Competition for alfalfa nodulation under metal stress by the metal-tolerant strain Ochrobactrum cytisi Azn6.2

Depósito de investigación de la Universidad de Sevilla https://idus.us.es/ “This is the peer reviewed version of the following article: Navarro- Torre S, Rodríguez-Llorente ID, Doukkali B, Caviedes MA, Pajuelo E. Competition for alfalfa nodulation under metal stress by the metal-tolerant strain Ochrobactrum cytisi Azn6.2. Ann Appl Biol. 2019; 175: 184–192., which has been published in final form at https://doi.org/10.1111/aab.12528 This article may be used for noncommercial purposes in accordance with Wiley Terms and Conditions for Use of Self-Archived Versions. This article may not be enhanced, enriched or otherwise transformed into a derivative work, without express permission from Wiley or by statutory rights under applicable legislation. Copyright notices must not be removed, obscured or modified. The article must be linked to Wiley’s version of record on Wiley Online Library and any embedding, framing or otherwise making available the article or pages thereof by third parties from platforms, services and websites other than Wiley Online Library must be prohibited." RESEARCH ARTICLE 1 2 Competition for alfalfa nodulation under metal stress by the 3 metal-tolerant strain Ochrobactrum cytisi Azn6.2 4 5 Salvadora Navarro-Torre, Ignacio D. Rodríguez-Llorente, Bouchra Doukkali, 6 Miguel A. Caviedes, Eloísa Pajuelo 7 8 Departamento de Microbiología, Facultad de Farmacia, Universidad de Sevilla, Seville, 9 Spain 10 11 Correspondence: Ignacio D. Rodríguez-Llorente, Departamento de Microbiología, 12 Facultad de Farmacia, Universidad de Sevilla, Calle Profesor García González 2, 41012 13 Seville, Spain. Email: [email protected] 14 15 16 17 ABSTRACT 18 19 Legume plants, in association with rhizobia, are gaining increasing interest for heavy 20 metal rhizoremediation. This symbiotic interaction combines the advantages of 21 rhizoremediation and soil nitrogen enrichment. In metal polluted soils, Ochrobactrum 22 cytisi can elicit non-fixing nodules on legumes, including Medicago sativa. Nodulation 23 kinetics was much slower when M. sativa plants were inoculated with O. cytisi Azn6.2 24 compared with the natural symbiont Ensifer meliloti 1021 and nodules were ineffective 25 in nitrogen fixation. A competition experiment was performed using alfalfa grown on 26 heavy metals, and co-inoculated with equal amounts of the metal-sensitive E. meliloti 27 1021 and the metal-resistant O. cytisi Azn6.2. When plants were inoculated in non-28 polluted substrates, all nodules were formed by E. meliloti 1021. Nevertheless, under 29 increasing metal concentrations, the number of nodules occupied by O. cytisi grew. At 30 the highest metal concentration, all nodules were elicited by O. cytisi, suggesting that the 31 resistant species can take the place of the natural symbiont. This fact has important 32 ecological and environmental implications when proposing legume–rhizobia symbioses 33 for rhizoremediation and highlights the need of selecting highly resistant rhizobia in order 34 to be competitive in polluted soils. 35 36 KEYWORDS: heavy metals, Medicago sativa, multiresistance, nodulation competitive 37 assay, α-proteobacteria 38 39 INTRODUCTION 40 41 Heavy metal pollution represents one of the most important environmental problems. 42 Toxic effects of metals and their accumulation throughout the food chain has serious 43 ecological, economic and health consequences (Vangronsveld et al., 2009). 44 Phytoremediation has emerged as a promising ecoremediation technology, which uses 45 plants and their associated rhizosphere microorganisms to remove pollutants from 46 contaminated environments (Ali et al., 2013; Ullah et al., 2015). This advantageous 47 system offers the combination of two approaches, phytoremediation and 48 bioaugmentation, resulting in rhizoremediation (Glick, 2010). Root exudates derived 49 from the plant can help to stimulate bacterial survival and action, leading to a more 50 efficient degradation of pollutants (Glick, 2010; Mishra et al, 2017). Anthropogenic 51 polluted soils usually contain low levels of organic matter, and hence plant growth on 52 such soils is often limited by nitrogen deficiency (Khan et al., 2009; Méndez & Maier, 53 2008). The Rhizobium–legume interaction is a process that allows the growth of the plant 54 in the absence of externally supplied nitrogen (Graham & Vance, 2003). This well-known 55 and important agronomical process is also used as a model to study general genetic 56 aspects of plant–microbe interactions (Oldroyd & Downie, 2008). In the last years, an 57 increasing interest has risen on the use of legume plants associated with microorganisms 58 for bioremediation of metals (Fagorzi et al., 2018; Hao et al., 2014; Pajuelo, et al., 2011; 59 Teng et al., 2015). Several metal tolerant legumes have been proposed for metal 60 phytostabilisation in polluted soils, since these plants are well-known root accumulators 61 (Dary et al., 2010). Phytostabilisation is the strategy of election in widely contaminated 62 areas, or when other removal techniques are unaffordable, since it prevents metal 63 dispersion in the environment (Méndez & Maier, 2008). Thus, strategies for revegetation 64 of contaminated soils could use tolerant leguminous species to improve the soil's nitrogen 65 content (Pajuelo et al., 2011). The establishment of a vegetal cover with pioneer plants 66 allows further colonisation by other metal-tolerant species (Lei &Duan, 2008). In order 67 to guarantee the effectiveness of the legume–rhizobia interactions in metal 68 phytostabilisation, it is important to select highly metal-resistant rhizobia, which in turn 69 must be highly competitive in the rhizosphere and symbiotically effective (El Aafi et al., 70 2015; Pajuelo et al., 2008).In the last two decades, nodulation of legumes by αandβ-71 proteobacteria has been reported (Rivas et al., 2003). These bacteria include, among 72 others, Burkholderia, Devosia and Ralstonia (β-proteobacteria) or Methylobacterium and 73 Rhizobium radiobacter (formerly Agrobacterium tumefaciens) (α-proteobacteria). The 74 genus Ochrobactrum, belonging to the Brucellaceae family, comprises several 75 rhizosphere bacteria, such as O. tritici and O. oryzae (Lebuhn et al., 2000; Tripathi et al., 76 2006) and legume endosymbionts such as O. lupini (Trujillo et al., 2005) and O. cytisi 77 (Zurdo-Piñeiro et al., 2007). In a previous work, more than 100 rhizobial strains from 78 nodules of native legumes grown on contaminated soils from southern Spain, affected by 79 a mine toxic spill, were isolated (Carrasco et al., 2005). Among them, several O. cytisi 80 strains were identified, being more frequently found in the most contaminated sites 81 (Rodríguez-Llorente et al., 2010; Zurdo-Piñeiro et al., 2007). Nodulation of legumes by 82 O. cytisi seems not to be host-specific, since these strains are able to nodulate Cytisus 83 scoparius, Medicago polymorpha and Phaseolus vulgaris (Rodríguez-Llorente et al., 84 2010; Zurdo-Piñeiro et al., 2007). The aim of this work was to determine whether O. cytisi 85 could compete with rhizobia for legume (Medicago sativa) nodulation in metal-86 contaminated soils and analyse the ecological implications of this competence. 87 88 MATERIALS AND METHODS 89 90 Bacterial strains 91 92 Ochrobactrum cytisi Azn6.2 was isolated from nodules of M. polymorpha plants 93 spontaneously growing on polluted soils after a toxic mine spill (Rodríguez-Llorente et 94 al., 2010). Ensifer meliloti 1021 is a widely used Medicago symbiont, whose nodulation 95 pattern is very well described (Gage, 2004). Bacteria were grown on TY (tryptone yeast 96 extract)-agar plates (Vincent, 1970) for 48–72 hr at 28 °C. 97 98 Bacterial resistance to metals and metalloids 99 100 The resistance of both bacteria to arsenic and heavy metals was determined on agar plates 101 containing solid TY medium supplemented with different concentrations of NaAsO2 102 (ranging from 1 to 10 mM), CdCl2 (from 1 to 5 mM), CoCl2 (from 1 to 5 mM), CuSO4 103 (from 1 to 5 mM), ZnSO4 (from 1 to 10 mM), Pb(NO3)2 (from 1 to 10 mM), NiCl2 (from 104 1 to 5 mM) or HgCl2 (from 10 to 100 μM). Plates were incubated for 72–96 hr at 28 °C. 105 The resistance was expressed as the maximum tolerable concentration for each element, 106 which is defined as the maximum concentration of a metal or metalloid not affecting 107 bacterial growth. 108 109 Plant growth promoting properties of O. cytisi Azn6.2 110 111 Bacterial plant growth promoting (PGP) traits were studied in absence and presence of 112 metals basically as described in Mesa et al. (2015). To test nitrogen fixation, bacterial 113 growth in nitrogen free broth medium was used (Dobereiner, 1995). Phosphate 114 solubilisation was recorded on National Botanical Research Institute's Phosphate Growth 115 medium plates (Nautiyal, 1999) when bacterial growth caused the appearance of 116 surrounding transparent halos. Similarly, orange halos revealed production of 117 siderophores on chrome azurol S plates (Schwyn &Neilands, 1987). Plates were always 118 incubated for 72 hr at 28 °C. The synthesis of IAA (indole-3-acetic acid) was 119 colorimetrically estimated as detailed in Mesa et al. (2015). Presence of 1-120 aminocyclopropan-1-carboxilic acid (ACC) deaminase enzyme was detected following 121 the method described in Penrose and Glick (2003). Finally, the ability to form biofilms in 122 vitro was assayed as described in Mesa et al. (2015). To test PGP properties in presence 123 of metals, plates were supplemented with NaAsO2 (1 mM), CdCl2 (0.5 mM), CuSO4 (1 124 mM), ZnSO4 (1 mM) or Pb(NO3)2 (1 mM). 125 126 Presence of megaplasmids and nod genes in O. cytisi Azn6.2 genome 127 128 The profile of megaplasmids in the bacterial strains was analysed by electrophoresis on 129 0.7% agarose gels following a modified version of the protocol described by Eckhardt 130 (1978). The lysis of the bacteria (30 μL of a109 cells/mL culture) was performed directly 131 in the wells of the gel in the presence of 1% sodium dodecyl sulphate (SDS) and lysozyme 132 (20 μg/μL). Samples were electrophoresed for 30 min at 100 mA and, further on, for5hr 133 at 25mA. Gel was stained in ethidium bromide (0.5μg/mL) for10 min, washed twice in 134 distilled water and photographed. Total O. cytisi Azn6.2 genomic DNA was extracted 135 using thei-genomic BYF and G spin Total DNA extraction kits (Intron Biotechnology 136 Ltd., Korea) according to the manufacturer's instructions. Amplification of the complete 137 nodA gene and a partial sequence of nodB (88 bp) were performed using nodA-1 and 138 nodA-2 primers as described in Haukka et al. (1998). Sequence of O. cytisi Azn6.2 nodA 139 gene has been deposited in the European Molecular Biology Laboratory data bank under 140 accession no. MH352352. Analysis of alfalfa nodulation by O. cytisiAlfalfa is the 141 prevalent forage legume in temperate climates (Russelle, 2001) for which commercial 142 seeds are available. Furthermore, this species was spontaneously growing at Aznalcóllar 143 (Spain) heavy metal polluted site (Carrasco et al., 2005). M. sativa L. ecotype Aragon 144 was used in this work. 145 146 Plant culture conditions and nodulation test 147 148 Surface sterilisation of alfalfa seeds and plant nodulation assays on squared plates were 149 developed basically as described in Pajuelo et al. (2008). Seeds were surface sterilised 150 with 70% ethanol for 10 min, exhaustively washed with distilled water and placed on 151 water-agar plates for 36 hr at 22 °C in the dark. Root pregerminated seedlings of 0.5–1 152 cm were carefully transferred to squared plates containing BNM-agar medium (buffered 153 nodulation medium, nitrogen free; Ehrhardt et al., 1992). At the time of transfer, seeds 154 (10 seeds per plate) were inoculated with 100 μL of an overnight culture of O. cytisi 155 Azn6.2 or E. meliloti 1021 (~105 CFU/mL). The lower part of the plate (up to the line of 156 the seeds) was covered with black paper and cultivated for 35 days in a plant growth 157 chamber at 22 °C:18 °C with a 16 hr light:8 hr dark photoperiod (120–130 μE m−2 s−1). 158 For each treatment, three replicates were performed (30 plants). Plants were observed 159 every day for counting nodules and primordia. The distance of nodules to plant 160 hypocotyls was alsomeasured. 161 162 Determination of nitrogen content in plant shoots 163 164 Since the medium used for plant culture is nitrogen-free, nitrogen content of the shoots is 165 considered as a parameter of nitrogen fixation efficiency in nodules (Graham & Vance, 166 2003). Shoots of 10 plants were mixed, washed in distilled water, dried for 48 hr at 60 °C 167 and powder using mortar and pestle. The determination of the nitrogen content was 168 performed using an InfrAlyzer 300 (Technicon, NY) according to Carrasco et al. (2005). 169 Nitrogen content was expressed as the mean± SD of the determination of three replicates. 170 171 Competitiveness assay in the presence of heavy metals 172 173 Labelling E. meliloti 1021 with a lacZ reporter gene 174 175 The plasmid containing the hemA::lacZ reporter gene fusion (Wopereis et al., 2000) was 176 transformed into electro-competent cells of E. meliloti 1021 using a Biorad E. coli pulser 177 electroporator (at 1.8 kV for 3.8 ms). Cells were plated on TY medium containing 178 tetracycline (5 μg/mL) and grown for 48 hr at 28 °C. Individual colonies were stricken 179 out on TY plates containing tetracycline (5 μg/mL) and X-gal (40 μg/mL) for selection 180 of blue colonies. 181 182 Competitiveness assay conditions 183 184 Seeds of M. sativa were surface sterilised as described in Pajuelo et al. (2008) potted in 185 sterile containers (20 seeds per container) of 14 cm high and 11.5 cm diameter provided 186 with a filter for gas exchange and containing a sterilised mix of vermiculite:sand (50:50) 187 and three different concentrations of a mix of heavy metals: M = (650 μM As + 600 μM 188 Cu + 30 μM Cd + 250 μM Pb + 750μM Zn); M/2 = half the concentration of each metal; 189 M/4 = one-fourth the concentration of each metal. M0 = control without metals was 190 included. Metals were diluted in BNM liquid medium and 50 mL per pot were added at 191 the beginning of the experiment. Seeds were inoculated with5 mL of 107 CFU/mL 192 cultures of E. meliloti 1021 (lacZ), O. cytisi Azn6.2or with a mix of both bacteria in the 193 proportion 1:1. Non-inoculated seeds were used as control. Containers were placed in a 194 growth chamber with 16 hr light:8 hr dark photoperiod at 22 °C:20 °C. At 35 days post-195 inoculation (dpi), plants were harvested, roots excised and total number of nodules 196 counted. Whole roots with nodules were stained for β-galactosidase activity as described 197 below. Both roots and shoots were dried in the oven at 60 C and weight was measured. 198 In all the experiments involving O. cytisi Azn6.2, bacteria were isolated rom white 199 nodules after surface disinfection as follows: first nodules were placed under running tap 200 water to remove big particles, followed by immersion in 70% (vol/vol) ethanol for 5 min 201 under slightly shaking, then soaking in 5% (vol/vol) sodium hypochlorite for10 min under 202 gentle agitation, and finally five rinses in sterile distilled water. Disinfected nodules were 203 then crushed separately using sterile mortars and pestles in a small volume of sterile 204 physiological saline solution and placed into eppendorf tubes. About 50 μL of the 205 resulting fluid were plated on TY and colonies corresponding to O. cytisi Azn6.2 206 identified. 207 208 β-Galactosidase staining 209 210 Roots with nodules were fixed overnight in 0.2 M sodium phosphate buffer pH 7.2 211 containing 1% formaldehyde and 2.5% glutaraldehyde. After exhaustive washing in 0.2 212 M sodium phosphate buffer pH 7.2, nodules were stained in the presence of X-gal as 213 described in Pajuelo et al. (2008) for 24 hr at 37 °C. Roots were washed five times in 214 distilled water, cleared in 5% sodium hypochlorite for 5 min and washed twice with 215 distilled water. Nodules were observed using an Olympus BX61 microscope (Olympus 216 Corporation, Japan) provided with a camera DP70. Blue nodules corresponded to 217 nodulation by E. meliloti 1021-lacZ, whereas white nodules corresponded to nodulation 218 by O. cytisi Azn6.2. 219 220 Statistical treatmentAnalysis of variance was performed using the software SPSS 13.0 221 (2004) for Windows (SPSS Inc. Chicago, IL). Results of N-content in shoots of plants is 222 given as means ± SDs of three independent replicates corresponding to 10 plants each (10 223 plants x three plates). 224 225 RESULTS 226 227 Resistance of Medicago symbionts to arsenic and heavy metals and presence of PGP 228 properties 229 230 The resistance of O. cytisi Azn6.2 towards arsenic and several heavy metals was 231 determined and compared to that of E. meliloti 1021(Table 1). O. cytisi was more resistant 232 than E. meliloti 1021 to As (six fold) and to all the metals tested (between 1.5-fold to 12-233 fold, depending on the metal). It is remarkable the high resistance of O. cytisi towards Zn 234 (12 times more resistant than E. meliloti). Concerning PGP properties, E. meliloti 1021 235 was able to form biofilms in absence of metals, but not in presence of them. This 236 bacterium was not able to fix nitrogen, solubilise phosphate or produce IAA, neither in 237 absence nor in presence of metals. On the other hand, O. cytisi Azn6.2 was able to form 238 biofilms in absence of metals and also in presence of Cd, Cu or Pb. Similarly, to E. 239 meliloti 1021, Azn6.2 was not able to fix nitrogen, solubilise phosphate or produce IAA 240 in absence of metals (Table 2). Nevertheless, O. cytisi Azn6.2 was able to fix nitrogen 241 and produce IAA in media supplemented with several metals (Table 2). Two PGP 242 properties, siderophores and ACC deaminase production were also assayed and not found 243 in O. cytisi Azn6.2 and E. meliloti 1021. 244 245 O. cytisi harbours a symbiotic megaplasmid and elicits ineffective nodules on alfalfa roots 246 247 Since strains of O. cytisi have been reported to nodulate several legumes, the presence of 248 a symbiotic megaplasmid in Azn6.2 was investigated. Visual analysis of the number and 249 size of plasmids seen on gels showed the presence of one single megaplasmid with a size 250 of 1.68 Mbp in Azn6.2, whereas E. meliloti harboured two megaplasmids of 1.35 Mbp 251 (pSymA) and 1.68 Mbp (pSymB) (Supplementary Figure 1). Amplification and 252 sequencing of nodA gene from O. cytisi Azn6.2 DNA showed 100% of identity with nodA 253 from E. meliloti. The ability of O. cytisi Azn6.2 to nodulate alfalfa (M. sativa) was tested. 254 O. cytisi Azn6.2 was able to elicit nodules on M. sativa roots. Nodulation kinetics was 255 much slower when plants were inoculated with Azn6.2 compared with E. meliloti 1021. 256 Thus, first nodule primordium appeared around 20–24 dpi in the distal part of the root, 257 whereas primordia elicited by 1021 appeared at 5–6 dpi at 1.5–2 cm from hypocotyls 258 (Table 3). Nitrogen content in shoots of plants inoculated withAzn6.2 was much lower 259 than that of plants inoculated with the natural symbiont 1021 (Figure 1a), suggesting that 260 these nodules were ineffective for nitrogen fixation. In fact, the nitrogen content was 261 similar to that of non-inoculated plants (nitrogen-starved). Nodules induced by Azn6.2 262 were small, white and ineffective (Figure 1c), whereas mature nodules elicited by 1021 263 were fully developed and pink coloured (Figure 1b). 264 265 Competitiveness between Ensifer and Ochrobactrum for nodulation of alfalfa plants 266 grownunder metal stress 267 268 The competitiveness of both Medicago symbionts for nodulation of alfalfa roots, in non-269 contaminated substrate (control) or in substrate with increasing levels of metals, was 270 Musarrat (Eds.), Biomanagement of metal-contaminated soils(pp. 95–123). Dordrecht, 475 the Netherlands: Springer. https://doi.org/10.1007/978-94-007-1914-9_4Penrose, D. M., 476 & Glick, B. R. (2003). Methods for isolating and characterizingACC deaminase 477 containing plant growth-promotingrhizobacteria. Physiologia Plantarum, 118, 10–478 15.Reichman, S. M. (2007). The potential use of the legume–rhizobium symbiosisfor the 479 remediation of arsenic contaminated sites. Soil Biology andBiochemistry, 39, 2587–480 2593. https://doi.org/10.1016/j.soilbio.2007.04.030Rivas, R., García-Fraile, P., & 481 Velázquez, E. (2009). Taxonomy of bacterianodulating legumes. Microbiology Insights, 482 2, 51.Rivas, R., Velázquez, E., Willems, A., Vizcaino, N., Subba-Rao, N. S.,Mateos, P. 483 F., … Martínez-Molina, E. (2002). A new species of Devosia that forms a unique 484 nitrogen-fixing root-nodule symbiosis with theaquatic legume Neptunia natans (L.f.) 485 Druce. Applied and EnvironmentalMicrobiology, 68, 5217–5222. 486 https://doi.org/10.1128/AEM.68.11.5217-5222.2002Rodríguez-Llorente, I. D., Dary, M., 487 Gamane, D., El Hamdaoui, A.,Doukkali, B., Lafuente, A., … Pajuelo, E. (2010). 488 Cadmium biosorptionproperties of the metal resistant Ochrobactrum cytisi Azn6.2. 489 Engineeringin Life Sciences, 10, 49–56. https://doi.org/10.1002/elsc.200900060Russelle, 490 M. (2001). Alfalfa. American Scientist, 89, 252–259.Sawada, H., Kuykendall, D., & 491 Young, J. M. (2003). Changing concepts inthe systematics of bacterial nitrogen-fixing 492 legume symbionts. Journalof General and Applied Microbiology, 49, 155–179. 493 https://doi.org/10.2323/jgam.49.155Schwyn, B., & Neilands, J. B. (1987). Universal 494 chemical assay for thedetection and determination of siderophores. Analytical 495 Biochemistry,160, 47–56.Teng, Y., Wang, X., Li, L., Li, Z., & Luo, Y. (2015). Rhizobia 496 and their biopartnersas novel drivers for functional remediation in contaminatedsoils. 497 Frontiers in Plant Science, 6, 32. https://doi.org/10.3389/fpls.2015.00032Tripathi, A. K., 498 Verma, S. C., Chowdhury, S. P., Lebuhn, M., Gattinger, A., &Schloter, M. (2006). 499 Ochrobactrum orizae, sp. nov., an endophytic bacterial species isolated from deep water 500 rice in India. International Journal of Systematic and Evolutionary Microbiology, 56, 501 1677–1680.https://doi.org/10.1099/ijs.0.63934-0Trujillo, M. E., Willems, A., Abril, A., 502 Planchuela, A. M., Rivas, R., Ludeña, D.,… Velázquez, E. (2005). Nodulation of Lupinus 503 albus by strains of Ochrobactrum lupini sp. nov. Applied and Environmental 504 Microbiology,71, 1318–1327. https://doi.org/10.1128/AEM.71.3.1318-1327.2005Ullah, 505 A., Heng, S., Munis, M. F. H., Fahad, S., & Yang, X. (2015).Phytoremediation of heavy 506 metals assisted by plant growth promoting(PGP) bacteria: A review. Environmental and 507 Experimental Botany, 117,28–40. https://doi.org/10.1016/j.envexpbot.2015.05.001van 508 Berkum, P., & Eardly, B. D. (2002). The aquatic budding bacteriumBlastobacter 509 denitrificans is a nitrogen-fixing symbiont ofAeschynomene indica. Applied and 510 Environmental Microbiology, 68,1132–1136. https://doi.org/10.1128/AEM.68.3.1132-511 1136.2002Vangronsveld, J., Herzig, R., Weyens, N., Boulet, J., Adriaensen, K.,Ruttens, 512 A., … Mench, M. (2009). Phytoremediation of contaminatedsoils and groundwater: 513 Lessons from the field. Environmental Scienceand Pollution Research, 16, 765–794. 514 https://doi.org/10.1007/s11356-009-0213-6Vincent, J. M. (1970). A manual for the 515 practical study of root-nodule bacteria.Oxford, England: Blackwell.Wani, P. A., Khan, 516 M. S., & Zaidi, A. (2008). Chromium reducing and plantgrowth promoting 517 Mesorhizobium improves chickpea growth in chromiumamended soil. Biotechnology 518 Letters, 30, 159–163. https://doi.org/10.1007/s10529-007-9515-2Wei, G., Fan, L., Zhu, 519 W., Fu, Y., Yu, F., & Tang, M. (2009). Isolation andcharacterization of the heavy metal 520 resistant bacteria CCNWRS33-2isolated from root nodule of Lespedeza cuneata in gold 521 mine tailings in China. Journal of Hazardous Materials, 162, 50–56. 522 https://doi.org/10.1016/j.jhazmat.2008.05.040Wopereis, J., Pajuelo, E., Dazzo, F. B., 523 Jiang, Q., Gresshoff, P. M., deBrujin, F. J., … Szczyglowski, K. (2000). Short root mutant 524 of Lotusjaponicus with a dramatically altered symbiotic phenotype. Plant Journal,23, 97–525 114. https://doi.org/10.1046/j.1365-313x.2000.00799.xZurdo-Piñeiro, J. L., Rivas, R., 526 Trujillo, M. E., Vizcaíno, N., Carrasco, J. A.,Chamber, M. A., … Velázquez, E. (2007). 527 Ochrobactrum cytisi sp. nov.,isolated from nodules of Cytisus scoparius in Spain. 528 International Journal of Systematic and Evolutionary Microbiology, 57, 784–788. 529 https://doi.org/10.1099/ijs.0.64613 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 Figures 545 546 547 Figure 1. Comparative nodulation of alfalfa by Ensifer meliloti 1021 and Ochrobactrum 548 cytisi Azn6.2. (a) Nitrogen content (%) in plant shoots compared with non-inoculated 549 control plants. Results are means ± SDs of three independent replicates. Different letters 550 indicate significant differen,ces at p < .01. (b) Aspect of a 35-dpimature pink nodule 551 formed by 1021. (c) Aspect of a 35-dpi non-fixing nodule induced by Azn6.2. Scale bars 552 = 1 mm (b, c). 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 Figure 2. Competitiveness of Ensifer meliloti 1021-lacZ and Ochrobactrum cytisi Azn6.2 568 for nodulation of Medicago sativa under metal stress. Example of a set of plants at the 569 end of the experiment. 570 571 572 573 574 575 576 577 Figure 3. Biomass of M. sativa plants non-inoculated, inoculated either with E. meliloti 578 1021-lacZ, O. cytisi Azn6.2, or coinoculated with both symbionts and grown under metal 579 stress. (A) Shoot biomass. (B) Root biomass. Different letters indicate significant 580 differences at p < 0.05 for each metal concentration. 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 Tables 596 597 Table 1. Resistance of Ochrobactrum cytisi Azn6.2 and Ensifer meliloti 1021 to arsenic and 598 heavy metals. 599 Medicago symbiont As Cd Co Cu Hg Ni Pb Zn E. meliloti 1021 1 0.25 1 1.5 0.01 0.5 4 0.5 O. cytisi Azn6.2 (x fold increased resistance) 6 (x6) 1.5 (x6) 2 (x2) 3.5 (x2.3) 0.05 (x5) 1 (x2) 6 (x1.5) 6 (x12) Note: The maximal tolerable concentration (mM) is presented. 600 601 602 603 Table 2. Plant growth promoting (PGP) properties of Ochrobactrum cytisi Azn6.2 in absence and 604 presence of heavy metals. 605 Without metal As (1 mM) Cd (0.5 mM) Cu (1 mM) Pb (1 mM) Zn (1 mM) Nitrogen fixation - ++ ++ + ND + Phosphate solubilization - - +/- - - - IAA production (mg/l) - 5.26 9.02 2.46 ND 1.55 Biofilm formation + - + + + - Note: − without growth or halo; + growth; ++ profuse growth; +/− small halo; ND, non-determined 606 Table 3. Comparison of the kinetics and effectiveness of alfalfa nodulation by Ensifer meliloti 607 1021 and Ochrobactrum cytisi Azn6.2. 608 Nodulation features E. meliloti 1021 O. cytisiAzn6.2 Number of nodules per plant 2-3 (at 35 dpi) 0.1-0.5 (at 35 dpi) Formation of nodule primordia 5-6 dpi 20-24 dpi Mature fixing nodules 12-15 dpi --- Distance of the nodules to the hypocotyls First nodule 1.5-2 cm 7-9 cm Second nodule 3-4.5 cm 8.5-13 cma Third nodule 8-12 cm ---- aVery occasionally, a second nodule on the same plant was found. 609 610 Table 4. Nodulation of M. sativa and plant survival in response to heavy metals and different symbionts. Inoculation [M] Number of plants Nº nodules/plant %Blue nodules %White nodules Non-inoculated 0 20 0 0 0 M/4 15a 0 0 0 M/2 6b 0 0 0 M 1-2c 0 0 0 E. meliloti 1021lacZ 0 18 2a 100 0 M/4 16a 0.9b 100 0 M/2 8d 0.4c 100 0 M 4e 0.1d 100 0 O. cytisi Azn6.2 0 19 0.2d 0 100 M/4 17a 0.1d 0 100 M/2 13a 0.1d 0 100 M 9d 0.1d 0 100 Coinoculation (1021+Azn6.2) 0 20 2.5a 100 0 M/4 17a 0.8b 32 68 M/2 8d 0.3c 25 75 M 4e 0.1d 0 100 LSD (5%), df = 32 0.03 0.001 - - 611 Note: [M], metals concentrations. M, M/2 and M/4 are defined in Section 2.6.2. Results are means ± SDs 612 of the number of plants or nodules counted in a container (20 plants per container, three independent 613 replicates). 614