Cloning, Expression Analysis and Functional Characterization of Squalene Synthase (SQS) from Tripterygium wilfordii

Celastrol is an active triterpenoid compound derived from Tripterygium wilfordii which is well-known as a traditional Chinese medicinal plant. Squalene synthase has a vital role in condensing two molecules of farnesyl diphosphate to form squalene, a key precursor of triterpenoid biosynthesis. In the present study, T. wilfordii squalene synthase (TwSQS) was cloned followed by prokaryotic expression and functional verification. The open reading frame cDNA of TwSQS was 1242 bp encoding 413 amino acids. Bioinformatic and phylogenetic analysis showed that TwSQS had high homology with other plant SQSs. To obtain soluble protein, the truncated TwSQS without the last 28 amino acids of the carboxy terminus was inductively expressed in Escherichia coli Transetta (DE3). The purified protein was detected by SDS-PAGE and Western blot analysis. Squalene was detected in the product of in vitro reactions by gas chromatograph-mass spectrometry, which meant that TwSQS did have catalytic activity. Organ-specific and inducible expression levels of TwSQS were detected by quantitative real-time PCR. The results indicated that TwSQS was highly expressed in roots, followed by the stems and leaves, and was significantly up-regulated upon MeJA treatment. The identification of TwSQS is important for further studies of celastrol biosynthesis in T. wilfordii.


Introduction
Tripterygium wilfordii (Celastraceae) is an important Chinese herbal medicine used for the treatment o f inflammation [1]. Active ingredients of T. wilfordii include sesquiterpenoid alkaloids, diterpenoid, and quinone-methide triterpenoid (QMTs) with varied pharmaceutical activities [2]. Many QMTs have already been isolated from Celastraceae, such as celastrol, maitenin, 22-hydroxytingenone and pristimerin [3]. They have similar chemical structures and biological activities. Celastrol is the first active compound isolated from the root bark of T. wilfordii and also is one of the major active ingredients of T. wilfordii [4]. As a representative compound of QMTs, Celastrol has been demonstrated to have promising activities including antioxidant [5], anti-inflammatory [6], immunomodulation [7], and anticancer activities [8]. Celastrol has shown promise as an anti-inflammatory compound for arthritis, lupus, amyotrophic lateral sclerosis, and Alzheimer's disease [6]. Recent studies have shown that celastrol is a potential therapeutic agent for anti-obesity [9,10]. Celastrol has attracted increased interest recently regarding its therapeutic mechanisms and clinical applications. Traditionally, celastrol is

Sequence Analysis of TwSQS
The open reading frame (ORF) cDNA of TwSQS was amplified by PCR using primers TwSQS-F and TwSQS-R. The ORF cDNA of TwSQS was 1242 bp, which encodes a polypeptide of 413 amino acids. The result of homologous domains search showed that TwSQS belonged to the superfamily of isoprenoid biosynthesis class 1 enzymes and had specific hit with trans-isoprenyl diphosphate synthase. The sequence comparison of TwSQS showed high identity to SQSs from other plants, including Glycyrrhiza glabra (BAA13083.1, 88%), P. notoginseng (ABA29019.1, 86%), Glycyrrhiza uralensis (ADG36719.1, 85%), Diospyros kaki (ACN69082.1), P. ginseng (ACV88718.1, 82%), Centella asiatica (AAV58897.1, 82%), Artemisia annua (AAR20329.1, 81%), respectively. The SQS from fungi Candida glabrata (BAB12207.1, 41%) was added into multiple sequence alignments as an outgroup. Multiple sequence alignments indicated that TwSQS had six homologous signature regions (I-VI, Figure 1) which were considered to be the binding, catalytic and regulatory sites of SQSs. Region II and IV contained two aspartate-rich motifs (DXXXD) which were deemed to coordinate and facilitate FPP binding by binding Mg 2+ [31,32]. Even though region VI showed low sequence similarity among the SQSs in alignment, they all had highly hydrophobic amino acid residues and performed the similar functions of membrane targeting and anchoring [32]. The TMHMM result showed that TwSQS has two transmembrane helices domains, containing residues 281-303 aa and 387-404 aa, respectively. A neighbour-joining phylogenetic tree was established to investigate the relations based on the sequences of TwSQS and other SQSs from different organisms ( Figure 2). The phylogenetic tree was divided into six main clusters: eudicots, monocots, gymnosperm, algae, animal, and fungi, and TwSQS clustered within the eudicots corresponding to taxonomic classification. The result showed that TwSQS was closely related to those from Glycyrrhiza glabra and Glycyrrhiza uralensis. TMHMM result showed that TwSQS has two transmembrane helices domains, containing residues 281-303 aa and 387-404 aa, respectively. A neighbour-joining phylogenetic tree was established to investigate the relations based on the sequences of TwSQS and other SQSs from different organisms ( Figure 2). The phylogenetic tree was divided into six main clusters: eudicots, monocots, gymnosperm, algae, animal, and fungi, and TwSQS clustered within the eudicots corresponding to taxonomic classification. The result showed that TwSQS was closely related to those from Glycyrrhiza glabra and Glycyrrhiza uralensis.

Purification and Identification of Recombinant Protein
To understand the catalytic function of TwSQS, it was essential to yield a certain amount of soluble protein. It was reported that removal of the C-terminal hydrophobic region could enhance the soluble expression of recombinant SQS [21]. Therefore, the truncated TwSQS without the last 28 amino acids of the C-terminal was inserted into vector pET-30a. The recombinant expression plasmid pET-30a-TwSQS was extracted and transformed into E. coli Transetta (DE3) to produce recombinant protein containing a His-tag at the N-terminus. The pET-30a empty vector was transformed into E. coli Transetta (DE3) and served as a control. After induced expression, extraction and purification, the crude protein and purified recombinant protein were analyzed by SDS-PAGE and Western blotting.
SDS-PAGE analysis showed that TwSQS was successfully expressed in the supernatant of lysates, with a molecular weight of around 49.41 kDa (Figure 3a). The corresponding protein band was also found at the same position (about 49.41 kDa) in the purified protein. Western blot detection showed that the TwSQS could specifically combine with anti-His-tag antibodies ( Figure  3b). Two single immunoreactive bands were detected from the crude and purified protein of pET-30a-TwSQS bacteria. No such band was found in the crude protein of the control strain. These results indicated that the recombinant protein was successfully expressed and purified.

Purification and Identification of Recombinant Protein
To understand the catalytic function of TwSQS, it was essential to yield a certain amount of soluble protein. It was reported that removal of the C-terminal hydrophobic region could enhance the soluble expression of recombinant SQS [21]. Therefore, the truncated TwSQS without the last 28 amino acids of the C-terminal was inserted into vector pET-30a. The recombinant expression plasmid pET-30a-TwSQS was extracted and transformed into E. coli Transetta (DE3) to produce recombinant protein containing a His-tag at the N-terminus. The pET-30a empty vector was transformed into E. coli Transetta (DE3) and served as a control. After induced expression, extraction and purification, the crude protein and purified recombinant protein were analyzed by SDS-PAGE and Western blotting.
SDS-PAGE analysis showed that TwSQS was successfully expressed in the supernatant of lysates, with a molecular weight of around 49.41 kDa (Figure 3a). The corresponding protein band was also found at the same position (about 49.41 kDa) in the purified protein. Western blot detection showed that the TwSQS could specifically combine with anti-His-tag antibodies (Figure 3b). Two single immunoreactive bands were detected from the crude and purified protein of pET-30a-TwSQS bacteria. No such band was found in the crude protein of the control strain. These results indicated that the recombinant protein was successfully expressed and purified.

Functional Characterization of the TwSQS
To verify the function of TwSQS, the conversion activity from FPP to squalene was assayed using the purified protein with the cofactor NADPH and Mg 2+ . The crude protein of the control strain was used as a control. The catalytic products were analyzed by GC-MS ( Figure 4). The peak at 20.3 min in the GC profile of TwSQS reaction products ( Figure 4B) corresponded to that observed in a standard squalene sample ( Figure 4A), but no such peak was detected in the control ( Figure 4C). Moreover, the mass spectrum characteristic peaks (m/z = 69, 81) of the 20.3-min peak of TwSQS ( Figure 4E) were consistent with those of standard squalene ( Figure 4D), while no corresponding mass fragments was observed in the control. These results illustrate that TwSQS encodes a functional squalene synthase, converting two molecules of FPP to squalene.

Functional Characterization of the TwSQS
To verify the function of TwSQS, the conversion activity from FPP to squalene was assayed using the purified protein with the cofactor NADPH and Mg 2+ . The crude protein of the control strain was used as a control. The catalytic products were analyzed by GC-MS ( Figure 4). The peak at 20.3 min in the GC profile of TwSQS reaction products ( Figure 4B) corresponded to that observed in a standard squalene sample ( Figure 4A), but no such peak was detected in the control ( Figure 4C). Moreover, the mass spectrum characteristic peaks (m/z = 69, 81) of the 20.3-min peak of TwSQS ( Figure 4E) were consistent with those of standard squalene ( Figure 4D), while no corresponding mass fragments was observed in the control. These results illustrate that TwSQS encodes a functional squalene synthase, converting two molecules of FPP to squalene.

Functional Characterization of the TwSQS
To verify the function of TwSQS, the conversion activity from FPP to squalene was assayed using the purified protein with the cofactor NADPH and Mg 2+ . The crude protein of the control strain was used as a control. The catalytic products were analyzed by GC-MS ( Figure 4). The peak at 20.3 min in the GC profile of TwSQS reaction products ( Figure 4B) corresponded to that observed in a standard squalene sample ( Figure 4A), but no such peak was detected in the control ( Figure 4C). Moreover, the mass spectrum characteristic peaks (m/z = 69, 81) of the 20.3-min peak of TwSQS ( Figure 4E) were consistent with those of standard squalene ( Figure 4D), while no corresponding mass fragments was observed in the control. These results illustrate that TwSQS encodes a functional squalene synthase, converting two molecules of FPP to squalene.

Expression Analysis of TwSQS
In order to study the expression pattern of TwSQS, we measured the TwSQS expression level in different tissues of T. wilfordii including roots, stems and leaves. The transcript levels of TwSQS were observed in these three tissues of T. wilfordii (Figure 5a). The expression level of TwSQS in stems was regarded as a control while the TwSQS expression in leaves and roots was evaluated compared with the stems. There were significant differences among the expression levels of the three tissues using Duncan's multiple range test at p < 0.05 in SPSS (IBM, Armonk, NY, USA). The expression level of TwSQS was significantly higher than that of stems and leaves (Figure 5a).
were observed in these three tissues of T. wilfordii (Figure 5a). The expression level of TwSQS in stems was regarded as a control while the TwSQS expression in leaves and roots was evaluated compared with the stems. There were significant differences among the expression levels of the three tissues using Duncan's multiple range test at p < 0.05 in SPSS (IBM, Armonk, NY, USA). The expression level of TwSQS was significantly higher than that of stems and leaves (Figure 5a).
In our previous studies, we found that 100 μM MeJA treatment could significantly increase the content of secondary metabolites and the expression level of related genes in Tripterygium wilfordii [12,33]. In order to determine the expression profile of TwSQS under MeJA, qRT-PCR analysis was performed. The result showed the TwSQS expression levels were significantly enhanced by 100 μM MeJA treatment in the hairy root cultures (Figure 5b). The TwSQS expression level before induction was regarded as a control and the TwSQS expression in other times of induction were calculated compared with the 0 h. After induced by MeJA, the levels of TwSQS expression raised rapidly at first, peaked at 6 h, and then gradually decreased. During the same period, there were no significant differences in the untreated control groups. The result indicated that MeJA can induce the expression of TwSQS.

Discussion
Celastrol is a representative molecule of quinone-methide triterpenoids which are deemed as the chemotaxonomic markers of Celastraceae [3]. Because of its important medicinal activity, many studies have focused on the identification of some specific enzymes involved in celastrol biosynthesis from the MVA pathway. Squalene synthase catalyzes two molecules of farnesyl diphosphate to form squalene, a key precursor of triterpenoid biosynthesis. Given its importance in terpenoid biosynthesis, SQSs were studied as key enzymes in sterol and triterpenoid biosynthesis. However, few literature studies have described the role of TwSQS in the biosynthetic pathway of celastrol in T. wilfordii. In the present study, we attempted to dissect the molecular biology of the celastrol biosynthetic pathway in T. wilfordii by cloning and characterizing TwSQS. In the present study, we cloned a squalene synthase gene that was involved in celastrol biosynthesis from T. wilfordii, which also was the first reported SQS of Celastraceae. Sequence alignment analysis of TwSQS with other plant SQS sequences showed that six highly homologous domains were present in the TwSQS, which were very important for the catalytic/functional activity of SQS [34,35]. SQS In our previous studies, we found that 100 µM MeJA treatment could significantly increase the content of secondary metabolites and the expression level of related genes in Tripterygium wilfordii [12,33]. In order to determine the expression profile of TwSQS under MeJA, qRT-PCR analysis was performed. The result showed the TwSQS expression levels were significantly enhanced by 100 µM MeJA treatment in the hairy root cultures (Figure 5b). The TwSQS expression level before induction was regarded as a control and the TwSQS expression in other times of induction were calculated compared with the 0 h. After induced by MeJA, the levels of TwSQS expression raised rapidly at first, peaked at 6 h, and then gradually decreased. During the same period, there were no significant differences in the untreated control groups. The result indicated that MeJA can induce the expression of TwSQS.

Discussion
Celastrol is a representative molecule of quinone-methide triterpenoids which are deemed as the chemotaxonomic markers of Celastraceae [3]. Because of its important medicinal activity, many studies have focused on the identification of some specific enzymes involved in celastrol biosynthesis from the MVA pathway. Squalene synthase catalyzes two molecules of farnesyl diphosphate to form squalene, a key precursor of triterpenoid biosynthesis. Given its importance in terpenoid biosynthesis, SQSs were studied as key enzymes in sterol and triterpenoid biosynthesis. However, few literature studies have described the role of TwSQS in the biosynthetic pathway of celastrol in T. wilfordii. In the present study, we attempted to dissect the molecular biology of the celastrol biosynthetic pathway in T. wilfordii by cloning and characterizing TwSQS. In the present study, we cloned a squalene synthase gene that was involved in celastrol biosynthesis from T. wilfordii, which also was the first reported SQS of Celastraceae. Sequence alignment analysis of TwSQS with other plant SQS sequences showed that six highly homologous domains were present in the TwSQS, which were very important for the catalytic/functional activity of SQS [34,35]. SQS genes from Arabidopsis thaliana, tobacco, human and yeast all have a single predicted transmembrane helix found in the C-terminal hydrophobic sequences identified by the TMHMM program. Interestingly, TwSQS not only have a transmembrane helix in domain VI, but also have an additional transmembrane helix near domain V. This second transmembrane helix was also reported in SQS from G. glabra, G. uralensis, P. ginseng and P. notoginseng, similar to T. wilfordii. The neighbor-joining phylogenetic tree also showed that TwSQS was very similar to those from G. glabra and G. uralensis.
Domain VI has highly hydrophobic amino acid residues, which is consistent with a membrane anchoring function. It were reported that deletion of the C-terminal membrane-anchoring could enhance the soluble expression of recombinant SQS [21]. This also was confirmed in our study. The soluble protein of TwSQS containing the putative full-length cDNA also was not obtained in this study (data not shown). In contrast, the soluble protein was observed in the supernatant of E. coli by removing the 28 amino acids of the C-terminal. In vitro enzyme reactions demonstrated that TwSQS could catalyze the reaction of FPP form to squalene under the coordination of Mg 2+ and NADPH. Somewhat similar results were also reported in M. officinalis SQS [21], S. miltiorrhiza SQS2 [22], P. notoginseng SQS [24], tobacco SQS [31], and W. somnifera SQS [14]. In addition, the recombinant AtSQS2 which was deemed to be a non-functional SQS, was unable to synthesize squalene from FPP in the presence of NADPH and either Mg 2+ or Mn 2+ [18]. These reports indicated that TwSQS encodes a functional squalene synthase. Results of qRT-PCR analyses of TwSQS expression demonstrated that TwSQS has tissue-specific expression with highest expression in roots followed by leaves and stems (Figure 5a). Some studies found that the tissue expression pattern of SQS varies greatly in different plants. Similar patterns of gene expression were found in S. miltiorrhiza [22], Euphorbia pekinensis [36], Taxus cuspidata [32], while some different expression patterns were found in B. platyphylla [37], W. somnifera [38], and Euphorbia tirucalli [39]. In addition, several studies have reported that the MeJA could upregulates the expression of SQS [21,36]. Therefore, we investigated the expression of TwSQS under MeJA treatment. The expression level of TwSQS rapidly increased after MeJA treatment and peaked at 6 h with an eight-fold increase compared with that of the control group (Figure 5b). It is reported that celastrol mainly accumulates in the root of T. wilfordii and MeJA treatment could significantly increase the content of celastrol [40]. Our results, thus, provide an additional evidence that TwSQS is positively involved in celastrol biosynthesis. Further work is to regulate the isoprenoid biosynthesis through the overexpression and MeJA induction of the TwSQS gene so as to enhance the production of celastrol in T. wilfordii.

Plant Material
The different fresh tissues of T. wilfordii were collected from the plants grown in the greenhouse of Northwest A & F University, China. The hairy root cultures were established as described previously [41]. The hairy root of T. wilfordii were cultured in 250 mL Erlenmeyer flasks containing 100 mL of Murashige-Skoog liquid medium, 30 g/L sucrose. Each flask was inoculated with 1 g fresh weight of roots and maintained in darkness at 26 • C. The hairy roots were elicited with 100 µM MeJA after 25 days of cultivation. The roots were harvested at 0, 1, 3, 6, 9, 12, 24 h after MeJA treatment. All the materials were rapidly frozen using liquid nitrogen and then kept at −80 • C for RNA isolation and chemical detection.

Total RNA Extraction and cDNA Synthesis
Total RNA was extracted from 100 mg of T. wilfordii roots using the plant total RNA extraction kit (TIANGEN Biotech Co., Ltd., Beijing, China). The RNA quality and concentration were measured by gel electrophoresis (1% agarose) and spectrophotometer analysis. The total RNA was reverse transcribed into cDNA template using Takara PrimeScript™ 1st strand cDNA synthesis kit (TaKaRa Biotechnology, Dalian, China).

Cloning and Bioinformatics Analyses of TwSQS
The ORF cDNA of TwSQS was cloned using the specific primers TwSQS-F and TwSQS-R (Table 1) which was designed according to the sequence from reported squalene synthase of T. wilfordii (KR401220.1). The PCR products of TwSQS were detected, purified and cloned into the pClone007 Blunt Simple Vector (Tsingke Biotech Co., Beijing, China) followed by sequencing.

Expression of Recombinant Protein
Squalene synthase is a membrane bound enzyme that has two transmembrane helices at the carboxy-terminal region. The second transmembrane helices is not located in the conserved domains. The C-terminal has highly hydrophobic amino acid residues, which is consistent with a membrane anchoring function. It was reported that removal of the C-terminal hydrophobic region could enhance the soluble expression of recombinant SQS [31]. It may be because removal of the C-terminal hydrophobic region could enhance the water-solubility of recombinant SQS and reduce the ability of SQS to be targeted to the membrane. So the carboxy-terminal truncated TwSQS was amplified using the designed site primers TwSQS-TF (BamHI) and TwSQS-TR (SacI) ( Table 1).
The PCR product and the expression vector pET-30a were separately double-digested with BamHI and SacI, and ligated by T4 ligase. The recombinant vector, pET-30a-TwSQS, was verified by sequencing. pET-30a-TwSQS was transformed into competent E. coli Transetta (DE3) to produce recombinant protein that containing a His-tag at N-terminus. The transformed bacteria was grown to A600 = 0.6 at 37 • C in LB medium. Expression of TwSQS was induced with 0.8 mM Isopropyl β-D-1-thiogalactopyranoside (IPTG) at 28 • C. After culturing for 6 h, the cells were collected by centrifuging at 6000× g for 10 min, and then washed 3 times with 0.05 M Tris-HCl (pH 7.5). The cells were resuspended in extraction buffer (50 mM Tris-HCl, pH 7.5, 10% glycerol, and 5 mM DTT) and lysed by sonication. The crude enzyme solution was harvested after centrifugation at 12,000× g at 4 • C for 30 min.

Purification of Expressed TwSQS and Western Blot Analysis
The crude enzyme solution was subjected to His-Tag purification using a Ni-NTA agarose gravity column (TransGen Biotech, Beijing, China) [42]. Ni-NTA agarose was equilibrated with balance buffer (10 mM imidazole, 50 mM Tris-HCl, pH 7.5, 50 mM NaH 2 PO 4 , and 0.3 M NaCl). The crude enzyme solution was added to the column and then washed with balance buffer.
The desired protein was eluted from Ni-NTA resin with increasing imidazole concentration (50-200 mM) in elution buffer. The eluted fractions were analyzed on a 10% SDS-PAGE gel and stained by Coomassie Blue R250 solution. The imidazole of purified protein was removed by centrifugal ultrafiltration (Amicon Ultra 10 kDa, Millipore, Burlington, MA, USA).
The recombinant protein was verified via Western blot analysis. Protein bands were transferred onto nitrocellulose membrane from SDS-PAGE gel. The membrane was blocked with 5% (w/v) skimmed milk in TBST (triethanolamine buffered saline adding 0.05% Tween 20) for 2 h, and then incubated with HRP Conjugated Anti His-Tag Mouse Monoclonal Antibody (1:3000, CWBIO, Beijing, China) at 4 • C overnight. After washing five times with TBST (5 min each time), the blots were revealed with a diaminobenzidine (DAB) kit (Solarbio, Beijing, China).

In Vitro Enzyme Reaction
The enzyme reaction of the recombinant protein TwSQS was conducted as follows. The purified protein incubated at 30 • C with 50 µM FPP triammonium salt (Sigma-Aldrich, St. Louis, MO, USA), 50 mM Tris-HCl (pH 7.5), 25 mM MgCl 2 , 1 mM DTT, 2% Glycine and 3 mM NADPH in a total reaction volume of 300 µL. After 8 h, the reaction product was extracted four times with 400 µL n-hexane and then analyzed by GC-MS. The GC-MS detection was performed on GCMS-QP2010 Ultra (Shimadzu, Kyoto, Japan) using the method described as Rong et al. [22].

Expression Analysis
Total RNA from different tissues and hairy roots treated by MeJA was extracted separately using the plant total RNA extraction kit (TIANGEN Biotech Co., Ltd., Beijing, China), and reversely transcribed into cDNA using the Prime Script™ RT reagent kit with gDNA Eraser (TaKaRa). The qRT-PCR analysis was performed using the SYBR Premix Ex Taq (TaKaRa) on a CFX96™ Real-time PCR system (Bio-Rad, Hercules, CA, USA). Elongation factor 1 α (EF1α) was used as a reference gene for gene expression normalization [43]. Primers used for expression analysis of TwSQS (TwSQS-qF and TwSQS-qR) and EF1α (Efα-F and Efα-R) were given in Table 1. Each analysis included three independent experiments and two biological replicates. The relative amounts of TwSQS were quantified by using 2 −∆∆Ct method [44].

Conclusions
We have successfully cloned and characterized a squalene synthase gene, which is involved in celastrol biosynthesis in the medicinal plant T. wilfordii. The recombinant TwSQS protein was expressed in E. coli and purified. After determination of squalene by GC-MS, we verified that the TwSQS encodes a functional squalene synthase. Furthermore, the expression profile of the SQS gene in different tissues and methyl jasmonate treatments were determined. The results indicate that TwSQS is positively involved in celastrol biosynthesis. The present study is helpful to understand celastrol in T. wilfordii at the molecular level. Further studies on TwSQS would be useful for understanding celastrol biosynthesis and would provide molecular resources for the biotechnological improvement of this important medicinal plant.