Introduction

Plant cell culture is an alternative to whole plant extraction for obtaining valuable secondary metabolites. Molecular diversity is a widely existing phenomenon in plant secondary metabolites (Sticher 1998), and intentional manipulation of the heterogeneity of secondary metabolites in cell cultures is a very important issue (Wang and Zhong 2002). Exogenously applied methyl jasmonate (MJA) could enhance production of secondary metabolites by a variety of plant species (Dong and Zhong 2001; Yu et al. 2002; Tabata 2004). Changes of the molecular diversity of ginsenosides in ginseng cell and root cultures were also observed (Wang et al. 2005; Yu et al. 2005). Recently, Qian et al. (2004a,b) reported that the hydroxylation of MJA at its C-1 position resulted in a higher stimulatory activity on taxane biosynthesis in cell cultures of Taxus chinensis. However, it remains unknown whether the newly synthesized jasmonates (Qian et al. 2004a,b) could change the heterogeneity of secondary metabolites as much as, or even more than, MJA.

Elicitation could induce the formation of defense compounds in plant cell cultures. This induced synthesis requires a series of signal molecules transmitting the message between the elicitor plant cell wall receptor and gene activation. The octadecanoid pathway leading to the synthesis of jasmonic acid (JA) has been identified as an integral part of the signaling cascades triggered by elicitation (Creelman and Mullet 1997). The oxidative burst, which corresponds to a rapid and transient production of active oxygen species (AOS) such as H2O2, is a typical early event in plant defense responses (e.g., Chen et al. 1993; Hu et al. 2003b; Zhao and Sakai 2003). It was reported that pathogens or elicitors could induce a rapid production of H2O2 and finally act on gene expression that caused programmed cell death and the activation of l-phenylalanine ammonia lyase (PAL) (Desikan et al. 1998). The rapid production of H2O2 could induce jasmonate signals and finally act on the biosynthesis of various secondary metabolites including ginsenosides (Hu et al. 2003a; Zhao et al. 2005). In other cases, JA induced by wounding could act on early signaling genes and the rapid production of H2O2, and H2O2 then induced the expression of late defense genes, but NO in this signal cascade inhibited the defense genes expression by inhibiting H2O2 accumulation (Orozco-Cárdenas and Ryan 2002). Qian et al. (2004a) investigated the response of H2O2 to 2,3-dihydroxypropyl jasmonate, a chemically synthesized elicitor, in cell cultures of T. chinensis. However, no detailed information is available about the studies on signal cascade for such synthetic elicitors, not to mention for the case of cell cultures of ginseng species.

Panax notoginseng (Sanchi-ginseng) is one of the most valuable traditional Chinese medicinal herbs. Cell culture of P. notoginseng is a promising technology to obtain ginsenoside, one of its major bioactive secondary metabolites (Woragidbumrunge et al. 2001; Zhang and Zhong 2004). Because different ginsenosides have different or even opposite pharmacological activities, engineering approach for manipulation of ginsenoside heterogeneity in cell cultures has a significant impact on practical application (Wang and Zhong 2002). We have demonstrated that an addition of 200 μM MJA could increase the content of individual ginsenosides and change their heterogeneity in cell cultures of P. notoginseng for both shake flask and bioreactor cultivations (Wang and Zhong 2002; Wang et al. 2005).

The present work investigates the effects of self-synthesized hydroxyl-containing jasmonate elicitor, 2-hydroxyethyl jasmonate (HEJA), on the biosynthesis and heterogeneity of ginsenosides in cell cultures of P. notoginseng. To understand the signal transduction from jasmonate elicitation to ginsenoside biosynthesis, early signal events in plant defense response (including oxidative burst and JA biosynthesis) were also studied together with the investigation on the activity of UDPG-ginsenoside Rd glucosyltransferase (UGRdGT), a new and important enzyme in the ginsenoside biosynthetic pathway (Yue and Zhong 2005a,b). In all experiments, MJA was also used for comparison and discussion.

Materials and methods

Chemicals

MJA was purchased from Tokyo Kasei Co. Ltd. (Tokyo, Japan). Silica gel (200–400 mesh, 60 Å, for column chromatography) was obtained from Aldrich. Organic solvents were obtained from commercial suppliers and were of the highest purity available; they were dried over 3Å molecular sieves for at least 48 h prior to use. Reagents for synthesis of HEJA were purchased from Sigma and used without further purification.

Chemical synthesis of HEJA

HEJA was synthesized and purified as described earlier (Qian et al. 2004b), and both MJA and HEJA contain the same ratio of stereoisomers (Qian et al. 2004b).

Cell subcultures

Suspension cells of P. notoginseng were grown in Murashige and Skoog (MS) medium and subcultured every 2 weeks. The details were described elsewhere (Woragidbumrunge et al. 2001; Wang and Zhong 2002).

Cultivation conditions

P. notoginseng cells (2 g of fresh weight) were incubated in a 250-ml Erlenmeyer flask containing 50-ml medium with the same culture conditions as in subcultures. MJA, HEJA, or JA biosynthesis inhibitors was dissolved in ethanol and sterilized by filtering through 0.22 μm polyvinylidenedifluoride (PVDF) syringe filters (Millipore) before adding to the cell cultures. The final ethanol concentration in cultures was 1 ml/l. Equal volumes of ethanol were added to all flask cultures. For all cultures, multiple flasks were run under each condition, and the cultivation data represent the mean values with the standard deviations from three independent samples.

Measurement of cell weight and ginsenoside content

For sampling, three identical shake flasks were used for each data point. The samples from flasks were filtered under vacuum and washed with several volumes of distilled water to remove residual medium. The analytical procedures for cell dry weight (DW) and ginsenoside content were the same as described previously (Zhang and Zhong 2004; Wang et al. 2005).

Enzyme extraction and assay of UGRdGT activity

UGRdGT, which catalyzes the formation of Rb1 from Rd in P. notoginseng cells (Yue and Zhong 2005a,b), was detected as described in Wang et al. (2005).

Analysis of H2O2

H2O2 produced by the cells and released into the medium was determined by the scopoletin fluorescence oxidative quenching method (excitation wavelength, 350 nm; emission, 460 nm) according to Cazalé et al. (1998). To measure the H2O2 concentration, samples were taken at various intervals after the elicitation. Aliquots of 3 ml cell-free broth were mixed with 90 μl 1 mM stock solution of scopoletin (Fluka) in DMSO (Sigma) and 90 μl 1 mg/ml stock solution of peroxidase (Sino-American Biotechnology Co., Shanghai, China), respectively. Scopoletin was oxidized, and the concentration of H2O2 was calculated from the fluorescence decrease monitored by a spectrofluorimeter (Varian Cary Eclipse, CA, USA) using a calibration curve established in the presence of H2O2. A standard curve by adding scopoletin to the solutions at different H2O2 concentrations was prepared by using cell-free medium.

The effects of MJA and HEJA on peroxidase-dependent assay for H2O2 determination were tested. Various chemicals were added to cell-free medium to obtain final concentrations of 50, 100, or 200 μM, respectively. In usual conditions of assay (see above), the addition of jasmonates had no obvious effect on the decrease of scopoletin fluorescence due to the addition of H2O2.

Assay of NO

NO in the medium was quantified by spectrophotometric measurement of the conversion of oxyhemoglobin (HbO2) to methemoglobin (metHb) (Murphy and Noack 1994; Delledonne et al. 2001). HbO2 was prepared according to Murphy and Noack (1994). Twenty-five milligrams of hemoglobin (Hb) (Sigma) was dissolved in 1 ml of phosphate buffer (50 mM, pH 7.4), and 1–2 mg of sodium hydrosulfite powder was added to the solution. Then, a light stream of O2 was continuously blown to the solution for about 15 min, and Hb was converted into HbO2 by reacting with O2. The resulting HbO2 solution is desalted and purified by passing it through a Sephadex G-25 (Pharmacia) column. HbO2 should be prepared fresh each day and stored on ice in dim light. NO concentration in the medium was analyzed as describe by Delledonne et al. (2001). Briefly, cell suspensions were filtrated, and HbO2 was added to the medium to a final concentration of 10 μM. After 2 min, the changes in absorbance of the medium at 421 nm and 401 nm were measured, and the NO levels were calculated by using an extinction coefficient of 77 mM−1 cm−1 [A401 (metHb)–A421 (HbO2)].

Enzyme extraction and assay of PAL activity

Fresh cells from the suspension culture were extracted for the assay of PAL as described by Qian et al. (2004b). Fresh cells of 1 g were frozen in liquid nitrogen and grounded with mortar and pestle. Enzyme in frozen powder was extracted by adding 50 mg polyvinylpolypyrroridone, 2-ml prechilled buffer of pH 7.2 (0.1 M phosphate buffer, 2 mM ethylenediaminetetraacetic acid, 4 mM dithiothreitol), and then homogenized at 4°C. The mixture was centrifuged at 10,000×g for 30 min at 4°C. The supernatant was used directly for PAL assay, using a method slightly modified from the one described by Heide et al. (1989). The enzyme extracts (200 μl) were incubated with 120 μl 0.1M l-phenylalanine (dissolved in 0.1 M borate buffer of pH 8.8) and 280 μl 0.1 M borate buffer of pH 8.8 at 30°C for 60 min. The reaction was stopped by adding 50 μl 5 N trichloroacetic acid. After centrifugation at 10,000×g for 3 min, 20 μl supernatant was analyzed by HPLC, using a Shimadzu LC-10ATVP HPLC apparatus equipped with a variable wavelength UV detector (Shimadzu, SPD-10AVP). A Shimadzu VP-ODS column (250×4.0 mm2; 5 μm) was used at 25°C. The mobile phase consisted of water, methanol, and acetic acid at a ratio of 40:60:1 (v/v/v). The flow rate was kept constant at 1 ml/min. Trans-cinnamic acid was monitored at 275 nm identified by comparison with its authentic sample (Sigma). One unit (U) of enzyme activity is defined as the amount of enzyme forming 1 pmol of trans-cinnamic acid from the substrate l-phenylalanine per minute. Protein was quantified with Bradford method.

Detection of JA

The cell suspensions were filtrated, and 10 g cells by fresh weight (FW) were extracted for the assay of JA using a method slightly modified from that of Gundlach et al. (1992). For the assay of JA, 10 g FW cells were shock-frozen in liquid N2, thawed in 20 ml of ethanol, and to which 9,10-dihydrojasmonic acid was added as internal standard. Samples were grinded, and the mixture was separated and extracted a second time with 20 ml ethanol. The ethanolic extracts were evaporated at 40°C to dryness, and the residue was dissolved in 40 ml water. The solution was acidified with 2.4 ml of 12 M HCl and was extracted with 40 ml CHCl3 for three times. The organic phase was dried over Na2SO4 and evaporated. The residue was dissolved in 1 ml of n-hexane and was applied to a silica solid-phase extraction column (Supelclean LC-Si SPE Tubes, 500 mg, 3 ml, Supelco, USA). The column was washed with 5 ml of n-hexane and eluted with 7 ml of n-hexane/diethyl ether, 2:1 (vol/vol). The sample was taken to dryness, dissolved in 100 μl methanol for analysis. JA was analyzed by gas chromatography/mass spectrometry (GC/MS) on a Micromass GCT unit (England), with an HP-5 fused silica capillary column of 30 m×0.32 mm ID and 0.25-μm film thickness. The column temperature was initially held at 80°C for 0.5 min, then shifted to 200°C at 4°C/min, and then increased to 280°C at 30°C/min, and the injector and detector temperature was set at 270 and 280°C, respectively. Hydrogen was used as the mobile phase at a flow rate of 1 ml/min.

Statistical analyses

Experiments were done in a completely randomized layout. Each experiment was carried out in triplicate, and three different sets of experiments reproduced the same result. Data were analyzed by one-way analysis of variance (ANOVA). Means of an experiment were analyzed using Tukey's Honestly Significant Difference multiple-comparison test with a family error rate of 0.05. All differences are significant unless otherwise indicated.

Results

Optimization of HEJA elicitation conditions

Exposure time and dosage of an elicitor are two main factors that affect cell growth and product yield for a specific culture system and elicitor (Wang and Zhong 2002; Ketchum et al. 1999; Tabata 2004). Thus, experiments on the timing of HEJA addition and dose response were carried out at first.

For the investigation of elicitation time, 100 μM HEJA was added to the cultures of P. notoginseng cells at different growth stages, i.e., at day 0 (lag phase), day 4 (beginning of log phase), day 7 (middle of log phase), and day 10 (end of log phase). Maximum DW and ginsenoside content on day 13 are shown in Table 1. DW on day 13 for the elicited cultures exposed at different log phases (day 4, 7, and 10) was almost the same as that of the control (without addition of elicitors). Whereas, cell growth was highly inhibited when HEJA was added just after inoculation, and there appeared to be about 20% decrease in DW on day 13 compared with the control. HEJA could increase each individual ginsenoside biosynthesis after its addition to the cell cultures at various growth stages. On the other hand, the cells showed different stimulatory responses of ginsenoside biosynthesis to elicitation at different growth stages. As shown in Table 1, relatively higher individual ginsenoside content was obtained with the cells exposed to the elicitor at early stage of cultivation (day 0 and 4). On the contrary, when HEJA was added on day 7 and 10, the stimulating effect on ginsenoside content was a little lower compared with the elicitation on day 0 and 4. With HEJA elicitation, the ginsenoside content of Rb group increased much more than Rg group compared to the control. In particular, Rb1 content increased much more than Rg1 and Re, whereas Rd was also detected in this case. Similar to MJA (Wang and Zhong 2002), HEJA could manipulate the ginsenoside heterogeneity in cell cultures of P. notoginseng.

Table 1 Effects of HEJA addition time on DW and the content of individual ginsenosides on day 13
Full size table

For the study on elicitation dosage, 4-day-old cell suspension cultures of P. notoginseng were elicited with different levels (100–500 μM) of HEJA. Maximum DW and ginsenoside content on day 13 are shown in Table 2. A lower HEJA concentration seemed to have no obvious effect on DW. The respective DW for 100 and 200 μM of HEJA on day 13 was 8.88±0.20 and 8.71±0.21 g/l, nearly the same as the control (9.11±0.18 g/l). Whereas, a higher HEJA concentration (i.e., 500 μM) resulted in a decreased DW (7.62±0.25 g/l) compared with the control. An increase of HEJA concentration from 100 to 200 μM could enhance the content of individual ginsenosides, for example, Rb1 content was increased from 1.86±0.05 to 2.31±0.05 mg/100 mg DW. However, an even higher HEJA concentration (i.e., 500 μM) showed a slight inhibition on ginsenoside content, and the optimum HEJA elicitation condition was identified to be 200 μM added on day 4.

Table 2 Effects of HEJA concentration on DW and the content of individual ginsenosides on day 13
Full size table

Comparison of HEJA with MJA

A comparison of the effects of MJA and HEJA on cell growth, ginsenoside biosynthesis, and UGRdGT activity was done. Both MJA and HEJA were added on day 4 to the cell cultures at 200 μM.

Time profiles of DW (a), total ginsenoside (Rg1+Re+Rb1+Rd) content (b), and the ratio of Rb to Rg group of ginsenosides (c) are shown in Fig. 1. Compared with the control, no obvious difference on cell growth was observed for both MJA and HEJA elicitation. The cells all reached their maximum DW on day 13, and the respective DW for control, MJA elicitation, and HEJA elicitation was 9.40±0.31, 9.01±0.37, and 8.94±0.46 g/l, respectively.

Fig. 1
figure 1

Time courses of dry cell weight (a), total ginsenoside (Rg1+Re+Rb1+Rd) content (b), and the ratio of Rb to Rg (c) of P. notoginseng cells with 200 μM of MJA or HEJA added on day 4. Control (dark circle), 200 μM MJA added on day 4 (open triangle), 200 μM HEJA added on day 4 (dark triangle)

Full size image

From Fig. 1b, we could see that for both MJA and HEJA elicitation, the content of total ginsenosides increased from day 4 to about day 11 after the addition, and then a slight fluctuation was observed. The stimulating effect of HEJA on ginsenoside biosynthesis was at a higher level than MJA during cultivation. Maximum ginsenoside content was 0.76±0.06 (day 11), 2.26±0.09 (day 15), and 3.54±0.23 (day 15) mg/100 mg DW for the control, MJA elicitation, and HEJA elicitation, respectively. The Rb/Rg ratio (Fig. 1c) of the control maintained at about 0.5 during the cultivation, whereas it increased sharply with MJA or HEJA elicitation. During the cultivation, the Rb/Rg ratio with HEJA elicitation was about 30% higher than that with MJA elicitation.

Time profiles of UGRdGT activity (a) and the content of ginsenoside Rb1 (b), and Rd (c) with addition of MJA and HEJA are shown in Fig. 2. This enzyme catalyzed the glucosylation of ginsenoside Rd into Rb1 in P. notoginseng cells (Yue and Zhong 2005a,b). The enzyme activity increased sharply after the addition of the elicitors on day 4 and reached a maximal level on day 7. Although a little decrease was seen after day 7, the enzyme activity still remained at a relatively higher level until day 15 compared with the control. In comparison to MJA, for HEJA, the increase of UGRdGT activity coincided with the higher content of ginsenoside Rb1 in its case. The maximum production of individual ginsenosides on day 13 is shown in Table 3. With HEJA elicitation, a maximal production titer of ginsenoside Rg1, Re, Rb1, and Rd was 47.4±4.8, 52.3±4.4, 190±18, and 12.1±2.5 mg/l, about 1.3-, 1.3-, 1.7-, and 2.1-fold than that with MJA elicitation, respectively.

Fig. 2
figure 2

Dynamic changes of UGRdGT activity (a) and the content of ginsenoside Rb1 (b) and Rd (c) of P. notoginseng cells with 200 μM of MJA or HEJA elicited on day 4. The symbols are the same as in Fig. 1

Full size image
Table 3 Effects of MJA and HEJA on the production of individual ginsenosides on day 13
Full size table

Effects of MJA and HEJA on oxidative burst in P. notoginseng cell cultures

Effect of the jasmonates (MJA and HEJA) on oxidative burst was studied by analyzing H2O2 concentration in the medium with 200 μM of the jasmonates added to the P. notoginseng cell cultures 4 days after inoculation. Both NO concentration in medium and PAL activity in cells related to oxidative burst were measured. Both MJA and HEJA had no obvious effects on H2O2 and NO levels in medium and PAL activity in cells until 24 h after elicitation (Fig. 3a–c).

Fig. 3
figure 3

Effect of 200 μM of MJA or HEJA on H2O2 (a) and NO (b) concentration in medium and PAL activity in cells (c) of P. notoginseng. The symbols are the same as in Fig. 1

Full size image

Effects of MJA and HEJA on JA biosynthesis

Three JA biosynthesis inhibitors, i.e., ibuprofen (Ibu), n-propylgallate (Pro), and phenylbutazone (Phyb) (Staswick et al. 1991; Farmer 1994), were tested for their possible effects on the jasmonate signaling pathway that led to enhanced ginsenoside biosynthesis (Table 4). Each inhibitor (at 100 μM) was added together with 200 μM of the jasmonates to the cell cultures of P. notoginseng on day 4, and the cells were sampled on day 13. Here, 100 μM of each inhibitor added alone on day 4 was proved to have no effects on cell growth and ginsenoside content (data not shown).

Table 4 Effects of JA biosynthesis inhibitors on the eliciting activities of MJA and HEJA in ginsenosides biosynthesis (day 13)
Full size table

When MJA was added simultaneously with the inhibitors, all three inhibitors (Ibu, Pro, and Phyb) did not affect MJA-mediated enhancement of ginsenoside biosynthesis, and the content of each ginsenoside was very similar to that of MJA-elicited cells. On the other hand, they could partially reduce the eliciting activity of HEJA on ginsenoside biosynthesis when added together with HEJA. For example, when Ibu was added with HEJA, total ginsenoside content was 2.44±0.13 mg/100 mg DW. It was lower than that of HEJA-elicited cells (2.89±0.15 mg/100 mg DW), but still much higher than that of control (0.61±0.06 mg/100 mg DW).

Muller et al. (1993) reported that JA level reached the maximum between 0.5 and 2 h after yeast cell wall elicitation and kept at a relatively high level until 12 h after elicitation. Thus, in this study, the endogenous JA content at 2 and 12 h after the elicitation was analyzed, and the results are shown in Table 5. Both MJA and HEJA significantly increased the endogenous JA content at 12 h after the addition. HEJA addition could increase JA content to a higher level than MJA addition (74.4±7.6 vs 20.8±3.9 ng/g FW). Ibu had no effects on JA content compared with the control, when it was added together with HEJA or MJA, whereas it could inhibit the increase of JA content induced by the jasmonates.

Table 5 JA content in the cells under different cultivation conditions
Full size table

Discussion

The optimal condition for HEJA elicitation was identified to be 200 μM added on day 4. The same phenomenon was observed for the MJA elicitation on P. notoginseng cultures (Wang and Zhong 2002). The inhibitory effect of high concentrations of elicitors on metabolites biosynthesis was also observed in JA-induced indole alkaloids biosynthesis (Rijhwani and Shanks 1998) and in MJA-induced taxol biosynthesis (Ketchum et al. 1999). The existence of an optimal dose of elicitors suggests that, at elicitor doses smaller than the optimum, the elicitor-binding sites in cells were still not fully utilized for activating the secondary metabolite synthesis, whereas excessive doses caused a deleterious effect on the cells' biosynthetic activity.

Stronger elicitation of ginsenoside biosynthesis by HEJA than MJA was demonstrated. It was confirmed that esterification did not change the distribution of jasmonate stereoisomers, as compared with the starting material (MJA) (Qian et al. 2004a,b). Therefore, it might be reasonable to neglect the effect of stereo configuration and take into account the difference in chemical structure. Compared with MJA, HEJA also had a higher stimulating activity on taxane biosynthesis in cell cultures of T. chinensis (Qian et al. 2004a,b). These results suggest that the esterification of the jasmonates at the C-1 position with glycol is essential to their stimulating activity on secondary metabolite biosynthesis.

A higher Rb/Rg ratio with HEJA elicitation was observed than that with MJA elicitation. The results indicate that compared to MJA, HEJA could lead to higher amounts of Rb group ginsenosides. This means that it can alter the distribution of heterogeneous ginsenosides more efficiently than MJA.

The increase of UGRdGT activity coincided with the higher content of ginsenoside Rb1. A higher UGRdGT activity with HEJA elicitation than that with MJA elicitation suggests that HEJA had higher stimulating activity than MJA on ginsenoside biosynthesis at a molecular level. In addition, by using HEJA, the Rd content increased more than Rb1. The fact implies that the jasmonate analogue might also strongly activate certain site(s) from 2,3-oxidosqualene to Rd in ginsenoside biosynthetic pathway (Fig. 4) besides its effect on UGRdGT. Future progress in the elucidation of the detailed ginsenoside biosynthetic pathway will help us to investigate and understand this phenomenon.

Fig. 4
figure 4

Proposed acting point of jasmonates on the biosynthetic pathway of ginsenoside Rb1 in P. notoginseng cells

Full size image

Exogenously applied MJA could induce the lipoxygenase and lead to endogenous JA biosynthesis as reported (Melan et al. 1993; Zhao and Sakai 2003). In tobacco Bright Yellow-2 cells, the exogenously applied MJA was hydrolyzed to JA and then metabolized to its glucose and gentiobiose esters (Swiatek et al. 2004). In this work, we found that the jasmonates could increase the endogenous JA level, and JA biosynthesis inhibitors could inhibit the stimulated JA level. It is possible that the exogenously applied jasmonates could stimulate JA biosynthesis in P. notoginseng cells, or they may be hydrolyzed to JA. JA biosynthesis inhibitors were found to only partially inhibit ginsenoside biosynthesis induced by HEJA, and they had no effects on MJA-induced ginsenoside biosynthesis. The results imply that in P. notoginseng cells, both exogenously applied jasmonates and endogenously elicited JA could induce ginsenoside biosynthesis via the induction of key enzymes involved in the ginsenoside biosynthetic pathway such as UGRdGT.

In this work, no obvious effects of both HEJA and MJA on H2O2, NO, and PAL were observed until 24 h after elicitation, although the UGRdGT activity was already enhanced (Fig. 3). This suggests that the AOS generation pathway might have no relationship with the jasmonates-induced ginsenoside biosynthesis in our P. notoginseng cell cultures. In cell cultures of P. ginseng as reported, oligosaccharide induced jasmonate signal through a rapid production of H2O2 and finally enhanced ginsenosides production, whereas JA itself had no effects on H2O2 (Hu et al. 2003a). Such a signal transduction was also shown in other systems like in β-thujaplicin production induced by a yeast elicitor (Zhao and Sakai 2003). However, different results have also been reported in some other cases. For example, in root cultures of P. ginseng and P. quinquefolium, the induction of antioxidant system and H2O2 accumulation was observed after MJA elicitation (Ali et al. 2005). In tomato leaves, JA or systemin induced early signaling genes and the rapid production of H2O2, and H2O2 then induced the expression of late defense genes, but NO in this signal cascade inhibited the defense genes expression by inhibiting H2O2 accumulation (Orozco-Cárdenas and Ryan 2002). Recently, Wang and Wu (2005) reported that NO inhibitors could suppress MJA-induced taxol accumulation, but enhance H2O2 and PAL. In suspension cultures of T. chinensis, an increased level of H2O2 followed by higher PAL activity and taxoid overproduction was observed after jasmonates elicitation (Qian et al. 2004a). Hu et al. (2003b) reported that NO mediated elicitor-induced saponin synthesis in cell cultures of Panax ginseng. All these reports including this work suggest that the signal transduction pathway from elicitation, gene activation to secondary metabolite synthesis, is a very complicated system. We think that even for the same elicitor, the signal cascade may be different in different cell culture systems. The information obtained here is considered helpful to further understand the signal transduction network in our case.

It is reported that different jasmonate elicitors in chemical structure might induce different levels of plant defense responses (Qian et al. 2004a,b; Tabata 2004; Staniszewska et al. 2003; Miersch et al. 1999; Koda et al. 1991). Lobler and Lee (1998) proposed that the exogenous jasmonate is recognized by a plasma membrane receptor. In this work, we found that the cells performed similar signal responses to the elicitation by MJA and chemically synthesized HEJA. The different levels of ginsenoside accumulation by their elicitation might be due to the structure relationship between the jasmonates and their receptors. However, no such receptors have been reported. Studies on the structural requirements for the stimulating activity of chemical elicitors may help to the illumination of the receptors and the rational design and synthesis of more potent elicitors.

Conclusions

Chemically synthesized HEJA was shown to stimulate ginsenosides biosynthesis and manipulate their heterogeneity more efficiently than MJA by leading to higher amounts of Rb group ginsenosides in cell cultures of P. notoginseng. Studies on the signal events, including JA biosynthesis and oxidative burst, confirmed that the cells had similar defense response to HEJA elicitation as to MJA elicitation. This is the first report on efficient manipulation of the heterogeneity of plant secondary metabolites by using chemically synthesized elicitor. The information obtained here is useful for the large-scale manipulation of the heterogeneity of valuable ginsenosides in plant cell cultures.