- Email: [email protected]
The effect of yeast elicitor on the growth and secondary metabolism of hairy root cultures of Salvia miltiorrhiza
Enzyme and Microbial Technology 28 (2001) 100 –105
www.elsevier.com/locate/enzmictec
The effect of yeast elicitor on the growth and secondary metabolism of hairy root cultures of Salvia miltiorrhiza Hui Chena, Feng Chena,*, Francis C. K. Chiub, Cindy M. Y. Lob a
Department of Botany, The University of Hong Kong, Pokfulam Road, Hong Kong, P. R. China Department of Pharmacy, The Chinese University of Hong Kong, Shatin, New Territories, Hong Kong, P. R. China
b
Received 20 March 2000; received in revised form 31 July 2000; accepted 2 August 2000
Abstract Hairy root cultures of Salvia miltiorrhiza transformed with Agrobacterium rhizogenes ATCC 15834 produced a tiny amount of tanshinones and a constituent level of phenolic acids under normal growth conditions. Upon elicitation with yeast elicitor, the production of both phenolic acids and tanshinones was enhanced. For example, the contents of two phenolic acids, rosmarinic acid and lithospermic acid B were elevated from 1.24% and 2.59% to 2.89% and 2.98% of dry wt, respectively while the intracellular content of cryptotanshinone increased from 0.001% to as much as 0.096% of dry wt. Yeast elicitor also improved the growth of hairy roots (from 3.9 g/l to 7.3 g/l on a dry wt basis). Liquid chromatography-mass spectrometry (LC-MS) was developed for simultaneous detection and identification of phenolic acids and tanshinones in the extracts of S. miltiorrhiza. Rosmarinic acid, lithospermic acid B, cryptotanshinone, tanshinone I, tanshinone IIA and tanshinone IIB were identified by comparison with standards available. Dihydrotanshinone I and methylenetanshiquinone were tentatively identified by the molecular weights and the elution comparable with the literature. An unknown compound with a molecular weight of 280 was found in yeast-elicitor treated hairy root cultures, which was one of the major tanshinones induced. © 2001 Elsevier Science Inc. All rights reserved. Keywords: Hairy roots; LC-MS; Phenolic acids; Salvia miltiorrhiza; Tanshinones; Yeast elicitor
1. Introduction Plants produce a high diversity of natural products or secondary metabolites, many of which have antimicrobial activity in vitro. Some of these compounds are synthesized during normal growth and development while others are absent from healthy plants, accumulating only in response to pathogen attack or stress conditions. These preformed and inducible antimicrobial compounds are called phytoanticipins and phytoalexins, respectively [1,2]. Elicitors are generally defined as molecules that stimulate any defense responses of plants, including the formation of phytoalexins [3]. The application of elicitors to plant cell cultures is not only useful for enhancing the biotechnological productivity of valuable secondary metabolites in fermentation systems but also important to the study of plant-microbe interactions. Salvia miltiorrhiza Bunge (Labiatae) is a famous traditional Chinese medicinal plant that contains two groups of biologically active compounds: caffeic acid-derived phe* Corresponding author. Tel.: ⫹852-2299 0309; fax: ⫹852-2299 0311. E-mail address: [email protected] (F. Chen).
nolic acids and various tanshinones belonging to diterpene quinones [4]. Our previous work demonstrated that phenolic acids and tanshinones may represent preformed and inducible defense compounds, respectively, in S. miltiorrhiza [4,5]. In a crown gall culture of S. miltiorrhiza, a large amount of phenolic acids and a negligible amount of tanshinones were produced under normal growth conditions [6,7]. However, upon elicitation with yeast elicitor, the level of tanshinones was enhanced greatly whereas the constituent level of phenolic acids was reduced [5]. Crown galls and hairy roots are plant tumors obtained after transformation by Agrobacterium tumefaciens and A. rhizogenes, respectively [8]. Hairy root cultures of S. miltiorrhiza have demonstrated the ability to synthesize both tanshinones [9] and phenolic acids [10]. Our crown galls and hairy roots of S. miltiorrhiza showed distinct morphologies. They both contained a comparable level of endogenous indole acetic acid (IAA) but crown galls had a much higher level of endogenous cytokinins than hairy roots (Chen and Chen, unpublished data). Since cytokinins play an important role in modulating plant defense response to environmental stress [11], it is interesting to know the re-
0141-0229/01/$ – see front matter © 2001 Elsevier Science Inc. All rights reserved. PII: S 0 1 4 1 - 0 2 2 9 ( 0 0 ) 0 0 2 8 4 - 2
H. Chen et al. / Enzyme and Microbial Technology 28 (2001) 100 –105
101
sponse of the growth and secondary metabolism of hairy roots of S. miltiorrhiza to the elicitation with yeast elicitor. This paper addresses this question and makes a comparison with that of crown galls. In addition, we have developed a liquid chromatographymass spectrometry (LC-MS) method for simultaneous detection and identification of phenolic acids and tanshinones in hairy root cultures.
2.5. Kinetics study
2. Materials and methods
2.6. Analyses
2.1. Hairy root culture
Cell dry weight determination and HPLC analysis for phenolic acids and tanshinones were carried out as previously described [4].
Experiments were carried out with a hairy root culture of S. miltiorrhiza transformed with Agrobacterium rhizogenes ATCC 15834 as previously described [10]. 2.2. Standards Cryptotanshinone, tanshinone I, tanshinone IIA and tanshinone IIB were obtained from Shanghai Institute of Materia Medica, Academia Sinica, Shanghai, China. Lithospermic acid B was a gift from Drs K. Kamata (Hoshi University, Tokyo, Japan) and Y. M. Xu (Shanghai Institute of Materia Medica, Academia Sinica, Shanghai, China). Rosmarinic acid was purchased from ICN (Costa Mesa, CA, USA). 2.3. Elicitor preparation A carbohydrate fraction isolated from the yeast extract was prepared by ethanol precipitation as described by Hahn and Albersheim [12]. Briefly, 50 g of the yeast extract was dissolved in 250 ml of distilled water. Ethanol was added to 80% (v/v). The precipitate was allowed to settle for 4 days at 6°C and the supernatant was decanted and discarded. The gummy precipitate was dissolved in 250 ml of distilled water. The ethanol precipitation was repeated. The second ethanol precipitate was dissolved in 200 ml of distilled water, yielding the crude preparation that was used without further purification. 2.4. Elicitor dosage response Hairy roots were subcultured with an inoculum of ca. 0.3 g fresh root segments in 50 ml of hormone-free liquid 6,7-V medium [13] containing 20 g/l sucrose in 250-ml shake flasks. Cultivation was performed on an orbital shaker at 140 rpm in darkness at 25°C. Twenty days after cultivation, hairy roots were transferred to fresh 6,7-V medium containing different volumes of yeast elicitor solution prepared above or 1 ml of distilled water (control). On day 7 of elicitation, hairy root cultures were harvested and the biomass and contents of phenolic acids and tanshinones were determined. The experiment was performed in triplicate.
For a time course study, ca. 0.3 g fresh root segments were inoculated into 50 ml of hormone-free liquid 6,7-V medium containing 20 g/l sucrose in 250-ml shake flasks. On day 20 of cultivation, 0.5 ml of elicitor solution or 0.5 ml of distilled water (control), having been sterilized by autoclaving at 121°C for 15 min, was added to each flask. Then, flasks were harvested daily and subsequent analyses were performed.
2.7. LC-MS conditions The HPLC-MS system consists of a Perkin Elmer series 200 micro pump, a Perkin Elmer series 200 autosampler, a Perkin Elmer 785A UV/Vis detector, and a Perkin Elmer SCIEX API-2000 mass spectrometer equipped with a TurbolonSpray interface. The whole system was controlled by the Mass Chrom 1.1 software. Separations were obtained at room temperature on a Waters Puresil 5 m C18 column (4.6 ⫻ 150 mm) with a guard column of the same material. The mobile phase consisted of 0.1% TFA in water (solvent A) and acetonitrile (solvent B). The following gradient procedure was used: 0 –20 min: 20 –70% B; 20 – 40 min: 70% B; 40 – 41 min, 70 –20% B. The absorbance was measured at 280 nm. The LC eluant after the UV-detector was diverted via a splitter such that 0.2 ml/min was delivered into the API-2000 mass spectrometer. The MS parameters were first optimized using representative reference compounds by direct infusion. Subsequent higher heater gas temperature (200°C) and gas flow (50 psi) were added to the LC-MS parameters. Tanshinone I ([M ⫹ H]⫹ 277) was used for positive mode and rosmarinic acid ([M ⫺ H]⫺ 359) for negative mode monitoring of tanshinones and phenolic acids respectively. The crude mixtures were screened by direct infusion to determine maximum scan range for the unknown samples. During LC-MS runs, the mass spectrometer was set up to monitor both positive and negative ions simultaneously using a loop sequence alternating between the positive and negative mode. A scan range of 100 to 400 amu and 100 to 750 amu were used for positive ions and negative ions respectively. A step size of 1 amu and dwell time of 1 ms were used in both modes. 3. Results 3.1. Separation and identification of phenolic acids and tanshinones by LC-MS Using the HPLC conditions developed in the present study, both the phenolic acids and the tanshinones in yeast-
102
H. Chen et al. / Enzyme and Microbial Technology 28 (2001) 100 –105
Table 1 HPLC-MS data on phenolic acids and tanshinones in S. miltiorrhiza hairy root extract Peak number
Retention time (min)
Major species (amu)
1 2 3 4 5 6 7 8 9
6.35 7.03 15.27 19.02 19.94 22.05 22.89 24.39 27.03
358.9 716.8 311.2 278.9 280.8 297.1 276.9 279.0 294.8
elicitor treated hairy roots were successfully separated and identified simultaneously (Table 1). Under the pH condition of the mobile phase (ca. 0.1% TFA), all of the tanshinones could be detected by the positive mode. The phenolic acids were found to give cleaner spectra with high sensitivity. Thus, the negative mode was used to monitor phenolic acids, while the positive mode was used for tanshinones. The identification and retention times of the major compounds are summarized in Table 1. Rosmarinic acid, lithospermic acid B, cryptotanshinone, tanshinone I, tanshinone IIA and tanshinone IIB were identified by comparison with standards available. Dihydrotanshinone I and methylenetanshiquinone were tentatively identified by the molecular weights and the elution comparable with the literature [14, 15]. The unknown compound (peak 5, retention time 19.94, MW 280) was probably the most abundant tanshinone induced. The further purification and identification of this compound is in progress in our laboratories.
Scan mode
MW
Identification
[M⫺H]⫹ [M⫺H]⫹ [M⫹H]⫹ [M⫹H]⫹ [M⫹H]⫹ [M⫹H]⫹ [M⫹H]⫹ [M⫹H]⫹ [M⫹H]⫹
360 718 310 278 280 296 276 278 294
rosmarinic acid lithospermic acid B tanshinone IIB dihydrotanshinone I unknown cryptotanshinone tanshinone I methylenetanshiquinone tanshinone IIA
2B). Among the known tanshinones, cryptotanshinone was most notable in its increase owing to elicitation as it was almost not produced in non-elicited hairy roots (Figs. 2C and 2D). 3.4. Time course of elicitation Time course of the effect of yeast elicitor on root growth and phenolic acid and tanshinone formation is shown in Fig.
3.2. Comparison of the HPLC profiles of the methanol extracts from hairy roots with and without elicitation As a typical chromatogram (Fig. 1) shows, the nonelicited hairy root contained a large amount of phenolic acids (peaks 1 and 2) and a negligible amount of tanshinones (Fig. 1A) whereas the elicited hairy roots contained elevated amounts of phenolic acids and tanshinones (peaks 3–7) (Fig. 1B). It was also found that phenolic acids were intracellular while tanshinones (peaks 3–7) existed both in the roots and in the culture medium (not shown). 3.3. Effect of elicitor dosage As shown in Fig. 2, the use of yeast elicitor at all four concentrations resulted in a great increase in biomass accumulation (from 3.9 g/l in the control to as much as 7.3 g/l) (Fig. 2A), rosmarinic acid (from 1.24% of dry wt in the control to as much as 2.97%) (Fig. 2B) and tanshinone (e.g. the intracellular cryptotanshinone content increased from 0.001% in the control to as much as 0.096% of dry wt) formation (Figs. 2C) whereas the content of lithospermic acid B increased slightly at lower elicitor dosages (0.1– 0.5 ml) (Fig.
Fig. 1. Typical HPLC profiles of the methanol extracts from non-elicited (A) and elicited (B) hairy roots of S. miltiorrhiza. Peaks: 1, rosmarinic acid; 2, lithospermic acid B; 3, tanshinone IIB; 4, dihydrotanshinone I; 5, unidentified tanshinone (MW 280); 6, cryptotanshinone; 7, tanshinone I; 8, methylenetanshinquinone; and 9, tanshinone IIA. The HPLC conditions were described previously [4].
H. Chen et al. / Enzyme and Microbial Technology 28 (2001) 100 –105
103
3. A rapid increase in biomass accumulation was seen 7 d after elicitation (Fig. 3A). At this moment, the dry weight of elicited culture (5.5 g/l) was much higher than that of the control (2.0 g/l). Fig. 3B shows the time course profile of phenolic acid formation. In the control, the content of lithospermic acid B was always higher than that of rosmarinic acid. Yeast elicitor caused a rapid increase in rosmarinic content whereas the content of lithospermic acid B in elicited roots was similar to that in the control during the first 6 d of elicitation. On day 7 of elicitation, when the growth of hairy roots was greatly improved due to the addition of yeast elicitor, a further increase in the contents of both rosmarinic acid and lithospermic acid B was observed and the content of rosmarinic acid was higher than that of lithospermic acid B (3.95% vs. 3.49% of dry wt) in the elicited hairy roots (Fig. 3B). As shown in Figs. 3C and 3D, the formation of cryptotanshinone started two days after elicitation and increased gradually with elicitation time. After 7 d of elicitation, the intracellular tanshinones increased more rapidly without a concomitant increase in extracellular tanshinone accumulation.
4. Discussion
Fig. 2. Effects of yeast elicitor concentration on biomass accumulation (A), phenolic acid (B), intracellular tanshinone (C) and extracellular tanshinone formation in hairy root cultures of S. miltiorrhiza. Symbols: y0, y0.1, y0.25, y0.5 and y1 ⫽ 0, 0.1, 0.25, 0.5 and 1.0 ml of yeast elicitor solution was added per flask, respectively. Symbols: RA ⫽ rosmarinic acid, LAB ⫽ lithospermic acid B, CT ⫽ cryptotanshinone, I ⫽ tanshinone I, and IIA ⫽ tanshinone IIA. Values are means of triplicate results and error bars show standard deviations.
LC-MS has been used in phytochemical analysis as a powerful tool for on-line identification and quantification of plant constituents, even in trace amounts [16]. This technique has been successfully applied to the identification of tanshinones and phenolic acids from natural root extracts of S. miltiorrhiza. Battistini et al. [14] identified dihydrotanshinone I, cryptotanshinone, tanshinone I, methylenetanshinquinone and tanshinone IIA while Kasimu et al. [17] identified lithospermic acid B, salvianolic acid K, salviaflaside and rosmarinic acid from natural root extracts of S. miltiorrhiza by LC-MS. However, there has been no report on simultaneous separation and identification of tanshinones and phenolic acids by LC-MS. Chemically the phenolic acids and the tanshinones are very different. Subsequently, the MS parameters under electrospray conditions for these two classes of compounds are very different. Due to the acidic nature of the phenolic acids, the negative ions in solution are readily detected under MS negative mode. In contrast, the tanshinones are poor proton acceptors and require strongly acidic conditions for the production of the protonated species to be detected in the positive mode. In order to detect both classes of compounds, and to avoid the use of time programming, the MS was set up to alternatively monitor the positive and negative ions in a loop sequence. Using this method, both phenolic acids and tanshinones were detected and positively identified in a single LC-MS run. This LC-MS method also offers a useful aid for the study of biosynthesis and regulation of phenolic acids and tanshinones in S. miltiorrhiza cell cultures. Phenolic acids e.g. rosmarinic acid exist constitutively in species of the Lamiaceae and Boraginaceae [18]. Due to the
104
H. Chen et al. / Enzyme and Microbial Technology 28 (2001) 100 –105 Table 2 Comparison between hairy roots and crown galls to the elicitation with yeast elicitor. The elicitor concentration was 2% v/v in each case. Hairy roots
Dry weight (g/l) RA (% dry wt) CT (in cells) (% dry wt) CT (in medium) (mg/l) CT (total) (mg/l)
Fig. 3. Time course of biomass accumulation (A), formation of phenolic acids (B), intracellular tanshinones (C) and extracellular tanshinones in non-elicited (Control) and elicited (Yeast) hairy roots of S. miltiorrhiza. 0.5 ml of yeast elicitor solution (Yeast) or 0.5 ml of distilled water (Control) was added to each flask. Symbols: RA ⫽ rosmarinic acid, LAB ⫽ lithospermic acid B, CT ⫽ cryptotanshinone, I ⫽ tanshinone I, and IIA ⫽ tanshinone IIA.
Crown galls [5]
Elicited
Control
Elicited
Control
7.3 2.97 0.095 3.9 10.8
3.9 1.24 0.001 0 0.04
6.1 2.3 0.13 11.8 19.7
10.2 3.4 0 0 0
antimicrobial activity of rosmarinic acid [18], this compound could serve as a phytoanticipin in plants [1]. Although preformed defense compounds present in healthy plants at levels that are anticipated to be antimicrobial, their levels may increase further in response to challenge by pathogens [19]. In the literature, fungal-elicitor enhanced accumulation of RA was reported often [18,20,21]. In hairy root cultures of S. miltiorrhiza, yeast elicitor enhanced the constituent level of RA and the induction was a rapid response as the levels of RA were quite similar on day 1 and day 6 of elicitation (Fig. 3B). The secondary phase of increase in rosmarinic acid and lithospermic acid B formation was most likely not caused by elicitation but by the improvement in root growth (Fig. 3C). We speculate that preformed phenolic acids are growth associated in S. miltiorrhiza. Consequently, culture conditions that favor the cell growth may also favor the phenolic acid formation and vise versa. However, in crown gall cultures of S. miltiorrhiza, the application of yeast elicitor reduced both the biomass accumulation and the constitutive level of RA (Table 2). The reason underlying the different responses observed is still not very clear, probably due to the differences in their endogenous cytokinin levels [11]. In contrast to the previous report by Hu and Alfermann [9] that about 160 mg/l cryptotanshinone was produced in hairy root cultures of S. miltiorrhiza without elicitation, our hairy root culture did not produce cryptotanshinone under normal growth conditions. Probably, Hu and Alfermann used a selected high-tanshinone producing hairy root line as we did in Ti transformed S. miltiorrhiza cell cultures. Without elicitation, the high-tanshinone-producing cell line (cell line B) from Ti transformed S. miltiorrhiza cells produced a high level of tanshinones with cryptotanshinone (150 mg/l) as the dominant compound [4]. However, our attempt to obtain a high-tanshinone producing hairy root line was unsuccessful. Although Ti transformed cells and Ri transformed roots produced different spectra of phenolic acids [7,10], they produced a quite similar profile of tanshinones after elicitation with yeast elicitor (Fig. 1B, [5]). The induced formation of cryptotanshinone was a slow response to elicitation as the level of cryptotanshinone increased gradually until day 6 after elicitation (Figs. 3C & 3D). The levels of cryptotanshinone induced in hairy root cultures (10.8 mg/l) and in crown gall cultures (19.7 mg/l) are
H. Chen et al. / Enzyme and Microbial Technology 28 (2001) 100 –105
summarized in Table 2. It is quite clear that the induced levels of tanshinones in both cultures are still much lower than those of the selected high-tanshinone-producing strains of hairy roots (160 mg/l, [9]) or crown galls (150 mg/l, [4]). Nevertheless, the finding that yeast elicitor could enhance both the biomass accumulation and the secondary metabolite formation in hairy root cultures of S. miltiorrhiza increases the usefulness of this culture system for the biotechnological production of pharmacologically important phenolic acids and tanshinones.
Acknowledgments We thank Drs K. Kamata (Hoshi University, Tokyo, Japan) and Y. M. Xu (Shanghai Institute of Materia Medica, Academia Sinica, Shanghai, China) for their generous gifts of authentic samples of lithospermic acid B. This work was supported by CRCG (the University of Hong Kong Committee on Research and Conference Grants) and the HKU Vice-Chancellor’s Development Fund on Chinese Medicine.
References [1] VanEtten HD, Mansfield JW, Bailey JA, Farmer EE. Two classes of plant antibiotics: phytoalexins versus “phytoanticipins”. Plant Cell 1994;6:1191–2. [2] Papadopoulou K, Melton RE, Leggett M, Daniels MJ, Osbourn AE. Compromised disease resistance in saponin-deficient plants. Proc Natl Acad Sci USA 1999;96:12923– 8. [3] Hahn MG. Microbial elicitors and their receptors in plants. Annu Rev Phytopathol 1996;34:387– 412. [4] Chen H, Chen F. Kinetics of cell growth and secondary metabolism of a high-tanshinone-producing line of the Ti transformed Salvia miltiorrhiza cells in suspension culture. Biotechnol Lett 1999;21:701–5. [5] Chen H, Chen F. Effect of yeast elicitor on the secondary metabolism of Ti transformed Salvia miltiorrhiza cell suspension cultures. Plant Cell Reports 2000;19:710 –7. [6] Chen H, Yuan JP, Chen F, Zhang YL, Song JY. Tanshinone production in Ti-transformed Salvia miltiorrhiza cell suspension cultures. J Biotechnol 1997;58:147–56.
105
[7] Chen H, Chen F, Zhang YL, Song JY. Production of rosmarinic acid and lithospermic acid B in Ti transformed Salvia miltiorrhiza cell suspension cultures. Process Biochem 1999;34:777– 84. [8] Gelvin SB. Crown gall disease and hairy root disease: A sledgehammer and a tackhammer. Plant Physiol 1990;92:281–5. [9] Hu ZB, Alfermann AW. Diterpenoid production in hairy root cultures of Salvia miltiorrhiza. Phytochemistry 1993;32:699 –703. [10] Chen H, Chen F, Zhang YL, Song JY. Production of lithospermic acid B and rosmarinic acid in hairy root cultures of Salvia miltiorrhiza. J Ind Microbiol Biotech 1999;22:133– 8. [11] Hare PD, Cress WA, van Staden J. The involvement of cytokinins in plant responses to environmental stress. Plant Growth Regul 1997; 23:79 –103. [12] Hahn MG, Albersheim P. Host-pathogen interactions. XIV. Isolation and partial characterization of elicitor from yeast extract. Plant Physiol 1978;62:107–11. [13] Veliky IA, Martin SM. A fermenter for plant cell suspension cultures. Can J Microbiol 1970;16:223– 6. [14] Battistini M, Griffini A, Peterlongo F. Identification of tanshinones from Salvia miltiorrhiza roots by liquid chromatography-thermospray mass spectrometry. Fitoterapia 1995;66:265–72. [15] Okamura N, Kobayashi K, Yagi A, Kitazawa T, Shimomura K. High-performance liquid-chromatography of abietane-type compounds. J Chromatogr 1991;542:317–26. [16] Wolfender JL, Maillard M, Hostettmann K. Thermospray liquidchromatography mass-spectrometry in phytochemical analysis. Phytochem Analysis 1994;5:153– 82. [17] Kasimu R, Tanaka K, Tezuka Y, Gong ZN, Li JX, Basnet P, Namba T, Kadota S. Comparative study of seventeen Salvia plants: Aldose reductase inhibitory activity of water and MeOH extracts and liquid chromatography mass spectrometry (LC-MS) analysis of water extracts. Chem Pharm Bull 1998;46:500 – 4. [18] Szabo E, Thelen A, Petersen M. Fungal elicitor preparations and methyl jasmonate enhance rosmarinic acid accumulation in suspension cultures of Coleus blumei. Plant Cell Reports 1999;18:485–9. [19] Morrissey JP, Osbourn AE. Fungal resistance to plant antibiotics as a mechanism of pathogenesis. Microbiol Mol Biol Rev 1999;63:708 – 24. [20] Sumaryono W, Proksch P, Hartmann T, Nimtz M, Wray V. Induction of rosmarinic acid accumulation in cell suspension cultures of Orthosiphon aristatus after treatment with yeast extract. Phytochemistry 1991;30:3267–71. [21] Mizukami H, Ogawa T, Ohashi H, Ellis BE. Induction of rosmarinic acid biosynthesis in Lithospermum erythrohizon cell suspension cultures by yeast extract. Plant Cell Reports 1992;11:480 –3.
www.elsevier.com/locate/enzmictec
The effect of yeast elicitor on the growth and secondary metabolism of hairy root cultures of Salvia miltiorrhiza Hui Chena, Feng Chena,*, Francis C. K. Chiub, Cindy M. Y. Lob a
Department of Botany, The University of Hong Kong, Pokfulam Road, Hong Kong, P. R. China Department of Pharmacy, The Chinese University of Hong Kong, Shatin, New Territories, Hong Kong, P. R. China
b
Received 20 March 2000; received in revised form 31 July 2000; accepted 2 August 2000
Abstract Hairy root cultures of Salvia miltiorrhiza transformed with Agrobacterium rhizogenes ATCC 15834 produced a tiny amount of tanshinones and a constituent level of phenolic acids under normal growth conditions. Upon elicitation with yeast elicitor, the production of both phenolic acids and tanshinones was enhanced. For example, the contents of two phenolic acids, rosmarinic acid and lithospermic acid B were elevated from 1.24% and 2.59% to 2.89% and 2.98% of dry wt, respectively while the intracellular content of cryptotanshinone increased from 0.001% to as much as 0.096% of dry wt. Yeast elicitor also improved the growth of hairy roots (from 3.9 g/l to 7.3 g/l on a dry wt basis). Liquid chromatography-mass spectrometry (LC-MS) was developed for simultaneous detection and identification of phenolic acids and tanshinones in the extracts of S. miltiorrhiza. Rosmarinic acid, lithospermic acid B, cryptotanshinone, tanshinone I, tanshinone IIA and tanshinone IIB were identified by comparison with standards available. Dihydrotanshinone I and methylenetanshiquinone were tentatively identified by the molecular weights and the elution comparable with the literature. An unknown compound with a molecular weight of 280 was found in yeast-elicitor treated hairy root cultures, which was one of the major tanshinones induced. © 2001 Elsevier Science Inc. All rights reserved. Keywords: Hairy roots; LC-MS; Phenolic acids; Salvia miltiorrhiza; Tanshinones; Yeast elicitor
1. Introduction Plants produce a high diversity of natural products or secondary metabolites, many of which have antimicrobial activity in vitro. Some of these compounds are synthesized during normal growth and development while others are absent from healthy plants, accumulating only in response to pathogen attack or stress conditions. These preformed and inducible antimicrobial compounds are called phytoanticipins and phytoalexins, respectively [1,2]. Elicitors are generally defined as molecules that stimulate any defense responses of plants, including the formation of phytoalexins [3]. The application of elicitors to plant cell cultures is not only useful for enhancing the biotechnological productivity of valuable secondary metabolites in fermentation systems but also important to the study of plant-microbe interactions. Salvia miltiorrhiza Bunge (Labiatae) is a famous traditional Chinese medicinal plant that contains two groups of biologically active compounds: caffeic acid-derived phe* Corresponding author. Tel.: ⫹852-2299 0309; fax: ⫹852-2299 0311. E-mail address: [email protected] (F. Chen).
nolic acids and various tanshinones belonging to diterpene quinones [4]. Our previous work demonstrated that phenolic acids and tanshinones may represent preformed and inducible defense compounds, respectively, in S. miltiorrhiza [4,5]. In a crown gall culture of S. miltiorrhiza, a large amount of phenolic acids and a negligible amount of tanshinones were produced under normal growth conditions [6,7]. However, upon elicitation with yeast elicitor, the level of tanshinones was enhanced greatly whereas the constituent level of phenolic acids was reduced [5]. Crown galls and hairy roots are plant tumors obtained after transformation by Agrobacterium tumefaciens and A. rhizogenes, respectively [8]. Hairy root cultures of S. miltiorrhiza have demonstrated the ability to synthesize both tanshinones [9] and phenolic acids [10]. Our crown galls and hairy roots of S. miltiorrhiza showed distinct morphologies. They both contained a comparable level of endogenous indole acetic acid (IAA) but crown galls had a much higher level of endogenous cytokinins than hairy roots (Chen and Chen, unpublished data). Since cytokinins play an important role in modulating plant defense response to environmental stress [11], it is interesting to know the re-
0141-0229/01/$ – see front matter © 2001 Elsevier Science Inc. All rights reserved. PII: S 0 1 4 1 - 0 2 2 9 ( 0 0 ) 0 0 2 8 4 - 2
H. Chen et al. / Enzyme and Microbial Technology 28 (2001) 100 –105
101
sponse of the growth and secondary metabolism of hairy roots of S. miltiorrhiza to the elicitation with yeast elicitor. This paper addresses this question and makes a comparison with that of crown galls. In addition, we have developed a liquid chromatographymass spectrometry (LC-MS) method for simultaneous detection and identification of phenolic acids and tanshinones in hairy root cultures.
2.5. Kinetics study
2. Materials and methods
2.6. Analyses
2.1. Hairy root culture
Cell dry weight determination and HPLC analysis for phenolic acids and tanshinones were carried out as previously described [4].
Experiments were carried out with a hairy root culture of S. miltiorrhiza transformed with Agrobacterium rhizogenes ATCC 15834 as previously described [10]. 2.2. Standards Cryptotanshinone, tanshinone I, tanshinone IIA and tanshinone IIB were obtained from Shanghai Institute of Materia Medica, Academia Sinica, Shanghai, China. Lithospermic acid B was a gift from Drs K. Kamata (Hoshi University, Tokyo, Japan) and Y. M. Xu (Shanghai Institute of Materia Medica, Academia Sinica, Shanghai, China). Rosmarinic acid was purchased from ICN (Costa Mesa, CA, USA). 2.3. Elicitor preparation A carbohydrate fraction isolated from the yeast extract was prepared by ethanol precipitation as described by Hahn and Albersheim [12]. Briefly, 50 g of the yeast extract was dissolved in 250 ml of distilled water. Ethanol was added to 80% (v/v). The precipitate was allowed to settle for 4 days at 6°C and the supernatant was decanted and discarded. The gummy precipitate was dissolved in 250 ml of distilled water. The ethanol precipitation was repeated. The second ethanol precipitate was dissolved in 200 ml of distilled water, yielding the crude preparation that was used without further purification. 2.4. Elicitor dosage response Hairy roots were subcultured with an inoculum of ca. 0.3 g fresh root segments in 50 ml of hormone-free liquid 6,7-V medium [13] containing 20 g/l sucrose in 250-ml shake flasks. Cultivation was performed on an orbital shaker at 140 rpm in darkness at 25°C. Twenty days after cultivation, hairy roots were transferred to fresh 6,7-V medium containing different volumes of yeast elicitor solution prepared above or 1 ml of distilled water (control). On day 7 of elicitation, hairy root cultures were harvested and the biomass and contents of phenolic acids and tanshinones were determined. The experiment was performed in triplicate.
For a time course study, ca. 0.3 g fresh root segments were inoculated into 50 ml of hormone-free liquid 6,7-V medium containing 20 g/l sucrose in 250-ml shake flasks. On day 20 of cultivation, 0.5 ml of elicitor solution or 0.5 ml of distilled water (control), having been sterilized by autoclaving at 121°C for 15 min, was added to each flask. Then, flasks were harvested daily and subsequent analyses were performed.
2.7. LC-MS conditions The HPLC-MS system consists of a Perkin Elmer series 200 micro pump, a Perkin Elmer series 200 autosampler, a Perkin Elmer 785A UV/Vis detector, and a Perkin Elmer SCIEX API-2000 mass spectrometer equipped with a TurbolonSpray interface. The whole system was controlled by the Mass Chrom 1.1 software. Separations were obtained at room temperature on a Waters Puresil 5 m C18 column (4.6 ⫻ 150 mm) with a guard column of the same material. The mobile phase consisted of 0.1% TFA in water (solvent A) and acetonitrile (solvent B). The following gradient procedure was used: 0 –20 min: 20 –70% B; 20 – 40 min: 70% B; 40 – 41 min, 70 –20% B. The absorbance was measured at 280 nm. The LC eluant after the UV-detector was diverted via a splitter such that 0.2 ml/min was delivered into the API-2000 mass spectrometer. The MS parameters were first optimized using representative reference compounds by direct infusion. Subsequent higher heater gas temperature (200°C) and gas flow (50 psi) were added to the LC-MS parameters. Tanshinone I ([M ⫹ H]⫹ 277) was used for positive mode and rosmarinic acid ([M ⫺ H]⫺ 359) for negative mode monitoring of tanshinones and phenolic acids respectively. The crude mixtures were screened by direct infusion to determine maximum scan range for the unknown samples. During LC-MS runs, the mass spectrometer was set up to monitor both positive and negative ions simultaneously using a loop sequence alternating between the positive and negative mode. A scan range of 100 to 400 amu and 100 to 750 amu were used for positive ions and negative ions respectively. A step size of 1 amu and dwell time of 1 ms were used in both modes. 3. Results 3.1. Separation and identification of phenolic acids and tanshinones by LC-MS Using the HPLC conditions developed in the present study, both the phenolic acids and the tanshinones in yeast-
102
H. Chen et al. / Enzyme and Microbial Technology 28 (2001) 100 –105
Table 1 HPLC-MS data on phenolic acids and tanshinones in S. miltiorrhiza hairy root extract Peak number
Retention time (min)
Major species (amu)
1 2 3 4 5 6 7 8 9
6.35 7.03 15.27 19.02 19.94 22.05 22.89 24.39 27.03
358.9 716.8 311.2 278.9 280.8 297.1 276.9 279.0 294.8
elicitor treated hairy roots were successfully separated and identified simultaneously (Table 1). Under the pH condition of the mobile phase (ca. 0.1% TFA), all of the tanshinones could be detected by the positive mode. The phenolic acids were found to give cleaner spectra with high sensitivity. Thus, the negative mode was used to monitor phenolic acids, while the positive mode was used for tanshinones. The identification and retention times of the major compounds are summarized in Table 1. Rosmarinic acid, lithospermic acid B, cryptotanshinone, tanshinone I, tanshinone IIA and tanshinone IIB were identified by comparison with standards available. Dihydrotanshinone I and methylenetanshiquinone were tentatively identified by the molecular weights and the elution comparable with the literature [14, 15]. The unknown compound (peak 5, retention time 19.94, MW 280) was probably the most abundant tanshinone induced. The further purification and identification of this compound is in progress in our laboratories.
Scan mode
MW
Identification
[M⫺H]⫹ [M⫺H]⫹ [M⫹H]⫹ [M⫹H]⫹ [M⫹H]⫹ [M⫹H]⫹ [M⫹H]⫹ [M⫹H]⫹ [M⫹H]⫹
360 718 310 278 280 296 276 278 294
rosmarinic acid lithospermic acid B tanshinone IIB dihydrotanshinone I unknown cryptotanshinone tanshinone I methylenetanshiquinone tanshinone IIA
2B). Among the known tanshinones, cryptotanshinone was most notable in its increase owing to elicitation as it was almost not produced in non-elicited hairy roots (Figs. 2C and 2D). 3.4. Time course of elicitation Time course of the effect of yeast elicitor on root growth and phenolic acid and tanshinone formation is shown in Fig.
3.2. Comparison of the HPLC profiles of the methanol extracts from hairy roots with and without elicitation As a typical chromatogram (Fig. 1) shows, the nonelicited hairy root contained a large amount of phenolic acids (peaks 1 and 2) and a negligible amount of tanshinones (Fig. 1A) whereas the elicited hairy roots contained elevated amounts of phenolic acids and tanshinones (peaks 3–7) (Fig. 1B). It was also found that phenolic acids were intracellular while tanshinones (peaks 3–7) existed both in the roots and in the culture medium (not shown). 3.3. Effect of elicitor dosage As shown in Fig. 2, the use of yeast elicitor at all four concentrations resulted in a great increase in biomass accumulation (from 3.9 g/l in the control to as much as 7.3 g/l) (Fig. 2A), rosmarinic acid (from 1.24% of dry wt in the control to as much as 2.97%) (Fig. 2B) and tanshinone (e.g. the intracellular cryptotanshinone content increased from 0.001% in the control to as much as 0.096% of dry wt) formation (Figs. 2C) whereas the content of lithospermic acid B increased slightly at lower elicitor dosages (0.1– 0.5 ml) (Fig.
Fig. 1. Typical HPLC profiles of the methanol extracts from non-elicited (A) and elicited (B) hairy roots of S. miltiorrhiza. Peaks: 1, rosmarinic acid; 2, lithospermic acid B; 3, tanshinone IIB; 4, dihydrotanshinone I; 5, unidentified tanshinone (MW 280); 6, cryptotanshinone; 7, tanshinone I; 8, methylenetanshinquinone; and 9, tanshinone IIA. The HPLC conditions were described previously [4].
H. Chen et al. / Enzyme and Microbial Technology 28 (2001) 100 –105
103
3. A rapid increase in biomass accumulation was seen 7 d after elicitation (Fig. 3A). At this moment, the dry weight of elicited culture (5.5 g/l) was much higher than that of the control (2.0 g/l). Fig. 3B shows the time course profile of phenolic acid formation. In the control, the content of lithospermic acid B was always higher than that of rosmarinic acid. Yeast elicitor caused a rapid increase in rosmarinic content whereas the content of lithospermic acid B in elicited roots was similar to that in the control during the first 6 d of elicitation. On day 7 of elicitation, when the growth of hairy roots was greatly improved due to the addition of yeast elicitor, a further increase in the contents of both rosmarinic acid and lithospermic acid B was observed and the content of rosmarinic acid was higher than that of lithospermic acid B (3.95% vs. 3.49% of dry wt) in the elicited hairy roots (Fig. 3B). As shown in Figs. 3C and 3D, the formation of cryptotanshinone started two days after elicitation and increased gradually with elicitation time. After 7 d of elicitation, the intracellular tanshinones increased more rapidly without a concomitant increase in extracellular tanshinone accumulation.
4. Discussion
Fig. 2. Effects of yeast elicitor concentration on biomass accumulation (A), phenolic acid (B), intracellular tanshinone (C) and extracellular tanshinone formation in hairy root cultures of S. miltiorrhiza. Symbols: y0, y0.1, y0.25, y0.5 and y1 ⫽ 0, 0.1, 0.25, 0.5 and 1.0 ml of yeast elicitor solution was added per flask, respectively. Symbols: RA ⫽ rosmarinic acid, LAB ⫽ lithospermic acid B, CT ⫽ cryptotanshinone, I ⫽ tanshinone I, and IIA ⫽ tanshinone IIA. Values are means of triplicate results and error bars show standard deviations.
LC-MS has been used in phytochemical analysis as a powerful tool for on-line identification and quantification of plant constituents, even in trace amounts [16]. This technique has been successfully applied to the identification of tanshinones and phenolic acids from natural root extracts of S. miltiorrhiza. Battistini et al. [14] identified dihydrotanshinone I, cryptotanshinone, tanshinone I, methylenetanshinquinone and tanshinone IIA while Kasimu et al. [17] identified lithospermic acid B, salvianolic acid K, salviaflaside and rosmarinic acid from natural root extracts of S. miltiorrhiza by LC-MS. However, there has been no report on simultaneous separation and identification of tanshinones and phenolic acids by LC-MS. Chemically the phenolic acids and the tanshinones are very different. Subsequently, the MS parameters under electrospray conditions for these two classes of compounds are very different. Due to the acidic nature of the phenolic acids, the negative ions in solution are readily detected under MS negative mode. In contrast, the tanshinones are poor proton acceptors and require strongly acidic conditions for the production of the protonated species to be detected in the positive mode. In order to detect both classes of compounds, and to avoid the use of time programming, the MS was set up to alternatively monitor the positive and negative ions in a loop sequence. Using this method, both phenolic acids and tanshinones were detected and positively identified in a single LC-MS run. This LC-MS method also offers a useful aid for the study of biosynthesis and regulation of phenolic acids and tanshinones in S. miltiorrhiza cell cultures. Phenolic acids e.g. rosmarinic acid exist constitutively in species of the Lamiaceae and Boraginaceae [18]. Due to the
104
H. Chen et al. / Enzyme and Microbial Technology 28 (2001) 100 –105 Table 2 Comparison between hairy roots and crown galls to the elicitation with yeast elicitor. The elicitor concentration was 2% v/v in each case. Hairy roots
Dry weight (g/l) RA (% dry wt) CT (in cells) (% dry wt) CT (in medium) (mg/l) CT (total) (mg/l)
Fig. 3. Time course of biomass accumulation (A), formation of phenolic acids (B), intracellular tanshinones (C) and extracellular tanshinones in non-elicited (Control) and elicited (Yeast) hairy roots of S. miltiorrhiza. 0.5 ml of yeast elicitor solution (Yeast) or 0.5 ml of distilled water (Control) was added to each flask. Symbols: RA ⫽ rosmarinic acid, LAB ⫽ lithospermic acid B, CT ⫽ cryptotanshinone, I ⫽ tanshinone I, and IIA ⫽ tanshinone IIA.
Crown galls [5]
Elicited
Control
Elicited
Control
7.3 2.97 0.095 3.9 10.8
3.9 1.24 0.001 0 0.04
6.1 2.3 0.13 11.8 19.7
10.2 3.4 0 0 0
antimicrobial activity of rosmarinic acid [18], this compound could serve as a phytoanticipin in plants [1]. Although preformed defense compounds present in healthy plants at levels that are anticipated to be antimicrobial, their levels may increase further in response to challenge by pathogens [19]. In the literature, fungal-elicitor enhanced accumulation of RA was reported often [18,20,21]. In hairy root cultures of S. miltiorrhiza, yeast elicitor enhanced the constituent level of RA and the induction was a rapid response as the levels of RA were quite similar on day 1 and day 6 of elicitation (Fig. 3B). The secondary phase of increase in rosmarinic acid and lithospermic acid B formation was most likely not caused by elicitation but by the improvement in root growth (Fig. 3C). We speculate that preformed phenolic acids are growth associated in S. miltiorrhiza. Consequently, culture conditions that favor the cell growth may also favor the phenolic acid formation and vise versa. However, in crown gall cultures of S. miltiorrhiza, the application of yeast elicitor reduced both the biomass accumulation and the constitutive level of RA (Table 2). The reason underlying the different responses observed is still not very clear, probably due to the differences in their endogenous cytokinin levels [11]. In contrast to the previous report by Hu and Alfermann [9] that about 160 mg/l cryptotanshinone was produced in hairy root cultures of S. miltiorrhiza without elicitation, our hairy root culture did not produce cryptotanshinone under normal growth conditions. Probably, Hu and Alfermann used a selected high-tanshinone producing hairy root line as we did in Ti transformed S. miltiorrhiza cell cultures. Without elicitation, the high-tanshinone-producing cell line (cell line B) from Ti transformed S. miltiorrhiza cells produced a high level of tanshinones with cryptotanshinone (150 mg/l) as the dominant compound [4]. However, our attempt to obtain a high-tanshinone producing hairy root line was unsuccessful. Although Ti transformed cells and Ri transformed roots produced different spectra of phenolic acids [7,10], they produced a quite similar profile of tanshinones after elicitation with yeast elicitor (Fig. 1B, [5]). The induced formation of cryptotanshinone was a slow response to elicitation as the level of cryptotanshinone increased gradually until day 6 after elicitation (Figs. 3C & 3D). The levels of cryptotanshinone induced in hairy root cultures (10.8 mg/l) and in crown gall cultures (19.7 mg/l) are
H. Chen et al. / Enzyme and Microbial Technology 28 (2001) 100 –105
summarized in Table 2. It is quite clear that the induced levels of tanshinones in both cultures are still much lower than those of the selected high-tanshinone-producing strains of hairy roots (160 mg/l, [9]) or crown galls (150 mg/l, [4]). Nevertheless, the finding that yeast elicitor could enhance both the biomass accumulation and the secondary metabolite formation in hairy root cultures of S. miltiorrhiza increases the usefulness of this culture system for the biotechnological production of pharmacologically important phenolic acids and tanshinones.
Acknowledgments We thank Drs K. Kamata (Hoshi University, Tokyo, Japan) and Y. M. Xu (Shanghai Institute of Materia Medica, Academia Sinica, Shanghai, China) for their generous gifts of authentic samples of lithospermic acid B. This work was supported by CRCG (the University of Hong Kong Committee on Research and Conference Grants) and the HKU Vice-Chancellor’s Development Fund on Chinese Medicine.
References [1] VanEtten HD, Mansfield JW, Bailey JA, Farmer EE. Two classes of plant antibiotics: phytoalexins versus “phytoanticipins”. Plant Cell 1994;6:1191–2. [2] Papadopoulou K, Melton RE, Leggett M, Daniels MJ, Osbourn AE. Compromised disease resistance in saponin-deficient plants. Proc Natl Acad Sci USA 1999;96:12923– 8. [3] Hahn MG. Microbial elicitors and their receptors in plants. Annu Rev Phytopathol 1996;34:387– 412. [4] Chen H, Chen F. Kinetics of cell growth and secondary metabolism of a high-tanshinone-producing line of the Ti transformed Salvia miltiorrhiza cells in suspension culture. Biotechnol Lett 1999;21:701–5. [5] Chen H, Chen F. Effect of yeast elicitor on the secondary metabolism of Ti transformed Salvia miltiorrhiza cell suspension cultures. Plant Cell Reports 2000;19:710 –7. [6] Chen H, Yuan JP, Chen F, Zhang YL, Song JY. Tanshinone production in Ti-transformed Salvia miltiorrhiza cell suspension cultures. J Biotechnol 1997;58:147–56.
105
[7] Chen H, Chen F, Zhang YL, Song JY. Production of rosmarinic acid and lithospermic acid B in Ti transformed Salvia miltiorrhiza cell suspension cultures. Process Biochem 1999;34:777– 84. [8] Gelvin SB. Crown gall disease and hairy root disease: A sledgehammer and a tackhammer. Plant Physiol 1990;92:281–5. [9] Hu ZB, Alfermann AW. Diterpenoid production in hairy root cultures of Salvia miltiorrhiza. Phytochemistry 1993;32:699 –703. [10] Chen H, Chen F, Zhang YL, Song JY. Production of lithospermic acid B and rosmarinic acid in hairy root cultures of Salvia miltiorrhiza. J Ind Microbiol Biotech 1999;22:133– 8. [11] Hare PD, Cress WA, van Staden J. The involvement of cytokinins in plant responses to environmental stress. Plant Growth Regul 1997; 23:79 –103. [12] Hahn MG, Albersheim P. Host-pathogen interactions. XIV. Isolation and partial characterization of elicitor from yeast extract. Plant Physiol 1978;62:107–11. [13] Veliky IA, Martin SM. A fermenter for plant cell suspension cultures. Can J Microbiol 1970;16:223– 6. [14] Battistini M, Griffini A, Peterlongo F. Identification of tanshinones from Salvia miltiorrhiza roots by liquid chromatography-thermospray mass spectrometry. Fitoterapia 1995;66:265–72. [15] Okamura N, Kobayashi K, Yagi A, Kitazawa T, Shimomura K. High-performance liquid-chromatography of abietane-type compounds. J Chromatogr 1991;542:317–26. [16] Wolfender JL, Maillard M, Hostettmann K. Thermospray liquidchromatography mass-spectrometry in phytochemical analysis. Phytochem Analysis 1994;5:153– 82. [17] Kasimu R, Tanaka K, Tezuka Y, Gong ZN, Li JX, Basnet P, Namba T, Kadota S. Comparative study of seventeen Salvia plants: Aldose reductase inhibitory activity of water and MeOH extracts and liquid chromatography mass spectrometry (LC-MS) analysis of water extracts. Chem Pharm Bull 1998;46:500 – 4. [18] Szabo E, Thelen A, Petersen M. Fungal elicitor preparations and methyl jasmonate enhance rosmarinic acid accumulation in suspension cultures of Coleus blumei. Plant Cell Reports 1999;18:485–9. [19] Morrissey JP, Osbourn AE. Fungal resistance to plant antibiotics as a mechanism of pathogenesis. Microbiol Mol Biol Rev 1999;63:708 – 24. [20] Sumaryono W, Proksch P, Hartmann T, Nimtz M, Wray V. Induction of rosmarinic acid accumulation in cell suspension cultures of Orthosiphon aristatus after treatment with yeast extract. Phytochemistry 1991;30:3267–71. [21] Mizukami H, Ogawa T, Ohashi H, Ellis BE. Induction of rosmarinic acid biosynthesis in Lithospermum erythrohizon cell suspension cultures by yeast extract. Plant Cell Reports 1992;11:480 –3.