Edited by A. Oren
Salt-tolerance genes involved in cation efflux and osmoregulation of Sinorhizobium fredii RT19 detected by isolation and characterization of Tn5 mutants
Version of Record online: 9 JAN 2006
FEMS Microbiology Letters
Volume 239, Issue 1, pages 139–146, October 2004
How to Cite
Jiang, J. Q., Wei, W., Du, B. H., Li, X. H., Wang, L. and Yang, S. S. (2004), Salt-tolerance genes involved in cation efflux and osmoregulation of Sinorhizobium fredii RT19 detected by isolation and characterization of Tn5 mutants. FEMS Microbiology Letters, 239: 139–146. doi: 10.1016/j.femsle.2004.08.029
- Issue online: 9 JAN 2006
- Version of Record online: 9 JAN 2006
- Received 25 May 2004, Revised 15 July 2004, Accepted 22 August 2004
- Sinorhizobium fredii;
- Tn5 mutagenesis;
- Cation efflux system protein;
- Na+ intracellular content;
Salt-tolerance genes of Sinorhizobium fredii RT19 were identified by the construction and screening of a transposon Tn5-1063 library containing over 30,000 clones. Twenty-one salt-sensitive mutants were obtained and five different genes were identified by sequencing. Eight mutants were found with disruptions in the phaA2 gene, which encodes a cation efflux system protein, while mutations in genes encoding other cation effux system proteins were found in seven (phaD2), two (phaF2) and two (phaG2) mutants. A mutation in the metH gene, encoding 5′ methyltetrahydrofolate homocysteine methyltransferase, was found in two of the salt sensitive strains. Growth experiments showed that phaA2, phaD2, phaF2 and phaG2 mutants were hypersensitive to Na+/Li+ and slightly sensitive to K+ and not sensitive to sucrose and that metH mutants were highly sensitive to any of Na+, Li+, K+ and sucrose. Na+ intracellular content measurements established that phaA2, phaD2, phaF2 and phaG2 are mainly involved in the Na+ efflux in S. fredii RT19. Recovery of growth of the metH mutants incubated with different concentrations of NaCl could be obtained by additions of methionine, choline and betaine, which showed that the metH gene is probably involved in osmoregulation in S. fredii RT19.
Rhizobium bacteria are microorganisms with an important economic value, since they can fix nitrogen in symbiotic association with leguminous plants . However, many environmental factors including high osmotic stress often limit the potential of these symbiotic systems and can have a negative effect on both growth and nitrogen fixation of rhizobia . Thus, investigations on the molecular mechanisms of osmoadaptation of Rhizobium bacteria would help us to rationally design and engineer better strains for field application.
To cope immediately with osmotic shocks, many bacteria have developed complex osmoregulatory systems to recover their osmotic balance. In Sinorhizobium meliloti, a complex osmoregulatory system, consisting of the betI-encoded regulator, the betB-encoded betaine aldehyde dehydrogenase, the betA-encoded choline dehydrogenase, the betC-encoded choline sulfatase and the betS-encoded major glycine/proline betaine transporter, is responsible for regulation, biosynthesis and transfer of glycine betaine [3,4]. In Rhizobium tropici, some genes involved in salt tolerance such as ntrY, greA, dnaJ, nifS, noeJ and kup were identified by transposon mutagenesis . The proteins encoded by ndvA and ndvB, which are involved in the synthesis and transfer of β-(1-2)-glucan, play an essential role in the adaptation of S. meliloti to hypo-osmotic stress .
Bacteria also employ cation efflux systems to counteract severe ionic toxicity. In Escherichia coli, three K+ efflux systems have been described and they include KefB, KefC and a third uncharacterized one that is present in the kefB–kefC double mutant . Further, three genes including nhaA, nhaB and chaA, were characterized to exchange Na+, Li+ for H+. In Vibrio alginolyticus, a K+/H+ antiporter was found to enhance the K+ efflux activity after the addition of a membrane-permeable amine . In Vibrio cholerae, it was reported that Na+(Li+)/H+ antiporters including NhaA, NhaB and NhaD and the electron-transport-linked pump (NADH-quinone oxidreductase), were required for the growth and survival in a saline environment .
Sinorhizobium fredii RT19, isolated from saline soil in Tianjin of China, is a halotolerant Rhizobium strain which can tolerate up to 0.6M NaCl, and thus strain RT19 could have developed sophisticated mechanisms to maintain its intracellular steady osmotic and ionic state. In this study, we identified several genes involved in salt tolerance of this strain. Such a genetic research should help us not only understand the halotolerant mechanisms but also characterize genes involved in salt tolerance, which can be important to improve salt tolerance of salt-sensitive rhizobia in the future.
2Materials and methods
2.1Strains, plasmids and growth condition
Sinorhizobium fredii RT19 which was isolated from the saline soil in Tianjin, China, is able to tolerate up to 0.6 mol l−1 NaCl. E. coli DH5α competent cells were purchased from China BioDev-Tech. Co, Ltd, Beijing. Plasmid pRK2013 has been described by Ditta et al. . Plasmid pRL1063a was kindly provided by Dr. C. P. Wolk.
Sinorhizobium fredii RT19 was grown at 28 °C in TY  or FY medium (modified minimal medium with the following composition: per litre: 0.1 g MgSO4; 0.22 g CaCl2; 1 g KNO3; 0.22 g K2HPO4; 0.02 g FeCl3; 10 g mannitol; 75 mg pantothenic acid; 75 mg biotin; 75 mg thiamine). A selective FY medium was used for screening S. fredii RT19 mutants and counter-selecting E. coli strains after mating experiments. E. coli strains were grown in Luria–Bertani medium . Kanamycin (Km) was added at the following final concentrations: 60 μg ml−1 for S. fredii RT19, and 100 μg ml−1 for E. coli strains.
2.2Tn5-1063 mutagenesis and selection of salt-sensitive mutants
Plasmid pRL1063a was introduced into the recipient S. fredii RT19 by three-parental mating, as described by de Bruijn and Rossbach . The only modification to this protocol is that the mating mixtures were plated on selective FY medium with 60 μg ml−1 Km. Salt sensitive mutants were selected by their inability to grow on FY medium containing 0.4 mol l−1 NaCl.
Sinorhizobium fredii RT19 and its salt-sensitive mutants were grown in TY media plus Km for two days, and then 1% of the cultures (5 ml) were added into 5-ml aliquots of FY media plus different agents (see below) and incubation was continued. After four days, growth was determined by measuring the optical density at 600 nm. The agents added were the following: 0–0.8 mol l−1 for NaCl (0–0.8 mol l−1); 0–0.8 mol l−1 for KCl; 0–0.6 mol l−1 for LiCl; 0–0.3 mol l−1for Na2SO4; 0–0.8 mol l−1 for sucrose. For each of sarcosine, methionine, choline and betaine, 1 mmol l−1 was added.
2.4Measurement of Na+ intracellular contents
After grown at 28 °C in liquid TY media for two days, harvested cells were incubated in 0.6 mol l−1 NaCl/10 mmol l−1 Tris–HCl (pH 7.2) at 28 °C for 0–180 min. As described by Nakamura et al. , incubation was terminated with 0.6 mol l−1 choline chloride/10 mmol l−1 Tris–HCl (pH 7.2), after which the cells were washed three times with this isotonic buffered solution. Na+ intracellular contents was measured using a Hitachi Z-5000 Polarized Zeeman atomic absorption spectrophotometer. The Na+ contents of the cells was expressed as nmol/mg protein. Cell protein concentrations were determined by a Coomassie brilliant blue G250 dye-binding assay , using bovine serum albumin as a reference standard.
2.5DNA manipulation and sequence analysis
Plasmid DNA preparation, extractions of total DNA, restriction enzyme digestions, ligations and Southern blotting were carried out as described by Sambrook et al. . Labeling of DNA probes and of DNA hybridizations were performed with DIG High Prime DNA Labeling and Detection Starter Kitl. Sequencing was performed by China Bioasia Bio-Technology Sequencing Co, Ltd, Beijing. Nucleotide–nucleotide blasts were done using the website http://www.ncbi.nlm.nih.gov/blast.
3.1Isolation of salt-sensitive mutants and identification of Tn5-1063 insertions in genomic DNA
A transposon library of over 30,000 clones was constructed and further screened. This library was a collection of clones obtained from a number of different experiments. The obtained library was subjected to screening for salt-sensitive mutants. At last, 21 salt-sensitive mutants, unable to grow on FY medium with 0.4 mol l−1 NaCl, were obtained. Southern blot analysis of genomic DNA of Tn5-1063-inserted mutants, probed with the biggest BglII-digested fragment of pRL1063a, showed single hybridizing bands of different sizes, which suggests a single insertion pattern for Tn5-1063 in S. fredii RT19 (Data not shown) .
3.2Isolation and sequence analysis of the genes in S. fredii involved in salt-tolerance
Due to the presence of an E. coli replication origin in Tn5-1063, the interrupted gene can be easily excised from genomic DNA, self-ligated and transformed into E. coli DH5α. All the interrupted genes were isolated and sequenced (Table 1). The GenBank Accession Number of phaA2, phaD2, phaF2 and phaG2 is AY496950 and that of metH is AY509252.
|Mutant strain||Gene (length)||Insertion site (bp)||Putative ORF function||Closest identity (%)|
|RTa-1||PhaA2 (2375 bp)||200–201||Putative cation efflux system protein, PhaA2, used for inorganic ion transfer and energy metabolism||S. meliloti 1021|
|RTd-1||PhaD2 (1560 bp)||19–20||Putative cation efflux system protein, PhaD2, used for inorganic ion transfer and energy metabolism||S. meliloti 1021|
|RTf-1||PhaF2 (386 bp)||76–77||Putative cation efflux system protein, PhaF2, used for inorganic ion transfer and energy metabolism||S. meliloti 1021|
|RTg-1||PhaG2 (359 bp)||222–223||Putative cation efflux system protein, PhaG2, used for inorganic ion transfer and energy metabolism||S. meliloti 1021|
|RTh-1||metH (3771 bp)||591–592||5′ Methyltetrahydrofolate homocysteine methyltransferase||S. meliloti 1021|
Based on the insertion locus of Tn5-1063, the 21 salt-sensitive mutants were divided into five categories. Since nucleotide–nucleotide blast analysis showed that the interrupted genes have the highest identity with the genes phaA2, phaD2, phaF2, phaG2 and metH of S. meliloti 1021 (Table 1), respectively, they were designated as phaA2 mutants: RTa-1 to 8; phaD2 mutants: RTd-1 to 7; phaF2 mutants: RTf-1, RTf-2; phaG2 mutants: RTg-1, RTg-2; metH mutants: RTh-1 and RTh-2. The genes phaA2, phaD2, phaF2 and phaG2 encode different putative cation efflux system proteins. The gene metH encodes methionine synthase II, also named 5′ methyltetrahydrofolate homocysteine methyltransferase. The genes phaA2, D2, F2 and G2 also have a very high identity with the corresponding genes of Agrobacterium tumefaciens strain C58: mnhA, mnhD, mnhF and mnhG, which encode subunits of putative multiunit Na+/H+ antiporter . Moreover, they share high similarity with the corresponding genes of the putative monovalent cation/proton antiporter of Coxiella burnetii RSA 493, nhaA, nhaD, nhaF and nhaG.
3.3Effect of osmotic agents on the growth of the mutants
In order to identify the sensitivity degree of all the mutants to Na+, we firstly performed growth test of the mutants at different concentrations from 0 to 0.8 mol l−1 NaCl. The results showed that all the mutants involved in putative cation efflux system proteins were unable to grow even in the presence of 0.1 mol l−1 NaCl, implying that these mutants were hypersensitive to NaCl (Fig. 1). In contrast, the two metH mutants showed high sensitivity to Na+ only at the higher concentrations of NaCl (0.4–0.6 mol l−1) (Fig. 2).
In order to evaluate if some osmoprotectants restore growth of the mutants, we tested growth of all the mutants in the presence of Na+ after the addition of choline, betaine, sarcosine. The results showed that choline and betaine almost restored the growth of the metH mutants to normal levels at the different concentrations of Na+ (0–0.6 mol l−1). In contrast to the wild-typed strain, choline and betaine even enhanced growth of the metH mutants even at concentrations of 0.7–0.8 mol l−1 Na+ (Fig. 2).
To determine whether Tn5 insertion indeed impaired the function of metH gene, we also tested the effect of methionine on the growth of the metH mutants at the different concentrations of Na+. The result showed that the addition of methionine restored the growth of the mutants at the different concentrations of Na+, especially below 0.4 mol l−1 Na+ (Fig. 2).
Moreover, to determine if the anion Cl− of NaCl was also responsible for growth arrest of all the salt-sensitive mutants, we tested the tolerance of the mutants to Na2SO4 instead of NaCl. As the results were similar to those of NaCl, this means that Cl− had no influence on the growth of the mutants (Fig. 3(a)).
To determine whether mutants mutated in the above mentioned five genes responded to other osmotic reagents, we tested the tolerance of the mutants RTa-1, RTd-1, RTf-1, RTg-1 and RTh-1 to other agents (Fig. 3(a)–(c)). As can be seen, the former four mutants showed normal growth in the presence of sucrose, while a certain growth reduction was observed in the higher concentrations of KCl (0.5–0.6 mol l−1). Moreover, these mutants were highly sensitive to Li+ since they were unable to grow even at 0.1 mol l−1 LiCl. However, the addition of sucrose had no effect on normal growth of these mutants. By comparison, the metH mutants showed high sensitivity to K+, Li+ and sucrose (Fig. 3).
3.4Difference of Na+ intracellular contents between wild-type strain and pha mutants
To identify to what extent the mutation in the pha genes affect Na+ contents, we measured Na+ intracellular contents of the wild-type strain and the pha mutants RTa-1, RTd-1, RTf-1 and RTg-1 (Fig. 4). With increasing time, Na+ intracellular contents of wild-type strain under high concentration of NaCl (0.6 mol l−1) remained relatively stable whereas four pha mutants under the same condition obviously accumulated more and more Na+ in the cells, which resulted in a much higher Na+ intracellular contents of the four pha mutants compared to the wild-type, especially after 180 min.
Sinorhizobium meliloti 1021 complete genome shows two kinds of pha genes, namely phaA1C1D1E1F1G1 (Accession No. SMc03179-84) and phaA2B2C2D2E2F2G2 (SMc00051-57) . The former genes encode different pH adaptation K+ efflux system transmembrane proteins, respectively, and have been identified to be membrane-spanning proteins and involved in pH adaptation and K+ efflux in S. meliloti 41 . And the latter ones encode different putative cation efflux system proteins, respectively, but have not still been characterized experimentally so far. In our paper, phaA2, phaD2, phaF2 and phaG2, for the first time, were identified experimentally to encode four different cation efflux system proteins.
The genes phaA2, phaD2, phaF2 and phaG2 have the highest identity with the corresponding genes of S. meliloti 1021 (Table 1) and also have very high identity with the corresponding genes mnhA, mnhD, mnhF and mnhG of A. tumefaciens strain C58 . A mnhABCDEFG operon in Staphylococcus aureus has been identified to recover the growth of an E. coli mutant lacking all of the major Na+/H+ antiporters on medium containing 0.2 mmol l−1 NaCl . Moreover, they share high similarity with the corresponding genes of Coxiella burnetii RSA 493 nhaA, nhaD, nhaF and nhaG. Na+/H+ antiporters NhaA, NhaD, NhaG have also been characterized to play a main role in the Na+ and Li+ efflux in many bacteria such as V. cholerea and Helicobacter pylori, Vibrio parahaemolyticus and Bacillus subtilis ATCC9372 [27–29]. Therefore, phaA2, phaD2, phaF2 and phaG2 are possibly closely related to salt-tolerance of S. fredii RT19.
It is worth noting that most salt-tolerant mutations took place in genes encoding phaA2, phaD2, phaF2 and phaG2 and that all the mutants involved in these genes were highly sensitive to Na+ and Li+. It is probable that they make great contributions to the Na+ and Li+ efflux in S. fredii RT19. Comparison of Na+ contents between wild-type strain and pha mutants showed mutation in pha genes resulted in an increased Na+ accumulation in the cells of these mutants, further demonstrating that pha genes play a predominant role in Na+ efflux in the wild-type strain. These mutants also showed growth arrest at the concentrations of 0.5–0.6 mol l−1 KCl, indicating that the four pha genes are involved, to some extent, in K+ efflux. In S. meliloti 41, pha A1B1C1D1E1F1G1 genes are involved in K+ efflux .
Since the mutants of S. fredii RT19, with mutations in the genes phaA2, phaD2, phaF2 and phaG2 showed the same degree of Na+ sensitivity (Fig. 1), it is possible that there is functional relationship or interaction between proteins PhaA2, PhaD2, PhaF2 and PhaG2. And the above proteins share very high identity with the “subunits” of a putative multiunit Na+/H+ antiporter . Therefore, we consider that these proteins play a role in the salt-tolerance as the subunits of a multiunit complex protein, in which a mutation in an individual subunit may lead to loss of function of the multiunit complex protein.
To our knowledge, this is the first report about the involvement of the gene metH in salt-tolerance. Since methionine addition restored growth of the metH mutants in cultures inhibited by salt, the function of the wild-type metH gene is defined. In E. coli, there are two kinds of synthases responsible for the last step of the methionine synthesis, namely MetE (5′ tetrahydropteroyltriglutamate homocysteine methyltransferase) and MetH. MetH was characterized to play a main role in the synthesis of methionine and the control of expression of metE. [30,31]. These two enzymes are also shown in the genome sequence of S. meliloti 1021 . The mutation in metH had some effect on growth of S. fredii RT19 on FY medium, possibly because of loss of control of metE expression by MetH (Fig. 2), which leads to a reduced methionine synthesis. S-adenosyl-l-methionine, called “active methionine”, is a direct metabolite of methionine. The methyl group of this molecule contributes to the synthesis of choline (the precursor of betaine), creatine, sarcosine, etc [32,33]. Therefore, the growth reduction of the metH mutants under osmotic stress is probably due to the insufficient supply of “methyl” groups of such osmolytes as choline and betaine. Interestingly, the addition of choline or betaine could also efficiently restore the growth of metH mutants under osmotic stress. The enzyme used for methionine synthesis with betaine as the methyl donor, betaine–homocysteine S-methyltransferase, is not shown in S. meliloti 1021 complete sequence . As shown in Fig. 2, betaine restore the growth of metH mutants even at higher concentrations of NaCl (0.4–0.8 mol l−1), but methionine could not under the same conditions, and choline could not restore the growth of metH mutants in the absence of NaCl. It is impossible that choline and betaine restore the growth of metH mutants by producing methyl groups for methionine synthesis. Therefore, both compounds probably function as osmoprotectants. These explain the involvement of metH in the osmoregulatory system.
Tn5-1063 was repeatedly found to be inserted at the same site of identical genes (Table 1). Since these mutants were obtained from different experiments, it was impossible that these strains are derived from cell multiplication. Therefore, it looks like a slight preference of transposon Tn5-1063 to insert into some special DNA sequences.
We thank Coleman Peter Wolk for donating the plasmid pRL1063a and some good advice. We are also grateful to Ton van Brussel (Leiden University the Netherlands) and Weimin Gao (Oak Ridge National Laboratory) for critical reading of the manuscript.
This work was supported by China Key Base Research Developing Project Programme (001CB108905), Chinese National Program for High Technology Research and Development (2003AA241150) and European Commission INCO-DC Programme (ICA4-CT-2001-10056).
- Drevon, J.J., Obaton, M. (1983) Technical handbook on symbiotic nitrogen fixation (legume/rhizobium). FAO/GRET, Rome.
- Stumpe, S., Schireser, A., Schleyer, M., Bakker, E.P. (1996) K+ circulation across the prokaryotic cell membrane: K+-uptake systems. In: Handbook of Biological Physics (Konings, W.N., Kaback, H.R., Lolkema, J.S., Eds.), pp.474–499 Elsevier, Amsterdam.
- Silhavy, T.J., Berman, M.L., Enquist, L.W. (1984) Experiments with Gene Fusions. Cold Spring Habor Laboratory, Cold Spring Harbor, NY.
- de Bruijn, F.J., Rossbach, S. (1994) Transposon mutagenesis. In: Methods for General and Molecular Biology (Gerhardt, P., Murray, G.E., Wood, W.A., Krieg, N.R., Eds.), pp.387–405 American Society for Microbiology, Washington DC.
- Sambrook, J., Fretsh, E.F., Maniatis, T. (1989) Molecular Cloning: A Laboratory Manual, 2nd edn. Cold Spring Habor Laboratory, Cold Spring Harbor, NY.
- Milcamps, A., Ragatz, D.M., Lim, P., Berger, K.A., de Bruijn, F.J. (1998) Isolation of carbon- and nitrogen-deprivation-induced loci of Sinorhizobium meliloti 1021 by Tn5-luxAB mutagenesis. 144, 3205–3218
- Shen, T., Wang, J. (1995) Biochemistry. Chinese High Education Press, Beijing.