- Email: [email protected]
Glutamine synthetase activity in Solanaceous cell suspensions accumulating alkaloids or not. 13C NMR and enzymatic assay
© 1999 Académie des sciences/Éditions scientifiques et médicales Elsevier SAS. Tous droits réservés
Biochemistry / Biochimie
Glutamine synthetase activity in Solanaceous cell suspensions accumulating alkaloids or not. 13 C NMR and enzymatic assay Activité de la glutamine synthétase dans des suspensions cellulaires de solanacées productrices ou non d’alcaloïdes. RMN du 13C et dosage enzymatique François Mesnarda, Danielle Martya, Françoise Gillet-Manceaub, Marc-André Fliniauxb, Jean-Pierre Montia* a
Laboratoire de biophysique, groupe de recherche des biomolécules : micro-environnement et métabolisme, faculté de pharmacie, 1, rue des Louvels, 80037 Amiens cedex 1, France b Laboratoire de phytotechnologie, faculté de pharmacie, 1, rue des Louvels, 80037 Amiens cedex 1, France (Received 1 March 1999; accepted 10 May 1999) Note communicated by Michel Thellier
Abstract — The metabolism of labelled pyruvate followed by 13C NMR and the measure of glutamine synthetase (GS) showed, according to previous results, a high activity of this enzyme in suspension cells of Nicotiana plumbaginifolia. This activity could derive glutamate from the alkaloid synthesizing pathways. However, a recent work showed that the rate of the GS gene transcription was inversely proportional to the Gln/Glu ratio. The measures of Gln and Glu concentrations in Nicotiana plumbaginifolia cells revealed that high GS activity correlates with the weak value of Gln/Glu ratio. Therefore, the hypothesis of GS dysfunctioning for the non-biosynthesis of alkaloids in N. plumbaginifolia suspension cells can be discarded. This conclusion is strengthened by the results obtained when using a GS inhibitor. © 1999 Académie des sciences/Éditions scientifiques et médicales Elsevier SAS alkaloids / glutamine synthetase / NMR / Solanaceous
Résumé — Le suivi métabolique du pyruvate marqué par RMN du carbone 13 et la mesure de l’activité enzymatique de la glutamine synthétase (GS) dans des cellules de Nicotiana plumbaginifolia ont montré une activité élévée de cette enzyme, confirmant des résultats précédemment obtenus. Ainsi, cette activité aurait pu dévier le glutamate de la voie de biosynthèse des alcaloïdes dans ces cellules. Cependant, un travail récent montre que le taux de transcription du gène codant la GS est inversement proportionnel à la valeur du rapport Gln/Glu. Les mesures de ces concentrations effectuées dans les cellules de Nicotiana plumbaginifolia montrent que cette forte activité enzymatique est en accord avec la faible valeur du rapport Gln/Glu. Ainsi l’hypothèse d’un dysfonctionnement de la GS, empêchant la synthèse d’alcaloïdes doit être écartée. Cette conclusion est renforcée par les résultats obtenus en utilisant un inhibiteur de la GS. © 1999 Académie des sciences/Éditions scientifiques et médicales Elsevier SAS alcaloïdes / glutamine-synthétase / RMN / solanacées
* Correspondence and reprints: [email protected],fr C. R. Acad. Sci. Paris, Sciences de la vie / Life Sciences 1999, 322, 743–748
743
F. Mesnard et al.
BAP, 6-benzylaminopurine; 2,4 D, 2,4 dichlorophenoxyacetic acid; GABA, 4-aminobutyric acid; GS, glutamine synthetase; MSX, L-methionine sulphoximine; NAA, 1-naphtylacetic acid.
Version abrégée Certaines plantes qui produisent dans la nature des alcaloïdes perdent cette propriété lorsqu’elles sont mises en culture in vitro. Une raison susceptible d’expliquer cette absence de production pourrait être un dysfonctionnement enzymatique au niveau du métabolisme primaire et/ou du métabolisme secondaire. Nous avons étudié cette hypothèse en comparant les métabolismes de deux suspensions cellulaires de solanacées, l’une productrice d’alcaloïdes (Datura stramonium), l’autre pas (Nicotiana plumbaginifolia). Le suivi par RMN du métabolisme du [2-13C] acétate, relaté dans un article précédent, a permis de montrer une haute activité de la glutamine synthétase chez Nicotiana plumbaginifolia, qui pourrait être à l’origine de l’absence de production d’alcaloïdes dans ces cellules. Le suivi par RMN du métabolisme du [3-13C] pyruvate ainsi que le dosage de l’activité enzymatique ont confirmé cette haute activité (act.GS N. plumbaginifolia = 158 nkat/mg de protéines; act.GS D. stramonium = 60 nkat/mg de protéines). Watanabe et al. ont récemment montré qu’il existe une relation entre la valeur du rapport Gln/Glu et le niveau de transcription du gène codant pour la GS. En effet, un rapport élevé implique un taux bas de transcription du gène, donc une faible quantité de GS produite et inversement. Les mesures des concentrations de glutamate et de glutamine par chromatographie échangeuse d’ions nous ont permis
1. Introduction In plant cells, the activity of the Krebs cycle leads to the biosynthesis of glutamate, which is then converted to arginine and ornithine, precursors in alkaloid biosynthesis (figure 1). Several authors have shown that the pyrrolidine ring of tobacco alkaloids and the tropane ring of tropane alkaloids derive from ornithine and/or arginine via putrescine [1–5]. In numerous undifferentiated plant cell cultures from Solanaceous species, this metabolic pathway is at least partially inhibited and does not lead to the biosynthesis of alkaloids. Several hypotheses have been suggested to explain this inhibition of the secondary metabolism [6]. Our previous work, using [2-13C] acetate, indicated a larger accumulation of glutamine in nonalkaloid-producing Nicotiana plumbaginifolia suspension culture cells than in tropane-alkaloid-accumulating Datura stramonium [7–9]. Thus, a first hypothesis to explain the non-biosynthesis of alkaloids by the N. plumbaginifolia cells would be that a high glutamine synthetase (GS, EC 6.3.1.2) activity leads to a very low concentration of glutamate [9]. Glutamine synthetase catalyses glutamine formation from glutamate and ammonium in the presence of ATP.
744
d’établir les rapports suivants: 5,9 pour N. plumbaginifolia et 9,1 pour D. stramonium. Les valeurs des deux rapports confirment les résultats du dosage enzymatique: le plus faible rapport, mesuré chez N. plumbaginifolia est en accord avec une plus forte transcription du gène codant pour la GS. L’hypothèse d’une déviation, par forte activité de la GS, du métabolisme du glutamate vers la glutamine, impliquant un faible taux de glutamate disponible pour la formation de précurseurs nécessaires à la synthèse d’alcaloïdes, n’est donc pas vérifiée. Pour conforter cette conclusion, nous avons effectué des suivis métaboliques en présence de méthionine sulfoximine (MSX), inhibiteur de la GS. Le rapport Gln/Glu a alors été inversé. Néanmoins, il n’a été mis en évidence ni production d’alcaloïdes (analyse par HPLC), ni formation d’intermédiaires du métabolisme secondaire (par résonance magnétique nucléaire du carbone 13). Ces résultats renforcent la conclusion que la haute activité de la GS chez N. plumbaginifolia n’est pas responsable de l’absence de synthèse d’alcaloïdes dans ces cellules. Les expériences effectuées avec l’inhibiteur ont permis, par ailleurs, de montrer qu’en accord avec les données de la littérature, des conditions de stress - présence de MSX dans notre cas - peuvent induire la formation de GABA à l’intérieur des cellules végétales, le GABA pouvant jouer un rôle de réserve d’azote utilisable pour la formation des acides aminés, lorsque le passage glutamate-glutamine est bloqué.
Glutamate/glutamine couple is involved in ammonia assimilation via glutamine synthetase and glutamine 2-oxoglutarate amino transferase, which reductively converts glutamine and 2-oxoglutarate to two molecules of glutamate. The role of glutamate dehydrogenase as another possible way of ammonia assimilation is at present very controversial (figure 1) [10, 11]. Watanabe et al. showed that the internal Gln/Glu ratio was a potential regulatory parameter for the expression of a glutamine synthetase gene [12]. Cells accumulating glutamine at high levels exhibited low levels of GS owing to low levels of gene transcript. The transcript level was, however, not regulated in response to the glutamine level alone but in response to the ratio of cellular glutamine to glutamate. Therefore, an unusually high GS activity due to modifications of the glutamine/glutamate ratio could be a first explanation for the non-biosynthesis of alkaloids. In order to investigate this hypothesis, we measured the GS enzymatic activity, followed the acetate and pyruvate metabolism by 13C NMR using [3-13C] pyruvate and [2-13C] acetate in suspension cells and realized cultures using L-methionine sulphoximine (MSX) [13, 14] to block the glutamate/glutamine conversion and to try to deviate glutamate towards alkaloid biosynthesis. C. R. Acad. Sci. Paris, Sciences de la vie / Life Sciences 1999, 322, 743–748
GS activity, 13C NMR and enzymatic assay
2.3. Feeding experiments
To follow metabolism in plant cells, culture media were supplemented with sodium [2-13C] acetate or sodium [3-13C] pyruvate (CEA, Saclay) or non-labelled sodium acetate or sodium pyruvate as controls, to be sure that the observed carbon resonances were due to the insertion of 13 C-labelled compounds into the various metabolites. Final acetate and pyruvate concentrations were 4 mM in all cases. In a previous work, it was verified that the 4-mM concentration has no inhibitory effect on cell growth. Samples were treated as described previously [9]. 2.4. NMR experiments
Carbon-13 NMR spectroscopy was performed on a Brucker AM 300 NMR spectrometer at 75.47 MHz. For cell extracts, all spectra were obtained in identical experimental conditions: temperature 22 °C, spectral width 18 kHz for 16 K data points, acquisition time 0.459 s, pulse delay 0.75 s and pulse angle equal to 60° with 1H composite pulse decoupling. Before Fourier transformation in absolute intensity mode, a zerofilling to 32 K was applied. An artificial line broadening of 10 Hz was used to improve the spectral signal-to-noise ratio. Spectra were acquired for 48 000 scans and resonance chemical shifts were relative to an external capillary solution of sodium formate (172.0 ppm). For culture media, the experimental conditions were identical except that only 2 000 scans were acquired. Insertion of 13C into the various metabolites was evaluated by differences between spectra with labelled and non-labelled precursors. The intensities of different resonances were expressed as signal-to-noise ratio (S/N) with the same vertical expansion relative to the external reference. 2.5. Extraction and enzyme assay of glutamine synthetase
Figure 1. Pathway showing the biosynthesis of alkaloids from 2-oxoglutarate. GDH = glutamate deshydrogenase; GOGAT = glutamine 2-oxoglutarate amino transferase; GS = glutamine synthetase; PMT = putrescine methyl transferase.
2. Materials and methods 2.1. Culture conditions
N. plumbaginifolia and D. stramonium cell suspensions were cultured as described previously [9]. In the GS inhibitor experiments, MSX (Sigma) was added to the growth medium after a 0.2-lm sterile filtration (Dynacard-Merck). To evaluate its toxicity, MSX was used at various concentrations. Optimal conditions were for 5 mM MSX. 2.2. Cell viability
To determine MSX toxicity, cell viability was assessed using fluorescein diacetate as described [15]. C.R. Acad. Sci. Paris, Sciences de la vie / Life Sciences 1999, 322, 743–748
GS cell-free extracts were prepared by grinding 10 g fresh matter of liquid-nitrogen-frozen material in 5 mL of buffer containing 50 mM TRIS-HCl (pH 7.8), 1 mM EDTA, 1 mM DTT, 10 mM MgSO4, 5 mM sodium glutamate and 10 % ethanediol, as described by Lea [10]. Following centrifugation at 10 000 g at 4 °C for 60 min, the supernatant was concentrated and desalted with CENTRIPLUST 100 (AMICON, INC, USA) concentrators at 4 °C and 3 000 g for 130 min. GS activity was measured by the formation of c-glutamyl hydroxamate in the synthetase reaction [16]. 2.6. Protein determination
The soluble proteins in cell-free extracts were determined according to Bradford, with bovine serum albumin as the standard [17]. 2.7. Amino-acid and alkaloid analysis
The amino-acids glutamine and glutamate were determined on a Beckman model automated analyser using
745
F. Mesnard et al.
Figure 2. Incorporation of label into (●) [4-13C] glutamine and (•) [4-13C] glutamate in N. plumbaginifolia cells as a function of time with 4 mM [2-13C] acetate (A) or with 4 mM [3-13C] pyruvate (B). [4-13C] glutamate is not detected in experiments with [3-13C] pyruvate because of a weak incorporation (see Discussion). Carbons 2-C and 3-C give similar curves; only variations in signal intensity have been noted. Examination of the inset (A) shows glutamate labelling to be later than for glutamine.
postcolumn derivatization with ninhydrin. The eluate was monitored at 570 and 440 nm, after elimination of proteins and nucleic acid [18]. The tobacco alkaloids were quantified using reversed-phase HPLC as described previously [19].
3. Results The in vitro activity of GS in desalted cell-free extracts was 158 ± 11 nkat mg–1 protein. The GS activity detected in extracts derived from D. stramonium cells was 60 ± 28 nkat mg–1 protein. The assays were performed immediately after extraction. The difference in activities is significant (Student’s paired t-test; P < 0.05). The quantitation of free amino-acids by HPLC showed an amino-acid level in N. plumbaginifolia cells much higher than in D. stramonium cells with Gln/Glu ratios of 5.9 ± 0.3 and 9.1 ± 0.4, respectively (Student’s paired t-test; P < 0.05). Note that the difference in the amino-acid level was in agreement with our previous results of a weak labelling from [2-13C] acetate in D. stramonium cells and a more intense incorporation by N. plumbaginifolia cells [9]. In the experimental conditions of cell culture, no tobacco alkaloid was detected by HPLC analysis. The labelled compounds observed in [3-13C] pyruvate experiments are identical to those obtained in [2-13C] acetate experiments [9] with, in addition, alanine, asparagine or aspartate and lactate (data not shown). Figure 2A and B shows the incorporation of label into C-4 glutamate and glutamine as a function of time after the uptake of
746
[2-13C] acetate and [3-13C] pyruvate, respectively, by N. plumbaginifolia cells. Note the weak labelling of glutamine when pyruvate was used as labelled precursor. In experiments with inhibitor, MSX treatment of N. plumbaginifolia cells with 4 mM [2-13C] acetate for 9 h (figure 3) resulted in labelling the pools of glutamine, glutamate, GABA, malate and citrate. The four resonances observed from MSX (54.0, 52.9, 42.0, 25.0 ppm) confirmed that the inhibitor had been taken up by the cells. The inset of figure 3 shows the incorporation of label into the C-4 of GABA as a function of time after the uptake of [2-13C] acetate by N. plumbaginifolia cells in the presence or absence of MSX. It should be emphasized that 5 mM MSX led to a slowing down of metabolism, characterized by weak labelling observed in 13C NMR spectra (inset figure 3) and a much reduced rate of acetate uptake, with an apparent uptake constant equal to (1.9 ± 0.3)10–3 h–1 versus (60 ± 10)10–3 h–1 in previous data [9]. The incorporation of label into [4-13C] glutamate and [4-13C] glutamine as a function of time after the uptake of [2-13C] acetate by N. plumbaginifolia cells cultured in presence of MSX is shown in figure 4.
4. Discussion The NMR studies presented here verify that, as previously reported [9], the experimental conditions used did not modify cellular viability and acetate or pyruvate uptake by N. plumbaginifolia suspension cultures. The advantage of exhibiting pyruvate compared to acetate is that it is incorporated more directly in other C. R. Acad. Sci. Paris, Sciences de la vie / Life Sciences 1999, 322, 743–748
GS activity, 13C NMR and enzymatic assay
Figure 3. 13C NMR spectrum (expansion from 10 to 60 ppm) of N. plumbaginifolia cells following treatment with 4 mM [2-13C] acetate and 5 mM L-methionine sulfoximine for 20 h showing labelled compound resonances. The four resonances of MSX inhibitor are detected (noted I). The inset shows the evolution of [4-13C] GABA as a function of time with (●) or without (•) MSX. Carbons 2-C and 3-C give similar curves; only variations in signal intensity have been noted.
metabolites, for example alanine by transamination or asparagine via oxaloacetate, permitting a more rapid and focused analysis of other metabolic pathways. Pyruvate enters the Krebs cycle, not only like acetate, via acetylCoA, but also via oxaloacetate or malate [20]. Thus, by using [3-13C] pyruvate we were able to observe the formation of alanine, asparagine or aspartate and lactate, produced in cells by reduction of pyruvate (data not shown). The minor disadvantage of using [3-13C] pyruvate was a relatively weak incorporation of label into the glutamate/glutamine pathway with no labelling of glutamate and only a weak labelling of glutamine (figure 2A versus B). Except for these observations, we observed similar incorporations with the two labelled precursors. Watanabe et al. [12] have proposed that the internal Gln/Glu ratio may act as a potential regulator of the expression of the GS gene. The measure of the GS activity in N. plumbaginifolia and D. stramonium cells shows an activity of GS from N. plumbaginifolia cells greater than that from D. stramonium cells, confirming the previous conclusions [9]. Thus, two hypotheses can be suggested in N. plumbaginifolia cells: 1) there is a high level of gene transcripts in response to a low Gln/Glu ratio, or 2) a low C.R. Acad. Sci. Paris, Sciences de la vie / Life Sciences 1999, 322, 743–748
Figure 4. Incorporation of label into (●) [4-13C] glutamine and (•) [4-13C] glutamate in N. plumbaginifolia cells as a function of time with 4 mM [2-13C] acetate and 5 mM L-methionine sulphoximine. Carbons 2-C and 3-C give similar curves, only variations in signal intensity have been noted.
747
F. Mesnard et al.
level of gene transcripts in response to a high Gln/Glu ratio. In the former case, the high activity of GS in N. plumbaginifolia cells should be considered as a normal response; in the latter it should be considered as an abnormal response with a dysfunctioning of the GS regulatory systems. The Gln/Glu ratio in N. plumbaginifolia cells compared to the ratio in D. stramonium cells is in agreement with the first hypothesis. In these conditions, the non-biosynthesis of alkaloids should not be due to the high enzymatic activity of GS, the level of glutamate being high enough to lead to the biosynthesis of arginine and ornithine, available to be converted into putrescine. To confirm these data, we induced metabolic disruption with the GS inhibitor, MSX. We can note the inversion of the glutamine/glutamate ratio (figure 2A versus figure 4) and a 13C enrichment of the GABA resonances (see inset figure 3). This observation is in agreement with previous
results [20, 21]. GABA is formed by the decarboxylation of glutamate by glutamate decarboxylase, and increases in GABA content and glutamate decarboxylase activity have been seen in plants under a variety of environmental stress conditions, including MSX treatment. Generally, it appears that GABA biosynthesis is enhanced when the glutamate to glutamine conversion cannot occur. GABA can have a role in nitrogen mobilization and can be viewed as a temporary nitrogen storage compound for protein amino-acids in plants [20–23]. From these results, the glutamate concentration is probably sufficiently elevated to be used for alkaloids biosynthesis. Therefore, the hypothesis of GS dysfunctioning for the non-biosynthesis of alkaloids in N. plumbaginifolia suspension cells can be discarded. Moreover, HPLC analyses confirmed the NMR studies, no tobacco alkaloids being detected by either method.
Acknowledgements: We thank Dr R.J. Robins (LAIEM, CNRS UPRES-A 6006, Nantes) for critical comments on the manuscript.
References [1] Leete E., Alkaloids derived from ornithine, lysine and nicotinic acid, in: Bell E.A., Charlwood B.V. (Eds.), Encyclopedia of Plant Physiology, Springer Verlag, Berlin, 1980, pp. 65–91. [2] Leete E., Recent developments in the biosynthesis of the tropane alkaloids, Planta Med. 56 (1990) 339–352. [3] Leete E., Endo T., Yamada Y., Biosynthesis of nicotine and scopolamine in a root culture of Duboisia lechhardtii, Phytochemistry 29 (1990) 1847–1851. [4] Robins R.J., Parr A.J., Bent E.G., Rhodes M.J.C., Studies on the formation of tropane alkaloids in Datura stramonium L. transformed root cultures, 1. The kinetics of alkaloid production and the influence of feeding intermediate metabolites, Planta 183 (1991) 185–195. [5] Robins R.J., Parr A.J., Walton N.J., Studies on the formation of tropane alkaloids in Datura stramonium L. transformed root cultures, 2. On the relative contributions of L-arginine and L-ornithine to the formation of the tropane ring, Planta 183 (1991) 196–201. [6] Berlin J., Secondary products from plant cell cultures, in: Rhem H.J., Reed G. (Eds.), Biotechnology, Verl. Chemie, Basel, 1986, pp. 629–658. [7] Manceau F., Fliniaux M.A., Jacquin A., Ability of a Nicotiana plumbaginifolia cell suspension to demethylate nicotine into nornicotine, Phytochemistry 28 (1989) 2671–2674. [8] Gontier E., Fliniaux M.A., Barbotin J.N., Sangwan-Norreel B.S., Tropane Alkaloid levels in the leaves of micropropagated Datura innoxia Mill., Plant. Planta Med. 59 (1993) 432–435. [9] Marty D., Mesnard F., Gillet-Manceau F., Fliniaux M.A., Monti J.P., Changes in primary metabolism in connection with alkaloid biosynthesis in Solanaceous cell suspensions. A 13C NMR study, Plant Sci. 122 (1997) 11–21. [10] Lea P.J., Enzymes of ammonia assimilation, in: Dey P.M., Harborne J.B. (Eds.), Methods in Plant Biochemistry, Academic Press, London, 1990, pp. 257–276. [11] Lea P.J., Primary nitrogen metabolism, in: Dey P.M., Harborne J.B. (Eds.), Plant Biochemistry, Academic Press, London, 1997, pp. 273–313.
748
[12] Watanabe A., Tagaki N., Hayashi H., Chino M., Watanabe, A., Internal Gln/Glu ratio as a potential regulatory parameter for the expression of a cytosolic glutamine synthetase gene of radish in cultured cells, Plant Cell Physiol. 38 (1997) 1000–1006. [13] Leason M., Cunliffe D., Parkin D., Lea P.J., Miflin, B.J., Inhibition of pea leaf glutamine synthetase by methionine sulphoximine, phosphinotricine and other glutamate analogues, Phytochemistry 21 (1982) 855–857. [14] Kumar P.A., Abrol, Y.P., Effect of L-methionine sulphoximine on the enzymes of nitrogen metabolism in barley leaves, Proc. Indian Acad. Sci. 100 (1990) 97–100. [15] Dixon R.A., Plant Cell Culture: A Practical Approach, IRL Press, Oxford, 1985. [16] Rhodes D., Rendon G.A., Stewart, G.R., The control of glutamine synthetase level in Lemna minor L., Planta 125 (1975) 201–211. [17] Bradford M.M., A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein dye binding, Anal. Biochem. 72 (1976) 248–254. [18] Acil Y., Müller P.K., Rapid method for the isolation of the mature collagen cross-links, hydroxylysylpyridinoline and lysylpyridinoline, J. Chrom. A 664 (1994) 183–188. [19] Manceau F., Fliniaux M.A., Jacquin, A., A high-performance liquid chromatographic procedure for the analysis of tobacco alkaloids – Application to the evaluation of tobacco alkaloids in plants and cell suspension cultures, Phytochem. Anal. 3 (1992) 223–226. [20] Robinson T., The Organic Constituents of Higher Plants, Cordus Press, North Amherst, 1991. [21] Desmaison A.M., Tixier M., Amino acids content in germinating seeds and seedlings from Castanea sativa L., Plant Physiol. 81 (1986) 692–695. [22] Jordan B.R., Givan C.V., Effects of light and inhibitors on glutamate metabolism in leaf disc of Vicia faba L sources of ATP for glutamine synthesis and photoregulation of tricarboxylic acid cycle metabolism, Plant Physiol. 64 (1979) 1043–1047. [23] Narayan V.S., Nair, P.M., Metabolism, enzymology and possible roles of 4-Aminobutyrate in higher plants, Phytochemistry 29 (1990) 367–375.
C. R. Acad. Sci. Paris, Sciences de la vie / Life Sciences 1999, 322, 743–748
Biochemistry / Biochimie
Glutamine synthetase activity in Solanaceous cell suspensions accumulating alkaloids or not. 13 C NMR and enzymatic assay Activité de la glutamine synthétase dans des suspensions cellulaires de solanacées productrices ou non d’alcaloïdes. RMN du 13C et dosage enzymatique François Mesnarda, Danielle Martya, Françoise Gillet-Manceaub, Marc-André Fliniauxb, Jean-Pierre Montia* a
Laboratoire de biophysique, groupe de recherche des biomolécules : micro-environnement et métabolisme, faculté de pharmacie, 1, rue des Louvels, 80037 Amiens cedex 1, France b Laboratoire de phytotechnologie, faculté de pharmacie, 1, rue des Louvels, 80037 Amiens cedex 1, France (Received 1 March 1999; accepted 10 May 1999) Note communicated by Michel Thellier
Abstract — The metabolism of labelled pyruvate followed by 13C NMR and the measure of glutamine synthetase (GS) showed, according to previous results, a high activity of this enzyme in suspension cells of Nicotiana plumbaginifolia. This activity could derive glutamate from the alkaloid synthesizing pathways. However, a recent work showed that the rate of the GS gene transcription was inversely proportional to the Gln/Glu ratio. The measures of Gln and Glu concentrations in Nicotiana plumbaginifolia cells revealed that high GS activity correlates with the weak value of Gln/Glu ratio. Therefore, the hypothesis of GS dysfunctioning for the non-biosynthesis of alkaloids in N. plumbaginifolia suspension cells can be discarded. This conclusion is strengthened by the results obtained when using a GS inhibitor. © 1999 Académie des sciences/Éditions scientifiques et médicales Elsevier SAS alkaloids / glutamine synthetase / NMR / Solanaceous
Résumé — Le suivi métabolique du pyruvate marqué par RMN du carbone 13 et la mesure de l’activité enzymatique de la glutamine synthétase (GS) dans des cellules de Nicotiana plumbaginifolia ont montré une activité élévée de cette enzyme, confirmant des résultats précédemment obtenus. Ainsi, cette activité aurait pu dévier le glutamate de la voie de biosynthèse des alcaloïdes dans ces cellules. Cependant, un travail récent montre que le taux de transcription du gène codant la GS est inversement proportionnel à la valeur du rapport Gln/Glu. Les mesures de ces concentrations effectuées dans les cellules de Nicotiana plumbaginifolia montrent que cette forte activité enzymatique est en accord avec la faible valeur du rapport Gln/Glu. Ainsi l’hypothèse d’un dysfonctionnement de la GS, empêchant la synthèse d’alcaloïdes doit être écartée. Cette conclusion est renforcée par les résultats obtenus en utilisant un inhibiteur de la GS. © 1999 Académie des sciences/Éditions scientifiques et médicales Elsevier SAS alcaloïdes / glutamine-synthétase / RMN / solanacées
* Correspondence and reprints: [email protected],fr C. R. Acad. Sci. Paris, Sciences de la vie / Life Sciences 1999, 322, 743–748
743
F. Mesnard et al.
BAP, 6-benzylaminopurine; 2,4 D, 2,4 dichlorophenoxyacetic acid; GABA, 4-aminobutyric acid; GS, glutamine synthetase; MSX, L-methionine sulphoximine; NAA, 1-naphtylacetic acid.
Version abrégée Certaines plantes qui produisent dans la nature des alcaloïdes perdent cette propriété lorsqu’elles sont mises en culture in vitro. Une raison susceptible d’expliquer cette absence de production pourrait être un dysfonctionnement enzymatique au niveau du métabolisme primaire et/ou du métabolisme secondaire. Nous avons étudié cette hypothèse en comparant les métabolismes de deux suspensions cellulaires de solanacées, l’une productrice d’alcaloïdes (Datura stramonium), l’autre pas (Nicotiana plumbaginifolia). Le suivi par RMN du métabolisme du [2-13C] acétate, relaté dans un article précédent, a permis de montrer une haute activité de la glutamine synthétase chez Nicotiana plumbaginifolia, qui pourrait être à l’origine de l’absence de production d’alcaloïdes dans ces cellules. Le suivi par RMN du métabolisme du [3-13C] pyruvate ainsi que le dosage de l’activité enzymatique ont confirmé cette haute activité (act.GS N. plumbaginifolia = 158 nkat/mg de protéines; act.GS D. stramonium = 60 nkat/mg de protéines). Watanabe et al. ont récemment montré qu’il existe une relation entre la valeur du rapport Gln/Glu et le niveau de transcription du gène codant pour la GS. En effet, un rapport élevé implique un taux bas de transcription du gène, donc une faible quantité de GS produite et inversement. Les mesures des concentrations de glutamate et de glutamine par chromatographie échangeuse d’ions nous ont permis
1. Introduction In plant cells, the activity of the Krebs cycle leads to the biosynthesis of glutamate, which is then converted to arginine and ornithine, precursors in alkaloid biosynthesis (figure 1). Several authors have shown that the pyrrolidine ring of tobacco alkaloids and the tropane ring of tropane alkaloids derive from ornithine and/or arginine via putrescine [1–5]. In numerous undifferentiated plant cell cultures from Solanaceous species, this metabolic pathway is at least partially inhibited and does not lead to the biosynthesis of alkaloids. Several hypotheses have been suggested to explain this inhibition of the secondary metabolism [6]. Our previous work, using [2-13C] acetate, indicated a larger accumulation of glutamine in nonalkaloid-producing Nicotiana plumbaginifolia suspension culture cells than in tropane-alkaloid-accumulating Datura stramonium [7–9]. Thus, a first hypothesis to explain the non-biosynthesis of alkaloids by the N. plumbaginifolia cells would be that a high glutamine synthetase (GS, EC 6.3.1.2) activity leads to a very low concentration of glutamate [9]. Glutamine synthetase catalyses glutamine formation from glutamate and ammonium in the presence of ATP.
744
d’établir les rapports suivants: 5,9 pour N. plumbaginifolia et 9,1 pour D. stramonium. Les valeurs des deux rapports confirment les résultats du dosage enzymatique: le plus faible rapport, mesuré chez N. plumbaginifolia est en accord avec une plus forte transcription du gène codant pour la GS. L’hypothèse d’une déviation, par forte activité de la GS, du métabolisme du glutamate vers la glutamine, impliquant un faible taux de glutamate disponible pour la formation de précurseurs nécessaires à la synthèse d’alcaloïdes, n’est donc pas vérifiée. Pour conforter cette conclusion, nous avons effectué des suivis métaboliques en présence de méthionine sulfoximine (MSX), inhibiteur de la GS. Le rapport Gln/Glu a alors été inversé. Néanmoins, il n’a été mis en évidence ni production d’alcaloïdes (analyse par HPLC), ni formation d’intermédiaires du métabolisme secondaire (par résonance magnétique nucléaire du carbone 13). Ces résultats renforcent la conclusion que la haute activité de la GS chez N. plumbaginifolia n’est pas responsable de l’absence de synthèse d’alcaloïdes dans ces cellules. Les expériences effectuées avec l’inhibiteur ont permis, par ailleurs, de montrer qu’en accord avec les données de la littérature, des conditions de stress - présence de MSX dans notre cas - peuvent induire la formation de GABA à l’intérieur des cellules végétales, le GABA pouvant jouer un rôle de réserve d’azote utilisable pour la formation des acides aminés, lorsque le passage glutamate-glutamine est bloqué.
Glutamate/glutamine couple is involved in ammonia assimilation via glutamine synthetase and glutamine 2-oxoglutarate amino transferase, which reductively converts glutamine and 2-oxoglutarate to two molecules of glutamate. The role of glutamate dehydrogenase as another possible way of ammonia assimilation is at present very controversial (figure 1) [10, 11]. Watanabe et al. showed that the internal Gln/Glu ratio was a potential regulatory parameter for the expression of a glutamine synthetase gene [12]. Cells accumulating glutamine at high levels exhibited low levels of GS owing to low levels of gene transcript. The transcript level was, however, not regulated in response to the glutamine level alone but in response to the ratio of cellular glutamine to glutamate. Therefore, an unusually high GS activity due to modifications of the glutamine/glutamate ratio could be a first explanation for the non-biosynthesis of alkaloids. In order to investigate this hypothesis, we measured the GS enzymatic activity, followed the acetate and pyruvate metabolism by 13C NMR using [3-13C] pyruvate and [2-13C] acetate in suspension cells and realized cultures using L-methionine sulphoximine (MSX) [13, 14] to block the glutamate/glutamine conversion and to try to deviate glutamate towards alkaloid biosynthesis. C. R. Acad. Sci. Paris, Sciences de la vie / Life Sciences 1999, 322, 743–748
GS activity, 13C NMR and enzymatic assay
2.3. Feeding experiments
To follow metabolism in plant cells, culture media were supplemented with sodium [2-13C] acetate or sodium [3-13C] pyruvate (CEA, Saclay) or non-labelled sodium acetate or sodium pyruvate as controls, to be sure that the observed carbon resonances were due to the insertion of 13 C-labelled compounds into the various metabolites. Final acetate and pyruvate concentrations were 4 mM in all cases. In a previous work, it was verified that the 4-mM concentration has no inhibitory effect on cell growth. Samples were treated as described previously [9]. 2.4. NMR experiments
Carbon-13 NMR spectroscopy was performed on a Brucker AM 300 NMR spectrometer at 75.47 MHz. For cell extracts, all spectra were obtained in identical experimental conditions: temperature 22 °C, spectral width 18 kHz for 16 K data points, acquisition time 0.459 s, pulse delay 0.75 s and pulse angle equal to 60° with 1H composite pulse decoupling. Before Fourier transformation in absolute intensity mode, a zerofilling to 32 K was applied. An artificial line broadening of 10 Hz was used to improve the spectral signal-to-noise ratio. Spectra were acquired for 48 000 scans and resonance chemical shifts were relative to an external capillary solution of sodium formate (172.0 ppm). For culture media, the experimental conditions were identical except that only 2 000 scans were acquired. Insertion of 13C into the various metabolites was evaluated by differences between spectra with labelled and non-labelled precursors. The intensities of different resonances were expressed as signal-to-noise ratio (S/N) with the same vertical expansion relative to the external reference. 2.5. Extraction and enzyme assay of glutamine synthetase
Figure 1. Pathway showing the biosynthesis of alkaloids from 2-oxoglutarate. GDH = glutamate deshydrogenase; GOGAT = glutamine 2-oxoglutarate amino transferase; GS = glutamine synthetase; PMT = putrescine methyl transferase.
2. Materials and methods 2.1. Culture conditions
N. plumbaginifolia and D. stramonium cell suspensions were cultured as described previously [9]. In the GS inhibitor experiments, MSX (Sigma) was added to the growth medium after a 0.2-lm sterile filtration (Dynacard-Merck). To evaluate its toxicity, MSX was used at various concentrations. Optimal conditions were for 5 mM MSX. 2.2. Cell viability
To determine MSX toxicity, cell viability was assessed using fluorescein diacetate as described [15]. C.R. Acad. Sci. Paris, Sciences de la vie / Life Sciences 1999, 322, 743–748
GS cell-free extracts were prepared by grinding 10 g fresh matter of liquid-nitrogen-frozen material in 5 mL of buffer containing 50 mM TRIS-HCl (pH 7.8), 1 mM EDTA, 1 mM DTT, 10 mM MgSO4, 5 mM sodium glutamate and 10 % ethanediol, as described by Lea [10]. Following centrifugation at 10 000 g at 4 °C for 60 min, the supernatant was concentrated and desalted with CENTRIPLUST 100 (AMICON, INC, USA) concentrators at 4 °C and 3 000 g for 130 min. GS activity was measured by the formation of c-glutamyl hydroxamate in the synthetase reaction [16]. 2.6. Protein determination
The soluble proteins in cell-free extracts were determined according to Bradford, with bovine serum albumin as the standard [17]. 2.7. Amino-acid and alkaloid analysis
The amino-acids glutamine and glutamate were determined on a Beckman model automated analyser using
745
F. Mesnard et al.
Figure 2. Incorporation of label into (●) [4-13C] glutamine and (•) [4-13C] glutamate in N. plumbaginifolia cells as a function of time with 4 mM [2-13C] acetate (A) or with 4 mM [3-13C] pyruvate (B). [4-13C] glutamate is not detected in experiments with [3-13C] pyruvate because of a weak incorporation (see Discussion). Carbons 2-C and 3-C give similar curves; only variations in signal intensity have been noted. Examination of the inset (A) shows glutamate labelling to be later than for glutamine.
postcolumn derivatization with ninhydrin. The eluate was monitored at 570 and 440 nm, after elimination of proteins and nucleic acid [18]. The tobacco alkaloids were quantified using reversed-phase HPLC as described previously [19].
3. Results The in vitro activity of GS in desalted cell-free extracts was 158 ± 11 nkat mg–1 protein. The GS activity detected in extracts derived from D. stramonium cells was 60 ± 28 nkat mg–1 protein. The assays were performed immediately after extraction. The difference in activities is significant (Student’s paired t-test; P < 0.05). The quantitation of free amino-acids by HPLC showed an amino-acid level in N. plumbaginifolia cells much higher than in D. stramonium cells with Gln/Glu ratios of 5.9 ± 0.3 and 9.1 ± 0.4, respectively (Student’s paired t-test; P < 0.05). Note that the difference in the amino-acid level was in agreement with our previous results of a weak labelling from [2-13C] acetate in D. stramonium cells and a more intense incorporation by N. plumbaginifolia cells [9]. In the experimental conditions of cell culture, no tobacco alkaloid was detected by HPLC analysis. The labelled compounds observed in [3-13C] pyruvate experiments are identical to those obtained in [2-13C] acetate experiments [9] with, in addition, alanine, asparagine or aspartate and lactate (data not shown). Figure 2A and B shows the incorporation of label into C-4 glutamate and glutamine as a function of time after the uptake of
746
[2-13C] acetate and [3-13C] pyruvate, respectively, by N. plumbaginifolia cells. Note the weak labelling of glutamine when pyruvate was used as labelled precursor. In experiments with inhibitor, MSX treatment of N. plumbaginifolia cells with 4 mM [2-13C] acetate for 9 h (figure 3) resulted in labelling the pools of glutamine, glutamate, GABA, malate and citrate. The four resonances observed from MSX (54.0, 52.9, 42.0, 25.0 ppm) confirmed that the inhibitor had been taken up by the cells. The inset of figure 3 shows the incorporation of label into the C-4 of GABA as a function of time after the uptake of [2-13C] acetate by N. plumbaginifolia cells in the presence or absence of MSX. It should be emphasized that 5 mM MSX led to a slowing down of metabolism, characterized by weak labelling observed in 13C NMR spectra (inset figure 3) and a much reduced rate of acetate uptake, with an apparent uptake constant equal to (1.9 ± 0.3)10–3 h–1 versus (60 ± 10)10–3 h–1 in previous data [9]. The incorporation of label into [4-13C] glutamate and [4-13C] glutamine as a function of time after the uptake of [2-13C] acetate by N. plumbaginifolia cells cultured in presence of MSX is shown in figure 4.
4. Discussion The NMR studies presented here verify that, as previously reported [9], the experimental conditions used did not modify cellular viability and acetate or pyruvate uptake by N. plumbaginifolia suspension cultures. The advantage of exhibiting pyruvate compared to acetate is that it is incorporated more directly in other C. R. Acad. Sci. Paris, Sciences de la vie / Life Sciences 1999, 322, 743–748
GS activity, 13C NMR and enzymatic assay
Figure 3. 13C NMR spectrum (expansion from 10 to 60 ppm) of N. plumbaginifolia cells following treatment with 4 mM [2-13C] acetate and 5 mM L-methionine sulfoximine for 20 h showing labelled compound resonances. The four resonances of MSX inhibitor are detected (noted I). The inset shows the evolution of [4-13C] GABA as a function of time with (●) or without (•) MSX. Carbons 2-C and 3-C give similar curves; only variations in signal intensity have been noted.
metabolites, for example alanine by transamination or asparagine via oxaloacetate, permitting a more rapid and focused analysis of other metabolic pathways. Pyruvate enters the Krebs cycle, not only like acetate, via acetylCoA, but also via oxaloacetate or malate [20]. Thus, by using [3-13C] pyruvate we were able to observe the formation of alanine, asparagine or aspartate and lactate, produced in cells by reduction of pyruvate (data not shown). The minor disadvantage of using [3-13C] pyruvate was a relatively weak incorporation of label into the glutamate/glutamine pathway with no labelling of glutamate and only a weak labelling of glutamine (figure 2A versus B). Except for these observations, we observed similar incorporations with the two labelled precursors. Watanabe et al. [12] have proposed that the internal Gln/Glu ratio may act as a potential regulator of the expression of the GS gene. The measure of the GS activity in N. plumbaginifolia and D. stramonium cells shows an activity of GS from N. plumbaginifolia cells greater than that from D. stramonium cells, confirming the previous conclusions [9]. Thus, two hypotheses can be suggested in N. plumbaginifolia cells: 1) there is a high level of gene transcripts in response to a low Gln/Glu ratio, or 2) a low C.R. Acad. Sci. Paris, Sciences de la vie / Life Sciences 1999, 322, 743–748
Figure 4. Incorporation of label into (●) [4-13C] glutamine and (•) [4-13C] glutamate in N. plumbaginifolia cells as a function of time with 4 mM [2-13C] acetate and 5 mM L-methionine sulphoximine. Carbons 2-C and 3-C give similar curves, only variations in signal intensity have been noted.
747
F. Mesnard et al.
level of gene transcripts in response to a high Gln/Glu ratio. In the former case, the high activity of GS in N. plumbaginifolia cells should be considered as a normal response; in the latter it should be considered as an abnormal response with a dysfunctioning of the GS regulatory systems. The Gln/Glu ratio in N. plumbaginifolia cells compared to the ratio in D. stramonium cells is in agreement with the first hypothesis. In these conditions, the non-biosynthesis of alkaloids should not be due to the high enzymatic activity of GS, the level of glutamate being high enough to lead to the biosynthesis of arginine and ornithine, available to be converted into putrescine. To confirm these data, we induced metabolic disruption with the GS inhibitor, MSX. We can note the inversion of the glutamine/glutamate ratio (figure 2A versus figure 4) and a 13C enrichment of the GABA resonances (see inset figure 3). This observation is in agreement with previous
results [20, 21]. GABA is formed by the decarboxylation of glutamate by glutamate decarboxylase, and increases in GABA content and glutamate decarboxylase activity have been seen in plants under a variety of environmental stress conditions, including MSX treatment. Generally, it appears that GABA biosynthesis is enhanced when the glutamate to glutamine conversion cannot occur. GABA can have a role in nitrogen mobilization and can be viewed as a temporary nitrogen storage compound for protein amino-acids in plants [20–23]. From these results, the glutamate concentration is probably sufficiently elevated to be used for alkaloids biosynthesis. Therefore, the hypothesis of GS dysfunctioning for the non-biosynthesis of alkaloids in N. plumbaginifolia suspension cells can be discarded. Moreover, HPLC analyses confirmed the NMR studies, no tobacco alkaloids being detected by either method.
Acknowledgements: We thank Dr R.J. Robins (LAIEM, CNRS UPRES-A 6006, Nantes) for critical comments on the manuscript.
References [1] Leete E., Alkaloids derived from ornithine, lysine and nicotinic acid, in: Bell E.A., Charlwood B.V. (Eds.), Encyclopedia of Plant Physiology, Springer Verlag, Berlin, 1980, pp. 65–91. [2] Leete E., Recent developments in the biosynthesis of the tropane alkaloids, Planta Med. 56 (1990) 339–352. [3] Leete E., Endo T., Yamada Y., Biosynthesis of nicotine and scopolamine in a root culture of Duboisia lechhardtii, Phytochemistry 29 (1990) 1847–1851. [4] Robins R.J., Parr A.J., Bent E.G., Rhodes M.J.C., Studies on the formation of tropane alkaloids in Datura stramonium L. transformed root cultures, 1. The kinetics of alkaloid production and the influence of feeding intermediate metabolites, Planta 183 (1991) 185–195. [5] Robins R.J., Parr A.J., Walton N.J., Studies on the formation of tropane alkaloids in Datura stramonium L. transformed root cultures, 2. On the relative contributions of L-arginine and L-ornithine to the formation of the tropane ring, Planta 183 (1991) 196–201. [6] Berlin J., Secondary products from plant cell cultures, in: Rhem H.J., Reed G. (Eds.), Biotechnology, Verl. Chemie, Basel, 1986, pp. 629–658. [7] Manceau F., Fliniaux M.A., Jacquin A., Ability of a Nicotiana plumbaginifolia cell suspension to demethylate nicotine into nornicotine, Phytochemistry 28 (1989) 2671–2674. [8] Gontier E., Fliniaux M.A., Barbotin J.N., Sangwan-Norreel B.S., Tropane Alkaloid levels in the leaves of micropropagated Datura innoxia Mill., Plant. Planta Med. 59 (1993) 432–435. [9] Marty D., Mesnard F., Gillet-Manceau F., Fliniaux M.A., Monti J.P., Changes in primary metabolism in connection with alkaloid biosynthesis in Solanaceous cell suspensions. A 13C NMR study, Plant Sci. 122 (1997) 11–21. [10] Lea P.J., Enzymes of ammonia assimilation, in: Dey P.M., Harborne J.B. (Eds.), Methods in Plant Biochemistry, Academic Press, London, 1990, pp. 257–276. [11] Lea P.J., Primary nitrogen metabolism, in: Dey P.M., Harborne J.B. (Eds.), Plant Biochemistry, Academic Press, London, 1997, pp. 273–313.
748
[12] Watanabe A., Tagaki N., Hayashi H., Chino M., Watanabe, A., Internal Gln/Glu ratio as a potential regulatory parameter for the expression of a cytosolic glutamine synthetase gene of radish in cultured cells, Plant Cell Physiol. 38 (1997) 1000–1006. [13] Leason M., Cunliffe D., Parkin D., Lea P.J., Miflin, B.J., Inhibition of pea leaf glutamine synthetase by methionine sulphoximine, phosphinotricine and other glutamate analogues, Phytochemistry 21 (1982) 855–857. [14] Kumar P.A., Abrol, Y.P., Effect of L-methionine sulphoximine on the enzymes of nitrogen metabolism in barley leaves, Proc. Indian Acad. Sci. 100 (1990) 97–100. [15] Dixon R.A., Plant Cell Culture: A Practical Approach, IRL Press, Oxford, 1985. [16] Rhodes D., Rendon G.A., Stewart, G.R., The control of glutamine synthetase level in Lemna minor L., Planta 125 (1975) 201–211. [17] Bradford M.M., A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein dye binding, Anal. Biochem. 72 (1976) 248–254. [18] Acil Y., Müller P.K., Rapid method for the isolation of the mature collagen cross-links, hydroxylysylpyridinoline and lysylpyridinoline, J. Chrom. A 664 (1994) 183–188. [19] Manceau F., Fliniaux M.A., Jacquin, A., A high-performance liquid chromatographic procedure for the analysis of tobacco alkaloids – Application to the evaluation of tobacco alkaloids in plants and cell suspension cultures, Phytochem. Anal. 3 (1992) 223–226. [20] Robinson T., The Organic Constituents of Higher Plants, Cordus Press, North Amherst, 1991. [21] Desmaison A.M., Tixier M., Amino acids content in germinating seeds and seedlings from Castanea sativa L., Plant Physiol. 81 (1986) 692–695. [22] Jordan B.R., Givan C.V., Effects of light and inhibitors on glutamate metabolism in leaf disc of Vicia faba L sources of ATP for glutamine synthesis and photoregulation of tricarboxylic acid cycle metabolism, Plant Physiol. 64 (1979) 1043–1047. [23] Narayan V.S., Nair, P.M., Metabolism, enzymology and possible roles of 4-Aminobutyrate in higher plants, Phytochemistry 29 (1990) 367–375.
C. R. Acad. Sci. Paris, Sciences de la vie / Life Sciences 1999, 322, 743–748