- Email: [email protected]
Metabolomic profiling of Vitis vinifera cell suspension culture elicited with silver nitrate by 1H NMR spectrometry and principal components analysis
Process Biochemistry 42 (2007) 271–274 www.elsevier.com/locate/procbio
Short communication
Metabolomic profiling of Vitis vinifera cell suspension culture elicited with silver nitrate by 1H NMR spectrometry and principal components analysis Hyung-Kyoon Choi *, Jung-Hye Yoon College of Pharmacy, Chung-Ang University, Seoul 156-756, Republic of Korea Received 6 April 2006; accepted 11 July 2006
Abstract The metabolomic profiling of Vitis vinifera cell suspension cultures with and without silver nitrate was performed by 1H NMR (nuclear magnetic resonance) spectrometry and principal components analysis (PCA), to assess the efficacy of this method for the characterization and monitoring of plant cell lines. The PCA of the 1H NMR spectra of the aqueous fractions allowed a clear discrimination of V. vinifera cell suspension culture samples with and without silver nitrate treatment by the first three principal components (PC1, PC2, and PC3), which cumulatively accounted for 95.9% of the variation in all variables. In particular, the score plots by the combining PC1 versus PC2 and PC2 versus PC3 facilitated an excellent separation of samples. In addition, the major peaks in 1H NMR spectra contributing to the discrimination were assigned to lactate, alanine, acetic acid, choline, fructose, a-glucose, and sucrose. This method based on metabolomic analysis allows the efficient monitoring and the differentiation of normal cell suspension system from elicited systems without any prepurification steps. # 2006 Elsevier Ltd. All rights reserved. Keywords: Metabolomic profiling; Vitis vinifera; 1H NMR; Principal component analysis
1. Introduction Plant cell cultures may represent as an attractive alternative for overcoming the limitations of extracting useful metabolites from natural resources. Many factors affect cell growth and the production of metabolites in plant cell cultures, such as the components of the basal medium, the source of carbon, phytohormones, antimetabolites, O2, pH, elicitors, temperature, stirring frequency, and light conditions [1]. The stability of a cell line is especially important for the commercialization of the plant cell culturing process. Large-scale production of useful secondary metabolites has been attempted by various research groups [2,3], which requires the monitoring and maintaining of the cell states, because plant cell culture system are labile to external stresses that could alter the bioactivity in producing useful compounds. Bx, pH, conductivity, and the levels of protein and peroxidase have been suggested as monitoring methods [4], with the levels of nitrate, phosphate, sucrose, and glucose, and respiration activity having been used
* Corresponding author. Tel.: +82 2 820 5605; fax: +82 2 816 7338. E-mail address: [email protected] (H.-K. Choi). 1359-5113/$ – see front matter # 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.procbio.2006.07.007
as monitoring tools [5,6]. Intra- and extra-metabolome analyses have recently been performed in a microbial cell culture system [7], but there has been little work on plant cell culture systems. The term ‘metabolome’ has been used to describe the observable chemical profile or fingerprint of the metabolites present in whole tissues [8]. The chemical analysis techniques applied to metabolites profiling should be rapid, reproducible, and stable over time, while requiring only simple sample preparation. A technique that potentially meets these demands is NMR, which has been widely used as a fingerprinting tool for the interpretation and quality assessment of industrial and natural products, with multivariate or pattern recognition techniques such as the well-known principal components analysis (PCA) having been specifically designed to analyze complex data sets [9]. Recently, the combination of NMR and PCA has been applied to the metabolic profiling of various types of coffee, juice, wine, and beer and some plants [10–17]. In this study, we applied a 1H NMR spectroscopy method coupled with PCA to the metabolomic profiling of Vitis vinifera cell suspension culture elicited with silver nitrate, which was used as an elicitor for enhancing the production of paclitaxel [18,19] as a model system, with the aim of suggesting the standard trajectory plots and elucidating the major metabolites
272
H.-K Choi, J.-H Yoon / Process Biochemistry 42 (2007) 271–274
contributing to the discrimination between normal and elicited samples. This method can be used to characterize a plant cell suspension by supplying the standard trajectory plots for the stabilized production of specific secondary metabolites.
CHCl3 extracts and TSP for aqueous extracts, and reduced to integrated regions of equal width (0.04 ppm) corresponding to d = 0.50–10.00. The region from 4.60 to 4.90 was excluded from the analysis because of the residual signal of water in aqueous extracts, whereas that from 7.00 to 7.50 was excluded because of the residual signal of CHCl3 in organic fractions. PCA was performed with SIMCA-P software (Umetrics, Umea˚, Sweden).
2. Materials and methods
3. Results and discussion 2.1. Plant cell line and media V. vinifera cells were obtained from the Korean Collection for Type Cultures (Daejeon, Korea) and grown under darkness at 24 1 8C on a shaking incubator (Jeiotech SI-900R, Korea) operating at 150 rpm. The liquid medium contained the inorganic salt formulation of Murashige and Skoog [20], 30 g/l sucrose, and 1 mg/l 2,4-dichlorophenoxyacetic acid. The cells were subcultured every 12 days by adding 25 ml of inoculum into 75 ml of fresh medium in 250 ml flasks. Flasks were capped with sterilized aluminum foil and taped with Micropore surgical tape (3M 1530-1, USA). To elicit a cell suspension, 4 mM AgNO3 was added on day 6. All reagents for cell cultivation were purchased from Sigma, and all measurement were madein triplicate.
2.2. Cell mass preparation To determine the dry cell weight (DCW) and internal metabolomes by 1H NMR, cell suspensions were filtered through Whatman No. 541 filter paper on a Buchner funnel under a slight vacuum, and then washed with distilled water. The cells were placed on an aluminum weighing dish and dried in an oven at 70 8C until the weight was constant (corresponding to the DCW).
2.3. Solvents and chemicals for NMR measurement First-grade chloroform, methanol, and D2O (99.9%) were purchased from Sigma (St. Louis, MO, USA), and CDCl3 (99.8%) and NaOD were purchased from Cambridge Isotope Laboratories (Miami, FL, USA) and Cortec (Paris, France), respectively.
2.4. Extraction of V. vinifera cell suspension culture materials for NMR measurement
3.1. Assignments of the compounds from visual inspection of 1H NMR spectra There were no significant differences in values of DCW, pH, and Bx. did not differ significantly between normal cells and cells elicited with silver nitrate during the cultivation period (data not shown). Therefore a further experiment was performed to investigate the metabolome profile inside the cells. Little difference was observed between the spectra of the CHCl3 extracts of the various samples (data not shown), and so only the aqueous fractions were further analyzed. Representative 1H NMR spectra of the aqueous extracts are shown in Fig. 1. The NMR signals were smaller in the aromatic region (d = 6.0–8.0) than in the aliphatic and sugar regions. The signals of the main aromatic compounds were assigned to fumaric acid (at d = 6.52 (s)) (Fig. 1(c)). In addition, the following signals were assigned based on comparisons with the chemical shifts of standard compounds and two-dimensional NMR using 1H-1H COSY (correlation spectroscopy), HMQC (heteronuclear multiple quantum coherence), and HMBC (heteronuclear multiple bond coherence): lactate at d = 1.34 (d, J = 6.6 Hz), alanine at d = 1.48 (d, J = 7.3 Hz), acetic acid at d = 1.90 (s), choline at d = 3.22 (s), fructose at d = 4.22 (d, J = 8.7 Hz), b-glucose at d = 4.64 (d, J = 7.9 Hz), a-glucose at d = 5.24 (d, J = 3.9 Hz), and sucrose at d = 5.42 (d, J = 3.9 Hz) (Fig. 1(a) and (b)).
One hundred milligrams of fresh cells were transferred into a centrifuge tube. Five milliliters of a 50% water–methanol mixture and 5 ml of chloroform were added to the antler sample in the tube and vortexed for 30 s and sonicated for 1 min. The materials were then centrifuged at 3000 rpm for 20 min. The extraction was performed twice. The aqueous and organic fractions were transferred separately into a 50 ml round-bottomed flask and dried with a rotary vacuum evaporator. Each experiment was performed in triplicate.
2.5. NMR measurements KH2PO4 was added to D2O as a buffering agent. The pH of the D2O used for NMR measurements was adjusted to 6.0 using a 1N NaOD solution. All spectra were obtained by a NMR spectrometer (Avance 600 FT-NMR, Bruker, Germany) operating at a proton NMR frequency of 600.13 MHz. For each sample, 128 scans were recorded with the following parameters: 0.155 Hz/point, pulse width of 4.0 ms (308), and relaxation delay of 1.0 s. Free induction decays were Fourier transformed with LB = 0.3 Hz, GB = 0, and PC = 1.0. The spectra were referenced to trimethyl silane propionic acid sodium salt (TSP) at 0.00 ppm for aqueous fractions and, for CHCl3 fractions, to residual solvent at 7.26 ppm. Hexamethyl disilane (HMDS, 0.01%, v/v) and TSP (0.01%, w/v) were used as internal standards for CDCl3 and D2O, respectively. The peak intensities in 0.04 ppm bins in the 1H NMR spectra for d = 0.50–10.00 were used as variables.
2.6. Data analysis The 1H NMR spectra were automatically reduced to ASCII files using AMIX (v. 3.7, Biospin, Bruker). Spectral intensities were scaled to HMDS for
Fig. 1. Representative 1H NMR spectra of the total (a), aliphatic (b), and aromatic (c) regions of the aqueous fraction of V. vinifera cell suspension culture samples. IS, internal standard; w, residual water; 1, lactate; 2, alanine; 3, acetic acid; 4, choline; 5, fructose; 6, b-glucose; 7, a-glucose; 8, sucrose; 9, fumaric acid. Values on the X-axis are the d values.
H.-K Choi, J.-H Yoon / Process Biochemistry 42 (2007) 271–274
273
Fig. 2. Score plots from PCA of aqueous extracts of V. vinifera cell suspension cultures with the cultivation time by the combination of PC1 and PC2 (a), PC1 and PC3 (b), and PC2 and PC3 (c). The ellipses represent confidences of 95% in Hotelling T2 tests.
3.2. Principal components analysis To ensure the objective interpretation of the results, the samples were also analyzed using PCA. PCA is an unsupervised clustering method (in that it does not require any knowledge of the data set) that attempts to reduce the dimensionality of multivariate data while preserving most of the variance therein [21]. The present study applied the covariance method for PCA because it produced a better separation than the correlation method (data not shown). As evident in Fig. 2, the elicited samples of the V. vinifera cells elicited with silver nitrate were clearly distinguishable from nonelicited cells, with the first two principal components cumulatively accounting for 82.1% of the variance. Silver nitrate has been used as an elicitor of the improved production of paclitaxel in Taxus chinensis suspension culture, with the mechanism of action known to be ethylene inhibition [18,19]. An excellent separation between samples in
score plots was achieved by combining principal component 1 (PC1) with principal component 2 (PC2). In the score plot of the combination of PC1 and PC2, samples from V. vinifera cells with and without elicitation with silver nitrate showed a different trajectory with the culture time. In addition, the combination of PC2 and principal component 3 (PC3) showed a clearly separate trajectory between cells with and without silver nitrate treatment. As shown in Fig. 2(c), the control samples formed a left-side trajectory plots, while the samples with silver nitrates treatment formed right-side trajectory plots. Elicitation with silver nitrate was conducted on day 6, and the plots at day 7 already started to show slight discrimination, with clear differentiation being present from day 11. PCA of the aliphatic region (from 0.50 to 3.3 ppm) produced no useful information (data not shown). The discriminating metabolites are clearly distinguishable in the loading plots for PC1, PC2, and PC3 (Fig. 3), with the score and loading plots complementing each other. The position of
Fig. 3. PCA loading plots of aqueous extracts of V. vinifera cell suspension cultures with the cultivation time by PC1 (a), PC2 (b), and PC3 (c).
274
H.-K Choi, J.-H Yoon / Process Biochemistry 42 (2007) 271–274
objects in a given direction in a score plot is influenced by variables lying in the same direction in the loading plot. The major compounds contributing to the discrimination by PC1 were lactate, acetic acid, choline, fructose, and sucrose; lactate, acetic acid, choline, and a-glucose contributed to the discrimination by PC2; acetic acid, choline, fructose and sucrose contributed to the discrimination by PC3 (Fig. 3). Although there were a few peaks in the aromatic region (such as for fumaric acid), the peaks did not contribute to the discrimination of the samples with the cultivation time. This study has shown that it is possible to discriminate elicited V. vinifera cells from nonelicited cells by applying PCA to 1H NMR spectra obtained from crude extracts, which suggests that this method can be used for the characterization and monitoring of plant cells. Acknowledgement This Research was supported by the Chung-Ang University Research Grants in 2004. References [1] Endress R. Plant cells as producers of secondary compounds. In plant cell biotechnology Berlin: Springer-Verlag; 1994. pp. 187–255. [2] Choi HK, Son JS, Na KH, Hong SS, Park YS, Song JY. Mass production of paclitaxel by plant cell culture. Kor J Plant Biotechnol 2002;29:59–62. [3] Roberts SC, Shuler ML. Large-scale plant cell culture. Curr Opin Biotechnol 1997;8:154–9. [4] Choi HK, Yun JH, Kim SI, Song JY, Kim JH, Choi HJ, et al. Monitoring of FCW/DCW ratio, production of protein and peroxidase activity during suspension culture of Taxus chinensis. Kor J Biotechnol Bioeng 2000;5:525–8. [5] Raval KN, Hellwig S, Prakash G, Ramos-Plasencia A, Srivastava A, Buchs J. Necessity of a two-stage process for the production of azadirachtin-related limonoids in suspension cultures of Azadirachta indica. J Biosci Bioeng 2003;96:16–22. [6] Anderlei T, Zang W, Buechs J. Online respiration activity measurement (OTR, CTR, RQ) in shake flasks. Biochem Eng J 2004;17:187–94. [7] Fredlund E, Broberg A, Boysen ME, Kenne L, Schnurer J. Metabolite profiles of the biocontrol yeast Pichia anomala J121 grown under oxygen limitation. Appl Microb Biotechnol 2004;64:403–9.
[8] Ott KH, Aranibar N, Singh B, Stockton QW. Metabonomic classifies pathway affected by bioactive compounds. Artificial neural network classification of NMR spectra of plant extracts. Phytochemistry 2003;62:971–85. [9] Sumner LW, Mandes P, Dixon RA. Plant metabolomics: large-scale phytochemistry in the functional genomics era. Phytochemistry 2003; 62:817–36. [10] Vogels J, Terwel L, Tas AC, van den Berg F, Dukel F, van der Greef J. Detection of adulteration in orange juices by a new screening method using proton NMR spectroscopy in combination with pattern recognition techniques. J Agric Food Chem 1996;44:175–80. [11] Belton PS, Colquhoun IJ, Kemsley EK, Delgadillo I, Roma P, Dennis MJ, et al. Application of chemometrics to the 1H NMR spectra of apple juices: discrimination between apple varieties. Food Chem 1998;61: 207–13. [12] Gall GL, Puaud M, Colquhoun IJ. Discrimination between orange juice and pulp wash by 1H nuclear magnetic resonance spectroscopy: identification of marker compounds. J Agric Food Chem 2001;49:580–8. [13] Brescia MA, Caldarola V, de Giglio A, Benedetti D, Fanizzi FP, Sacco A. Characterization of the geographical origin of Italian red wine based on traditional and nuclear magnetic resonance spectrometric determinations. Anal Chim Acta 2002;458:177–86. [14] Charlton AJ, Farrington HH, Brereton P. Application of 1H NMR and multivariate statistics for screening complex mixtures: quality control and authenticity of instant coffee. J Agric Food Chem 2002;50:3098– 103. [15] Duarte I, Barros A, Belton P, Righelato R, Spraul M, Humpfer E, et al. High resolution nuclear magnetic resonance spectroscopy and multivariate analysis for the characterization of beer. J Agric Food Chem 2002;50:2475–81. [16] Ward JL, Harris C, Lewis J, Beale MH. Assessment of 1H NMR spectroscopy and multivariate analysis as a technique for metabolite fingerprinting of Arabidopsis thaliana. Phytochemistry 2003;62:949–58. [17] Choi HK, Choi YH, Verberne M, Lefeber AWM, Erkelens C, Verpoorte R. Metabolic fingerprinting of wild type and transgenic tobacco plants by 1H NMR and multivariate analysis technique. Phytochemistry 2004;65:857– 64. [18] Choi HK, Kim SI, Son JS, Hong SS, Lee HS, Lee HJ. Enhancement of paclitaxel production by temperature shift in suspension culture of Taxus chinensis. Enzyme Microb Technol 2000;27:593–8. [19] Zhang CH, Wu JY. Ethylene inhibitors enhance elicitor-induced paclitaxel production in suspension cultures of Taxus spp. cells. Enzyme Microb Technol 2003;32:71–7. [20] Murashige T, Skoog F. A revised medium for rapid growth and bioassays with tobacco tissue cultures. Physiol Plant 1962;15:473–9. [21] Eriksson L, Johansson E, Kettaneh-Wold N, Wold S. Multi- and megavariate data analysis Umea˚, Sweden: Umetrics Academy; 2001.
Short communication
Metabolomic profiling of Vitis vinifera cell suspension culture elicited with silver nitrate by 1H NMR spectrometry and principal components analysis Hyung-Kyoon Choi *, Jung-Hye Yoon College of Pharmacy, Chung-Ang University, Seoul 156-756, Republic of Korea Received 6 April 2006; accepted 11 July 2006
Abstract The metabolomic profiling of Vitis vinifera cell suspension cultures with and without silver nitrate was performed by 1H NMR (nuclear magnetic resonance) spectrometry and principal components analysis (PCA), to assess the efficacy of this method for the characterization and monitoring of plant cell lines. The PCA of the 1H NMR spectra of the aqueous fractions allowed a clear discrimination of V. vinifera cell suspension culture samples with and without silver nitrate treatment by the first three principal components (PC1, PC2, and PC3), which cumulatively accounted for 95.9% of the variation in all variables. In particular, the score plots by the combining PC1 versus PC2 and PC2 versus PC3 facilitated an excellent separation of samples. In addition, the major peaks in 1H NMR spectra contributing to the discrimination were assigned to lactate, alanine, acetic acid, choline, fructose, a-glucose, and sucrose. This method based on metabolomic analysis allows the efficient monitoring and the differentiation of normal cell suspension system from elicited systems without any prepurification steps. # 2006 Elsevier Ltd. All rights reserved. Keywords: Metabolomic profiling; Vitis vinifera; 1H NMR; Principal component analysis
1. Introduction Plant cell cultures may represent as an attractive alternative for overcoming the limitations of extracting useful metabolites from natural resources. Many factors affect cell growth and the production of metabolites in plant cell cultures, such as the components of the basal medium, the source of carbon, phytohormones, antimetabolites, O2, pH, elicitors, temperature, stirring frequency, and light conditions [1]. The stability of a cell line is especially important for the commercialization of the plant cell culturing process. Large-scale production of useful secondary metabolites has been attempted by various research groups [2,3], which requires the monitoring and maintaining of the cell states, because plant cell culture system are labile to external stresses that could alter the bioactivity in producing useful compounds. Bx, pH, conductivity, and the levels of protein and peroxidase have been suggested as monitoring methods [4], with the levels of nitrate, phosphate, sucrose, and glucose, and respiration activity having been used
* Corresponding author. Tel.: +82 2 820 5605; fax: +82 2 816 7338. E-mail address: [email protected] (H.-K. Choi). 1359-5113/$ – see front matter # 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.procbio.2006.07.007
as monitoring tools [5,6]. Intra- and extra-metabolome analyses have recently been performed in a microbial cell culture system [7], but there has been little work on plant cell culture systems. The term ‘metabolome’ has been used to describe the observable chemical profile or fingerprint of the metabolites present in whole tissues [8]. The chemical analysis techniques applied to metabolites profiling should be rapid, reproducible, and stable over time, while requiring only simple sample preparation. A technique that potentially meets these demands is NMR, which has been widely used as a fingerprinting tool for the interpretation and quality assessment of industrial and natural products, with multivariate or pattern recognition techniques such as the well-known principal components analysis (PCA) having been specifically designed to analyze complex data sets [9]. Recently, the combination of NMR and PCA has been applied to the metabolic profiling of various types of coffee, juice, wine, and beer and some plants [10–17]. In this study, we applied a 1H NMR spectroscopy method coupled with PCA to the metabolomic profiling of Vitis vinifera cell suspension culture elicited with silver nitrate, which was used as an elicitor for enhancing the production of paclitaxel [18,19] as a model system, with the aim of suggesting the standard trajectory plots and elucidating the major metabolites
272
H.-K Choi, J.-H Yoon / Process Biochemistry 42 (2007) 271–274
contributing to the discrimination between normal and elicited samples. This method can be used to characterize a plant cell suspension by supplying the standard trajectory plots for the stabilized production of specific secondary metabolites.
CHCl3 extracts and TSP for aqueous extracts, and reduced to integrated regions of equal width (0.04 ppm) corresponding to d = 0.50–10.00. The region from 4.60 to 4.90 was excluded from the analysis because of the residual signal of water in aqueous extracts, whereas that from 7.00 to 7.50 was excluded because of the residual signal of CHCl3 in organic fractions. PCA was performed with SIMCA-P software (Umetrics, Umea˚, Sweden).
2. Materials and methods
3. Results and discussion 2.1. Plant cell line and media V. vinifera cells were obtained from the Korean Collection for Type Cultures (Daejeon, Korea) and grown under darkness at 24 1 8C on a shaking incubator (Jeiotech SI-900R, Korea) operating at 150 rpm. The liquid medium contained the inorganic salt formulation of Murashige and Skoog [20], 30 g/l sucrose, and 1 mg/l 2,4-dichlorophenoxyacetic acid. The cells were subcultured every 12 days by adding 25 ml of inoculum into 75 ml of fresh medium in 250 ml flasks. Flasks were capped with sterilized aluminum foil and taped with Micropore surgical tape (3M 1530-1, USA). To elicit a cell suspension, 4 mM AgNO3 was added on day 6. All reagents for cell cultivation were purchased from Sigma, and all measurement were madein triplicate.
2.2. Cell mass preparation To determine the dry cell weight (DCW) and internal metabolomes by 1H NMR, cell suspensions were filtered through Whatman No. 541 filter paper on a Buchner funnel under a slight vacuum, and then washed with distilled water. The cells were placed on an aluminum weighing dish and dried in an oven at 70 8C until the weight was constant (corresponding to the DCW).
2.3. Solvents and chemicals for NMR measurement First-grade chloroform, methanol, and D2O (99.9%) were purchased from Sigma (St. Louis, MO, USA), and CDCl3 (99.8%) and NaOD were purchased from Cambridge Isotope Laboratories (Miami, FL, USA) and Cortec (Paris, France), respectively.
2.4. Extraction of V. vinifera cell suspension culture materials for NMR measurement
3.1. Assignments of the compounds from visual inspection of 1H NMR spectra There were no significant differences in values of DCW, pH, and Bx. did not differ significantly between normal cells and cells elicited with silver nitrate during the cultivation period (data not shown). Therefore a further experiment was performed to investigate the metabolome profile inside the cells. Little difference was observed between the spectra of the CHCl3 extracts of the various samples (data not shown), and so only the aqueous fractions were further analyzed. Representative 1H NMR spectra of the aqueous extracts are shown in Fig. 1. The NMR signals were smaller in the aromatic region (d = 6.0–8.0) than in the aliphatic and sugar regions. The signals of the main aromatic compounds were assigned to fumaric acid (at d = 6.52 (s)) (Fig. 1(c)). In addition, the following signals were assigned based on comparisons with the chemical shifts of standard compounds and two-dimensional NMR using 1H-1H COSY (correlation spectroscopy), HMQC (heteronuclear multiple quantum coherence), and HMBC (heteronuclear multiple bond coherence): lactate at d = 1.34 (d, J = 6.6 Hz), alanine at d = 1.48 (d, J = 7.3 Hz), acetic acid at d = 1.90 (s), choline at d = 3.22 (s), fructose at d = 4.22 (d, J = 8.7 Hz), b-glucose at d = 4.64 (d, J = 7.9 Hz), a-glucose at d = 5.24 (d, J = 3.9 Hz), and sucrose at d = 5.42 (d, J = 3.9 Hz) (Fig. 1(a) and (b)).
One hundred milligrams of fresh cells were transferred into a centrifuge tube. Five milliliters of a 50% water–methanol mixture and 5 ml of chloroform were added to the antler sample in the tube and vortexed for 30 s and sonicated for 1 min. The materials were then centrifuged at 3000 rpm for 20 min. The extraction was performed twice. The aqueous and organic fractions were transferred separately into a 50 ml round-bottomed flask and dried with a rotary vacuum evaporator. Each experiment was performed in triplicate.
2.5. NMR measurements KH2PO4 was added to D2O as a buffering agent. The pH of the D2O used for NMR measurements was adjusted to 6.0 using a 1N NaOD solution. All spectra were obtained by a NMR spectrometer (Avance 600 FT-NMR, Bruker, Germany) operating at a proton NMR frequency of 600.13 MHz. For each sample, 128 scans were recorded with the following parameters: 0.155 Hz/point, pulse width of 4.0 ms (308), and relaxation delay of 1.0 s. Free induction decays were Fourier transformed with LB = 0.3 Hz, GB = 0, and PC = 1.0. The spectra were referenced to trimethyl silane propionic acid sodium salt (TSP) at 0.00 ppm for aqueous fractions and, for CHCl3 fractions, to residual solvent at 7.26 ppm. Hexamethyl disilane (HMDS, 0.01%, v/v) and TSP (0.01%, w/v) were used as internal standards for CDCl3 and D2O, respectively. The peak intensities in 0.04 ppm bins in the 1H NMR spectra for d = 0.50–10.00 were used as variables.
2.6. Data analysis The 1H NMR spectra were automatically reduced to ASCII files using AMIX (v. 3.7, Biospin, Bruker). Spectral intensities were scaled to HMDS for
Fig. 1. Representative 1H NMR spectra of the total (a), aliphatic (b), and aromatic (c) regions of the aqueous fraction of V. vinifera cell suspension culture samples. IS, internal standard; w, residual water; 1, lactate; 2, alanine; 3, acetic acid; 4, choline; 5, fructose; 6, b-glucose; 7, a-glucose; 8, sucrose; 9, fumaric acid. Values on the X-axis are the d values.
H.-K Choi, J.-H Yoon / Process Biochemistry 42 (2007) 271–274
273
Fig. 2. Score plots from PCA of aqueous extracts of V. vinifera cell suspension cultures with the cultivation time by the combination of PC1 and PC2 (a), PC1 and PC3 (b), and PC2 and PC3 (c). The ellipses represent confidences of 95% in Hotelling T2 tests.
3.2. Principal components analysis To ensure the objective interpretation of the results, the samples were also analyzed using PCA. PCA is an unsupervised clustering method (in that it does not require any knowledge of the data set) that attempts to reduce the dimensionality of multivariate data while preserving most of the variance therein [21]. The present study applied the covariance method for PCA because it produced a better separation than the correlation method (data not shown). As evident in Fig. 2, the elicited samples of the V. vinifera cells elicited with silver nitrate were clearly distinguishable from nonelicited cells, with the first two principal components cumulatively accounting for 82.1% of the variance. Silver nitrate has been used as an elicitor of the improved production of paclitaxel in Taxus chinensis suspension culture, with the mechanism of action known to be ethylene inhibition [18,19]. An excellent separation between samples in
score plots was achieved by combining principal component 1 (PC1) with principal component 2 (PC2). In the score plot of the combination of PC1 and PC2, samples from V. vinifera cells with and without elicitation with silver nitrate showed a different trajectory with the culture time. In addition, the combination of PC2 and principal component 3 (PC3) showed a clearly separate trajectory between cells with and without silver nitrate treatment. As shown in Fig. 2(c), the control samples formed a left-side trajectory plots, while the samples with silver nitrates treatment formed right-side trajectory plots. Elicitation with silver nitrate was conducted on day 6, and the plots at day 7 already started to show slight discrimination, with clear differentiation being present from day 11. PCA of the aliphatic region (from 0.50 to 3.3 ppm) produced no useful information (data not shown). The discriminating metabolites are clearly distinguishable in the loading plots for PC1, PC2, and PC3 (Fig. 3), with the score and loading plots complementing each other. The position of
Fig. 3. PCA loading plots of aqueous extracts of V. vinifera cell suspension cultures with the cultivation time by PC1 (a), PC2 (b), and PC3 (c).
274
H.-K Choi, J.-H Yoon / Process Biochemistry 42 (2007) 271–274
objects in a given direction in a score plot is influenced by variables lying in the same direction in the loading plot. The major compounds contributing to the discrimination by PC1 were lactate, acetic acid, choline, fructose, and sucrose; lactate, acetic acid, choline, and a-glucose contributed to the discrimination by PC2; acetic acid, choline, fructose and sucrose contributed to the discrimination by PC3 (Fig. 3). Although there were a few peaks in the aromatic region (such as for fumaric acid), the peaks did not contribute to the discrimination of the samples with the cultivation time. This study has shown that it is possible to discriminate elicited V. vinifera cells from nonelicited cells by applying PCA to 1H NMR spectra obtained from crude extracts, which suggests that this method can be used for the characterization and monitoring of plant cells. Acknowledgement This Research was supported by the Chung-Ang University Research Grants in 2004. References [1] Endress R. Plant cells as producers of secondary compounds. In plant cell biotechnology Berlin: Springer-Verlag; 1994. pp. 187–255. [2] Choi HK, Son JS, Na KH, Hong SS, Park YS, Song JY. Mass production of paclitaxel by plant cell culture. Kor J Plant Biotechnol 2002;29:59–62. [3] Roberts SC, Shuler ML. Large-scale plant cell culture. Curr Opin Biotechnol 1997;8:154–9. [4] Choi HK, Yun JH, Kim SI, Song JY, Kim JH, Choi HJ, et al. Monitoring of FCW/DCW ratio, production of protein and peroxidase activity during suspension culture of Taxus chinensis. Kor J Biotechnol Bioeng 2000;5:525–8. [5] Raval KN, Hellwig S, Prakash G, Ramos-Plasencia A, Srivastava A, Buchs J. Necessity of a two-stage process for the production of azadirachtin-related limonoids in suspension cultures of Azadirachta indica. J Biosci Bioeng 2003;96:16–22. [6] Anderlei T, Zang W, Buechs J. Online respiration activity measurement (OTR, CTR, RQ) in shake flasks. Biochem Eng J 2004;17:187–94. [7] Fredlund E, Broberg A, Boysen ME, Kenne L, Schnurer J. Metabolite profiles of the biocontrol yeast Pichia anomala J121 grown under oxygen limitation. Appl Microb Biotechnol 2004;64:403–9.
[8] Ott KH, Aranibar N, Singh B, Stockton QW. Metabonomic classifies pathway affected by bioactive compounds. Artificial neural network classification of NMR spectra of plant extracts. Phytochemistry 2003;62:971–85. [9] Sumner LW, Mandes P, Dixon RA. Plant metabolomics: large-scale phytochemistry in the functional genomics era. Phytochemistry 2003; 62:817–36. [10] Vogels J, Terwel L, Tas AC, van den Berg F, Dukel F, van der Greef J. Detection of adulteration in orange juices by a new screening method using proton NMR spectroscopy in combination with pattern recognition techniques. J Agric Food Chem 1996;44:175–80. [11] Belton PS, Colquhoun IJ, Kemsley EK, Delgadillo I, Roma P, Dennis MJ, et al. Application of chemometrics to the 1H NMR spectra of apple juices: discrimination between apple varieties. Food Chem 1998;61: 207–13. [12] Gall GL, Puaud M, Colquhoun IJ. Discrimination between orange juice and pulp wash by 1H nuclear magnetic resonance spectroscopy: identification of marker compounds. J Agric Food Chem 2001;49:580–8. [13] Brescia MA, Caldarola V, de Giglio A, Benedetti D, Fanizzi FP, Sacco A. Characterization of the geographical origin of Italian red wine based on traditional and nuclear magnetic resonance spectrometric determinations. Anal Chim Acta 2002;458:177–86. [14] Charlton AJ, Farrington HH, Brereton P. Application of 1H NMR and multivariate statistics for screening complex mixtures: quality control and authenticity of instant coffee. J Agric Food Chem 2002;50:3098– 103. [15] Duarte I, Barros A, Belton P, Righelato R, Spraul M, Humpfer E, et al. High resolution nuclear magnetic resonance spectroscopy and multivariate analysis for the characterization of beer. J Agric Food Chem 2002;50:2475–81. [16] Ward JL, Harris C, Lewis J, Beale MH. Assessment of 1H NMR spectroscopy and multivariate analysis as a technique for metabolite fingerprinting of Arabidopsis thaliana. Phytochemistry 2003;62:949–58. [17] Choi HK, Choi YH, Verberne M, Lefeber AWM, Erkelens C, Verpoorte R. Metabolic fingerprinting of wild type and transgenic tobacco plants by 1H NMR and multivariate analysis technique. Phytochemistry 2004;65:857– 64. [18] Choi HK, Kim SI, Son JS, Hong SS, Lee HS, Lee HJ. Enhancement of paclitaxel production by temperature shift in suspension culture of Taxus chinensis. Enzyme Microb Technol 2000;27:593–8. [19] Zhang CH, Wu JY. Ethylene inhibitors enhance elicitor-induced paclitaxel production in suspension cultures of Taxus spp. cells. Enzyme Microb Technol 2003;32:71–7. [20] Murashige T, Skoog F. A revised medium for rapid growth and bioassays with tobacco tissue cultures. Physiol Plant 1962;15:473–9. [21] Eriksson L, Johansson E, Kettaneh-Wold N, Wold S. Multi- and megavariate data analysis Umea˚, Sweden: Umetrics Academy; 2001.