- Email: [email protected]
Rapid and sensitive NMR method for osmolyte determination
Journal of Microbiological Methods 58 (2004) 289 – 294 www.elsevier.com/locate/jmicmeth
Note
Rapid and sensitive NMR method for osmolyte determination Andrea Motta *, Ida Romano, Agata Gambacorta* Istituto di Chimica Biomolecolare del CNR, Comprensorio Olivetti, Edificio 70, via Campi, Flegrei 34, Pozzuoli I-80078, Italy Received 16 March 2004; received in revised form 24 March 2004; accepted 8 April 2004 Available online 7 June 2004
Abstract We propose a rapid and sensitive method for osmolyte determination, based on one-dimensional and two-dimensional 1H NMR spectroscopy applied directly on culture of haloalkalophilic Halomonas pantelleriensis and acidothermophilic archaeon Sulfolobus solfataricus, without any extraction procedure. The osmoprotectants hydroxyectoine, ectoine, glutamate, glycinebetaine and treahalose can easily be quantified by integrating the peak areas with respect to an internal standard, and the concentrations evaluated with this method are in excellent agreement with the values previously reported. Furthermore, trace amount of osmoprotectants, often undetectable after extraction procedures, can also be evaluated. D 2004 Elsevier B.V. All rights reserved. Keywords: NMR spectroscopy; Halomonas pantelleriensis; Sulfolobus solfataricus
Nuclear magnetic resonance (NMR) spectroscopy is a powerful technique for the determination of molecular structure, conformation and dynamics of biomolecules (Evans, 1996), including osmolytes. It is based on the fact that atomic nuclei oriented by a strong magnetic field absorb radiation at characteristic frequencies (typically a few hundred megahertz, in the radiofrequency part of the spectrum). Most structure determinations by NMR have used the spin of 1H and 13 C, both of which constitute the most important nuclei for the study of biomolecules. When a sample dissolved in a solvent is placed in a strong magnetic field, the spin of their hydrogen atoms aligns along the field. This equilibrium alignment can be changed to an
* Corresponding authors. E-mail addresses: [email protected] (A. Motta), [email protected] (A. Gambacorta). 0167-7012/$ - see front matter D 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.mimet.2004.04.012
excited state by applying radio frequency pulses to the sample. When the nuclei of the molecule revert to their equilibrium state, they emit radiofrequency radiation that can be measured. The exact frequency of the emitted radiation from each nucleus depends on the molecular environments of the nucleus and is different for each atom. These different frequencies are obtained relative to a reference signal and are called chemical shifts. The nature, duration and combination of applied radiofrequency pulses can be varied enormously, and different properties of the sample can be probed by selecting appropriate combination of pulses. In principle, it is possible to obtain a unique signal (chemical shift) for each hydrogen atom in a molecule, except those that are chemically equivalent, for example, the protons of the C(6)H of ectoine and hydroxyectoine (Fig. 1). In practice, however, such one-dimensional NMR spectra contain overlapping signals because the differences in chem-
290
A. Motta et al. / Journal of Microbiological Methods 58 (2004) 289–294
Fig. 1. 1H spectrum of cells of H. pantelleriensis grown on glucose and 3.0 M NaCl.
ical shifts are often smaller than the resolving power of the experiment. In recent years, this problem has been bypassed by designing experimental conditions that yield two-dimensional NMR spectra, the results of which are usually plotted in a diagram as shown in Fig. 2. The diagonal in such a diagram corresponds to a normal one-dimensional NMR spectrum. The off-diagonal peaks result from interactions between hydrogen atoms that are close to each other in the molecular skeleton. By varying the nature of the applied radio frequency pulses, these off-diagonal peaks can reveal different types of interactions. The TOCSY (total correlation spectroscopy) experiment reported in Fig. 2 gives correlations between hydrogen atoms that are covalently connected through one, two or more chemical bonds. This experiment yields large portion of a molecule identifying horizontally (or vertically) off-diagonal peaks corresponding to the chemical shifts of the hydrogen involved (for example, the hydrogen atoms attached to the C(4) – C(5) – C(6) of hydroxyectoine). Assignment (that is, identification) of NMR peaks in one-dimensional spectra is possible by comparing the observed chemical shifts with reference data, followed by small addition of the identified biomolecule, when commercially available. The increase of
the corresponding spectral resonances warrants that the identification is correct. When no reference data are available, identification can only be achieved after the molecular structure has been obtained. Halotolerant and halophilic microorganisms synthesize and accumulate organic solutes, referred to as compatible solutes or osmolytes, in order to maintain osmotic balance between their cytoplasm and the saline environment (Da Costa et al., 1998; Oren, 2002). These low molecular mass compounds do not belong to a defined chemical class, being sugars, polyols, organic sulfur compounds and phosphates, primary and secondary amino acids (Da Costa et al., 1998). These compounds, fully compatible with the metabolism of the cells, exert a general stabilizing effect on polymers, enzymes and whole cells against external injuries and low water activity (Da Costa et al., 1998; Oren, 2002), and are accumulated either by de novo synthesis or by transport from the medium (Da Costa et al., 1998; Oren, 2002). Several methods are used to identify and quantify osmolytes. The most commonly used is the extraction of wet cells with boiling 80% ethanol either at room temperature for 2 h (Reed et al., 1984) or for 1 h in sealed tubes (Romano et al., 2001). The extraction is usually repeated twice and the solvent from the com-
A. Motta et al. / Journal of Microbiological Methods 58 (2004) 289–294
291
Fig. 2. Homonuclear TOCSY experiment on cells of H. pantelleriensis grown on glucose and 3.0 M NaCl.
bined supernatants is removed by rotary evaporation. The residue is then dissolved in CHCl3/MeOH/H2O (65:25:4 by volume) until lipid components are eliminated and their presence tracked by TLC eluted in the above solvent mixture (Romano et al., 2001). The amount of solute present in the extract (Amol) can be determined either from the resonances areas in 13 C-NMR spectra with respect to internal dioxane or by HPLC (Lai et al., 1991). These values are then normalized (Amol/mg protein) to the total protein content of the cells, obtained from Bio-Rad assay on the suspended pellet in distilled water and subjected to ultrasonic treatment after ethanol extraction (Desmarais et al., 1997). Bernard et al. (1993) and Nagata et al. (1996) used 1 H NMR spectroscopy to study intracellular solutes in Brevibacterium linens. The authors reported the 1HNMR identification of ectoine (1,4,5,6-tetrahydro-2methyl-4-pyrimidinecarboxylic acid) and its 5-hydroxy derivative in ethanol extracts of cells exposed to hyperosmotic shock, and of extracellular medium.
Recently, Riis et al. (2003) reported a highly sensitive method for the determination of ectoine and other compatible solutes by anion-exchange chromatography and pulsed amperometric detection. The compatible solutes were analyzed after extraction of cells or cell suspension, and ectoine and its 5-hydroxy derivative were determined after hydrolytic cleavage of the pyrimidine ring. All the methods described are time-consuming in that they require extensive manipulation of the sample. Here, we propose a rapid method for analyzing osmolytes by using 1H-NMR directly on cultures without any extraction procedure. As a test, the method has been applied to a halo- and alkalitolerant bacterial strain, Halomonas pantelleriensis (DSM 9661) (Romano et al., 1996), and an acidothermophilic archaeon, Sulfolobus solfataricus (DSM 5833) (De Rosa et al., 1975). Cells of H. pantelleriensis were grown in standard growth conditions at 35 jC, pH 9.0, 1.8 M NaCl in the following medium (g/l): yeast extract (10.0), Na3-
292
A. Motta et al. / Journal of Microbiological Methods 58 (2004) 289–294
citrate (3.0), KCl 4.(2.0), MgSO47H2O (1.0), Na2CO3 (3.0), (mg/l) MnCl24H2O (0.36) and FeSO4 (50) (complex medium). Na2CO3 and NaCl were autoclaved separately. The effects of organic components on the accumulation of osmoprotectants were examined by replacing the yeast extract with glucose or trehalose (10 g/l) in the following saline solution (g/l): K2HPO4 (7.0), KH2PO4 (2.0), MgSO47H2O (0.1), (NH4)2SO4 (1.0), Na2CO3 (3.0), 500 Al of thiamine hydrochloride (100 mg/l) (minimal media). S. solfataricus was grown in the medium containing the following components (g/l): yeast extract (2.0), KH2PO4 (3.0), (NH4)2SO4 (2.5), MgSO47H2O (0.2), CaCl27H2O (0.25), pH 3.5 with concentrated H2SO4. The cells from 1.5 ml of culture with A540 of 1430 were collected after 24, 48 or 72 h of incubation by centrifugation at 8500 g for 20 min, then washed with isoosmotic solutions; and after centrifugation, they were weighed (9.5 mg wet cells) and resuspended in isoosmotic saline solution with 30 Al of D2O and 3.0 Al of sodium 3-(trimethylsilyl)-(2,2,3,3-2H4)propionate (TSP, 26 mM), as internal standard (800 Al final volume). The cells and the supernatants were analyzed by 1H NMR at 300 K on a Bruker Avance-400 spectrometer operating at 400 MHz, using an inverse multinuclear probehead fitted with gradient along the z axis. Spectra were referenced to the internal TSP standard. One-dimensional experiments were obtained by using the excitation sculpting sequence (Hwang and Shaka, 1995) for water suppression. We used a pulsed-field gradient double echo with a soft square pulse of 4 – 8 ms at the water resonance frequency, with the gradient pulses of 1 ms each in duration. The 256 – 1024 transients were acquired for each experiment, with 4-s pulse repetition time. Resonances were identified by a homonuclear two-dimensional clean TOCSY (Griesinger et al., 1988) experiment incorporating the excitation sculpting sequence for water suppression. In isoosmotic saline solution, the NMR spectra of supernatants (not shown) contain only signals stemming from unused yeast, glucose or trehalose media, while signals due to osmolytes are not observed. This indicates that in these experimental conditions, cell lysis for both microorganisms did not occur, and that osmoprotectants were not released during the growth and/or during washing. Fig. 1 reports 1H-NMR spectrum of H. pantelleriensis cells in isoosmotic
1
H2O/1H2O (90/10 v/v) 3.0 M NaCl, grown in 3.0 M NaCl and with glucose as a sole carbon source. Peaks assignment for the most abundant osmoprotectants was achieved by TOCSY experiment, and the identification of the spin systems is indicated in Fig. 2. In particular, the signals at 2.07 and 2.10 ppm originate from C(5)H2 protons of ectoine, as they correlate between them and with the C(4)H2 at 3.27 and 3.42 ppm, and the C(6)H at 4.04 ppm. Interestingly, the C(2)CH3 at 2.22 ppm correlates with the C(4) and C(6) protons. The hydroxyectoine protons on C(4) and C(6) are partially superimposed with the corresponding ectoine signals, but the C(5)H and the C(2)CH3 are well identified at 4.57 and 2.27 ppm, respectively. In analogy with the C(2) ectoine methyl, the latter shows small cross-peaks with the C(6) and C(4) protons. The glutamate spin system is clearly observed at 3.72 ppm (CaH), 2.50 ppm (CgH2) and at 2.05 ppm (ChH2). The methyl signals of glycine betaine were barely visible at 3.40 ppm, superimposed with C(4) protons of hydroxyectoine and ectoine, while the CH2 resonated at 3.52 ppm. After acquisition of NMR spectra, the cells were incubated in their optimal growth conditions, and the growth was positive, confirming that no lysis occurred during the above experiments. This is an important result in that the same cultures can be used for further studies by changing only the growth conditions. The cells were then centrifuged and resuspended in 2 H2O without salt and analyzed by proton NMR (Fig. 3). Due to the absence of 3 M NaCl, the signals’ resolution increases, but the peaks’ chemical shifts correspond to those observed above (Fig. 1), with only small variations. Therefore, peak integration could safely be applied to establish the Amoles of osmoprotectants present in the amount of wet cells under investigation. From experiments carried out at least in triplicate, we obtained a total amount of osmoprotectants of 0.322 Amol/mg wet cells, corresponding to 0.186 Amol for ectoine, 0.07 Amol for hydroxyectoine, 0.06 Amol for glutamate and 0.005 Amol for glycine betaine. When the cells were grown with trehalose as sole carbon source, except for the absence of glycine betaine, the results were similar to those reported above for glucose. Furthermore, no internalization of trehalose was observed, confirming the results previously reported (Romano et al., 2001). The ethanolic extracts
A. Motta et al. / Journal of Microbiological Methods 58 (2004) 289–294
293
Fig. 3. 1H spectrum of cells of H. pantelleriensis grown on glucose and 3.0 M NaCl, and treated with 1H2O without salt.
of H. pantelleriensis grown in standard growth conditions showed the presence of glycine-betaine as the major component, and ectoine and glutamate, while hydroxyectoine was not observed (Romano et al., 2001). On the contrary, the 1H NMR spectra of cells grown on yeast (not shown here) detected the pres-
ence of hydroxyectoine in trace amount, and the amount of glutamate was higher than that previously (Romano et al., 2001). We believe that during the extraction procedure these osmolytes were either lost or underestimated. Corresponding 1H NMR studies on S. solfataricus cells (Fig. 4) indicate the presence of trehalose as sole
Fig. 4. 1H spectrum of S. solfataricus cells. The signals from trehalose are observed at 5.18 ppm and in the interval 3.9 – 3.4 ppm.
294
A. Motta et al. / Journal of Microbiological Methods 58 (2004) 289–294
osmolyte, in agreement with extraction procedure (De Rosa et al., 1975). In conclusion, differently from the most used methodology for osmoprotectant analyses, the proposed NMR method requires a small volume of cells and little manipulation of samples. In particular, the proposed method is rapid and well reproducible and allows the detection of a trace amount of osmoprotectants often undetectable after extraction procedures. We believe that NMR may become routinely accepted in microbiological laboratory because NMR spectrometers are widely spread and one-dimensional and two-dimensional TOCSY spectra, as those reported here, are commonly run in any NMR Service laboratory.
References Bernard, T., Jebbar, M., Rassouli, Y., Himdi-Kabbab, S., Hamelim, J., Blanco, C., 1993. Ectoine accumulation and osmotic regulation in Brevibacterium linens. J. Gen. Microbiol. 139, 129 – 136. Da Costa, M.S., Santos, H., Galinski, E.A., 1998. An overview of the role and diversity of compatible solutes in bacteria and archaea. Adv. Biochem. Eng. Biotechnol. 61, 117 – 153. De Rosa, M., Gambacorta, A., Bu’Lock, J.D., 1975. Extremely thermophilic acidophilic bacteria convergent with Sulfolobus acidocaldarius. J. Gen. Microbiol. 86, 156 – 164. Desmarais, D., Jablonski, P.E., Fedarko, N.S., Roberts, M.F., 1997. Sulfotrehalose, a novel osmolyte in haloalkaliphilic archaea. J. Bacteriol. 179, 3146 – 3153.
Evans, J.N.S., 1996. Biomolecular NMR Spectroscopy. Oxford Univ. Press, Oxford. Griesinger, C., Otting, G., Wu¨thrich, K., Ernst, R.R., 1988. Clean TOCSY for proton spin system identification in macromolecules. J. Am. Chem. Soc. 110, 7870 – 7872. Hwang, T.-L., Shaka, A.J., 1995. Water suppression that works. Excitation sculpting using arbitrary wave-forms and pulsed-field gradients. J. Magn. Reson. 112, 275 – 279. Lai, M.C., Sowers, K.R., Robertson, D.E., Roberts, M.F., Gunsalus, R.P., 1991. Distribution of compatible solutes in the halophilic methanogenic archaebacteria. J. Bacteriol. 173, 5352 – 5358. Nagata, S., Adachi, K., Sano, H., 1996. NMR analyses of compatible solutes in a halotolerant Brevibacterium sp. Microbiology 142, 3355 – 3362. Oren, A., 2002. Intracellular salt concentrations and ion metabolism in halophilic microorganisms. Halophilic Microorganisms and their Environments. Kluver Academic Publishing, Boston, pp. 207 – 231. Reed, R.H., Richardson, S.R., Warr, S.R.C., Stewart, W.D.P., 1984. Carbohydrate accumulation and osmotic stress in cyanobacteria. J. Gen. Microbiol. 130, 1 – 4. Riis, V., Maskow, T., Babel, W., 2003. Highly sensitive determination of ectoine and other compatible solutes by anion-exchange chromatography and pulsed amperometric detection. Anal. Bioanal. Chem. 377, 203 – 207. Romano, I., Nicolaus, B., Lama, L., Manca, M.C., Gambacorta, A., 1996. Characterization of a haloalkalophilic strictly aerobic bacterium, isolated from Pantelleria island. Syst. Appl. Microbiol. 19, 326 – 333. Romano, I., Nicolaus, B., Lama, L., Trabasso, D., Caracciolo, G., Gambacorta, A., 2001. Accumulation of osmoprotectants and lipid pattern modulation in response to growth conditions by Halomonas pantelleriense. Syst. Appl. Microbiol. 24, 342 – 352.
Note
Rapid and sensitive NMR method for osmolyte determination Andrea Motta *, Ida Romano, Agata Gambacorta* Istituto di Chimica Biomolecolare del CNR, Comprensorio Olivetti, Edificio 70, via Campi, Flegrei 34, Pozzuoli I-80078, Italy Received 16 March 2004; received in revised form 24 March 2004; accepted 8 April 2004 Available online 7 June 2004
Abstract We propose a rapid and sensitive method for osmolyte determination, based on one-dimensional and two-dimensional 1H NMR spectroscopy applied directly on culture of haloalkalophilic Halomonas pantelleriensis and acidothermophilic archaeon Sulfolobus solfataricus, without any extraction procedure. The osmoprotectants hydroxyectoine, ectoine, glutamate, glycinebetaine and treahalose can easily be quantified by integrating the peak areas with respect to an internal standard, and the concentrations evaluated with this method are in excellent agreement with the values previously reported. Furthermore, trace amount of osmoprotectants, often undetectable after extraction procedures, can also be evaluated. D 2004 Elsevier B.V. All rights reserved. Keywords: NMR spectroscopy; Halomonas pantelleriensis; Sulfolobus solfataricus
Nuclear magnetic resonance (NMR) spectroscopy is a powerful technique for the determination of molecular structure, conformation and dynamics of biomolecules (Evans, 1996), including osmolytes. It is based on the fact that atomic nuclei oriented by a strong magnetic field absorb radiation at characteristic frequencies (typically a few hundred megahertz, in the radiofrequency part of the spectrum). Most structure determinations by NMR have used the spin of 1H and 13 C, both of which constitute the most important nuclei for the study of biomolecules. When a sample dissolved in a solvent is placed in a strong magnetic field, the spin of their hydrogen atoms aligns along the field. This equilibrium alignment can be changed to an
* Corresponding authors. E-mail addresses: [email protected] (A. Motta), [email protected] (A. Gambacorta). 0167-7012/$ - see front matter D 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.mimet.2004.04.012
excited state by applying radio frequency pulses to the sample. When the nuclei of the molecule revert to their equilibrium state, they emit radiofrequency radiation that can be measured. The exact frequency of the emitted radiation from each nucleus depends on the molecular environments of the nucleus and is different for each atom. These different frequencies are obtained relative to a reference signal and are called chemical shifts. The nature, duration and combination of applied radiofrequency pulses can be varied enormously, and different properties of the sample can be probed by selecting appropriate combination of pulses. In principle, it is possible to obtain a unique signal (chemical shift) for each hydrogen atom in a molecule, except those that are chemically equivalent, for example, the protons of the C(6)H of ectoine and hydroxyectoine (Fig. 1). In practice, however, such one-dimensional NMR spectra contain overlapping signals because the differences in chem-
290
A. Motta et al. / Journal of Microbiological Methods 58 (2004) 289–294
Fig. 1. 1H spectrum of cells of H. pantelleriensis grown on glucose and 3.0 M NaCl.
ical shifts are often smaller than the resolving power of the experiment. In recent years, this problem has been bypassed by designing experimental conditions that yield two-dimensional NMR spectra, the results of which are usually plotted in a diagram as shown in Fig. 2. The diagonal in such a diagram corresponds to a normal one-dimensional NMR spectrum. The off-diagonal peaks result from interactions between hydrogen atoms that are close to each other in the molecular skeleton. By varying the nature of the applied radio frequency pulses, these off-diagonal peaks can reveal different types of interactions. The TOCSY (total correlation spectroscopy) experiment reported in Fig. 2 gives correlations between hydrogen atoms that are covalently connected through one, two or more chemical bonds. This experiment yields large portion of a molecule identifying horizontally (or vertically) off-diagonal peaks corresponding to the chemical shifts of the hydrogen involved (for example, the hydrogen atoms attached to the C(4) – C(5) – C(6) of hydroxyectoine). Assignment (that is, identification) of NMR peaks in one-dimensional spectra is possible by comparing the observed chemical shifts with reference data, followed by small addition of the identified biomolecule, when commercially available. The increase of
the corresponding spectral resonances warrants that the identification is correct. When no reference data are available, identification can only be achieved after the molecular structure has been obtained. Halotolerant and halophilic microorganisms synthesize and accumulate organic solutes, referred to as compatible solutes or osmolytes, in order to maintain osmotic balance between their cytoplasm and the saline environment (Da Costa et al., 1998; Oren, 2002). These low molecular mass compounds do not belong to a defined chemical class, being sugars, polyols, organic sulfur compounds and phosphates, primary and secondary amino acids (Da Costa et al., 1998). These compounds, fully compatible with the metabolism of the cells, exert a general stabilizing effect on polymers, enzymes and whole cells against external injuries and low water activity (Da Costa et al., 1998; Oren, 2002), and are accumulated either by de novo synthesis or by transport from the medium (Da Costa et al., 1998; Oren, 2002). Several methods are used to identify and quantify osmolytes. The most commonly used is the extraction of wet cells with boiling 80% ethanol either at room temperature for 2 h (Reed et al., 1984) or for 1 h in sealed tubes (Romano et al., 2001). The extraction is usually repeated twice and the solvent from the com-
A. Motta et al. / Journal of Microbiological Methods 58 (2004) 289–294
291
Fig. 2. Homonuclear TOCSY experiment on cells of H. pantelleriensis grown on glucose and 3.0 M NaCl.
bined supernatants is removed by rotary evaporation. The residue is then dissolved in CHCl3/MeOH/H2O (65:25:4 by volume) until lipid components are eliminated and their presence tracked by TLC eluted in the above solvent mixture (Romano et al., 2001). The amount of solute present in the extract (Amol) can be determined either from the resonances areas in 13 C-NMR spectra with respect to internal dioxane or by HPLC (Lai et al., 1991). These values are then normalized (Amol/mg protein) to the total protein content of the cells, obtained from Bio-Rad assay on the suspended pellet in distilled water and subjected to ultrasonic treatment after ethanol extraction (Desmarais et al., 1997). Bernard et al. (1993) and Nagata et al. (1996) used 1 H NMR spectroscopy to study intracellular solutes in Brevibacterium linens. The authors reported the 1HNMR identification of ectoine (1,4,5,6-tetrahydro-2methyl-4-pyrimidinecarboxylic acid) and its 5-hydroxy derivative in ethanol extracts of cells exposed to hyperosmotic shock, and of extracellular medium.
Recently, Riis et al. (2003) reported a highly sensitive method for the determination of ectoine and other compatible solutes by anion-exchange chromatography and pulsed amperometric detection. The compatible solutes were analyzed after extraction of cells or cell suspension, and ectoine and its 5-hydroxy derivative were determined after hydrolytic cleavage of the pyrimidine ring. All the methods described are time-consuming in that they require extensive manipulation of the sample. Here, we propose a rapid method for analyzing osmolytes by using 1H-NMR directly on cultures without any extraction procedure. As a test, the method has been applied to a halo- and alkalitolerant bacterial strain, Halomonas pantelleriensis (DSM 9661) (Romano et al., 1996), and an acidothermophilic archaeon, Sulfolobus solfataricus (DSM 5833) (De Rosa et al., 1975). Cells of H. pantelleriensis were grown in standard growth conditions at 35 jC, pH 9.0, 1.8 M NaCl in the following medium (g/l): yeast extract (10.0), Na3-
292
A. Motta et al. / Journal of Microbiological Methods 58 (2004) 289–294
citrate (3.0), KCl 4.(2.0), MgSO47H2O (1.0), Na2CO3 (3.0), (mg/l) MnCl24H2O (0.36) and FeSO4 (50) (complex medium). Na2CO3 and NaCl were autoclaved separately. The effects of organic components on the accumulation of osmoprotectants were examined by replacing the yeast extract with glucose or trehalose (10 g/l) in the following saline solution (g/l): K2HPO4 (7.0), KH2PO4 (2.0), MgSO47H2O (0.1), (NH4)2SO4 (1.0), Na2CO3 (3.0), 500 Al of thiamine hydrochloride (100 mg/l) (minimal media). S. solfataricus was grown in the medium containing the following components (g/l): yeast extract (2.0), KH2PO4 (3.0), (NH4)2SO4 (2.5), MgSO47H2O (0.2), CaCl27H2O (0.25), pH 3.5 with concentrated H2SO4. The cells from 1.5 ml of culture with A540 of 1430 were collected after 24, 48 or 72 h of incubation by centrifugation at 8500 g for 20 min, then washed with isoosmotic solutions; and after centrifugation, they were weighed (9.5 mg wet cells) and resuspended in isoosmotic saline solution with 30 Al of D2O and 3.0 Al of sodium 3-(trimethylsilyl)-(2,2,3,3-2H4)propionate (TSP, 26 mM), as internal standard (800 Al final volume). The cells and the supernatants were analyzed by 1H NMR at 300 K on a Bruker Avance-400 spectrometer operating at 400 MHz, using an inverse multinuclear probehead fitted with gradient along the z axis. Spectra were referenced to the internal TSP standard. One-dimensional experiments were obtained by using the excitation sculpting sequence (Hwang and Shaka, 1995) for water suppression. We used a pulsed-field gradient double echo with a soft square pulse of 4 – 8 ms at the water resonance frequency, with the gradient pulses of 1 ms each in duration. The 256 – 1024 transients were acquired for each experiment, with 4-s pulse repetition time. Resonances were identified by a homonuclear two-dimensional clean TOCSY (Griesinger et al., 1988) experiment incorporating the excitation sculpting sequence for water suppression. In isoosmotic saline solution, the NMR spectra of supernatants (not shown) contain only signals stemming from unused yeast, glucose or trehalose media, while signals due to osmolytes are not observed. This indicates that in these experimental conditions, cell lysis for both microorganisms did not occur, and that osmoprotectants were not released during the growth and/or during washing. Fig. 1 reports 1H-NMR spectrum of H. pantelleriensis cells in isoosmotic
1
H2O/1H2O (90/10 v/v) 3.0 M NaCl, grown in 3.0 M NaCl and with glucose as a sole carbon source. Peaks assignment for the most abundant osmoprotectants was achieved by TOCSY experiment, and the identification of the spin systems is indicated in Fig. 2. In particular, the signals at 2.07 and 2.10 ppm originate from C(5)H2 protons of ectoine, as they correlate between them and with the C(4)H2 at 3.27 and 3.42 ppm, and the C(6)H at 4.04 ppm. Interestingly, the C(2)CH3 at 2.22 ppm correlates with the C(4) and C(6) protons. The hydroxyectoine protons on C(4) and C(6) are partially superimposed with the corresponding ectoine signals, but the C(5)H and the C(2)CH3 are well identified at 4.57 and 2.27 ppm, respectively. In analogy with the C(2) ectoine methyl, the latter shows small cross-peaks with the C(6) and C(4) protons. The glutamate spin system is clearly observed at 3.72 ppm (CaH), 2.50 ppm (CgH2) and at 2.05 ppm (ChH2). The methyl signals of glycine betaine were barely visible at 3.40 ppm, superimposed with C(4) protons of hydroxyectoine and ectoine, while the CH2 resonated at 3.52 ppm. After acquisition of NMR spectra, the cells were incubated in their optimal growth conditions, and the growth was positive, confirming that no lysis occurred during the above experiments. This is an important result in that the same cultures can be used for further studies by changing only the growth conditions. The cells were then centrifuged and resuspended in 2 H2O without salt and analyzed by proton NMR (Fig. 3). Due to the absence of 3 M NaCl, the signals’ resolution increases, but the peaks’ chemical shifts correspond to those observed above (Fig. 1), with only small variations. Therefore, peak integration could safely be applied to establish the Amoles of osmoprotectants present in the amount of wet cells under investigation. From experiments carried out at least in triplicate, we obtained a total amount of osmoprotectants of 0.322 Amol/mg wet cells, corresponding to 0.186 Amol for ectoine, 0.07 Amol for hydroxyectoine, 0.06 Amol for glutamate and 0.005 Amol for glycine betaine. When the cells were grown with trehalose as sole carbon source, except for the absence of glycine betaine, the results were similar to those reported above for glucose. Furthermore, no internalization of trehalose was observed, confirming the results previously reported (Romano et al., 2001). The ethanolic extracts
A. Motta et al. / Journal of Microbiological Methods 58 (2004) 289–294
293
Fig. 3. 1H spectrum of cells of H. pantelleriensis grown on glucose and 3.0 M NaCl, and treated with 1H2O without salt.
of H. pantelleriensis grown in standard growth conditions showed the presence of glycine-betaine as the major component, and ectoine and glutamate, while hydroxyectoine was not observed (Romano et al., 2001). On the contrary, the 1H NMR spectra of cells grown on yeast (not shown here) detected the pres-
ence of hydroxyectoine in trace amount, and the amount of glutamate was higher than that previously (Romano et al., 2001). We believe that during the extraction procedure these osmolytes were either lost or underestimated. Corresponding 1H NMR studies on S. solfataricus cells (Fig. 4) indicate the presence of trehalose as sole
Fig. 4. 1H spectrum of S. solfataricus cells. The signals from trehalose are observed at 5.18 ppm and in the interval 3.9 – 3.4 ppm.
294
A. Motta et al. / Journal of Microbiological Methods 58 (2004) 289–294
osmolyte, in agreement with extraction procedure (De Rosa et al., 1975). In conclusion, differently from the most used methodology for osmoprotectant analyses, the proposed NMR method requires a small volume of cells and little manipulation of samples. In particular, the proposed method is rapid and well reproducible and allows the detection of a trace amount of osmoprotectants often undetectable after extraction procedures. We believe that NMR may become routinely accepted in microbiological laboratory because NMR spectrometers are widely spread and one-dimensional and two-dimensional TOCSY spectra, as those reported here, are commonly run in any NMR Service laboratory.
References Bernard, T., Jebbar, M., Rassouli, Y., Himdi-Kabbab, S., Hamelim, J., Blanco, C., 1993. Ectoine accumulation and osmotic regulation in Brevibacterium linens. J. Gen. Microbiol. 139, 129 – 136. Da Costa, M.S., Santos, H., Galinski, E.A., 1998. An overview of the role and diversity of compatible solutes in bacteria and archaea. Adv. Biochem. Eng. Biotechnol. 61, 117 – 153. De Rosa, M., Gambacorta, A., Bu’Lock, J.D., 1975. Extremely thermophilic acidophilic bacteria convergent with Sulfolobus acidocaldarius. J. Gen. Microbiol. 86, 156 – 164. Desmarais, D., Jablonski, P.E., Fedarko, N.S., Roberts, M.F., 1997. Sulfotrehalose, a novel osmolyte in haloalkaliphilic archaea. J. Bacteriol. 179, 3146 – 3153.
Evans, J.N.S., 1996. Biomolecular NMR Spectroscopy. Oxford Univ. Press, Oxford. Griesinger, C., Otting, G., Wu¨thrich, K., Ernst, R.R., 1988. Clean TOCSY for proton spin system identification in macromolecules. J. Am. Chem. Soc. 110, 7870 – 7872. Hwang, T.-L., Shaka, A.J., 1995. Water suppression that works. Excitation sculpting using arbitrary wave-forms and pulsed-field gradients. J. Magn. Reson. 112, 275 – 279. Lai, M.C., Sowers, K.R., Robertson, D.E., Roberts, M.F., Gunsalus, R.P., 1991. Distribution of compatible solutes in the halophilic methanogenic archaebacteria. J. Bacteriol. 173, 5352 – 5358. Nagata, S., Adachi, K., Sano, H., 1996. NMR analyses of compatible solutes in a halotolerant Brevibacterium sp. Microbiology 142, 3355 – 3362. Oren, A., 2002. Intracellular salt concentrations and ion metabolism in halophilic microorganisms. Halophilic Microorganisms and their Environments. Kluver Academic Publishing, Boston, pp. 207 – 231. Reed, R.H., Richardson, S.R., Warr, S.R.C., Stewart, W.D.P., 1984. Carbohydrate accumulation and osmotic stress in cyanobacteria. J. Gen. Microbiol. 130, 1 – 4. Riis, V., Maskow, T., Babel, W., 2003. Highly sensitive determination of ectoine and other compatible solutes by anion-exchange chromatography and pulsed amperometric detection. Anal. Bioanal. Chem. 377, 203 – 207. Romano, I., Nicolaus, B., Lama, L., Manca, M.C., Gambacorta, A., 1996. Characterization of a haloalkalophilic strictly aerobic bacterium, isolated from Pantelleria island. Syst. Appl. Microbiol. 19, 326 – 333. Romano, I., Nicolaus, B., Lama, L., Trabasso, D., Caracciolo, G., Gambacorta, A., 2001. Accumulation of osmoprotectants and lipid pattern modulation in response to growth conditions by Halomonas pantelleriense. Syst. Appl. Microbiol. 24, 342 – 352.