Osmoprotection Mechanisms in Rhizobia Isolated from Vicia Faba Var

Osmoprotection mechanisms in rhizobia isolated from Vicia faba var. major and Cicer arietinum Fatiha Brhada, M. Poggi, G. van de Sype, Daniel Le Rudulier

To cite this version:

Fatiha Brhada, M. Poggi, G. van de Sype, Daniel Le Rudulier. Osmoprotection mechanisms in rhizobia isolated from Vicia faba var. major and Cicer arietinum. Agronomie, EDP Sciences, 2001, 21 (6-7), pp.583-590. 10.1051/agro:2001148. hal-00886135

HAL Id: hal-00886135 https://hal.archives-ouvertes.fr/hal-00886135 Submitted on 1 Jan 2001

HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non, lished or not. The documents may come from émanant des établissements d’enseignement et de teaching and research institutions in France or recherche français ou étrangers, des laboratoires abroad, or from public or private research centers. publics ou privés. Agronomie 21 (2001) 583Ð590 583 © INRA, EDP Sciences, 2001

Original article

Osmoprotection mechanisms in rhizobia isolated from Vicia faba var. major and Cicer arietinum

Fatiha BRHADAa*, M.C. POGGIb, G. VAN DE SYPEb, Daniel LE RUDULIERb

a Laboratoire de Microbiologie et Biologie moléculaire, Faculté des Sciences, Université Mohammed V, BP 1014, Rabat, Morocco b Laboratoire de Biologie Végétale et Microbiologie, Faculté des Sciences et Techniques, Université de Nice-Sophia Antipolis, France

(Received 22 January 2001; revised 9 May 2001; accepted 13 July 2001)

Abstract Ð Research of mechanisms involved in osmoprotection of two rhizobia strains isolated from nodules of Vicia faba var. major and one strain nodulating Cicer arietinum, showed that choline and glycine-betaine improved growth in salt stress conditions. Provided radioactive choline was converted into glycine betaine catabolized at low osmolarity and accumulated under osmotic stress. Enzyme activities involved in the synthesis of glycine betaine from choline were not modified by salt whereas addition of choline to the growth medium enhanced them. Exogenous radioactive glycine betaine was accumulated in salt stressed bacteria after one hour incubation but catabolized four hours later, suggesting a transient accumulation. Growth experiments indicated that betaine and its derivatives could be used as nitrogen and carbon sources. An investigation, by liquid phase chromatography, revealated accumulation of glutamate and alanine at different levels according to the strain and degree of stress. rhizobia / osmoprotection / choline / glycine betaine

Résumé Ð Mécanismes d’osmoprotection chez les rhizobia isolés de Vicia faba var. major et de Cicer arietinum. La recherche des mécanismes impliqués dans l’osmoprotection chez deux souches de rhizobium isolées des nodosités de Vicia faba var. major et une souche nodulant Cicer arietinum, a montré que la glycine bétaïne et la choline améliorent la croissance bactérienne sous stress salin. Lors d’études métaboliques, la choline intracellulaire radioactive est transformée en glycine bétaïne et ce indépendamment de l’osmolarité du milieu. En effet, les activités choline déshydrogénase et glycine bétaïne aldéhyde déshydrogénase ne sont pas modi- fiées par le sel. Néanmoins, la présence de la choline dans le milieu augmente les deux activités enzymatiques. L’incubation en pré- sence de la glycine bétaïne durant une heure est suivie d’une accumulation de ce composé chez les cellules cultivées à 0,15 M NaCl. Cette accumulation n’a plus lieu quatre heures plus tard. Des expériences de croissance avec la glycine bétaïne ou ses métabolites montrent son utilisation comme source de carbone et d’azote. La recherche, par chromatographie en phase liquide a révélé l’accumulation du glutamate et de l’alanine à des niveaux variables en fonction de la souche et du degré de stress. rhizobia / choline / glycine betaine / osmoprotection

1. INTRODUCTION Due to their excessive salt concentrations and high pH values, saline and alkali soils are unsuitable for the Salinity is one of the major factors responsible for growth of the crop legume plants and their root-nodule deterioration of soil and making it unfit for agriculture. bacteria. The osmotic stress prevailing in these areas and

Communicated by Jean-Jacques Drevon (Montpellier, France)

* Correspondence and reprints [email protected] 584 F. Brhada et al.

non-adaptability of both symbionts may be important Growth of this bacterium at elevated osmolarity results limiting factors. It has been shown in many reports that in inhibition of this catabolism [1, 25]. Although the soil salinity may have an adverse effect on the establish- choline-betaine pathway has been well characterized in ment of functional symbiosis [24]. Hence, it is necessary S. meliloti, there is little information concerning other to inoculate with strains selected for their osmotic toler- members of Rhizobiaceae. ance. rhizobia differ in their ability to tolerate salt [8, 12]. The upper limit of salt concentration tolerated by We have recently examined salt tolerance of two rhi- species of rhizobia depends on strains, varying from zobia, RlF12 and RlF16, isolated from nodules of Vicia 0.2% to 10% of NaCl [31]. El Sheikh and Wood [8] faba and one, Rch60, from Cicer arietinum cultivated in reported that fast-growing rhizobia were more salt-toler- arid and semi-arid areas of Morocco. Increasing the salt ant than the slow growers. concentration in the medium reduced both the growth rate and the final yield of all strains but less in the chick- An increase in extracellular osmolarity inhibits a pea strain, Rch60, more salt-tolerant [5]. Addition of number of physiological and biochemical activities in glycine betaine or choline at 1 mM in salt-added medium the bacteria, hence growth is generally reduced. had a beneficial role on the growth of the three strains. However, bacteria have involved a variety of adaptive However, the choline effect was delayed and less pro- mechanisms in order to restore the cell turgor pressure nounced. Both molecules were taken up by cells grown and then reduce the osmotic potential between the cell at low osmolarity, and whereas glycine betaine uptake and the environment [4]. Among these mechanisms, one activity was stimulated in cells grown in the presence of of the most clearly established is the accumulation of 0.15 M NaCl, choline uptake was strongly inhibited by ions such as potassium [30] and organic compatible salt in all tested strains. In cells grown with exogenous solutes like amino acids, sugars and betaines [16]. choline, the uptake inhibition exerted by salt was The most universally adopted compatible solute is relieved, mainly in the strain isolated from nodules of glycine betaine, which can be accumulated during stress Cicer arietinum. On the basis of kinetic determinations, by a large variety of bacteria [7]. This was reported for in control cells as well as in salt-stressed cells, only Sinorhizobium meliloti in 1986 by Le Rudulier and high-affinity uptake activities were observed for glycine Bernard [17]. Recently, Boncompagni et al. [2] reported betaine and choline. Periplasmic proteins that bound [ ] the utilization of glycine betaine as osmoprotectant by glycine betaine or choline were identified 5 . Rhizobium tropici, R. galegea, Sinorhizobium fredii, In this paper we provide information on the biosyn- Mesorhizobium loti and M. hualkuii. The accumulation thesis and catabolism of glycine betaine in strains RlF12, of glycine betaine, which enhances growth at high osmo- RlF16 and Rch60, and results of research into other larity, results either via transport from the environment osmo-protectants by liquid phase chromatography or via biosynthesis from precursors, choline or glycine (HPLC). betaine aldehyde [7]. In a few bacteria such as the halophilic Ectothiorhodospora, de novo biosynthesis is observed [11]. 2. MATERIALS AND METHODS Biosynthesis of betaine results from oxidation of choline via a two-step reaction with betaine aldehyde as 2.1. Bacteria and medium intermediate. In microorganisms two choline oxidase systems are possible. A soluble choline oxidase system Rhizobium strains RlF12 and RlF16, isolated from capable of catalysing both choline and betaine aldehyde Vicia faba, strain Rch60 nodulating Cicer arietinum and oxidations in vitro has been identified in Arthrobacter [ ] [ ] the Sinorhizobium meliloti strain 102F34 were main- globiformis 14 and Alcaligenes sp. 23 and a mem- tained on Yeast Extract Mannitol medium, pH 6.8 [28]. brane-bound choline dehydrogenase employed in conju- gation with a soluble betaine aldehyde dehydrogenase is In osmoprotection assays, strains RlF12, RlF16 and used in Pseudomonas aeruginosa [19], in E. coli [15] Rch60 were grown at 30 ¡C in a mannitol-aspartate-salts and also in Sinorhizobium meliloti [25]. Furthermore, the medium (MAS). Increased osmolarity was obtained by actual pathway of biosynthesis of glycine betaine from adding NaCl at 0.15 and 0.3 M. The osmolarity of the choline and its degradation by this species is well known different media was measured by using a microosmome- and evidence of their osmomodulation has been provided ter (H. Roebling) and corresponds to 52, 327 and [25]. Effectively, if glycine betaine is a metabolic end 600 miliosmoles/kg water respectively for MAS medium product in Escherichia coli and other enteric bacteria 0 M, 0.15 M and 0.3 M NaCl. When added, glycine [21] this compound is catabolized by S. meliloti in media betaine and choline were used at 1 mM. S. meliloti was of low osmolarity, salvaging both carbon and nitrogen. cultured at 30 ¡C in LAS, a lactate-aspartate-salts Osmoprotection mechanisms in rhizobia 585

medium [22]. Cultures were grown aerobically at this substrate and its suspected metabolites, dimethyl- 200 rpm and growth was monitored at 600 nm. glycine, monomethylglycine and serine as sole carbon and/or nitrogen sources was tested. Cells grown on YEM 2.2. Glycine betaine biosynthesis and catabolism medium were harvested, washed in S medium (salt ele- ments of the MAS) and then used to inoculate the To determine the effect of salt on the cytosolic fate of S medium alone or with mannitol (5 g/l) or NH4Cl choline and glycine betaine, RlF12, RlF16 and Rch60 (5 mM) with glycine betaine or its derivatives at 1, 5 and cells suspensions (3 ml, optical density of 0.6 to 1 unit, 10 mM. Growth was monitored at 600 nm. middle of the exponential phase) from MAS at 0 and 0.15 M NaCl were incubated in two Warburg vials, with 1 µM of [1,2-14C]-choline or [1,2-14C]-glycine betaine. 2.5. Intracellular amino acid analysis 14 CO2 was trapped with KOH. After one hour’s incubation with shaking at 30 ¡C, bacteria were collected on Strains RlF12 and RlF16 were grown in YEM medi- cellulose-nitrate filter (0.45 µm pore size; Millipore) and um at 0 and 0.15 M NaCl. The strain Rch60 was grown extracted with 70% (v/v) ethanol [1]. Radioactivity of at 0, 0.15 and 0.3 M NaCl. Cells were harvested at the each fraction, CO2, ethanol-insoluble, ethanol-soluble end of log phase. Extraction was carried out at 75 ¡C in and filtered medium, was measured. Soluble compounds ethanol, once at 95% and twice at 80%. The extracts were separated by high-voltage paper electrophoresis, were evaporated and clarified with chloroform at 4 ¡C and the radioactivity of choline and glycine betaine was and the pH supernatent was adjusted to 7 before determi- determined in a liquid scintillation spectrometer nation of protein concentration. Aliquots (70 mg protein) (LS 6000, Beckman Instruments). were passed through an SP-Sephadex column. Elution of the cationic fraction, corresponding to the amino acids, 2.3. Choline dehydrogenase and betaine aldehyde was done with NH4OH 0.2 M. The eluate was evaporat- dehydrogenase assays ed at 35 ¡C and resuspended in sterile water. Amino acids were measured in 100 µl of the purified sample These activities were measured in strains RlF16 and using the PicoTag method [26]. The high-pressure liquid Rch60 grown in MAS medium at 0 M and 0.15 M NaCl chromatography (HPLC) system was a Waters Milford added or not with 10 mmol/l of choline or glycine Mass with column PIN 10950. Identification and quan- betaine and, as a control in S. meliloti from LAS 0 M tification of the amino acids was established by co-chro- NaCl. Cells were collected at the log phase, washed and matography with standards (Sigma, A 9906). resuspended in a 100 mM potassium phosphate buffer pH 7.6. For the determination of choline dehydrogenase activity, a cell volume suspension equivalent to 200 µg 2.6. Analysis of protein content of total proteins permeabilised with toluene (0.5%) 10 min at 200 rpm, 30 ¡C. Enzymic activity was immedi- The protein concentrations of cell suspensions or ately assayed after toluene treatment by measuring crude enzymic extracts were assayed by the Lowry [methyl-14C]betaine aldehyde produced from [methyl- method [18]. 14C]choline [15]. [14C]betaine aldehyde was isolated by ion-exchange chromatography and quantified by liquid scintillation. To determine the betaine aldehyde dehydro- 3. RESULTS AND DISCUSSION genase activity, 3 ml of potassium phosphate buffer resuspended bacteria were disrupted in with a French 3.1. Effect of salt stress on glycine betaine press at 10000 psi (700 bars) and enzyme activity was biosynthesis and catabolism followed in 200 µl of the supernatant (about 400 µg protein) at 340 nm via NADH produced from betaine alde- When strain RlF16 cells were grown in low osmolari- hyde and NAD+. The mixture contained 10 mM-NAD+ ty medium (Fig. 1A), the ethanol-soluble and ethanol- and 1 mM-glycine betaine aldehyde, suggested condi- insoluble fractions were nearly equal after one hour’s tions to ovoid interference with dehydrogenase NADH incubation and only 14% of the radioactivity taken up activity [25]. remained as choline, whereas 24 and 18% corresponded to glycine betaine and other soluble compounds. In salt- 2.4. Glycine betaine and its derivatives as substrates stressed cells (Fig. 1B) the insoluble fraction represented only 10% while the ethanol-soluble fraction doubled To determine the pathway of glycine betaine catabo- with a greater percentage of 14C label appearing in lism, the ability of the strains RlF16 and Rch60 to use glycine betaine (38%) and its metabolites (35%). Similar 586 F. Brhada et al.

(Fig. 1B) was followed by a temporary inhibition of the betaine catabolism: after 1 h, this compound still accounted for 70% of the absorbed radioactivity and the insoluble-ethanol fraction represented only 5%. However, when incubation time was longer (5 h), the radioactivity of the insoluble fraction strongly increased and betaine and its products were incorporated into the cell macromolecules. Similarly, in RlF12 and Rch60 bacteria, even if osmotic stress delayed the glycine betaine degradation, it was not really blocked. These results were confirmed using separate cultures.

It is clear from experiments on the fate of choline that patterns were influenced by the salt concentration of the medium: at low osmolarity, 14C accumulated appeared in the ethanol-insoluble fraction but at high osmolarity it remained in the soluble one. Similar results have been obtained with S. meliloti [1, 25]. In this species, salt treatment modified the distribution of the radioactivity within the soluble fraction: at high osmolarity a highest label glycine betaine suggesting a more efficient synthesis. Interestingly, in the bacteria studied here the percentage of glycine betaine did not change, suggesting that its biosynthesis seems not to be salt-stimulated.

Furthermore, results of the fate of glycine betaine in tested strains allowed us to conclude that degradation of this substrate occured in bacteria grown either at low or highosmolarity. Nevertheless, when cells were from Figure 1. Fate of choline and glycine betaine after 1 and NaCl-added medium this catabolism was delayed, sug- 5 hours in strain RlF16 isolated from Vicia faba and grown in gesting a temporary inhibition by salt in all strains. In MAS medium 0 M NaCl (A) and 0.15 M NaCl (B). CO2 (), contrast, a high osmotic stress in S. meliloti grown medi- Ethanol soluble fraction (), Ethanol insoluble fraction ( ). um enhances the glycine betaine accumulation [1, 25]. Ch: choline, GB: glycine betaine, M: metabolites of GB, 100% However, Talibard et al. [27] followed the fate of glycine = CO2 + Ethanol Soluble Fraction (FST) + Ethanol Insoluble betaine throughout the growth cycle of S. meliloti and Fraction (FIT). demonstrated that bacteria accumulate betaine as a cyto- plasmic osmolyte only during lag and early exponential phases of growth in young cultures and this compound disappears when stressed cells reach the stationary phase. Concerning rhizobia studied here, as choline and data was obtained with RlF12, the second strain isolated glycine betaine added at 1 mM to the synthetic medium from Vicia faba, and also with strain Rch60 nodulating did not change growth of bacteria [5], the enhancement Cicer arietinum (not shown). These results indicate that of growth observed in salt-added medium is mainly due all strains transform choline into glycine betaine which is to a transient osmoprotectant effect of glycine betaine. catabolized mainly at low osmolarity. So, differences observed between 1 and 5 hours’ incubation could be attributed to the physiological state of bac- When cells of RlF16 grown at low osmolarity were teria. Effectively, even if bacteria were at the middle of incubated with radioactive glycine betaine (Fig. 1A) the exponential phase growth at the begining of incubation insoluble-ethanol fraction contained 65% of the absorbed with radioactive substrates, they were harvested with radioactivity after 1 h incubation and increased after 5 h. 4 hours delay. In addition, Talibart et al. [27] revealated The 14C label recovered into glycine betaine and its the uncertainty about the S. meliloti to accumulate metabolites also decreased and after 5 h, the radioactivi- glycine betaine as osmolyte during long periods of ty remaining as glycine betaine and its derivatives was osmotic stress and evoked the possibility that high levels extremely low. Addition of NaCl to the medium of cytosolic betaine in this rhizobia may be regulatory Osmoprotection mechanisms in rhizobia 587

and could reverse the osmotic inhibition of betain catab- increased when RlF16 and Rch60 were sub-cultured olism. with choline. In addition, betaine aldehyde dehydrogenase activity was salt-stimulated in Rch60 choline sub- cultured. Interestingly, we previously demonstrated that 3.2. Enzyme activities in the glycine betaine choline uptake activity was inhibited by salt in the two biosynthetic pathway strains but in cells grown with exogenous choline, the uptake inhibition by salt was relieved. Furthermore, on the basis of Km determination we showed, in both RlF16 For both strains, choline dehydrogenase activity was and Rch60, a constitutive choline uptake activity, partial- low in free choline medium, about 10 nmol/min/mg pro- ly inhibited in salt-added medium and in the strain tein (Tab. I) and addition of salt did not change this Rch60, an inducible activity salt-stimulated [5]. Data amount. The presence of choline at 10 mM in the growth suggest that low enzyme activities in 0.15 M NaCl medi- medium increased this enzyme activity two- to three-fold um may be related to a low availability of substrate. time whereas addition of glycine betaine at the same molarity had no effect. The measured betaine aldehyde dehydrogenase activity was also very low in the two species and became undetectable in cells grown in salt- 3.3. Bacterial growth with glycine betaine added medium but was stimulated ten-fold in cells and its derivatives grown with choline 10 mM. However, compared to the choline dehydrogenase and the betainal dehydrogenase activities determined in S. meliloti grown in 7 mM added choline medium (146.5 and 56 nmol/min/mg protein), To determine if the glycine betaine catabolism path- those measured for RlF16 and Rch60 were negligible. way was the same as that proposed for S. meliloti by Smith et al. [25], the enzyme assays were complemented Enzymic studies indicated that a choline and a betaine by growth experiments in order to see if the glycine aldehyde dehydrogenases exist in RlF16 and Rch60, betaine and its expected derivatives can be used as respectively isolated from Vicia faba and Cicer ariet- growth substrate. Cells were grown in nitrogen and car- inum. Previously, Smith et al. [25] reported for bon-free medium containing at 1, 5 and 10 mM glycine S. meliloti the existence of a membrane-bound CDH and betaine, dimethylglycine, monomethylglycine (sarco- a soluble BDH. Comparison of enzymic activities in sine) or serine as sole carbon and/or nitrogen source. For control and salt-stressed bacteria of studied strains pro- all substrates, no growth occured at 1 mM, 5 mM and vide confirmation about the absence of any salt stimula- 10 mM gave comparable results. RlF16 had poor growth tion on the glycine betaine biosynthesis from choline. with glycine betaine alone and also when ammonium The authors above showed that in S. meliloti choline chloride was added. Addition of mannitol allowed cells dehydrogenase activity was not modified by salt but the to grow as well as in MAS medium (Tab. II). Thus, this betainal dehydrogenase activity increased. Although, as strain can use glycine betaine mainly as a nitrogen was reported for S. meliloti, both enzymes’ activities source.

Table I. Effect of choline, betaine and salt addition on choline and glycine betaine aldehyde dehydrogenases activities of RlF16 and Rch60.

Culture conditionsa Strain RlF16 Strain Rch60

Choline Betainal Choline Betainal dehydrogenasea dehydrogenaseb dehydrogenasea dehydrogenaseb

Control 14.7 1.4 9.5 2.5 Control + Ch 20.8 12.1 31.3 16.9 Control + GB 10.5 3.2 8.9 NL 0.15 M NaCl 10.7 NLc 14.2 NL 0.15 M NaCl + Ch 25.7 12.2 29 24.7 0.15 M NaCl + GB 12 NL 10.8 NL a Cells were grown in MAS medium; choline (Cho) and glycine betaine (GB) were at 10 mM. b Enzymic activities are expressed in nmol/min/mg protein. c NL: nearly null (results are means of two duplicate assays, standard deviations never exceeded 10%). 588 F. Brhada et al.

Table II. Use of glycine betaine and its metabolites as carbon or nitrogen source by RlF16 and Rch60 respectively isolated from Vicia faba and Cicer arietinum.

Culture conditionsa Strain RlF16 Strain Rch60

Gb ODc Gb ODc

MAS 13 ± 1 2.8 ± 0.2 15 ± 1 2.5 ± 0.5 S + M + GB 16 ± 1 2.4 ± 0.2 26 ± 1 2.1 ± 0.3 S + M + DMG 14 ± 2 2.6 ± 0.2 25 ± 1 2.3 ± 0.3 S + M + SAR 25 ± 3 1.5 ± 0.4 40 ± 4 0.8 ± 0.2 S + M + SER 12 ± 1 3.1 ± 0.3 16 ± 1 2.8 ± 0.4 + S + NH4 + GB Ð 0.3 ± 0.1 48 ± 3 0.9 ± 0.2 + S + NH4 + DMG Ð 0.4 ± 0.1 50 ± 2 0.7 ± 0.2 + S + NH4 + SAR Ð 0.2 ± 0.1 Ð 0.3 ± 0.1 + S + NH4 + SER Ð 0.4 ± 0.1 54 ± 2 0.6 ± 0.1 a Cells were grown in minimal medium without added NaCl (MAS), in nitrogen and carbon-free MAS medium (S), in S + 5 g/l of mannitol (M), or in S + 5 mmol/l of ammonium chloride (NH4); added by 5 mM glycine betaine (GB) or dimethylglycine (DMG), sarcosine (SAR), serine (SER). b Generation time in hours. c Final optical density was determined after 6 days.

Strain Rch60 showed the same capability but was also 0.15 M NaCl in the case of strain RlF16 (three fold), this able to use the betaine as a carbon source, although man- molarity had no effect on this parameter in strain Rch60, nitol was a better substrate. Similar results have been more salt-tolerant. In this strain, the amino acids’ con- obtained with the suspected intermediates degradative centration increase at 0.3 M NaCl (more than four-fold). pathway. However, if the glycine betaine, dimethyl- Salt also had an effect on the relative composition of the glycine and serine were used as nitrogen source by amino compounds (Tab. III). In RlF16 glutamate RlF16 and as nitrogen and carbon sources by Rch60, sar- increased (× 3.5) at 0.15 M NaCl. This salt molarity did cosine was less effective while serine gave the best not modify the glutamic concentration in Rch60, more results. So, there is evidence that as in S. meliloti, the salt-tolerant. However, in 0.3 M NaCl medium the glycine betaine is probably catabolized by demethylation amount of glutamate increased about 5.7 times. The in both RlF16 and Rch60. amide, glutamine, increased about 7 times in RlF16 and 22.5 times in Rch60 respectively, grown at 0.15 and Within the rhizobia, the pathway of glycine betaine 0.3 M NaCl. At those molarities, alanine concentration degradation is well known, but so far, only in free-living tripled in RlF16 and doubled in Rch60. On the other S. meliloti [25] and in its bacteroids [9]. In S. meliloti hand, in both strains studied proline does not seem to glycine betaine substrate was catabolized by series of play any osmoprotecting role. demethylation giving dimethylglycine, monomethylglycine, glycine and serine. The large investigation of Studies have indicated that many rhizobia adapt to Boncompagni et al. [2] within varoius Rhizobium, high levels of osmolarity by increasing their intracellular Sinorhizobium, Mesorhizobium, Bradyrhizobium and level of glutamate. This was observed in Rhizobium sp. Agrobacterium reference strains, revealed that except nodulating Prosopis [13] and in Bradyrhizobium japon- B. japonicum, all strains were able to use glycine betaine icum [29]. Glutamic acid is also accumulated under and choline as sole carbon and nitrogen sources. The osmotic stress in R. meliloti [3], R. leguminosarum [6] autors suggested that this catabolic function, reported for and R. fredii [10]. The accumulation of this acidic amino few soil bacteria, could increase competitiveness of rhi- acid may act as counter ion to the accumulated K+ [7]. zobial species in the rhizosphere. An increased alanine level in high osmotic medium was reported in the case of four strains of rhizobia nodulating Acacia [12]. 3.4. Intracellular amino acid analysis Interestingly, an investigation into the dynamics of the endogenous osmolytes throughout the growth cycle Growth in salt-added medium increased the total con- of salt-stressed S. meliloti strains showed that bacteria centration of amino acids in the two tested strains accumulated glycine betaine at the lag and the early (Tab. III). However, if this increase was significant at exponential phase. Then betaine decreased during the Osmoprotection mechanisms in rhizobia 589

Table III. Effect of NaCl on the intracellular concentration of amino acids in strains RlF16 and Rch60, respectively isolated from Vicia faba and Cicer arietinum.

Amino acid Strain RlF16 Strain Rch60 (nm⋅mgÐ1 proteins) NaCl 0 M NaCl 0.15 M NaCl 0 M NaCl 0.15 M NaCl 0.3 M

Aspartic A. 5.9 13.5 3.9 1.7 10 Glutamic A. 86 303.9 30 39.2 170 Serine 3.1 4.7 1.1 4.1 14.9 Asparagine 1.4 2.4 0.7 1.1 2.8 Glycine 17.4 11.7 7.6 7.8 32.8 Glutamine 3.9 27 2.5 15.1 56.3 §-alanine 1.8 7.5 0.8 0.5 35.9 Alanine 33 95 77.9 107.8 146.5 A. aminobutyric 2.1 9.2 0.1 0.6 15.1 Arginine ND 7 0.9 2.5 32.9 Proline 5.1 3.3 2.6 2.1 ND Tyrosine 2.1 2.6 0.6 0.7 16.2 Valine 3.2 3.6 7.7 1.6 20.8 Methionine 1.5 1.8 0.1 0.1 2.1 Cystine 4 7.7 3.1 3.6 40.8 Leucine+Isoleucine 6.8 7.3 1.3 1.5 9 Phenylalanine 1.9 3 0.5 0.6 2.7 Tryptophane 1.2 5.3 ND ND 4.3 Ornitine 0.8 3.1 1.5 1.2 17.6 Lysine 4.9 14.1 7.9 6.4 41.2 Total 186 534 151 198 672

ND: Not detected. Values obtained with duplicate, deviation was about 1/8.

second half of the exponential phase while levels of glu- ria while it was catabolized at low osmolarity. The fact tamate and the N-acetylglutaminylglutamine amide that the accumulation of glycine betaine seems to be increased supplanting glycine betaine in ageing cultures transitory led us to search for other accumulated solutes [27]. In the case of strains RlF16 and Rch60, amino acid by HPLC. Under salt stress, both strains accumulated analysis used extracts from cells harvested at the end of glutamate and alanine but at different levels according to log phase while betaine tests were conducted with bacte- the strain and to the degree of stress. ria in the middle of exponential phase growth. So, it seems that as in S. meliloti, betaine acts as osmoprotec- Acknowledgements: F. Brhada’s sabbatical leave at the tant in young bacteria and is replaced by glutamate in University of Nice-Sophia Antipolis was sponsored by a grant older cultures. from the European Community (TS 2A 0107). This work was supported by the Centre National de la Recherche Scientifique.

4. CONCLUSION REFERENCES

Research of mechanisms involved in osmoprotection [1] Bernard T., Pocard J.A., Perroud B., Le Rudulier D., of rhizobia isolated from nodules of Vicia faba var. Variation in the response of salt-stressed Rhizobium strains to major and rhizobia from Cicer arietinum showed that betaines, Arch. Microbiol. 143 (1986) 359Ð364. glycine-betaine, or its precursor choline, improved [2] Boncompagni E., Osteras M., Poggi M.C., Le Rudulier growth in salt stress conditions. Intracellular choline was D., Occurrence of choline and glycine betaine uptake and converted into glycine betaine, which was catabolized in metabolism in the family Rhizobiaceae and their roles in osmo- medium of low-osmolarity and accumulated when bacte- protection, Appl. Environ. Microbiol. 65 (1999) 2072Ð2077. ria are cultivated in media with high osmolarity. When [3] Botsford L., Thomas L., Osmoregulation in Rhizobium exogenous glycine betaine was provided, a temporary meliloti production of glutamic acid in response to osmotic accumulation was observed in salt-stressed young bacte- stress, Appl. Environ. Microbiol. 56 (1990) 488Ð494. 590 F. Brhada et al.

[4] Brown A.D., Microbial water stress, Bacteriol. Rev. 40 [18] Lowry O.H., Rosebrough N.J., Farr A.L., Randall R.J., (1976) 803. Protein measurement with the Folin phenol reagent, J. Biol. [5] Brhada F., Poggi M.C., Le Rudulier D., Choline and Chem. 193 (1951) 265Ð275. glycine betaine uptake in various strains of rhizobia isolated [19] Nagasawa T., Kawabata Y., Tani Y., Ogata K., from nodules of Vicia faba var. major and Cicer arietinum L.: Purification and characterization of betaine aldehyde dehydro- Modulation by salt, choline and glycine betaine, Current genase from Pseudomonas aeruginosa A-16, Agric. Biol. Microbiol. 34 (1997) 167Ð172. Chem. 40 (1976) 1743Ð1749. [6] Chien C.T., Maundu J., Cavaness J., Dandurand L.M., [20] Nagasawa T., Mori N., Tani Y., Ogata K., Cindy S.O., Characterisation of salt tolerant and salt sensitive Characterization of choline dehydrogenase from Pseudomonas mutants of R. leguminosarum bv. viciae strain C1204b, FEMS aeruginosa A-16, Agric. Biol. Chem. 40 (1976) 2077Ð2084. Microbiol. Lett. 90 (1992) 135Ð140. [21] Perroud B., Le Rudulier D., Glycine betaine transport [7] Csonka L.N., Physiological and genetic responses of in Escherichia coli: osmotic modulation, J. Bacteriol. 161 bacteria to osmotic stress, Microbiol. Rev. 53 (1989) 121Ð147. (1985) 393Ð401. [8] El Sheikh E.A.E., Wood M., Salt effects on survival and [ ] multiplication of chikpea and soybean rhizobia, Soil Biol. 22 Pocard J.A., Bernard T., Smith L.T., Le Rudulier D., Biochem. 22 (1990) 343Ð347. Characterization of three choline transport activities in Rhizobium meliloti: Modulation by choline and osmotic stress, [9] Fougere F., Le Rudulier D., Glycine betaine biosynthe- J. Bacteriol. 171 (1989) 531Ð537. sis and catabolism in bacteroids of Rhizobium meliloti: effect of salt stress, J. Gen. Microbiol. 136 (1990) 2503Ð2510. [23] Ohta-Fukuyama M., Miyake Y., Emi S., Yamano T., Identification and properties of the prosthetic group of choline [10] Fujihara S., Yoneyama T., Effects of pH and osmotic oxidase from Alcaligenes sp., J. Biochem. 88 (1980) 197Ð203. stress on cellular polyamine contents in the soybean rhizobia, Rhizobium freedi P220 and Bradyrhizobium japonicum A1017, [24] Singleton P.W., Ben Bohlool B.B., Effect of salinity on Appl. Environ. Microbiol. 59 (1993) 1104Ð1109. nodule formation by soybean, Plant Physiol. 74 (1984) 74Ð76. [11] Galinski E., Truper H.G., Betaine, a compatible solute [25] Smith L.T., Pocard J.A., Bernard T., Le Rudulier D., in: the extremely halophilic phototrophic bacterium Osmotic control of Glycine betaine biosynthesis and degrada- Ectothiorhodospira halochloris, FEMS Microbiol. Lett. 13 tion in Rhizobium meliloti, J. Bacteriol. 170 (1988) 3142Ð3149. (1982) 357Ð360. [26] Streeter J.G., Carbohydrate, organic acid and amino [12] Graig G.F., Atkins C.A., Bell D.T., Effect of salinity on acid composition of bacteroids and cytosol from soybean nod- grouth of four strains of Rhizobium and their infectivity and ules, Plant Physiol. 85 (1987) 768Ð773. effectiveness on two species of Acacia, Plant and Soil 133 (1991) 253Ð262. [27] Talibard R., Jebbar M., Gouffi K., Pichereau V., [ ] Gouesbet G., Blanco C., Bernard T., Pocard J.A., Transient 13 Hua S.T., Tsai V.Y., Lichens G.M., Noma A.T., accumulation of glycine betaine and dynamics of endogenous Accumulation of amino acids in Rhizobium sp. strain WR1001 osmolytes in salt-stressed cultures of Sinorhizobium meliloti, in response to sodium chloride salinity, Appl. Environ. Appl. Environ. Microbiol. 63 (1997) 4657Ð4663. Microbiol. 44 (1982) 135Ð140. [28] Vincent J.M., A manual for the practical study of the [14] Ikuta S., Maatuura K., Imamura S., Misaki H., Horiuti root-nodule bacteria, Blackwell Scientific Publications, Y., Oxidative pathway of choline to betaine in soluble fraction Oxford, England, 1970. prepared from Arthrobacter globiformis, J. Biochem. 82 (1977) 157Ð163. [29] Yap S.F., Lim S.T., Response of Rhizobium sp. UMKL [15] Landfald B., Strom A.R., Choline-glycine betaine path- 20 to sodium chloride stress, Arch. Microbiol. 135 (1983) way confers a high level of osmotic tolerance in Escherichia 224Ð228. coli, J. Bacteriol. 165 (1986) 849Ð855. [30] Yelton M.M., Yang S.S., Edie S.A., Lim S.T., [16] Le Rudulier D., Strom A.R., Danekar A.M., Smith Characterisation of an effective salt tolerant, fast-growingb L.T., Valentine R.C., Molecular biology of osmoregulation, strain of Rhizobium japonicum, J. Gen. Microbiol. 129 (1983) Science 244 (1984) 1064Ð1068. 1547Ð1573. [17] Le Rudulier D., Bernard T., Salt tolerance in [31] Zahran H.H., Conditions of successful Rhizobium- Rhizobium: a possible role for betaines, FEMS Microbiol. Rev. legume symbiosis in saline environments, Biol. Fertil. Soils 12 39 (1986) 67Ð72. (1991) 73Ð80.