Chinese Journal of Natural Medicines  2019, Vol. 17Issue (8): 575-584  
0

Cite this article as: 

TONG Yu-Ru, ZHANG Yi-Feng, ZHAO Yu-Jun, HU Tian-Yuan, WANG Jia-Dian, HUANG Lu-Qi, GAO Wei. Differential expression of the TwHMGS gene and its effect on triptolide biosynthesis in Tripterygium wilfordii[J]. Chinese Journal of Natural Medicines, 2019, 17(8): 575-584.
[Copy]

Research funding

This work was supported by the National Natural Science Foundation of China (No. 81773830), Beijing Natural Science Foundation Program and Scientific Research Key Program of Beijing Municipal Commission of Education (No. KZ201710025022), the Support Project of High-level Teachers in Beijing Municipal Universities in the Period of 13th Five-year Plan (No. CIT & TCD20170324) and National Program for Special Support of Eminent Professionals, and Key project at central government level:The ability establishment of sustainable use for valuable Chinese medicine resources (No. 2060302)

Corresponding author

Tel: 86-10-64014411-2955, Fax: 86-10-64013996, E-mail: huangluqi01@126.com (HUANG Lu-Qi)
Tel/Fax: 86-10-83916572, E-mail: weigao@ccmu.edu.cn (GAO Wei)

Article history

Received on: 22 Apr., 2019
Available online: 20 Aug., 2019
Differential expression of the TwHMGS gene and its effect on triptolide biosynthesis in Tripterygium wilfordii
TONG Yu-Ru1,2,3 Δ, ZHANG Yi-Feng3,4 Δ, ZHAO Yu-Jun2 , HU Tian-Yuan3,4 , WANG Jia-Dian3,4 , HUANG Lu-Qi1,2 , GAO Wei3,4,5     
1 School of Pharmacy Traditional Chinese Materia Medica, Shenyang Pharmaceutical University, Shenyang 110016, China;
2 National Resource Center for Chinese Materia Medica, China Academy of Chinese Medical Sciences, Beijing 100700, China;
3 School of Pharmaceutical Science, Capital Medical University, Beijing 100069, China;
4 School of Traditional Chinese Medicine, Capital Medical University, Beijing 100069, China;
5 Advanced Innovation Center for Human Brain Protection, Capital Medical University, Beijing 100069, China
[Abstract]: 3-Hydroxy-3-methylglutaryl-CoA synthase (HMGS) is the first committed enzyme in the MVA pathway and involved in the biosynthesis of terpenes in Tripterygium wilfordii. The full-length cDNA and a 515 bp RNAi target fragment of TwHMGS were ligated into the pH7WG2D and pK7GWIWG2D vectors to respectively overexpress and silence, TwHMGS was overexpressed and silenced in T. wilfordii suspension cells using biolistic-gun mediated transformation, which resulted in 2-fold increase and a drop to 70% in the expression level compared to cells with empty vector controls. During TwHMGS overexpression, the expression of TwHMGR, TwDXR and TwTPS7v2 was significantly upregulated to the control. In the RNAi group, the expression of TwHMGR, TwDXS, TwDXR and TwMCT visibly displayed downregulation to the control. The cells with TwHMGS overexpressed produced twice higher than the control value. These results proved that differential expression of TwHMGS determined the production of triptolide in T. wilfordii and laterally caused different trends of relative gene expression in the terpene biosynthetic pathway. Finally, the substrate acetyl-CoA was docked into the active site of TwHMGS, suggesting the key residues including His247, Lys256 and Arg296 undergo electrostatic or H-bond interactions with acetyl-CoA.
[Key words]: Overexpression    RNAi    HMGS    Triptolide    Acetyl-CoA    
Introduction

Tripterygium wilfordii (Hook. f.) is a perennial vine plant belonging to the family Celastraceae, which has been widely used to treat rheumatoid arthritis and other inflammatory diseases [1-2]. The medicinal properties of T. wilfordii are attributed to terpenes which mainly include di- and triterpenes. Triptolide is a diterpene triepoxide isolated from T. wilfordii that exhibits strong antitumor activity. The potential mechanism involves the inhibition of the general RNAPII-mediated transcription [3]. A representative T. wilfordii triterpene is the structurally unique celastrol, which hold promise as a powerful anti-obesity agent, a potent anti-inflammatory drug, and a breast cancer treatment [4-6].

Terpenes are derived from the universal C5 precursors isopentenyl diphosphate (IPP) and dimethylallyl diphosphate (DMAPP) [7], which are provided by the mevalonate (MVA) pathway and the 2-methyl-Derythritol-4-phosphate (MEP) pathway [8]. The enzyme 3-hydroxy-3-methylglutaryl-CoA synthase (HMGS) is responsible for the first committed step in the MVA pathway, catalyzing the aldol-type condensation of acetoacetyl-CoA and acetyl-CoA to 3-hydroxy-3-methylglutaryl-CoA (HMG-CoA) [9]. HMGS plays a critical role in the biosynthesis of terpenes in medicinal plants and provides the possibility of regulating the production of target terpene metabolites. The HMGS of Ginkgo biloba has been isolated and characterized via functional complementation of GbHMGS1 in Saccharomyces cerevisiae [10]. An Escherichia coli strain that was transformed with a codon-optimized HMGS gene exhibited significantly improved bisabolene production over the control bacteria [11]. LIU et al. cloned the full-length cDNA of TwHMGS gene and proved that the enzyme participated in yeast terpene biosynthesis in a functional complement assay [12].

However, the unambiguous correlation between TwHMGS expression and terpene biosynthesis in T. wilfordii has not yet been studied in detail. Overexpression and RNA interference (RNAi) genes encoding the presumptive enzymes directly regulate the transcription levels and change the flux towards the desired end-products. Overexpressing the rate-limiting enzyme HMGR in tobacco (Nicotiana benthamiana) led to a 20- to 40-fold increase of sesquiterpenes and a 6-fold increase of triterpenes. The production of α-Bisabolol and β-amyrin finally reached 28.8 and 9.8 mg·g–1 dry weight, respectively [13]. Manipulation of Brassica juncea HMGS1 in transgenic tomato not only up-regulated several genes related to terpene biosynthesis, but also increased the accumulation of MVA- derived squalene and phytosterols, as well as MEP-derived α-tocopherol and carotenoids [14]. It stands to reason that the overexpression of HMGS endowed the transgenic tomato plants with a higher terpene accumulation capacity through both MVA and MEP pathways. RNAi provides an effective post-transcriptional silencing method to inhibit the expression of endogenous genes. In this approach, the RNA-induced silencing complex, an siRNA-containing effector complex, recognizes and cleaves the complementary target RNAs [15].

In this study, we analyzed the phylogenetic position of TwHMGS. Furthermore, TwHMGS was overexpressed and silenced using RNAi in T. wilfordii suspension cells. The expression of relevant genes in the terpene biosynthetic pathway and accumulation of triptolide and celastrol were analyzed in the TwHMGS overexpression and RNAi groups. The increase and decrease of triptolide content was consistent with the up- and downregulation of TwHMGS, suggesting that the encoded enzyme plays an important role in the biosynthesis of this important bioactive molecule. On the basis of molecular docking, we obtained the binding state of the TwHMGS complex with acetyl-CoA and analyzed key residues around the substrate which were suggested to participate in the acetylation/deacetylation, condensation/cleavage and hydrolysis/dehydration processes of the catalytic reaction [16]. Thess findings add to the fundamental knowledge of HMGS and the production of terpenes with pharmacological properties.

Materials and Methods Suspension cells

The T. wilfordii suspension cells used in this study were cultured as previously reported [17].

Bioinformatics analysis

The phylogenetic analysis of TwHMGS and other HMGSs from diverse species was conducted using MEGA 5.1 software (Arizona State University, USA). The multiple alignment analysis of TwHMGS and nine other HMGSs was especially performed using ClustalW2 (https://www.ebi.ac.uk/Tools/msa/clustalw2/) and ESPript 3 (http://espript.ibcp.fr/ESPript/ESPript/). The nine HMGSs from different species were obtained in PDB (http://www.rcsb.org/).

Construction of the entry- and expression vectors

The entry vectors were constructed using pENTRTM/D- TOPO® Cloning Kit (Invitrogen, USA). The overexpression fragment corresponded to the open reading frame (ORF) of TwHMGS, while the RNAi fragment contained a specific 515 bp fragment from 880-1394 bp of the TwHMGS ORF. All the primers were designed using Primer Premier 5.0 software (PREMIER Biosoft, Canada). Both the overexpression fragment and RNAi fragment were amplified using Phusion High-Fidelity DNA Polymerase (New England Biolabs, USA), with the TwHMGS plasmid being as the template. Subsequently, the purified target fragments were ligated into the pENTRTM/D-TOPO entry vector according to the manufacturer's instructions (Invitrogen, USA). The entry vectors pENTR-TwHMGSOE and pENTR-TwHMGSRi were confirmed by sequencing.

The expression vectors were constructed using Gateway LR ClonaseTM Ⅱ Enzyme Mix (Invitrogen, USA). The reaction mixture was composed of pENTR-TwHMGSOE (100 ng), pH7WG2D (150 ng), 1 μL of LR Clonase, and 0.6 μL of TE buffer (pH 8.0). The reaction was carried out at 25 ℃ for 2 h, after which 0.5 μL of Proteinase K Solution was added and incubated at 37 ℃ for 10 min. pENTR-TwHMGSRi and pK7GWIWG2D were used to construct the RNAi expression vector. After screening and sequencing, the recombinant expression vectors pH7WG2D-TwHMGSOE and pK7GWIWG 2D-TwHMGSRi were enriched and extracted using EZNA® Plasmid Maxi Kit (Omega Bio-tek Inc., USA).

Genetic transformation of T. wilfordii suspension cells using a biolistic-gun

The T. wilfordii suspension cells were cultured in MS solid medium. We used a PDS, 100/He gene gun (Bio-Rad, USA) to deliver the expression vectors pH7WG2D- TwHMGSOE and pK7GWIWG2D-TwHMGSRi mixed with 1 μm gold micro-particles into the T. wilfordii suspension cells. The empty vectors pH7WG2D and pK7GWIWG2D were also performed as negative controls. The detailed protocol has been published previously [18]. Each vector transformation was performed in five biological replicates.

Verification of successful transformation with the expression vectors

After genetic transformation, the suspension cells were cultured for 5-7 days to allow sufficient gene expression and metabolite accumulation. To verify the successful transformation, total RNA of suspension cells was isolated from the suspension cells using the Total RNA Extraction Kit (Promega, Shanghai, China), after which the RNAs were reverse-transcribed to first-strand cDNAs as templates for PCR using the FastQuant RT kit (with gDNase, Tiangen Biotech, Beijing, China). The pH7WG2D vector encodes hygromycin resistance (Hyg) and spectinomycin resistance, while the pK7G WIWG2D vector contains a kanamycin resistance (Kan) cassette. Specific primers were designed to amplify the Hyg fragment and the Kan fragment with the cDNA as templates. The PCR products were detected by gel electrophoresis to verify the successful transformation of the suspension cells with the expression vectors.

Transcriptional expression analysis of TwHMGS and other selected genes

Transcript levels of TwHMGS and a number of other relevant genes were analyzed using real-time quantitative polymerase chain reactions (RT-qPCR). The LightCycler 480 Ⅱ (Roche, Switzerland) and KAPA SYBR® FAST qPCR Master Mix Kit (KAPA Biosystems, USA) were used to carry out the expression analysis. The house-keeping gene β-actin was used as internal control in each reaction. The RT-qPCR primers for TwHMGS, TwHMGR, TwDXS, TwDXR, TwMCT, TwGGPS, and TwTPS7v2 are listed in Supplementary Table. The RT-qPCR reaction mixture contained 10 μL of 2 × qPCR Master Mix, 1 μL of cDNA (10 ng), and 0.4 μL of primers (10 μmol·L-1) and 8.2 μL of ddH2O. The RT-qPCR conditions encompassed enzyme activation for 3 min at 95 ℃, followed by denaturation for 3 s at 95 ℃, annealing, extension and data acquisition for 30 s at 60 ℃. These conditions were repeated for 40 cycles. Each sample was run in triplicate to reduce data errors, and the relative expression levels were calculated using the 2–△△Ct method.

Extraction and determination of biologically active terpenes in T. wilfordii

The remaining suspension cells were ground into powder in liquid nitrogen and freeze-dried for 36 h. Approximately 20 mg dry powder were weighed and soaked in 1 mL of 80% (V/V) methanol overnight at 4 ℃, followed ultrasonication at 60 kHz and 25 ℃ for 30 min. Subsequently, the supernatant obtained by centrifugation at 12 000 g for 10 min was filtered through a 0.22 μm PTFE microporous membrane. The contents of triptolide and celastrol were determined using an ACQUITY UPLCTM I-Class System (Waters, Milford, MA, USA) that included a PDA detector. An ACQUITY UPLC HSS T3 chromatographic column (1.8 μm, 2.1 mm ×100 mm, Waters, USA) was used to separate ingredients, keeping the column temperature at 40 ℃. The mobile phase was a mixture containing 0. 1% (V/V) formic acid in acetonitrile (mobile phase A) and 0.1% (V/V) formic acid in water (mobile phase B). The elution programme comprised 30% (A) at the beginning, 30%-35% (A) at 0-5 min, 35% (A) at 5-8 min, 35%-70% (A) at 8-15 min, 70%-90% (A) at 15-21 min, 90% (A) until 24 min, and the flow rate was set at 0.4 mL·min–1. The standard triptolide (> 98%, DESITE, China) and celastrol (> 98%, DESITE, China) were used to generate standard curves to quantify their contents in each suspension cells. The standard curve was plotted by corrected peak area (Y) for every standard against its concentration (X/mg·L1), generating relative equations to quantify the triptolide and celastrol.

Molecular docking of TwHMGS-ligand complexes

The HMGS from Brassica juncea is the only plant-derived HMGS with an available crystal structure. Thus, the model structure of TwHMGS was constructed using the SWISS Model (https://swissmodel.expasy.org/) with BjHMGS (PDB: 2FA3) as the template. The value of GMQE and identity calculated by SWISS Model proved the reliability of modeling. The 2D structure of the substrate acetyl-CoA was downloaded from PubChem (https://pubchem.ncbi.nlm.nih.gov/compound/acetyl-CoA) and refined using the elbow program in the PHENIX suite. Then, the acetyl-CoA was prepared and docked to the receptor grid of the prepared TwHMGS as described previously [19]. The analysis of the docking result was performed in PyMOL (DeLano Scientific LLC, USA).

Results Sequence analysis of TwHMGS

TwHMGS catalyzes the first committed step of the MVA pathway for terpene biosynthesis in T. wilfordii. The MVA pathway genes which were derived from archaeal evolved through lateral transfer from bacteria and other domain of life [20]. The evolutionary position of TwHMGS was exhibited in a phylogenetic tree of the HMGSs from various species (Fig. 1). The TwHMGS has arisen from a common ancestral gene that evolved into the five groups of enzymes from bacteria, animals, fungi, angiosperms and gymnosperms, which suggests an evolutionary relationship from lower organisms to higher organisms. In the angiosperm group, TwHMGS was 81.74% identical to Camptotheca acuminata HMGS [21].

Fig. 1 Phylogenetic analysis of the amino acid sequence of HMGS from different species. The Neighbor-Joining phylogenetic tree was constructed using the bootstrap method in MEGA 5.1 and the number of Bootstrap replications was 1000. The protein sequences used in the tree are as follows. Bacteria: Borreliella burgdorferi CA382 (AGS66682.1), Nocardia brasiliensis ATCC 700358 (AFU04599.1), Actinotignum schaalii (AIE81946.1); Animals: Homo sapiens (AAB72036.1), Bombyx mori (BAF62107.1), Rattus norvegicus (AAA41336.1); Fungi: Schizosaccharomyces japonicus yFS275 (XP_002175595.2), Komagataella phaffii CBS 7435 (SCV12104.1), Metarhizium robertsii (EXV00357.1); Angiosperms: Camptotheca acuminata (ACD87446.1), Matricaria chamomilla (AMN10095.1), Nicotiana benthamiana (BBE00752.1), Artemisia annua (PWA96733.1); Gymnosperms: Taxus x media (AAT73206.1), Ginkgo biloba (AUW64606.1), Pinus sylvestris (CAA65250.1)

To date, the crystal structures of nine HMGSs have been reported, including those from Homo sapiens, bacteria and plant [22-24]. To explore the probable relationship of function and secondary structure, we aligned the sequence of TwHMGS with the nine known HMGSs. The details of secondary structures elements were shown in Fig. 2. Multiple sequence alignment among the HMGS from Brassica juncea showed 81% identity, from bacteria for Myxococcus xanthus (strain DK 1622), Enterococcus faecalis, Moorea producens 3L, showed 28.36%, 24.25% and 19.20% identity, respectively, from Homo sapiens exhibited 46.65% identity.

Fig. 2 Sequence alignment and secondary structure of TwHMGS and nine HMGSs with reported crystal structures. The protein sequences are as follows. Brassica juncea (PDB: 2FA3), Staphylococcus aureus (PDB: 1TXT), Enterococcus faecalis (PDB: 1X9E); Staphylococcus aureus (strain MW2) (PDB: 1XPK), Streptococcus mutans serotype c (strain ATCC 700610/UA159) (PDB: 3SQZ), Myxococcus xanthus (strain DK 1622) (PDB: 5HWO), Moorea producens 3L (PDB: 5KP5), Methanothermococcus thermolithotrophicus (PDB: 6ESQ), Homo sapiens (PDB: 2P8U)
Verification of successful transformation with the expression vectors

The TwHMGS overexpression vector harbored the ORF fragment of TwHMGS with 1398 bp, while the RNAi expression vector contained partial fragment of TwHMGS with 515 bp. Electrophoresis of the target bands was shown in Fig. 3B. Specific primers were designed to amplify the cDNA of each sample. The pH7WG2D and pH7WG2D-TwHMGSOE were identified via the Hyg fragment of 1787 bp, while pK7GWIWG2D and pK7GWIWG2D-TwHMGSRi were verified by the Kan fragment of 1397 bp (Fig. 3). The electrophoresis bands were shown in Fig. 3C. These results indicated that the suspension cells were successfully transformed with the pH7WG2D-TwHMGSOE, pK7GWIWG2D- TwHMGSRi, pH7WG2D and pK7GWIWG2D vectors, respectively.

Fig. 3 Agarose gel electrophoresis of target fragments. (A) Schematic diagram of vector construction. Hyg stands for the hygromycin resistance cassette, Kan for kanamycin. (B) OE: TwHMGS overexpression fragment (1398 bp), Ri: TwHMGS RNAi fragment (515 bp). (C) Partial agarose gel electrophoresis of the Hyg fragment (1787 bp) and Kan fragment (1397 bp). OE stands for overexpression; Ri is RNAi
Expression analysis of TwHMGS and the related genes in the terpene biosynthetic pathway

The relative transcription level of TwHMGS was determined by RT-qPCR. Compared to the suspension cells transformed with the empty vector, the overexpression group exhibited a 2-fold increase (Fig. 4A). Conversely, the expression of TwHMGS in the RNAi group was reduced to approximately 70% of the control (Fig. 4B).

Fig. 4 Expression analysis of related genes in overexpression and RNAi suspension cells. (A) Relative expression levels in the TwHMGS overexpression group with empty vector pH7WG2D as control. (B) Relative expression levels in the TwHMGS RNAi group with empty vector pK7GWIWG2D as control. The data represented the averages of five independent biological samples with three technical replicates. **P < 0.01, n = 5. The difference between the means of two samples is tested by independent samples t-test method

Apart from the expression analysis of TwHMGS, we focused on the expression of genes in the MEP pathway and further downstream in the biosynthesis of terpene (Fig. 4). HMG-CoA reductase (HMGR) and HMGS are adjacent enzymes in the cytosolic MVA pathway. The HMGR converts HMG-CoA, the product of HMGS, to mevalonate. In the TwHMGS overexpression group, the TwHMGR mRNA was induced, reaching an approximately 1.9-fold high value that that of the control. In the RNAi group, the TwHMGR mRNA level decreased to about half that of the control (Fig. 4). The 1-deoxy-D-xylulose-5-phosphate synthase (DXS) and 1-deoxy-D-xylulose-5-phosphate reductoisomerase (DXR) are the first two rate-limiting enzymes in the plastidic MEP pathway. Additionally, 2-C-methyl-D-erythritol-4-phosphate cytidylyltransferase (MCT or IspD) is another key enzyme in the MEP pathway. The expression of TwDXS, TwDXR and TwMCT decreased in all RNAi suspension cells, whereas the levels of TwDXS and TwMCT in the overexpression group remained unaffected. Among these, the mRNA of TwDXR had the largest increment during TwHMGS overexpression (Fig. 4A). TwTPS7v2 is the class I diterpene cyclase in the downstream pathway of terpene synthesis [25]. Approximately 1.8-fold increasing was observed in the expression of TwTPS7v2 when TwHMGS was overexpressed (Fig. 4). The expression of TwGGPS did not change significantly during TwHMGS overexpression and RNAi.

Accumulation of triptolide and celastrol in the suspension cells

Triptolide and celastrol are pharmacologically active diterpene and triterpene natural products of T. wilfordii, respectively. We detected the content of triptolide and celastrol in the overexpression groups, RNAi groups and the control groups. The methanol extracts from independent samples were analyzed by UPLC. The chromatogram of triptolide and celastrol was shown in Figs. 5A and 5B, where peak 1 detected at 425 nm corresponded to triptolide and peak 2 detected at 219.5 nm to celastrol. The standard curve equations for quantifying triptolide and celastrol were Y = 10 849X + 6075.8 (R2 = 0.9964) and Y = 2678.4X – 261.29 (R2 = 0.9998), respectively. Overexpression of TwHMGS promoted the synthesis of triptolide by 2-fold and had no significant effect on the celastrol content compared to the control groups (Fig. 5C). In the RNAi and pK7GWIWG2D groups, the contents of triptolide were 5.51 ± 1.87 and 14.80 ± 2.17 μg·g–1, respectively, while those of celastrol were 41.50 ± 5.76 and 50.56 ± 10.95 μg·g–1. The limited culture time of the suspension cells that were transformed with expression vectors may be a bit short for the accumulation of celastrol. These results supported the vital function of TwHMGS in the biosynthesis of terpenes in T. wilfordii suspension cells.

Fig. 5 UPLC chromatograms and contents of triptolide and celastrol in T. wilfordii suspension cells after transformation with expression vectors. (A) UPLC chromatograms of samples and celastrol standard monitored at 425 nm. Peak 1: celastrol. (B) UPLC chromatograms of samples and triptolide standard monitored at 219.5 nm. Peak 2: triptolide. (C) The data represented the averages of five independent biological samples with three technical replicates. OE stands for overexpression; Ri is RNAi. **P < 0.01, n = 5. The difference between the means of two samples is tested by independent samples t-test method
Ligand docking suggesting the key residues for catalysis

The crystal structures of BjHMGS in the apo-form and BjHMGS in complex with acetyl-CoA exhibited an open- closed change of conformation [22]. Therefore, the structure of BjHMGS in complex with acetyl-CoA structure was used as template to construct the three-dimensional homology model of TwHMGS. The values of GMQE and identity were 0.92% and 83.93%, respectively, which support the reliability of modeling. The overall structure of TwHMGS was composed of helices, sheets and loops, and as similar to the reported structures BjHMGS and bacterial HMGS with the conserved αβαβα catalytic fold (Fig. 6A) [26-29]. Depending on the ligand docking result, the substrate acetyl-CoA and some residues, including the Arg296, Lys256, and His247, potentially existed mutual polar contacts excluding solvent. The His247 was reported to be a highly conserved site (Fig. 2), serving as a hydrogen donor in the deacetylation process of BjHMGS [22]. In TwHMGS, the distance between His-247 and the acetyl oxygen atom was 2.9 Å, which may share the same mechanism reported for BjHMGS. The phosphate groups of acetyl- CoA were bound to basic residues, such as K256, K34 and R296 through electrostatic interactions (Fig. 6B). In addition, the CoA was anchored by an H-bond interaction between the adenine moiety and the hydroxyl side chain of Ser31 (Fig. 6B).

Fig. 6 Ligand docking of acetyl-CoA in the active site of TwHMGS. (A) The overall structure of TwHMGS according to homology modeling with BjHMGS (PDB: 2FA3) as the template. The structure of TwHMGS is colored by secondary structure, red for helices, yellow for sheets and green for loops. The ligand acetyl-CoA is shown in stick. (B) Key residues around the acetyl-CoA are labeled, the yellow dotted line linkes the residues with potential polar contacts and the distances are less than 3.5 Å
Discussion

HMGS is the rate-limiting enzyme in the biosynthesis of isoprenoids, the C5 precursor of all terpenes. The crystal structures of HMGS from Staphylococcus aureus [26] and from Brassica juncea [22], in conjunction with site-directed mutagenesis studies, have revealed details of the mechanism of catalysis. The reaction catalyzed by HMGS consists of three steps namely acetylation, condensation and dehydration. We noticed that key residue R296 was not conserved among the seven bacteria including Myxococcus xanthus and Enterococcus faecalis (Fig. 2). Furthermore, the residue K256 of TwHMGS was not completely conserved in other HMGS, especially that from Moorea producens 3 L [30]. This phenomenon suggested that the residues may be responsible for some of the structural differences between bacterial and plant enzymes. Overexpression of Brassica juncea HMGS has been studied in Arabidopsis. The stigmasterol and sitosterol contents were significantly increased in leaves and seedlings of the overexpression lines [31]. A similar result was shown in Fig. 5. However, neither the overexpression group nor the RNAi group showed significant changes in the accumulation of celastrol, the main bioactive triterpene of T. wilfordii. Generally speaking, triterpene biosynthesis relies on the cytosolic MVA pathway, while diterpene biosynthesis mainly uses the plastidial MEP pathway. However, metabolic cross-talk has been reported to occur between the MVA and MEP pathways, suggesting that isoprenoids flow between them [32-34]. This may explain the significant changes in the content of triptolide. In addition, the ratio of DMAPP/IPP influences the isoprene flux, and the differential expression of TwHMGS may interfere with the dynamic equilibrium between IPP and DMAPP [35], which we assumed to affect triptolide and celasol biosyntehsis. ZHANG et al. and SU et al. measured the terpenes of T. wilfordii suspension cells and found that the content of celastrol was higher than that of triptolide [25, 36-37]. The differential expression of single gene may prefer metabolite flux to compounds with low content. It was reported that the overexpression and RNAi of TwDXR and TwIDI affected the feedback regulation of other genes in the terpene synthesis pathway [36, 38]. The TwHMGS in our research just exhibited a synergistic regulation (Fig. 7).

Fig. 7 The relative genes in the biosynthesis pathway of triptolide and celastrol. OE indicates TwHMGS overexpression and Ri indicated TwHMGS RNAi. The upregulating genes and product are labeled with red arrows during TwHMGS overexpression. The downregulating genes and product are labeled with blue arrows during TwHMGS RNAi

The His188 and Ser359 of BjHMGS play a particularly important role in the catalytic activity of this enzyme [39]. The H188N mutation abrogated the substrate inhibition by acetoacetyl-CoA, while the S359A mutation highly increased BjHMGS activity. We noticed that the Ser359 in TwHMGS was close to the substrate acetyl-CoA (Fig. 6B). Overexpression of the wild type BjHMGS, the H188N mutant, S359A mutant and H188N/S359A double mutant in Arabidopsis improved both sterol production and stress tolerance [31]. Furthermore, the introduction of BjHMGS: S359A into transgenic tomato obviously enhanced the α-tocopherol, carotenoid, squalene and phytosterol contents in the fruits [14].

Conclusion

TwHMGS enzyme catalyzes the condensation of acetoacetyl-CoA to form HMG-CoA in the MVA pathway of terpene synthesis. RNAi and overexpression vectors targeting TwHMGS gene were respectively introduced into T. wilfordii suspension cells via biolistic-gun transformation and verified by PCR. TwHMGS overexpression in T. wilfordii not only elevated the expression of adjacent TwHMGR, but also TwDXR in the MEP pathway, and further downstream, the terpene cyclase gene TwTPS7v2 related to diterpene biosynthesis was also upregulated. Meanwhile, a synergistic downregulation of TwDXS, TwDXR, TwMCT and TwHMGR occurred in the TwHMGS RNAi group (Fig. 7). Subsequently, changes in the diterpene end-product triptolide were investigated, providing an important reference for regulating the key genes in the biosynthesis pathway to generate pharmacologically active ingredients. The potential mechanism of HMGS catalytic activity was analyzed on the basis of molecular docking, exploring the binding state of TwHMGS in complex with acetyl-CoA.

Acknowledgments

The authors acknowledge XIANG Yu-Hong in Capital Normal University for supporting SYBYL-based computing services.

References
[1]
Goldbach-Mansky R, Wilson M, Fleischmann R, et al. Comparison of Tripterygium wilfordii Hook F versus sulfasalazine in the treatment of rheumatoid arthritis: a randomized trial[J]. Ann Intern Med, 2009, 151(4): 229-240. DOI:10.7326/0003-4819-151-4-200908180-00005
[2]
Wang M, Huang J, Fan H, et al. Treatment of rheumatoid arthritis using combination of methotrexate and tripterygium glycosides tablets-A quantitative plasma pharmacochemical and pseudotargeted metabolomic approach[J]. Front Pharmacol, 2018, 9: 1051. DOI:10.3389/fphar.2018.01051
[3]
He QL, Minn I, Wang Q, et al. Targeted delivery and sustained antitumor activity of triptolide through glucose conjugation[J]. Angew Chem Int Ed Engl, 2016, 55(39): 12035-12039. DOI:10.1002/anie.201606121
[4]
Liu J, Lee J, Salazar Hernandez MA, et al. Treatment of obesity with celastrol[J]. Cell, 2015, 161(5): 999-1011. DOI:10.1016/j.cell.2015.05.011
[5]
Hu M, Luo Q, Alitongbieke G, et al. Celastrol-induced Nur77 interaction with TRAF2 alleviates inflammation by promoting mitochondrial ubiquitination and autophagy[J]. Mol Cell, 2017, 66(1): 141-153. DOI:10.1016/j.molcel.2017.03.008
[6]
Dandawate PR, Subramaniam D, Jensen RA, et al. Targeting cancer stem cells and signaling pathways by phytochemicals: Novel approach for breast cancer therapy[J]. Semin Cancer Biol, 2016, 40-41: 192-208. DOI:10.1016/j.semcancer.2016.09.001
[7]
Kuzuyama T, Seto H. Two distinct pathways for essential metabolic precursors for isoprenoid biosynthesis[J]. Proc Jpn Acad Ser B Phys Biol Sci, 2012, 88(3): 41-52. DOI:10.2183/pjab.88.41
[8]
Lombard J, Moreira D. Origins and early evolution of the mevalonate pathway of isoprenoid biosynthesis in the three domains of life[J]. Mol Biol Evol, 2011, 28(1): 87-99.
[9]
Puisac B, Marcos-Alcalde I, Hernandez-Marcos M, et al. Human mitochondrial HMG-CoA synthase deficiency: Role of enzyme dimerization surface and characterization of three new patients[J]. Int J Mol Sci, 2018, 19(4): E1010. DOI:10.3390/ijms19041010
[10]
Meng X, Song Q, Ye J, et al. Characterization, function, and transcriptional profiling analysis of 3-hydroxy-3-methylglutaryl-CoA synthase gene (GbHMGS1) towards stresses and exogenous hormone treatments in Ginkgo biloba[J]. Molecules, 2017, 22(10): E1706. DOI:10.3390/molecules22101706
[11]
Peralta-Yahya PP, Ouellet M, Chan R, et al. Identification and microbial production of a terpene-based advanced biofuel[J]. Nat Commun, 2011, 2: 483. DOI:10.1038/ncomms1494
[12]
Liu YJ, Zhao YJ, Zhang M, et al. Cloning and characterisation of the gene encoding 3-hydroxy-3-methylglutaryl-CoA synthase in Tripterygium wilfordii[J]. Molecules, 2014, 19(12): 19696-19707. DOI:10.3390/molecules191219696
[13]
Lee AR, Kwon M, Kang MK, et al. Increased sesqui- and triterpene production by co-expression of HMG-CoA reductase and biotin carboxyl carrier protein in tobacco (Nicotiana benthamiana)[J]. Metab Eng, 2019, 52: 20-28. DOI:10.1016/j.ymben.2018.10.008
[14]
Liao P, Chen X, Wang M, et al. Improved fruit alpha-tocopherol, carotenoid, squalene and phytosterol contents through manipulation of Brassica juncea 3-HYDROXY-3-METHYLGLUTARYL-COA SYNTHASE1 in transgenic tomato[J]. Plant Biotechnol J, 2018, 16(3): 784-796. DOI:10.1111/pbi.12828
[15]
Meister G, Tuschl T. Mechanisms of gene silencing by double-stranded RNA[J]. Nature, 2004, 431(7006): 343-349. DOI:10.1038/nature02873
[16]
Miziorko HM, Clinkenbeard KD, Reed WD, et al. 3-Hydroxy- 3-methylglutaryl coenzyme A synthase. Evidence for an acetyl- S-enzyme intermediate and identification of a cysteinyl sulfhydryl as the site of acetylation[J]. J Biol Chem, 1975, 250(15): 5768-5773.
[17]
Tong Y, Su P, Zhao Y, et al. Molecular cloning and characterization of DXS and DXR genes in the terpenoid biosynthetic pathway of Tripterygium wilfordii[J]. Int J Mol Sci, 2015, 16(10): 25516-25535. DOI:10.3390/ijms161025516
[18]
Zhao Y, Zhang Y, Su P, et al. Genetic transformation system for woody plant Tripterygium wilfordii and its application to product natural celastrol[J]. Front Plant Sci, 2017, 8: 2221.
[19]
Tong YR, Su P, Guan HY, et al. Eudesmane-type sesquiterpene diols directly synthesized by a sesquiterpene cyclase in Tripterygium wilfordii[J]. Biochem J, 2018, 475(17): 2713-2725. DOI:10.1042/BCJ20180353
[20]
Boucher Y, Doolittle WF. The role of lateral gene transfer in the evolution of isoprenoid biosynthesis pathways[J]. Mol Microbiol, 2000, 37(4): 703-716. DOI:10.1046/j.1365-2958.2000.02004.x
[21]
Kai G, Li S, Wang W, et al. Molecular cloning and expression analysis of a gene encoding 3-hydroxy-3-methylglutaryl-CoA synthase from Camptotheca acuminata[J]. Rus J Plant Physiol, 2013, 60(1): 131-138. DOI:10.1134/S102144371206009X
[22]
Pojer F, Ferrer JL, Richard SB, et al. Structural basis for the design of potent and species-specific inhibitors of 3-hydroxy- 3-methylglutaryl CoA synthases[J]. Proc Natl Acad Sci U S A, 2006, 103(31): 11491-11496. DOI:10.1073/pnas.0604935103
[23]
Fu Z, Runquist JA, Forouhar F, et al. Crystal structure of human 3-hydroxy-3-methylglutaryl-CoA lyase insights into catalysis and the molecular basis for hydroxymethylglutaric aciduria[J]. J Biol Chem, 2006, 281(11): 7526-7532. DOI:10.1074/jbc.M506880200
[24]
Skaff DA, Ramyar KX, McWhorter WJ, et al. Biochemical and structural basis for inhibition of Enterococcus faecalis hydroxymethylglutaryl-CoA synthase, mvaS, by hymeglusin[J]. Biochemistry, 2012, 51(23): 4713-4722. DOI:10.1021/bi300037k
[25]
Su P, Guan H, Zhao Y, et al. Identification and functional characterization of diterpene synthases for triptolide biosynthesis from Tripterygium wilfordii[J]. Plant J, 2018, 93(1): 50-65. DOI:10.1111/tpj.13756
[26]
Theisen MJ, Misra I, Saadat D, et al. 3-Hydroxy-3-methylglutaryl-CoA synthase intermediate complex observed in preal- timeq[J]. Proc Natl Acad Sci U S A, 2004, 101(47): 16442-16447. DOI:10.1073/pnas.0405809101
[27]
Steussy CN, Vartia AA, Burgner JW, et al. X-ray crystal structures of HMG-CoA synthase from Enterococcus faecalis and a complex with its second substrate/inhibitor acetoacetyl-CoA[J]. Biochemistry, 2005, 44(43): 14256-14267. DOI:10.1021/bi051487x
[28]
Steussy CN, Robison AD, Tetrick AM, et al. A structural limitation on enzyme activity: the case of HMG-CoA synthase[J]. Biochemistry, 2006, 45(48): 14407-14414. DOI:10.1021/bi061505q
[29]
Vogeli B, Engilberge S, Girard E, et al. Archaeal acetoacetyl- CoA thiolase/HMG-CoA synthase complex channels the intermediate via a fused CoA-binding site[J]. Proc Natl Sci U S A, 2018, 115(13): 3380-3385. DOI:10.1073/pnas.1718649115
[30]
Maloney FP, Gerwick L, Gerwick WH, et al. Anatomy of the beta-branching enzyme of polyketide biosynthesis and its interaction with an acyl-ACP substrate[J]. Proc Natl Acad Sci U S A, 2016, 113(37): 10316-10321. DOI:10.1073/pnas.1607210113
[31]
Wang H, Nagegowda DA, Rawat R, et al. Overexpression of Brassica juncea wild-type and mutant HMG-CoA synthase 1 in Arabidopsis up-regulates genes in sterol biosynthesis and enhances sterol production and stress tolerance[J]. Plant Biotechnol J, 2012, 10(1): 31-42. DOI:10.1111/j.1467-7652.2011.00631.x
[32]
Mendoza-Poudereux I, Kutzner E, Huber C, et al. Metabolic cross-talk between pathways of terpenoid backbone biosynthesis in spike lavender[J]. Plant Physiol Biochem, 2015, 95: 113-120. DOI:10.1016/j.plaphy.2015.07.029
[33]
Opitz S, Nes WD, Gershenzon J. Both methylerythritol phosphate and mevalonate pathways contribute to biosynthesis of each of the major isoprenoid classes in young cotton seedlings[J]. Phytochemistry, 2014, 98: 110-119. DOI:10.1016/j.phytochem.2013.11.010
[34]
Hemmerlin A, Hoeffler JF, Meyer O, et al. Cross-talk between the cytosolic mevalonate and the plastidial methylerythritol phosphate pathways in tobacco bright yellow-2 cells[J]. J Biol Chem, 2003, 278(29): 26666-26676. DOI:10.1074/jbc.M302526200
[35]
Weise SE, Li Z, Sutter AE, et al. Measuring dimethylallyl diphosphate available for isoprene synthesis[J]. Anal Biochem, 2013, 435(1): 27-34.
[36]
Zhang Y, Zhao Y, Wang J, et al. Overexpression and RNA interference of TwDXR regulate the accumulation of terpenoid active ingredients in Tripterygium wilfordii[J]. Biotechnol Lett, 2018, 40(2): 419-425.
[37]
Su P, Cheng Q, Wang X, et al. Characterization of eight terpenoids from tissue cultures of the Chinese herbal plant, Tripterygium wilfordii, by high-performance liquid chromatography coupled with electrospray ionization tandem mass spectrometry[J]. Biomed Chromatogr, 2014, 28(9): 1183-1192. DOI:10.1002/bmc.3140
[38]
Wang J, Zhao Y, Zhang Y, et al. Overexpression and RNAi- mediated downregulation of TwIDI regulates triptolide and celastrol accumulation in Tripterygium wilfordii[J]. Gene, 2018, 679: 195-201. DOI:10.1016/j.gene.2018.08.072
[39]
Nagegowda DA, Bach TJ, Chye ML. Brassica juncea 3-hydroxy-3-methylglutaryl (HMG)-CoA synthase 1: expression and characterization of recombinant wild-type and mutant enzymes[J]. Biochem J, 2004, 383(Pt. 3): 517-527.