Lipopolysaccharide Enhances Tanshinone Biosynthesis via a Ca²⁺-Dependent Manner in Salvia miltiorrhiza Hairy Roots

Abstract

Tanshinones, the major bioactive components in Salvia miltiorrhiza Bunge (Danshen), are synthesized via the mevalonic acid (MVA) pathway or the 2-C-methyl-D-erythritol-4-phosphate (MEP) pathway and the downstream biosynthesis pathway. In this study, the bacterial component lipopolysaccharide (LPS) was utilized as a novel elicitor to induce the wild type hairy roots of S. miltiorrhiza. HPLC analysis revealed that LPS treatment resulted in a significant accumulation of cryptotanshinone (CT) and dihydrotanshinone I (DTI). qRT-PCR analysis confirmed that biosynthesis genes such as SmAACT and SmHMGS from the MVA pathway, SmDXS and SmHDR from the MEP pathway, and SmCPS, SmKSL and SmCYP76AH1 from the downstream pathway were markedly upregulated by LPS in a time-dependent manner. Furthermore, transcription factors SmWRKY1 and SmWRKY2, which can activate the expression of SmDXR, SmDXS and SmCPS, were also increased by LPS. Since Ca²⁺ signaling is essential for the LPS-triggered immune response, Ca²⁺ channel blocker LaCl₃ and CaM antagonist W-7 were used to investigate the role of Ca²⁺ signaling in tanshinone biosynthesis. HPLC analysis demonstrated that both LaCl₃ and W-7 diminished LPS-induced tanshinone accumulation. The downstream biosynthesis genes including SmCPS and SmCYP76AH1 were especially regulated by Ca²⁺ signaling. To summarize, LPS enhances tanshinone biosynthesis through SmWRKY1- and SmWRKY2-regulated pathways relying on Ca²⁺ signaling. Ca²⁺ signal transduction plays a key role in regulating tanshinone biosynthesis in S. miltiorrhiza.

1. Introduction

Salvia miltiorrhiza Bunge, also known as danshen, is a widely used Chinese herbal medicine for treating cardiovascular and cerebrovascular diseases [1]. The primary bioactive ingredients of S. miltiorrhiza comprises salvianolic acids and tanshinones. In these secondary metabolites, tanshinones have received extensive attention for their multiple pharmacological properties, including cardioprotective effects, antitumor activity, anti-inflammatory activity and antibacterial activity [1,2]. The valuable tanshinones consist of tanshinone I (TI), tanshinone IIA (TIIA), tanshinone IIB (TIIB), cryptotanshinone (CT), dihydrotanshinone I (DTI), etc.

In S. miltiorrhiza, tanshinone biosynthesis experiences a complicated process. Based on metabonomics and genomics research, isopentenyl diphosphate (IPP) and dimethylallyl diphosphate (DMAPP) have been identified as the precursors of tanshinones. These two compounds can be generated either from the mevalonic acid (MVA) pathway or the 2-C-methyl-D-erythritol-4-phosphate (MEP) pathway [2]. In the MVA pathway, two molecules of acetyl-CoA are formed to acetoacetyl-CoA by acetyl-CoA C-acetyltransferase (AACT), firstly. Then, 3-hydroxy-3-methylglutaryl-CoA (HMG-CoA) is synthesized through adding another acetyl-CoA by 3-hydroxy-3-methylglutaryl-CoA synthase (HMGS). HMG-CoA is further reduced to MVA by 3-hydroxy-3-methylglutaryl-CoA reductase (HMGR). Subsequently, MVA is catalyzed in turn by mevalonate kinase (MK), 5-phosphomevalonate kinase (PMK) and MVAPP decarboxylase (MDC) to generate key intermediate IPP. IPP can be transformed to another precursor DMAPP by isopentenyl-diphosphate deltaisomerase (IPPI). In the MEP pathway, the initial reactants pyruvate and glyceraldehyde 3-phosphate (GA-3P) are catalyzed by 1-deoxy-D-xylulose 5-phosphate synthase (DXS) to form 1-deoxy-D-xylulose 5-phosphate (DXP). Then, DXP is reduced to MEP catalyzed by DXP reductoisomerase (DXR). In the rest of the reactions, there are five enzymes contributing to IPP and DMAPP synthesis, including 2-C-methyl-D-erythritol-4-phosphate cytidyl transferase (MCT), 4-(cytidine 5-diphospho)-2-C-methyl-D-erythritol kinase (CMK), 2-C-methyl-D-erythritol-2,4-cyclodiphosphate synthase (MDS), 1-hydroxy-2-methyl-2-(E)-butenyl-4-diphosphate (HMBPP) synthase (HDS), and HMBPP reductase (HDR) [2,3].

In the subsequent cyclization reactions, IPP and DMAPP are transformed into ferruginol catalyzed in turn by copalyl diphosphate synthase (CPS), kaurene synthase-like (KSL) and cytochrome P450 monooxygenase (CYP76AH1). During the last stage, ferruginol is eventually transformed into different tanshinones through some undefined reactions [4].

Recently, researchers have improved the content of tanshinones in S. miltiorrhiza through various strategies, including elicitor treatment, hormone signal regulation, overexpression of key biosynthesis genes, and transcriptional regulation [3,5,6]. Nevertheless, few studies have illustrated the role of Ca²⁺ signaling in tanshinone biosynthesis. In plant cells, calcium acts not only as an essential nutrient, but also as a crucial second messenger. When confronted with diverse abiotic and biotic stresses, plant cells generate a cytoplasmic Ca²⁺ signal, which can be decoded by calcium sensors such as calmodulin (CaM), calmodulin-like proteins (CMLs), Ca²⁺-dependent protein kinases (CDPKs), and calcineurin B-like proteins (CBLs) to regulate numerous downstream metabolic reactions [7,8].

Notably, Ca²⁺ signaling is closely related to secondary metabolism in plants [9,10,11,12]. For instance, through binding with a Ca²⁺-CaM complex, transcription factor CAMTA3 promotes the production of glucosinolates, which are defensive compounds against herbivores [9,10]. The biosynthesis of salicylic acid (SA) is controlled by CaM-binding transcription factors such as CBP60g, SARD1 and CAMTA3 [11,12]. Our previous studies have shown that Ca²⁺ signaling is essential to the SA-induced rosmarinic acid accumulation in S. miltiorrhiza [13]. Therefore, Ca²⁺ signaling can act as a vital player to regulate secondary metabolite biosynthesis.

Here, we focused on the regulation of secondary metabolism by Ca²⁺ signaling. In this study, bacterial endotoxin lipopolysaccharide (LPS), which is a characteristic glycolipid component of a Gram-negative bacteria cell wall and is a stimulator of pathogen-associated molecular pattern (PAMP) triggered immunity (PTI) [14,15,16], was utilized as an elicitor to treat the wild type hairy roots of S. miltiorrhiza. We found that LPS treatment significantly upregulated the expression of key tanshinone biosynthesis genes and enhanced the accumulation of tanshinones in hairy roots. Due to LPS-triggered PTI depending on Ca²⁺ signaling in plant cells [15], the Ca²⁺ channel blocker LaCl₃ and CaM antagonist W-7 were also used to analyze the role of Ca²⁺ signaling in tanshinone biosynthesis. These results demonstrate that LPS enhances tanshinone biosynthesis in a Ca²⁺-dependent manner and suggest that Ca²⁺ signal transduction is essential for modulating secondary metabolism in S. miltiorrhiza.

2. Results

2.1. LPS Enhances Tanshinone Accumulation in the Wild Type Hairy Roots of S. miltiorrhiza

Since the biosynthesis of secondary metabolites might be induced by microorganisms, the bacterial component LPS was applied as a novel elicitor to treat the wild type (WT) hairy roots of S. miltiorrhiza. After being treated by 50 μg/mL LPS for 10 days, the hairy roots and the culture medium showed a deep red color, which is the characteristic color of tanshinones (Figure 1A). Compared to the control, LPS did not obviously affect the growth of hairy roots (Figure 1A). Further, the content of the tanshinones, including dihydrotanshinone I (DTI), cryptotanshinone (CT), tanshinones I (TI) and tanshinones IIA (TIIA), was analyzed by HPLC. When the hairy roots were treated by LPS, the content of DTI significantly increased from 0.38 mg/g to 0.86 mg/g in contrast to the control (Figure 1B), and the content of CT increased from 0.62 mg/g to 0.9 mg/g (Figure 1C). However, the content of TI and TIIA showed no significant change (Figure 1D,E). These results indicate that LPS can induce the accumulation of DTI and CT without markedly inhibiting the growth of S. miltiorrhiza WT hairy roots.

2.2. LPS Upregulates Key Gene’s Expression in Tanshinone Biosynthesis Pathways

To elucidate the regulation mechanism of LPS, the key biosynthesis genes of tanshinones were analyzed by qRT-PCR. In the biosynthesis pathways of tanshinones, SmAACT and SmHMGS are from the MVA pathway, and SmDXS and SmHDR are from the MEP pathway (Figure 2A). After being induced by LPS, the transcripts levels of SmAACT and SmHMGS were obviously upregulated at 6 h (Figure 2B,C). Similarly, SmDXS and SmHDR showed the same response to LPS treatment (Figure 2D,E). In the confirmed biosynthesis pathway of tanshinones, SmCPS, SmKSL and SmCYP76AH1 are located downstream the MVA and MEP pathways (Figure 2A). Notably, the expression of these three genes also increased along with LPS treatment (Figure 2F–H). SmCPS and SmCYP76AH1 were especially upregulated by LPS in a time-dependent manner (Figure 2F,H).

To further explore the stimulation mechanism of these biosynthesis genes by LPS, the expression of transcription factors SmWRKY1 and SmWRKY2 was analyzed by qRT-PCR. In S. miltiorrhiza, SmWRKY1 can bind with the promoter of SmDXR, and SmWRKY2 can bind with SmDXS and SmCPS, to positively regulate tanshinone biosynthesis [5,17,18]. Our further analysis indicated that LPS upregulated the transcript levels of SmWRKY1 and SmWRKY2 in the same time-dependent manner. The expression of these transcription factors responded to LPS and reached a peak at 6 h (Figure 3A,B). Hence, SmDXS, SmCPS and SmDXR can be highly transcribed due to the activation of SmWRKY1 and SmWRKY2.

Taken together, the secondary metabolite tanshinones can be induced by the immune regulator LPS. LPS enhances tanshinone accumulation through stimulating SmWRKY1- and SmWRKY2-regulated gene expression in tanshinone biosynthesis pathways.

2.3. Ca²⁺ Inhibitors Affect Tanshinone Accumulation

Ca²⁺ signal transduction is essential for the LPS-triggered plant immune response [15]. Thus, three Ca²⁺ signal inhibitors, including Ca²⁺ channel blocker LaCl₃, CaM antagonist W-7 and Ca²⁺ chelator EGTA, were applied to analyze the role of Ca²⁺ signaling in tanshinone biosynthesis [15,19]. Since tanshinones can generate a deep red color in the roots of S. miltiorrhiza, we preliminarily observed the color of the hairy roots treated by different Ca²⁺ reagents. As shown in Figure 4A,B, the hairy roots treated by 1 mmol/L LaCl₃ apparently showed a light color compared to the H₂O control. Similarly, 100 μmol/L W-7 also led to light color in contrast to the DMSO control. Nevertheless, 1mmol/L EGTA did not obviously affect the color of the hairy roots compared to the H₂O control. These results suggest that Ca²⁺ signaling is closely associated with tanshinone biosynthesis. The accumulation of tanshinones might be inhibited by blocking Ca²⁺ influx or repressing CaM-mediated signaling in the hairy roots of S. miltiorrhiza.

2.4. Ca²⁺ Channel Blocker Inhibits LPS-Induced Tanshinone Accumulation

LaCl₃ is capable of suppressing cytoplasmic Ca²⁺ elevation via blocking Ca²⁺ influx [15,19]. Thus, LaCl₃ was synergistically utilized with LPS to analyze the role of Ca²⁺ signaling in tanshinone biosynthesis. As shown in Figure 5A, the LPS-treated hairy roots showed the deepest color and LaCl₃ treatment resulted in the lightest color. LPS-induced deep red was apparently decreased by LaCl₃ synergetic treatment (Figure 5A). Further, the content of tanshinones was examined by HPLC. Compared to the LPS treatment, the content of DTI in the LaCl₃+LPS-treated sample significantly reduced from 0.71 mg/g to 0.28 mg/g, and CT reduced from 0.88 mg/g to 0.63 mg/g (Figure 5B,C). The LPS+LaCl₃ treatment also led to a significant reduction in TI and TIIA in a similar way (Figure 5D,E). These results confirmed that with the inhibition of Ca²⁺ influx by LaCl₃, LPS-induced tanshinone accumulation was accordingly diminished. Therefore, the Ca²⁺ influx signal is involved in regulating tanshinone accumulation.

2.5. CaM Antagonist Inhibits LPS-Induced Tanshinone Accumulation

CaM serves as a crucial sensor in Ca²⁺ signal transduction. Through binding with Ca²⁺, the Ca²⁺-CaM complex interacts with target proteins such as CNGC, CDPK, and MAPK to regulate numerous metabolism reactions [8,20]. Hence, the CaM antagonist W-7 was utilized to corporately treat hairy roots with LPS. As shown in Figure 6A, W-7 treatment partly decreased the LPS-induced deep red of the hairy roots and generated the lightest color, while showing no obvious growth inhibition. Compared to LPS treatment, the content of DTI, CT and TI significantly declined from 0.52 mg/g to 0.23 mg/g, 1.15 mg/g to 0.52 mg/g, and 0.29 mg/g to 0.20 mg/g in LPS+W-7 treated hairy roots (Figure 6B–D). Notably, the separate W-7 treatment resulted in extreme inhibition of these four tanshinones, especially CT and TI (Figure 6B–E). Taken together, CaM-mediated signaling is essential for LPS-induced tanshinone accumulation in S. miltiorrhiza hairy roots.

2.6. LPS Induces the Expression of Key Tanshinone Biosynthesis Genes in a Ca²⁺-Dependent Manner

To further investigate the role of Ca²⁺ signaling in tanshinone biosynthesis, the expression levels of key genes in LPS and LaCl₃ treated WT hairy roots were analyzed by qRT-PCR. When synergistically treated by LaCl₃ and LPS for 6 h, the expression levels of SmCPS and SmCYP76AH1 reduced approximately 40-fold and 37-fold, respectively, compared to the LPS separate treatment (Figure 7A,B). However, SmHDR and SmDXS did not show apparent reduction by LaCl₃+LPS treatment (Figure 7C,D). Comparatively, LaCl₃ preferentially inhibits the downstream genes (SmCPS and SmCYP76AH1) in tanshinone biosynthesis pathways. This suggests that the downstream genes of the tanshinone biosynthesis pathway are more likely to be regulated by Ca²⁺ signaling than the MEP pathway genes.

Therefore, we present the mechanism of LPS-induced tanshinone biosynthesis in Figure 8. Firstly, LPS induces the generation of Ca²⁺ signaling in the cytoplasm, which is accordingly decoded by the Ca²⁺-dependent regulators. Then, SmWRKY1 and SmWRKY2 are upregulated and activated by some undefined Ca²⁺-dependent regulators. Eventually, the key biosynthesis genes of tanshinones such as SmCPS, SmDXS, SmDXR and SmCYP76AH1 are transcribed in a high level that in turn synthesizes the tanshinones in the hairy roots of S. miltiorrhiza.

3. Discussion

The dry roots of S. miltiorrhiza (Danshen) have been used in Traditional Chinese Medicine (TCM) since 200–300AD [2]. Because of slow growth and a low content of bioactive components, the wild resources of S. miltiorrhiza cannot meet the growing requirements from pharmaceutical markets. Therefore, improving the content of pharmacological ingredients is the main purpose of metabolic research. Up to now, many approaches have been applied to enhance the content of phenolic acids and tanshinones in S. miltiorrhiza [21]. In this study, the bacterial component lipopolysaccharide (LPS) was utilized as a novel elicitor to induce the wild type hairy roots of S. miltiorrhiza. According to the biosynthesis pathway of tanshinones, cryptotanshinone (CT) is the first tanshinone to be generated, and then tanshinone IIA (TIIA), tanshinone IIB (TIIB), tanshinone I (TI), and dihydrotanshinone I (DTI) [3,22]. We have found that LPS significantly enhances the accumulation of tanshinones CT and DTI. Furthermore, the gene expression analysis has shown that key genes from the MVA pathway (SmAACT, SmHMGS), the MEP pathway (SmDXS, smHDR) and the downstream biosynthesis pathway (SmCPS, SmKSL, SmCYP76AH1) respond to LPS treatment in a time-dependent manner. These results demonstrate that LPS is capable of activating key genes’ expression in the tanshinone biosynthesis process. It is worth noting that LPS does not obviously inhibit the growth of hairy roots. Thus, LPS can be applied as a positive elicitor to enhance the content of tanshinones without affecting the growth of the S. miltiorrhiza hairy roots. This is valuable for increasing the content of metabolites.

In S. miltiorrhiza, researchers have promoted the content of tanshinones via pathway engineering such as SmGGPPS-SmDXS2 [23], SmHMGR-SmDXR [24]; or overexpression of key transcription factors including SmMYB98 [25], SmMYB36 [26], SmWRKY1 [17], SmWRKY2 [5], and SmbHLH3 [6]. Previous studies have confirmed that SmDXS2 and SmCPS are the target genes of SmWRKY2 [5], and SmDXR is the target of SmWRKY1 [17]. Our results have further shown that SmWRKY1 and SmWRKY2 were upregulated by LPS in the same manner as SmCPS, SmDXS2, SmCYP76AH1, etc. Therefore, we deduce that LPS promotes tanshinone biosynthesis through activating the SmWRKY1,2-SmCPS, SmDXS2, SmDXR pathway. In plants, WRKY transcription factors act as a key regulator in response to diverse abiotic and biotic stresses [18,27], including PAMP triggered immunity (PTI) [28,29]. WRKYs can interact with Ca²⁺-related regulators to regulate immune reactions, such as CPK and calmodulin [30,31,32,33]. These suggest that tanshinone metabolism is closely related with Ca²⁺ signaling and might be regulated in a similar way by immunity responding reactions.

The bacterial component LPS is an immune activator. It is capable of inducing cytoplasm Ca²⁺ elevation, which is essential for the plant innate immune response [15,34]. To analyze the role of Ca²⁺ signaling in tanshinone biosynthesis, S. miltiorrhiza hairy roots were collaboratively treated by LPS and Ca²⁺ inhibitors. Both Ca²⁺ channel blocker LaCl₃ and CaM antagonist W-7 can significantly inhibit the accumulation of tanshinones. Further analysis has shown that the downstream biosynthetic genes (SmCPS, SmCYP76AH1) are presumably regulated by Ca²⁺ signaling in priority. Based on these data, we present the pathway of LPS-induced tanshinone biosynthesis as Ca²⁺ signal-Ca²⁺-dependent regulators-SmWRKY1,2-downstream genes axis in S. miltiorrhiza. Our study provides a new insight into the essential role of Ca²⁺ signaling in tanshinone biosynthesis. However, the exact mechanism of how SmWRKY1 and SmWRKY2 are modulated by Ca²⁺-dependent regulators remains unresolved. In the future, searching for the Ca²⁺-dependent master regulators, which are capable of activating SmWRKY1 and SmWRKY2, might be the key to uncovering the mechanism of Ca²⁺-mediated tanshinone biosynthesis in S. miltiorrhiza. For the purpose of promoting the content of valuable metabolites, the Ca²⁺ transduction pathway might be the potential regulation target.

Based on our findings, the LPS-induced Ca²⁺ signal is highly associated with ion influx sourced from apoplast. In plant tissues, the arabinogalactan proteins (AGPs), negatively charged and anchored to the extracellular side of the plasma membrane, can reversibly bind with Ca²⁺ and hypothetically serve as the calcium capacitor [35]. Triggered by a low pH related to plasma membrane (PM) H⁺-ATPases, the AGPs-Ca²⁺ complex can release free Ca²⁺ into the cell-surface apoplast and in turn lead to [Ca²⁺]_cyt signal generation [35]. In recent in-depth research, knockouts of the key β-glucuronosyltransferases (GlcATs), which are responsible for adding glucuronic acid (GlcA) to AGPs, resulted in reduced AGPs glucuronidation, impaired Ca²⁺ signaling and consequent deficient plant development [36,37]. This AGPs-Ca²⁺ interaction model highlights the crucial role of the proton pump in modulating Ca²⁺ signaling. The post-translational regulation, especially phosphorylation, is central to alternating PM H⁺-ATPases between the auto-inhibited state and active state [38]. For instance, fusicoccin, the secreta of fungi Fusicoccum amygdali, is able to activate plant PM H⁺-ATPases by increasing the phosphorylation level [38,39]. Notably, LPS-induced phosphorylation of key proteins such as AMPK [40], p53 [41] and mTOR [42], has been proved by massive studies in animals. In plants, LPS might similarly regulate phosphorylation of crucial proteins and might be a potential activator of PM H⁺-ATPases. Thus, we further hypothesize that LPS induces Ca²⁺ influx via the regulation of PM H⁺-ATPases. The phosphorylation modification of PM H⁺-ATPases could be the key to uncover LPS-generated Ca²⁺ influx in S. miltiorrhiza.

In addition, in comparison with the elaborated studies in animals, LPS-regulated pathways in plants are still elusive. Up to date, several proteins including AtLBR1,2 (LPS binding protein) and OsCERK1 (LysM-type receptor-like kinase) have been determined as the key players in LPS-induced immune responses [43,44]. However, the potential correlations between these LPS-related regulators and secondary metabolism have not been deeply elucidated. Consequently, the regulatory network of LPS in plants still needs to be illuminated by more comprehensive research in the future.

4. Materials and Methods

4.1. Reagents

Lipopolysaccharides (L9143) and LaCl₃ (449830) were from Sigma-Aldrich (St. Louis, MO, USA) and were dissolved in sterile water. W-7 (N-(6-Aminohexyl)-5-chloro-1-naphthalenesulfonamide Hydrochloride) (N136431) was from Aladdin (Shanghai, China) and was dissolved in DMSO. The SteadyPure Plant RNA Extraction Kit (AG21019), the Evo M-MLV RT Kit with gDNA Clean for qPCR (AG11601), and the SYBR^® Green Premix Pro Taq HS qPCR Kit (AG11701) were from Accurate Biotechnology(Changsha, China). Acetonitrile and methyl alcohol for HPLC analysis were from TEDIA (Fairfield, OH, USA). The standards of DT, CT, TI and TIIA were from Herbpurify (Chengdu, China).

4.2. Hairy Roots Culture and Treatment

The S. miltiorrhiza wild type (WT) hairy roots were generated by Agrobacterium rhizogenes (ATCC15834). The generation and culture of hairy roots were based on previous research [45]. Before analysis, hairy roots weighing 0.3 g were cultured in 50 mL 6, 7-V liquid medium [46] containing an amount of 30 g/L sucrose for 21 days at 25 °C. LPS and other reagents were added into the culturing medium on the 10th day, and then the hairy roots were harvested on the 21st day. The hairy roots were dried at 45 °C for 4 days before HPLC analysis.

4.3. Reverse Transcription and Quantitative Real-Time PCR Analysis

To investigate the expression of key biosynthesis genes, the WT hairy roots were treated by 50 μg/mL LPS in a time gradient, and the 0 h treatment was used as a control. The total RNAs of the control and LPS-treated hairy roots were extracted by a SteadyPure Plant RNA Extraction Kit (AG21019). Total RNA (1 μg) was reversely transcribed by an Evo M-MLV RT Kit (with gDNA Clean) (AG11601). The gene expression was analyzed by a SYBR^® Green Premix Pro Taq HS qPCR Kit (AG11701). qRT-PCR was conducted on a real-time PCR system (Bio-RAD CFX96, Hercules, CA, USA). SmACT was used as a reference gene. The relative expression level of a gene was calculated by the 2^−ΔΔCt method. The gene’s expression level of 0 h was set to 1. Gene-specific primers were shown in the Appendix A

Table A1. Primers for qRT-PCR.

Primer	Sequence (5′–3′)
SmACT-F	GGTGCCCTGAGGTCCTGTT
SmACT-R	AGGAACCACCGATCCAGACA
SmAACT1-F	TGAAGGACGGACTCTGGGATGT
SmAACT1-R	CCTTGTCAACAATGGTGGATGG
SmCPS1-F	CCACATCGCCTTCAGGGAAGAAAT
SmCPS1-R	TTTATGCTCGATTTCGCTGCGATCT
SmCYP76AH1-F	ACGCATCACTTCACCCATCTCA
SmCYP76AH1-R	ATTGCCGACTCATCCACGAT
SmDXS2-F	CTCACGGTCGCATTGCATCAT
SmDXS2-R	CGCTTTCGTCTCGTTTAGGGA
SmHDR1-F	GGATTTGACCCGGACAAGGAT
SmHDR1-R	CCGCCAATGACTAGGATGAGA
SmHMGS1-F	TTAGGGCGAATCACATGGCTCA
SmHMGS1-R	TCGGCATCCAAGATCGAGAAC
SmKSL1-F	TGGAAACAGTGTGACCCTTCTGCT
SmKSL1-R	GCTTGCATACAAATAACACCCAATCCT
SmWRKY1-F	ACCTACAACGGCCAACACACT
SmWRKY1-R	TCGTCCGGTGTTTTCATTTG
SmWRKY2-F	ACTCATCCAAGCTGTCCGGT
SmWRKY2-R	ATTCATTGTTCCGTTTGAGCC

.

To investigate the regulation of Ca²⁺ signaling in gene expression, the WT hairy roots were treated by 50 μg/mL LPS and 50 μg/mL LPS + 1 mmol/L LaCl₃ for 6 h, and H₂O was used as a control. The total RNAs of the control and the different treated hairy roots were extracted, reversely transcribed and analyzed by qRT-PCR, as mentioned above. The gene’s expression level of control at 0 h was set to 1.

4.4. HPLC Analysis

The dried hairy roots powder (0.02 g) was extracted by 70% methyl alcohol (4 mL) overnight and treated by ultrasonic for 45 min. Then, the mixture was centrifuged at 10,000 rpm for 10 min. The supernatant was filtered through a 0.45 µm membrane before HPLC analysis. The content of DTI, CT, TI and TIIA was determined by the Waters 1525/2489 HPLC system (Milford, MA, USA) equipped with an InertSustain^® C18 column (5 um, 250 mm × 4.6 mm, SHIMADZU-GL, Tokyo, Japan).

The HPLC operation software was Empower 2. The detection wavelength for tanshinones was 270 nm. Elution gradients are as follows (A: acetonitrile; B: 0.02% phosphoric acid solution): 0–10 min, 5–20% A; 10–15 min, 20–22% A; 15–20 min, 22–25% A; 20–28 min, 25–30% A; 28–40 min, 30–35% A; 40–45 min, 35–45% A; 45–50 min, 45–50% A; 50–58 min, 50–58% A; 58–67 min, 58–50% A; 67–70 min, 50–60% A; 70–80 min, 60–70% A; 80–85 min, 70–100% A; 85–95 min, 100–5% A. The column temperature was set at 30 °C. The flow rate was set as 1 mL/min.

4.5. Statistical Analysis

All experiments were conducted over three times. The results were described as the mean ± standard deviation (SD). The significant differences between different groups were calculated by the Student’s t-test. (*) indicates a significant difference (0.01 < p < 0.05). (**) indicates a very significant difference (p ≤ 0.01).

5. Conclusions

On the basis of the data in this study, we proposed the model of LPS-enhanced tanshinone biosynthesis in S. miltiorrhiza. Firstly, LPS induces a cytoplasmic Ca²⁺ signal which consequently activates the expression of the transcription factors SmWRKY1 and SmWRKY2 via some undefined Ca²⁺ sensors. Then, the key biosynthesis genes of tanshinones are upregulated by SmWRKY1 and SmWRKY2. To summarize, the Ca²⁺ signal-Ca²⁺-dependent regulators-SmWRKY1,2-downstream genes axis might be central to regulate tanshinone biosynthesis in S. miltiorrhiza.