Abstract Natural habitats containing high amounts of heavy metals provide a valuable source of bacteria adapted to deal with metal toxicity. A functional analysis of the population of legume endosymbiotic bacteria in an ultramafic soil was undertaken by studying a collection of Rhizobium leguminosarum bv viciae (Rlv) isolates obtained using pea as trap plant. One of the isolates, Rlv UPM1137, was selected on the basis of its higher tolerance to nickel and cobalt and presence of inducible mechanisms for such tolerance. A random transposon mutagenesis of Rlv UPM1137 allowed the generation of 14 transposant derivatives with increased nickel sensitivity; five of these transposants were also more sensitive to cobalt. Sequencing of the insertion sites revealed that one of the transposants (D2250) was affected in a gene homologous to the cation diffusion facilitator gene dmeF first identified in the metal-resistant bacterium Cupriavidus metallidurans CH34. The symbiotic performance of D2250 and two other transposants bearing single transposon insertions was unaffected under high-metal conditions, suggesting that, in contrast to previous observations in other Rlv strain, metal tolerance in UPM1137 under symbiotic conditions might be supported by functional redundancy between several mechanisms. rhizobia, nickel, cobalt, metal tolerance, ultramafic soil, Pisum sativum INTRODUCTION Nickel and cobalt are essential microelements for microbial nutrition as they participate in a variety of cellular processes in bacteria (Li and Zamble 2009; Eitinger 2013). These two elements are usually present at low concentrations in soils, and bacteria have developed high-affinity metal uptake systems to take them up (Rodionov et al.2006). At the same time, these two elements can become toxic, even at moderate concentrations, by displacing other metals from the active site of metalloenzymes (Macomber and Hausinger 2011). Ultramafic or serpentine rocks are igneous rocks originated from earth upper mantle, with high concentrations of nickel, cobalt and chromium. Weathering of these rocks results in ultramafic soils with the chemical characteristics of the original igneous rock. This kind of soils presents deficiencies in essential nutrients such as phosphorus and potassium, and also low Ca/Mg ratios (Brady, Kruckeberg and Bradsahw 2005; Ryan, Dell and Claassen 2009). Ultramafic sites occupy about 1% of the earth crust and are considered as ‘ecological islands’ from an evolutionary point of view with flora specifically adapted to this environment (Sobczyk et al. 2017). Soils with high concentrations of metals are affected in crop production, microbial activity and biodiversity (Stefanowicz, Niklinska and Laskowski 2008), and bacteria inhabiting them have developed different mechanisms to maintain low intracellular metal concentrations (Haferburg and Kothe 2007; Mirete, de Figueras and Gonzalez-Pastor 2007). Metal efflux systems are the most prevalent group within such tolerance mechanisms (Bruins, Kapil and Oehme 2000). The potential relevance of bacteria thriving in high-metal soils was illustrated by a recent report (Ma et al.2015) showing that inoculation of Brassica juncea and Ricinus communis with plant-related bacteria isolated from a serpentine environment improved plant growth and translocation and accumulation of nickel, zinc and iron. High-metal niches have been exploited for the identification of novel determinants for nickel resistance through isolation of bacteria (Stoppel and Schlegel 1995) and metagenomic analysis of the rhizosphere of plants adapted to metal-rich habitats (Mirete, de Figueras and Gonzalez-Pastor 2007). More recently, two novel species of rhizospheric Burkholderia adapted to high-metal soils have been described (Guentas et al.2016). Several studies show that bacteria isolated from ultramafic soils or from the rhizosphere of plants growing in this kind of soils contain several metal tolerance mechanisms (Marrero et al.2006; Pal, Wauters and Paul 2007). A relevant fraction of these microorganisms are endosymbiotic bacteria (Abou-Shanab, Berkum and Angle 2007; Klonowska et al.2012), collectively designated as rhizobia. Rhizobia are legume endosymbiotic bacteria able to carry out the biological nitrogen fixation process in association with Fabaceae plants. In this symbiosis, rhizobia occupy special structures called nodules and fix molecular nitrogen to ammonium by means of the nitrogenase enzyme, thus making nitrogen available to the plant (Oldroyd et al.2011). The information on genetic determinants involved in cobalt and nickel tolerance in legume endosymbiotic bacteria is scarce, and restricted to a few reports related to the level of metal tolerance in rhizobia (Pereira, Lima and Figueira 2006; Grison et al. 2015) and descriptions of efflux systems in Ensifer, Mesorhizobium and Bradyrhizobium strains (Chaintreuil et al.2007; Zielazinski et al.2013; Xie et al.2015). As for beta-rhizobia, some endosymbiotic Cupriavidus strains with high levels of metal resistance have been also described (Chen et al.2008; Platero et al.2016). A recent genome-wide association study identified candidate genes for adaptation of Mesorhizobium strains to nickel in ultramafic soils (Porter et al.2017). Such adaptation was previously shown to trade-off with growth under low-metal conditions (Porter and Rice 2013). Following our previous study on the DmeRF nickel and cobalt efflux system from Rhizobium leguminosarum bv. viciae UPM791 (Rubio-Sanz et al.2013), in the current study we undertook the analysis of metal tolerance determinants in R. leguminosarum isolates obtained from an ultramafic soil and analyze their relevance in free-living and symbiotic conditions. MATERIALS AND METHODS Soil sampling and analysis The site studied in this work was located in Gorro (Borgo Val di Taro, Italy, coordinates N 44°31΄24″, E 9°52΄41″) near to a place with characteristic inclusions of ultramafic rocks. Soil samples were collected from the A horizon (5–30 cm) in three different spots (separated 5 m), homogenized and stored at 4°C until its use for the isolation of the rhizobial strains. Total metal content was determined after digestion of 1 g of soil with a mixture of acids (10 ml HF + 10 ml HCl) in a Teflon tube using a microwave oven (Mars Microwave System, CEM Corporation) at 1.2 MPa pressure. The available fraction of metal cations was determined by the DTPA-TEA method (Lindsay and Norvell 1978). All the extractions were made in triplicate, and metal content was determined in a flame atomic absorption spectrophotometer (AAnalyst200, PerkinElmer). Methodologies to determine other physicochemical characteristics of the soils are included as supplementary material. Bacterial growth conditions Rhizobium strains were routinely grown at 28°C in tryptone-yeast extract (TY), yeast-mannitol broth (YMB) or Rhizobium minimum (Rmin) media (Brito et al.2002). Escherichia coli strains were grown in LB medium (Sambrook and Russell, 2002). When required, the antibiotic spectinomycin was added to a final concentration of 50 μg ml−1. Minimal inhibitory concentrations (MIC) for nickel and cobalt were estimated by the ability of the Rhizobium strains to grow on TY plates containing different concentrations of NiCl2 (0.5, 1.0, 1.5, 2.0 and 2.5 mM) or CoCl2 (0.1, 0.25, 0.75 and 1.0 mM). For disk diffusion susceptibility tests, bacterial cultures grown to exponential phase (OD ≈ 0.6) were mixed with warm TY-agar medium to reach a final equivalent OD ≈ 0.05 and poured into Petri plates. Antibiogram-type disks (6 mm) with 15 μl of solutions containing different amounts of NiCl2 (200 mM), CoCl2 (100 mM), ZnSO4 (500 mM) and CuSO4 (500 mM) were placed on the plate surface. The diameter of the inhibition zone was measured after 24 h of incubation at 28°C. Growth curves of bacterial cultures in liquid media were determined by incubation at 28°C with continuous shaking in an incubator-spectrophotometer Bioscreen C instrument (Growth Curves Ltd). In these experiments, cultures were grown in TY medium supplemented with metals (1.5 mM NiCl2 or 0.75 mM CoCl2). DNA manipulation techniques Genomic DNA was isolated from Rhizobium cultures grown in TY for 2 days at 28°C using DNeasy Blood and Tissue Kit columns (Qiagen Ltd). To carry out the random mutagenesis in the Rlv UPM1137 strain, plasmid pSS240, containing the minitransposon Tn5SSoriRgusA (mTn5), was introduced by conjugation (Di Gregorio et al.2004). Spectinomycin-resistant transposants showing increased sensitivity to nickel were selected by replica plating the transposants on TY plates and on TY supplemented with 1.5 mM NiCl2. The region affected by the insertion on each Ni-sensitive transposant was identified by sequencing. To do this we cloned, as PstI fragments, the plasmid replication origin (oriR) present in the mTn5 along with the regions adjacent in the genome. Self-ligated PstI fragments were transformed into E. coli, and transformants were selected in LB plates supplemented with spectinomycin. Plasmid DNA preparations, restriction enzyme digestions and DNA transformations into E. coli were carried out by standard protocols (Sambrook and Russell 2002). Sequencing of the mTn5-flanking region was carried out using the aad1846 oligonucleotide (5΄-GCTGGCTTTTTCTTGTTATCG-3΄). The sequences obtained were compared with the NCBI database using BLAST program at the NCBI (National Centre for Biotechnology Information). Additional information about the identified genes was obtained from RhizoBase (http://genome.kazusa.or.jp/rhizobase/), JGI (Joint Genomics Institute) and NCBI databases. Plant tests Pea (Pisum sativum L. cv. Frisson) trap plants were set up by inoculating surface-sterilized seedlings with 3 ml of a soil/water suspension (25 g/45 ml). Plants (four replicate pots per treatment, three plants per pot) were grown in sterile Leonard jar-type assemblies filled with vermiculite and maintained in a greenhouse with 16/8 h light/dark cycle and 25/23°C day/night temperature under bacteriologically controlled conditions. Nodules from 28-day-old plants were superficially disinfected and crushed, and nodule suspensions were streaked into YMB plates using standard procedures. Isolates were purified by repeated streaking on YMB plates, and their ability to induce effective nodulation in pea was confirmed. Plant shoot dry weight was determined after drying at 60°C for 24 h using at least four samples (three plants each) per condition. Total nitrogen content of the dried shoot samples was determined using a TruMac C/N analyzer (Leco Corporation). To test the symbiotic performance under high-metal conditions, inoculated pea plants were watered (from day 10 after inoculation) with nutrient solutions containing medium or high levels of Ni2+ (85 or 170 μM, respectively) or Co2+ (42.5 or 85 μM, respectively). Plants were grown in vermiculite jars (six plants per condition except for D2250 with three plants). Statistical significance of the data was determined by applying ANOVA followed by the post hoc F-test with MS Excel (Microsoft, Redmond, WA) version 15.13.4 add-in software. The resulting values for F and p parameters are included in Tables S2 and S3 in supplementary material (Supporting Information). RESULTS AND DISCUSSION Physico-chemical analysis of soil Analysis of soil indicated a loamy texture (48/36/16% of sand/silt/clay), with a medium-basic pH (7.2), and a low salt content as deduced from the low level of electric conductivity (101 μS/cm). Analysis of exchangeable cations (Ca2+, Mg2+, Na+, K+) showed a high concentration of magnesium as compared to calcium (10:1 Ca/Mg ratio). This characteristic is usual in ultramafic soils (Brady, Kruckeberg and Bradsahw 2005). The concentrations of heavy metals nickel, cobalt and chromium were high (Table 1), with values of total and available levels of these elements comparable to those reported for ultramafic soils (Kabata-Pendias and Pendias 1999; Kabata-Pendias 2011; Klonowska et al.2012). Other heavy metals such as copper, zinc and lead were also present at high concentrations. Further details on physico-chemical properties are shown in Table S1 (supplementary material). Table 1. Concentrations of heavy metals in soil. Cation Metal fraction Total Available Ni2+ 145 ± 28 2.10 ± 0.60 Co2+ 45.4 ± 6.7 0.31 ± 0.08 Cr6+ 153 ± 7 0.15 ± 0.01 Cu2+ 60.3 ± 8.8 2.61 ± 0.86 Zn2+ 66.2 ± 13.1 0.67 ± 0.25 Pb2+ 36.6 ± 4.9 0.85 ± 0.09 Cation Metal fraction Total Available Ni2+ 145 ± 28 2.10 ± 0.60 Co2+ 45.4 ± 6.7 0.31 ± 0.08 Cr6+ 153 ± 7 0.15 ± 0.01 Cu2+ 60.3 ± 8.8 2.61 ± 0.86 Zn2+ 66.2 ± 13.1 0.67 ± 0.25 Pb2+ 36.6 ± 4.9 0.85 ± 0.09 Values (expressed in mg kg−1) are the average of three replicate determinations ± S.E. View Large Analysis of metal tolerance in R. leguminosarum isolates A total of 23 rhizobial isolates were obtained from root nodules excised from pea trap plants inoculated with a suspension of the ultramafic soil. All isolates were ascribed to R. leguminosarum bv viciae on the basis of partial recA sequencing and, based on RAPD profiles (see Fig. S1, Supporting Information), a total of 8 representative isolates were chosen for metal tolerance analysis through disk diffusion tests (Table 2). In this analysis, one of the isolates (UPM1137) showed a higher level of tolerance to nickel and cobalt, whereas three other isolates (UPM1283, UPM1284 and UPM1286) showed consistently higher levels of tolerance to cobalt as deduced from their smaller inhibition halos. Similar levels of tolerance to divalent cations copper and zinc were observed in all isolates (Table 2). Based on the previous results, Rlv UPM1137 strain was selected for further analysis. Table 2. Metal tolerance of Rlv isolates from Borgo soil. Rlv strain Diameter of inhibition zone (mm) NiCl2 CoCl2 CuSO4 ZnSO4 UPM1283 19 14 18 28 UPM1284 20 15 19 26 UPM1286 20 16 18 29 UPM1287 23 18 22 25 UPM1288 23 19 20 28 UPM1289 27 19 21 29 UPM1290 18 22 18 25 UPM1137 13 15 18 25 UPM791* 23 21 ND ND 3841* 20 20 ND ND Rlv strain Diameter of inhibition zone (mm) NiCl2 CoCl2 CuSO4 ZnSO4 UPM1283 19 14 18 28 UPM1284 20 15 19 26 UPM1286 20 16 18 29 UPM1287 23 18 22 25 UPM1288 23 19 20 28 UPM1289 27 19 21 29 UPM1290 18 22 18 25 UPM1137 13 15 18 25 UPM791* 23 21 ND ND 3841* 20 20 ND ND The final concentration of metal in disks was as follows: NiCl2, 200 mM; CoCl2, 100 mM; CuSO4, 500 mM; ZnSO4, 500 mM. Data are the average of at least two replicate measurements. Standard errors were below 5%. *Reference laboratory strains. View Large We first tested the effect of metals on bacterial growth in liquid culture (Fig. 1). When Rlv UPM1137 was grown in TY liquid media supplemented with high-metal concentrations (1.5 mM NiCl2 or 0.75 mM CoCl2), a clear delay in growth was observed. Pre-incubation with low metal concentration (0.1 mM NiCl2 or 0.05 mM CoCl2) resulted in improved growth when cells were subsequently grown under high levels of nickel or cobalt (Fig. 1). These results suggest the existence of inducible metal tolerance mechanisms in Rlv UPM1137, as previously described for a Serratia marcescens strain isolated from an ultramafic soil (Marrero et al.2006). Nickel-dependent induction of expression has been described for other metal tolerance systems, such as E. coli rcnRA (Koch, Nies and Grass 2007). In all these cases, metal sensor proteins couple specific metal ion binding with a change in their DNA-binding affinity and/or specificity, thus translating the concentration of a certain metal ion into a change in gene expression (Musiani et al.2015). Figure 1. View largeDownload slide Effect of Ni2+ and Co2+ on growth of R. leguminosarum bv. viciae UPM1137 strain. Graphs correspond to growth curves in TY medium with/without low metal pre-incubation (squares/diamonds, respectively), or in TY medium supplemented with the indicated amount of metal with/without low metal pre-incubation (circles/triangles, respectively). Pre-incubation was carried out in the presence of 0.1 mM NiCl2 or 0.05 mM CoCl2. Values are the average of three replicates. Error bars are of the same size as symbols, and are not included. Figure 1. View largeDownload slide Effect of Ni2+ and Co2+ on growth of R. leguminosarum bv. viciae UPM1137 strain. Graphs correspond to growth curves in TY medium with/without low metal pre-incubation (squares/diamonds, respectively), or in TY medium supplemented with the indicated amount of metal with/without low metal pre-incubation (circles/triangles, respectively). Pre-incubation was carried out in the presence of 0.1 mM NiCl2 or 0.05 mM CoCl2. Values are the average of three replicates. Error bars are of the same size as symbols, and are not included. Identification of genetic systems involved in nickel tolerance A general mutagenesis approach was carried out by random insertion of a Tn5-derived minitransposon, bearing a spectinomycin-resistance cassette, followed by selection on TY plates supplemented with NiCl2. Out of 4313 transposants screened for nickel tolerance, 14 were unable to grow on TY plates supplemented with 1.5 mM NiCl2 (Table 3). Six out of the 14 transposants (D0362, D2248, D2249, D2250, D4632, D8134) were also strongly affected in tolerance to cobalt. No major changes in tolerance to neither Cu2+, Zn2+ or Mn2+ ions in any of the 14 transposants were detected in this analysis. Table 3. Characterization of Ni-sensitive Rlv UPM1137 mTn5 derivatives. Strain Codea Protein size (# aa) Annotation Metal sensitivity (ZI)b Symbiotic phenotype NiCl2 CoCl2 CuSO4 ZnSO4 MnCl2 Nod Fix Total Nc UPM1137 - - Wild type 13 15 20 23 28 + + 15.46 ± 0.82 D0013 RL2862 318 Conserved hypothetical exported protein 16 15 19 20 27 – – 3.44 ± 0.06** D0362 RL2436 149 Putative transmembrane protein 24 26 17 22 ND + + 15.09 ± 0.65 D2248 RL2322 213 Conserved hypothetical protein 18 22 19 23 27 + + 13.01 ± 0.32 D2249 pRL110066 572 Putative adenylate cyclase 23 23 18 22 30 + + 14.34 ± 0.81 D2250 RL1351 323 Cation efflux system protein (dmeF) 16 30 17 24 29 + + 15.09 ± 0.96 D3466 RL4539 253 Putative heme exporter protein C (CycZ) 20 15 ND ND ND + – 4.27 ± 0.15** D4239 pRL90287 303 Putative AraC family transcriptional regulatory protein 20 16 ND ND ND + – 4.08 ± 0.14** D4632 RL4188 230 Conserved hypothetical exported protein 22 21 18 24 26 + + 8.08 ± 0.11** D4719 RL2793 363 Putative ATP-binding component of ABC transporter 14 16 24 26 29 + + 15.15 ± 1.26 D4740 RL2100 355 Conserved hypothetical exported protein 16 18 21 20 29 + + 18.48 ± 0.47 D5217 RL0615 161 Putative arsenate reductase 16 16 21 28 28 + + 11.76 ± 0.30* D5451 RL1589 211 Putative ropB outer membrane protein 18 18 ND ND ND + + 15.05 ± 0.15 D6426 pRL110071 276 Conserved hypothetical protein 14 15 20 22 33 + + 10.06 ± 0.59* D8134 RL1553 328 Putative transmembrane protein 24 24 ND ND 31 – – 3.61 ± 0.05** Strain Codea Protein size (# aa) Annotation Metal sensitivity (ZI)b Symbiotic phenotype NiCl2 CoCl2 CuSO4 ZnSO4 MnCl2 Nod Fix Total Nc UPM1137 - - Wild type 13 15 20 23 28 + + 15.46 ± 0.82 D0013 RL2862 318 Conserved hypothetical exported protein 16 15 19 20 27 – – 3.44 ± 0.06** D0362 RL2436 149 Putative transmembrane protein 24 26 17 22 ND + + 15.09 ± 0.65 D2248 RL2322 213 Conserved hypothetical protein 18 22 19 23 27 + + 13.01 ± 0.32 D2249 pRL110066 572 Putative adenylate cyclase 23 23 18 22 30 + + 14.34 ± 0.81 D2250 RL1351 323 Cation efflux system protein (dmeF) 16 30 17 24 29 + + 15.09 ± 0.96 D3466 RL4539 253 Putative heme exporter protein C (CycZ) 20 15 ND ND ND + – 4.27 ± 0.15** D4239 pRL90287 303 Putative AraC family transcriptional regulatory protein 20 16 ND ND ND + – 4.08 ± 0.14** D4632 RL4188 230 Conserved hypothetical exported protein 22 21 18 24 26 + + 8.08 ± 0.11** D4719 RL2793 363 Putative ATP-binding component of ABC transporter 14 16 24 26 29 + + 15.15 ± 1.26 D4740 RL2100 355 Conserved hypothetical exported protein 16 18 21 20 29 + + 18.48 ± 0.47 D5217 RL0615 161 Putative arsenate reductase 16 16 21 28 28 + + 11.76 ± 0.30* D5451 RL1589 211 Putative ropB outer membrane protein 18 18 ND ND ND + + 15.05 ± 0.15 D6426 pRL110071 276 Conserved hypothetical protein 14 15 20 22 33 + + 10.06 ± 0.59* D8134 RL1553 328 Putative transmembrane protein 24 24 ND ND 31 – – 3.61 ± 0.05** aProtein annotation corresponds to Rlv 3841 genome. bZone of inhibition: diameter of the inhibition zone (mm) produced by the halos around disks embedded in metal solution (200 mM NiCl2, 100 mM CoCl2, 500 mM CuSO4, 500 mM ZnSO4 and 100 μM MnCl2). Data are the average of at least two replicate measurements. Standard errors were below 5%. cValues are expressed in mg N/plant and are the average of at least two replicate measurements ± S.E. Asterisks denote significant differences as compared to wild type (*P < 0.05; **P < 0.01). View Large Genes affected by the mTn5 insertion on each transposant were identified by sequencing the DNA region flanking each insertion. All genes identified in this analysis showed high similarity to genetic determinants present in the Rlv 3841 reference genome (Table 3). Most of the genes affected in the nickel-sensitive transposants encoded membrane proteins. Specific annotations were found in only four cases: a cycZ-like gene (D3466), an AraC-like transcriptional regulator (D4239), a RopB-like outer membrane protein (D5451) and a divalent metal cation efflux (DmeF) protein (D2250). The cycZ gene has been previously studied in Ensifer meliloti. In that bacterium, inactivation of the cycZ-like gene ccmC led to the inability to synthesize c-type cytochromes. Ensifer meliloti ccmC mutant exhibited a pleiotropic phenotype, including a reduced ability to grow in rich media (Yurgel et al.2007). A similar phenoptype was previously reported in a Paracoccus denitrificans mutant (Page and Ferguson 1999). A lower competence for growth in rich media could be partially responsible for the increased sensitivity to metals observed in D3466 transposant. A similar condition affects the mutant D4239. Further analysis of this transposant, bearing a mTn5-mutation in a AraC-like transcriptional regulator, revealed that the increased sensitivity to nickel only occurred in TY medium, but not in Rmin of YMB media (data not shown). The gene encoding the outer membrane protein RopB, interrupted in transposant D5451, was previously studied in Rlv VF39SM. Mutants affected in this gene have an increased sensitivity to detergents, hydrophobic antibiotics and organic acids, reflecting a pleiotropic effect related to outer membrane stability (Foreman et al.2010). Further analysis of transposant Rlv D5451 revealed increased sensitivity to erythromycin and to SDS as compared to Rlv UPM1137 strain (data not shown). These phenotypes suggest that the increased metal sensitivity found in D5451 transposant could be due to an indirect effect associated with the alteration of the outer membrane. Transposant D2250 carries a mTn5 insertion in a gene encoding a protein highly similar to a metal efflux protein (DmeF) previously identified in Cupriavidus metallidurans (Munkelt, Grass and Nies 2004) and Rlv SPF25 (Rubio-Sanz et al.2013). Expression of Rlv SPF25 dmeRF system is inducible by Ni2+ and Co2+ through the action of metalloregulator DmeR (Rubio-Sanz et al.2013). Sequence analysis of the region surrounding UPM1137 dmeF revealed that this gene is also preceded by a dmeR-like gene (92% identity in amino acid residues) encoding a metal-responsive regulator of the RcnR/CsoR family. Furthermore, the dmeRF promoter region is highly conserved in both Rlv UPM1137 and Rlv SPF25 strains, and contains a potential DmeR-binding sequence (data not shown), which is consistent with the inducible mechanism for nickel and cobalt tolerance observed in this work. Symbiotic performance of Ni-sensitive transconjugants The symbiotic phenotype of the 14 nickel-sensitive transposants was tested by analyzing their ability to induce effective nodules in pea plants. Symbiotic performance was estimated through the determination of nitrogen accumulation in plants inoculated with the different strains (Table 3). Four transposants were impaired in symbiosis, resulting in chlorotic plants containing levels of nitrogen accumulation similar to those in uninoculated controls. Two of these transposants (D0013 and D8134) were unable to induce nodules in pea roots, thus suggesting a defect affecting the initial steps of the symbiosis. In other two transposants (D3466 and D4239), the nodules formed were small, white and ineffective for nitrogen fixation. We attribute these drastic differences to potential pleiotropic effects not directly linked to the effect of metals, since plants were grown in standard nutrient solution (i.e. without addition of further metals). On the other hand, transposants D4632, D6426 and D5217 were partially impaired in their nitrogen fixation capability, leading to shoot nitrogen accumulation levels substantially lower than those corresponding to the wild type strain. The remaining seven transposants did not show significant differences in symbiotic performance when compared to the wild-type strain under normal, low metal conditions (Table 3). We had previously shown that a dmeRF-deficient mutant of Rlv SPF25 was symbiotically impaired when pea plants inoculated with this mutant were grown in the presence of high nickel and cobalt concentrations (Rubio-Sanz et al.2013). Rlv SPF25 derives from UPM791, a strain with a medium level of metal tolerance (Table 2). In order to check whether the dmeRF system may also have a relevant role in defining symbiotic metal tolerance in metal-tolerant strain Rlv UPM1137, the dmeF-deficient transposant D2250 was used as inoculum of pea plants that were irrigated with either standard nutrient solutions or with nutrient solutions supplemented with medium/high amounts of metals (85/170 μM NiCl2 and 42.5/85 μM CoCl2, respectively). Two other nickel-sensitive transposants (D2249 and D4740) were also included in the assay, along with the wild-type strain. Metal additions caused the accumulation of high levels of available metal ions in the substrate at the end of the experiment (3.9/6.7 mg kg−1 for Ni2+ and 3.1/4.6 mg kg−1 Co2+, respectively, for the medium/high-metal doses) well over those determined in the original ultramafic soil (Table 2). Growth of pea plants under these conditions did not reveal significant impairment of the symbiotic performance associated with the mutations (Table 4). For unknown reasons, one of the transconjugants (D2249) showed a slight increase on N accumulation under standard conditions (Table 4). Table 4. Symbiotic performance of Rlv strains in pea plants grown in nutrient solutions containing different metal concentrations. Strain No extra Nickel Cobalt metal added 85 μM 170 μM 42.5 μM 85 μM Control 1.72 ± 0.23 ** 2.47 ± 0.68 ** 1.53 ± 1.36 ** 1.49 ± 0.19 ** 1.65 ± 0.66 ** UPM1137 15.25 ± 4.28 20.89 ± 6.86 17.83 ± 3.07 21.65 ± 7.46 21.90 ± 5.01 D2250 18.36 ± 1.99 18.59 ± 3.55 20.74 ± 2.87 22.37 ± 4.48 21.73 ± 2.07 D2249 20.64 ± 3.78 * 20.55 ± 4.67 23.98 ± 7.52 27.99 ± 4.18 25.04 ± 4.41 D4740 19.89 ± 7.13 24.28 ± 5.52 22.22 ± 4.62 20.88 ± 6.46 20.38 ± 6.39 Strain No extra Nickel Cobalt metal added 85 μM 170 μM 42.5 μM 85 μM Control 1.72 ± 0.23 ** 2.47 ± 0.68 ** 1.53 ± 1.36 ** 1.49 ± 0.19 ** 1.65 ± 0.66 ** UPM1137 15.25 ± 4.28 20.89 ± 6.86 17.83 ± 3.07 21.65 ± 7.46 21.90 ± 5.01 D2250 18.36 ± 1.99 18.59 ± 3.55 20.74 ± 2.87 22.37 ± 4.48 21.73 ± 2.07 D2249 20.64 ± 3.78 * 20.55 ± 4.67 23.98 ± 7.52 27.99 ± 4.18 25.04 ± 4.41 D4740 19.89 ± 7.13 24.28 ± 5.52 22.22 ± 4.62 20.88 ± 6.46 20.38 ± 6.39 Values correspond to shoot nitrogen content (mg N/plant) of pea plants 25 days after inoculation with the indicated strains. Control values correspond to uninoculated plants and are the average of six replicates, except for D2250 (n = 3). Asterisks denote the presence of significant differences regarding values corresponding to wild-type strain UPM1137 (*P < 0.05; **P < 0.01). View Large The results obtained in the analysis of transconjugant D2250 differ from what was previously observed in a dmeRF-deficient mutant derived from Rlv UPM791 (Rubio-Sanz et al.2013). Although both mutations are of different nature, the transposon insertion in D2250 is assumed to be equally stable, since it was made using a minitransposon that did not include the transposase gene, and no tnpA-like genes are found in the Rlv UPM1137 genome draft (data not shown). The lack of effect of this mutation when nodulated plants were exposed to high-metal conditions suggests the existence of redundancy in metal tolerance mechanisms operating in bacteroids of this strain. It is interesting to note that dmeRF was the only metal tolerance system known as such identified in our mutagenesis screening. BLAST search of the draft genome of Rlv UPM1137 for homologs to known nickel tolerance determinants (CnrA, NreB) revealed the presence of a cnrA-like gene (YSWDRAFT_01386, 33% identity, 52% similarity with C. metallidurans protein) that has not been identified here, whereas no nreB homologs were found. Search of Mesorhizobium metallidurans STM2638T genome revealed the absence of proteins with significant similarity to DmeF or NreB, whereas four proteins (WP_008877233.1, CCV08990.1, WP_040594136.1 and WP_008874404.1) showing moderate levels of sequence conservation to CnrA (23 to 33% identity, 41 to 52% similarity) were identified in this species. No significant similarities to neither Rlv dmeF, C. metallidurans cnrA or nreB were detected when this search was carried out in the genome of metal-tolerant Cupriavidus strain UYMMa02A (Iriarte et al.2016, data not shown). The reason for the lack of identification of cnrA in the screening of UPM1137 transposants could be either an insufficient number of clones analyzed or the fact that an insertion on this gene could cause only a slight reduction in tolerance that might not be detected. In order to check whether other Ni tolerance mechanisms could be still operating in mutant D2250, we determined the growth kinetics of wild-type and mutant strains in the presence of lower amounts of nickel. In these assays, Rlv strain UPM1289, showing the lowest degree in Ni tolerance in Table 2, was included as negative control. Analysis of the growth curves (Fig. S2, Supporting Information) revealed that D2250 maintained almost wild-type levels of Ni tolerance at 0.75 mM NiCl2. The mutant also showed partial nickel tolerance in TY supplemented with 1.0 mM NiCl2, being clearly more tolerant than the nickel-sensitive strain UPM1289 at these two nickel concentrations. These data suggest that D2250 still possesses nickel tolerance mechanisms, such as CnrA inducing a lower level of tolerance. The partial phenotype of D2250 also indicates that the level of tolerance of Rlv UPM1137 is the result of the synergy between cation diffusion facilitator (DmeRF) and RND (CnrA) efflux systems. We do not have a CnrA mutant available for comparison, but from the results obtained we can also hypothesize that such mutant might not be detected in our screening. We conclude that dmeRF is likely a relevant tolerance mechanism for free-living cells of R. leguminosarum UPM1137, whereas other systems could be masked either by lower expression or by redundancy with other genes. The lack of difference in growth between UPM1137 and D2250 in a medium with no Ni added (Fig. S2, Supporting Information) suggests that there is not a trade-off between the presence of nickel tolerance system dmeRF and growth in the absence of nickel similar to that previously shown in nickel-adapted populations of other endosymbiotic bacteria (Porter and Rice 2013). Our data also suggest the presence of redundant mechanisms giving lower levels of tolerance that could mask the effect of mutation on dmeRF under symbiotic, high-metal conditions. The synergistic action of different tolerance systems has been described previously on Pseudomonas aeruginosa (Teitzel et al. 2006). In our case, combined mutations in several of the identified genes are likely required to elucidate their actual role in metal tolerance under symbiotic conditions. Such studies could allow the identification of new mechanisms mediating the adaptation of bacteria to high levels of metal, a key aspect in the definition of strategies for the development of rhizobial inoculants adapted to sustain the Rhizobium–legume symbiosis in high-metal soils. SUPPLEMENTARY DATA Supplementary data are available at FEMSLE online. Acknowledgements We are grateful to Dr David Durán for his help with the construction of the phylogenetic tree and to Dr Anita Toffanin for her work on selection of the field site and isolation of strains. FUNDING This work was supported by a project from Spain's Ministerio de Economía y Competitividad (MINECO BIO2013-43040-P to José Palacios). Conflict of interest. None declared. REFERENCES Abou-Shanab RI, Berkum P, Angle JS. Heavy metal resistance and genotypic analysis of metal resistance genes in gram-positive and gram-negative bacteria present in Ni-rich serpentine soil and in the rhizosphere of Alyssum murale. Chemosphere 2007; 68: 360– 7. Google Scholar CrossRef Search ADS PubMed Brady KU, Kruckeberg AR, Bradsahw HD. Evolutionary ecology of plant adaptation to serpentine soils. Annu Rev Ecol Evol Syst 2005; 36: 243– 66. Google Scholar CrossRef Search ADS Brito B, Palacios JM, Imperial J et al. Engineering the Rhizobium leguminosarum bv. viciae hydrogenase system for expression in free-living microaerobic cells and increased symbiotic hydrogenase activity. Appl Environ Microb 2002; 68: 2461– 7. Google Scholar CrossRef Search ADS Bruins MR, Kapil S, Oehme FW. Microbial resistance to metals in the environment. Ecotox Environ Safe 2000; 45: 198– 207. Google Scholar CrossRef Search ADS Chaintreuil C, Rigault F, Moulin L et al. Nickel resistance determinants in Bradyrhizobium strains from nodules of the endemic New Caledonia legume Serianthes calycina. Appl Environ Microb 2007; 73: 8018– 22. Google Scholar CrossRef Search ADS Chen WM, Wu CH, James EK et al. Metal biosorption capability of Cupriavidus taiwanensis and its effects on heavy metal removal by nodulated Mimosa pudica. J Hazard Mater 2008; 151: 364– 71. Google Scholar CrossRef Search ADS PubMed Di Gregorio S, Zocca C, Sidler S et al. Identification of two new sets of genes for dibenzothiophene transformation in Burkholderia sp. DBT1. Biodegradation 2004; 15: 111– 23. Google Scholar CrossRef Search ADS PubMed Eitinger T. Transport of nickel and cobalt in prokaryotes. In: Culotta V, Scott RA (eds). Metals in Cells . Chichester, UK: Wiley, 2013, 145– 54. Google Scholar CrossRef Search ADS Foreman DL, Vanderlinde EM, Bay DC et al. Characterization of a gene family of outer membrane proteins (ropB) in Rhizobium leguminosarum bv. viciae VF39SM and the role of the sensor kinase ChvG in their regulation. J Bacteriol 2010; 192: 975– 83. Google Scholar CrossRef Search ADS PubMed Grison CM, Jackson S, Merlot S et al. Rhizobium metallidurans sp. nov., a symbiotic heavy metal resistant bacterium isolated from the Anthyllis vulneraria Zn-hyperaccumulator. Int J Syst Evol Microbiol 2015; 65: 1525– 30. Google Scholar CrossRef Search ADS PubMed Guentas L, Gensous S, Cavaloc Y et al. Burkholderia novacaledonica sp. nov. and B. ultramafica sp. nov. isolated from roots of Costularia spp. pioneer plants of ultramafic soils in New Caledonia. Syst Appl Microbiol 2016; 39: 151– 9. Google Scholar CrossRef Search ADS PubMed Haferburg G, Kothe E. Microbes and metals: interactions in the environment. J Basic Microb 2007; 47: 453– 67. Google Scholar CrossRef Search ADS Iriarte A, Platero R, Romero V et al. Draft genome sequence of Cupriavidus UYMMa02A, a novel beta-rhizobium species. Appl Environ Microb 2016; 4: e01258– 16. Kabata-Pendias A. Trace Elements In Soils and Plants . 4th edn. Boca Ratón, FL: CRC Press LLC, 2011. Kabata-Pendias A, Pendias H. Biogeochemistry of Trace Elements . 2nd edn. Warszawa, Polish: Wyd Nauk PWN, 1999. Klonowska A, Chaintreuil C, Tisseyre P et al. Biodiversity or Mimosa pudica rhizobial symbionts (Cupriavidus taiwanensis, Rhizobium mesoamericanum) in New Caledonia and their adaptation to heavy metal-rich soils. FEMS Microbiol Lett 2012; 81: 618– 35. Google Scholar CrossRef Search ADS Koch D, Nies DH, Grass G. The RcnRA (YohLM) system of Escherichia coli: a connection between nickel, cobalt and iron homeostasis. Biometals 2007; 20: 759– 71. Google Scholar CrossRef Search ADS PubMed Li Y, Zamble DB. Nickel homeostasis and nickel regulation: an overview. Chem Rev 2009; 109: 4617– 43. Google Scholar CrossRef Search ADS PubMed Lindsay WL, Norvell WA. Development of a DTPA soil test for zinc, iron, manganese and copper. Soil Sci Soc Am J 1978; 42: 421– 8. Google Scholar CrossRef Search ADS Ma Y, Rajkumar M, Rocha I et al. Serpentine bacteria influence metal translocation and bioconcentration of Brassica juncea and Ricinus communis grown in multi-metal polluted soils. Front Plant Sci 2015; 5: 757. Google Scholar CrossRef Search ADS PubMed Macomber L, Hausinger RP. Mechanisms of nickel toxicity in microorganisms. Metallomics 2011; 3: 1153– 62. Google Scholar CrossRef Search ADS PubMed Marrero J, Auling G, Coto O et al. High-level resistance to cobalt and nickel but probably no transenvelope efflux: metal resistance in the Cuban Serratia marcescens strain C-1. Microb Ecol 2006; 53: 123– 33. Google Scholar CrossRef Search ADS PubMed Mirete S, de Figueras CG, Gonzalez-Pastor JE. Novel nickel resistance genes from the rhizosphere metagenome of plants adapted to acid mine drainage. Appl Environ Microb 2007; 73: 6001– 11. Google Scholar CrossRef Search ADS Munkelt D, Grass G, Nies DH. The chromosomally encoded cation diffusion facilitator proteins DmeF and FieF from Wautersia metallidurans CH34 are transporters of broad metal specificity. J Bacteriol 2004; 186: 8036– 43. Google Scholar CrossRef Search ADS PubMed Musiani F, Zambelli B, Bazzani M et al. Nickel-responsive transcriptional regulators. Metallomics 2015; 7: 1305– 18. Google Scholar CrossRef Search ADS PubMed Oldroyd GE, Murray JD, Poole PS et al. The rules of engagement in the legume-rhizobial symbiosis. Annu Rev Genet 2011; 45: 119– 44. Google Scholar CrossRef Search ADS PubMed Page MD, Ferguson SJ. Mutational analysis of the Paracoccus denitrificans c-type cytochrome biosynthetic genes ccmABCDG: disruption of ccmC has distinct effects suggesting a role for CcmC independent of CcmAB. Microbiology 1999; 145: 3047– 57. Google Scholar CrossRef Search ADS PubMed Pal A, Wauters G, Paul AK. Nickel tolerance and accumulation by bacteria from rhizosphere of nickel hyperaccumulators in serpentine soil ecosystem of Andaman, India. Plant Soil 2007; 293: 37– 48. Google Scholar CrossRef Search ADS Pereira SI, Lima AI, Figueira EM. Screening possible mechanisms mediating cadmium resistance in Rhizobium leguminosarum bv. viciae isolated from contaminated Portuguese soils. Microb Ecol 2006; 52: 176– 86. Google Scholar CrossRef Search ADS PubMed Platero R, James EK, Rios C et al. Novel Cupriavidus strains isolated from root nodules of native Uruguayan Mimosa species. Appl Environ Microb 2016; 82: 3150– 64. Google Scholar CrossRef Search ADS Porter SS, Chang PL, Conow CA et al. Association mapping reveals novel serpentine adaptation gene clusters in a population of symbiotic Mesorhizobium. ISME J 2017; 11: 248– 62. Google Scholar CrossRef Search ADS PubMed Porter SS, Rice KJ. Trade-offs, spatial heterogeneity, and the maintenance of microbial diversity. Evolution 2013; 67: 599– 608. Google Scholar CrossRef Search ADS PubMed Rodionov DA, Hebbeln P, Gelfand MS et al. Comparative and functional genomic analysis of prokaryotic nickel and cobalt uptake transporters: evidence for a novel group of ATP-binding cassette transporters. J Bacteriol 2006; 188: 317– 27. Google Scholar CrossRef Search ADS PubMed Rubio-Sanz L, Prieto RI, Imperial J et al. Functional and expression analysis of the metal-inducible dmeRF system from Rhizobium leguminosarum bv. viciae. Appl Environ Microb 2013; 79: 6414– 22. Google Scholar CrossRef Search ADS Ryan E, Dell O, Claassen VP. Serpentine revegetation: a review. Northeast Nat 2009; 16: 253– 71. Google Scholar CrossRef Search ADS Sambrook J, Russell DW. Molecular Cloning: A Laboratory Manual . 3rd edn. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory, 2002. Sobczyk M, Smith JAC, Pollard AJ et al. Evolution of nickel hyperaccumulation and serpentine adaptation in the Alyssum serpyllifolium species complex. Heredity 2017; 118: 31– 41. Google Scholar CrossRef Search ADS PubMed Stefanowicz AM, Niklinska M, Laskowski R. Metals affect soil bacterial and fungal functional diversity differently. Environ Toxicol Chem 2008; 27: 591– 8. Google Scholar CrossRef Search ADS PubMed Stoppel R, Schlegel HG. Nickel-resistant bacteria from anthropogenically nickel-polluted and naturally nickel-percolated ecosystems. Appl Environ Microb 1995; 61: 2276– 85. Teitzel GM, Geddie A, de Long SK et al. Survival and growth in the presence of elevated copper: transcriptional profiling of copper-stressed Pseudomonas aeruginosa. J Bacteriol 2006; 188: 7242– 56. Google Scholar CrossRef Search ADS PubMed Xie P, Hao X, Herzberg M et al. Genomic analyses of metal resistance genes in three plant growth promoting bacteria of legume plants in Northwest mine tailings, China. J Environ Sci 2015; 27: 179– 87. Google Scholar CrossRef Search ADS Yurgel SN, Berrocal J, Wilson C et al. Pleiotropic effects of mutations that alter the Sinorhizobium meliloti cytochrome c respiratory system. Microbiology 2007; 153: 399– 410. Google Scholar CrossRef Search ADS PubMed Zielazinski EL, Gonzalez-Guerrero M, Subranmanian P et al. Sinorhizobium meliloti Nia is a P1B-5-ATPase expressed in the nodule during plant symbiosis and is involved in Ni and Fe transport. Metallomics 2013; 5: 1614– 23. Google Scholar CrossRef Search ADS PubMed © FEMS 2018. All rights reserved. For permissions, please e-mail: firstname.lastname@example.org
FEMS Microbiology Letters – Oxford University Press
Published: Feb 1, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
All the latest content is available, no embargo periods.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud