower expression. Other AtFLS isoforms were expressed at much lower levels or were undetectable in all tissues [23]. Therefore, Owens et al. [23] suggested that only AtFLS1 contributes to flavonol synthesis in Arabidopsis. Transcription of Citrus unshiu FLS was higher in young leaves than in old leaves and increased in the peel during fruit maturation [28]. Based on the transcription patterns in several plants, we conclude that FLS is differentially regulated in different plants and plant tissues.

Accumulation of Flavonol in Different
Tissues. Kaempferol, quercetin, and myricetin were detected in almost all tissues (i.e., flowers, stems, leaves, and roots) by HPLC ( Figure 5). The myricetin content was significantly higher than that of kaempferol and quercetin. The transcription pattern of SbFLS was similar to the accumulation pattern of kaempferol. Roots showed the highest kaempferol content (0.97 mg g −1 dry weight (DW)); the leaves and flowers contained similar amounts (0.62 and 0.51 mg g −1 DW), but no kaempferol was detected in the stem. Kaempferol and some glycoside derivatives have a wide range of pharmacological activities [29] and account for the utility of dry Scutellaria root as a multipurpose herb in oriental medicine. The leaves contained the highest levels of quercetin (1.26 mg g −1 DW), whereas the lowest amount of quercetin was in the flowers (0.64 mg g −1 DW). Interestingly, roots displayed the lowest amount of myricetin (0.8 mg g −1 DW), whereas the highest myricetin content was found in the flowers (6.05 mg g −1 DW). The myricetin content in the flowers was 2.2-and 7.5-fold higher than that in the leaves and roots, respectively. Feng and Liu [30] reported that the quercetin content of leaves was significantly higher than the content in pseudostems in different tissues of Welsh onion (Allium fistulosum L.). Quercetin, kaempferol, morin, rutin, and myricetin act as antioxidants and possess anti-inflammatory, antiallergic, antiviral, and anticancer properties [31]. Quercetin is a free radical scavenger that protects against liver reperfusion ischemic tissue damage [32]. The scavenging activity of flavonols decreases as follows: myricetin > quercetin > kaempferol [33].
SbFLS transcription was the highest in the roots, correlating with kaempferol content in this organ. Thus, SbFLS might regulate the biosynthesis of kaempferol in S. baicalensis. Unlike kaempferol, quercetin and myricetin were most highly concentrated in the leaves and flowers, respectively. The expression level of SbFLS was not consistent with the contents of myricetin and quercetin. Owens and his colleagues [23] reported that Arabidopsis uses different isoforms of FLS with different substrate specificities to mediate the production of the quercetin and kaempferol in different tissue or cell types. In addition, Lillo et al. [34] described that FLS activity of the ANS enzyme may contribute to the differential accumulation of kaempferol and quercetin. Moreover, it was pointed out that differential expression of the F3 H enzyme could also 6 The Scientific World Journal mediate kaempferol and quercetin ratio in petunia flowers [35]. Stracke et al. [36] reported that FLS1 could be activated by flavonol-specific transcription factors (TFs) MYB11, MYB12, and MYB111 of A. thaliana and these TFs caused different spatial accumulation of specific flavonol derivatives in leaves, stems, inflorescences, siliques, and roots [23]. Several isoforms of FLS have been isolated in other plants such as Arabidopsis [23] and maize [24]. Therefore, we presume that SbFLS isoforms, SbANS, and SbF3 H may be contributed to the biosynthesis of quercetin and myricetin.

Conclusion
We have described the molecular cloning and characterization of a S. baicalensis gene encoding SbFLS. Our results suggest that SbFLS is a key potential target for the flux of the flavonol biosynthesis pathway in S. baicalensis. To explain adequately the flavonol biosynthesis mechanisms in S. baicalensis, in vitro enzyme assay of SbFLS isoforms, SbANS, and SbF3 H should be examined in the near future. Our study may help to determine the role of FLS and facilitate metabolic engineering of flavonol biosynthesis in S. baicalensis.