Effect of Manganese on Growth and Nodulation of Mimosa caesalpiniaefolia (Benth.)

Soil pollution by heavy metals is mainly caused by poor management of tailings generated by mining during exploration and processing of metals. Given the concern about the soil environment, this study aimed to evaluate the tolerance and effectiveness of isolates of rhizobium strains native to soils of Mn-mining areas in association with Mimosa caesalpiniaefolia (Benth.) cultivated under increasing concentrations of Mn. Isolates of rhizobium strains, obtained from Mn-mining area, were authenticated and morphologically characterized. Effect of Mn on the growth and nodulation of M. caesalpiniaefolia was subsequently evaluated. We obtained 14 isolates which were similar, differing only in the form of colonies. Mn did not inhibit the symbiotic process between isolates and plant species. The isolates from mining area showed higher efficiency than native strains. All isolates are Mn-tolerant, and those from mining area can increase the tolerance of M. caesalpiniaefolia to excess Mn in the soil.


Introduction
Soil contamination by heavy metals has become a serious issue worldwide, generating risks to human health due to the entry of these toxic elements in the food chain, as well as major risk for environmental safety (Glick, 2010). Activities of mineral extraction, or mining, constitute one of the main sources of release of heavy metals in the environment. Areas degraded by this activity, compared to others such as deforestation and livestock farming, are considered as relatively small, but have a great potential to degrade the environment, especially due to the negligence in the treatment of the tailings generated in the mining process, the actual sources of contaminants, such as the heavy metals that are released in the environment and can cause serious ecological damage, affecting the quality of waters and microorganisms, as well as the soil and its microbial and faunal components, besides the plant communities (Abdu, Abdullahi, & Abdulkadir, 2017).
approximately 10 ms of length and 30 cm of DAP (diameter to the height of the chest) (Carvalho, 2007). Sabiá is considered among the native species of the Northeast the most recommended and used for programs of recovery of degraded areas, for having high rates of growth and survival (P. C. F. Lima, A. Q. Lima, & Drumond, 2003).
In an attempt to revegetate mining areas, several techniques have been sought, including phytoremediation, which is basically the use of association between plants and soil microorganisms to reduce the toxic effects of contaminants such as heavy metals or organic pollutants on the environment (Ali, Khan, & Sajad, 2013). It is considered a rather effective, low-cost technique in the remediation of polluted soils (Baker, Ernst, Ent, Malaisse, & Ginocchio, 2010).
According to Foy (1983), high Mn levels in plants may affect the activity of some enzymes and production of some hormones (auxin and gibberellin), as well as the proportion of amino acids and root development. Leguminous species appear to be more sensitive to high levels of Mn, since its excess affects the number of nodules of rhizobia and consequently the efficiency of biological N 2 fixation (Amara, Mohjadi, & Miller, 1986;Dobereiner, 1966;Kabata-Pendias, 2011). Some effects of Mn toxicity on nodulating nitrogen-fixing bacteria (NNFB), according to Hungria and Araújo (1994), are mutation and effect on cell division of the bacteria, reducing their effectiveness.
Given the above, the present study aimed to evaluate the tolerance and effectiveness of isolates of strains native to soils of Mn +2 -mining areas in association with sabiá (Mimosa caesalpiniaefolia Benth.) cultivated under increasing concentrations of Mn 2+ .

Location and Description of the Areas
Soil samples were collected in a Mn-mining area situated in the municipality of Ocara, Ceará, Brazil. Soil samples were collected in three different areas: Area 1 (Lat.: 4º30′17.82″ S; Long.: 38º31′22.89″ W), considered as preserved native forest and adjacent to the mining area; Area 2 (Lat.: 4º30′14.05″ S; Long.: 38º31′17.65″ W), degraded by mineral extraction; and Area 3 (Lat.: 4º30′22.36″ S; Long.: 38º31′20.50″ W), where the sample was collected directly from the tailings generated by the mineral extraction process.
After selecting and georeferencing the points, soil samples of approximately 30 kg were collected in the 0-20 cm layer. After collection, the soil was sieved through a 4-mm mesh and placed in plastic bags. Soil samples were pounded to break up clods and sieved through a 2-mm mesh to obtain ADFE (Air-Dried Fine Earth) and then used for analyses of chemical and physical characteristics according to the methodologies proposed by the Brazilian Agricultural Research Corporation (EMBRAPA, 2017). The results of the analyses are presented in Table 1. Then, this species was tested for symbiotic efficiency with the isolated and selected strains of NNFB, in soil with increasing levels of Mn. Sabiá seeds were purchased from the company Biosementes.

Isolation, Authentication and Cultural Characterization of Isolates of Native Strains of NNFB From Manganese Mining Areas
The isolation of native NNFB strains was conducted in greenhouse, using 500 mL plastic pots which contained soil from the three areas evaluated, sown with four seeds. The seeds used in the planting were disinfected by immersion in aqueous solution of ethanol (95%) for 50 seconds, followed by immersion in an aqueous solution of sodium hypochlorite (5%) for 5 minutes. After disinfection, the seeds were washed six times in sterile distilled water to remove hypochlorite residues.
Plants were daily irrigated with distilled water sterilized by autoclaving at 121 ºC and 1 atm for 2 hours. Every week, 150 mL of the nutrient solution of Hoagland and Armon (1950), without N, were applied.
Sabiá plants were harvested 45 days after sowing to collect the most representative nodules from the root region. The collected nodules were superficially disinfected by immersion in 95% ethanol solution for 10 seconds, followed by immersion in a 10% sodium hypochlorite solution for 10 minutes, and then washed in sterilized water 5 times to remove the excess sodium hypochlorite (Dionísio et al., 2016).
Disinfected nodules were open under aseptic conditions in a laminar flow chamber using a surgical scalpel, and a sterile needle was used to extract a portion of the mass of Bacteroides from the nodule center, which showed a reddish color, characterized by the presence of leghemoglobin, typical color of nodules that are efficient at N 2 fixation (Hungria & Araujo, 1994). The drop's content was subcultured to another Petri dish containing solid culture medium of Vincent (1970)-YMA, with pH adjusted to 6.8, mixed with Congo Red dye at 1% in aqueous solution to reveal any contamination.
The plates were incubated for a period of 12 days in the absence of light in a bacterial growth chamber regulated at 28 ºC until the formation of visible, light-colored colonies due to the non-absorption of Red Congo, an indication of a possible rhizobium colony without contamination. After 12 days, the light-colored colonies were subcultured to a new Petri dish containing YMA solid medium of Vincent (1970), mixed with Bromothymol blue indicator and incubated again under the same conditions.
To authenticate the isolated strains, they were inoculated again in Sabiá plants using 500-mL plastic pots containing 400 g of washed sand (to remove salts), which was autoclaved for 2 hours, at pressure of 1 atm and temperature of 121 ºC. Three superficially disinfected seeds were sown in each plastic pot, with four replicates for each isolate.
Strains inoculated in the seeds were multiplied in 125-mL Erlenmeyer flasks containing 50 mL of YMB culture medium of Vincent (1970), mixed with Bromothymol blue and kept under shaking until reaching a concentration of 10 9 cells per mL of broth, quantified in Neubauer chamber. Then, 3 mL of the rhizobium broth were aseptically applied to the seeds at planting and five days after germination, as a follow-up application (3 mL) at the collar region of each recently emerged seedling. After observing the formation of nodules in the roots, the strains resulting from this authentication were again isolated and incubated to be used in the subsequent experiments.
After a 12-day period, these colonies were morphologically characterized according to Vincent (1970), based on the following parameters: reaction to Gram stain, pH of the medium, growth time-GT; colony color-CC; transparency-TR; mucus elevation-EL; mucus formation-MF; mucus elasticity-ME; colony form-CF.

Effect of Manganese on Growth and Nodulation of Mimosa caesalpiniaefolia Benth.
The experiment was conducted in a greenhouse, belonging to the Department of Soil Science of the Federal University of Ceará (UFC). Substrate consisted of a soil with low content of organic matter and low Mn concentration, to allow the evaluation of the effect of increasing doses of this metal on the development of symbiosis between nodulating N 2 -fixing bacteria (NNFB) and leguminous species.
The soil used was collected from the 0-20 cm layer, at the Lavoura Seca farm (Lat.: 5º003′734″ S; Long.: 38º995′421″ W), situated in the municipality of Quixadá, Ceará, belonging to UFC. The soil, classified as Typic Eutrophic HAPLIC PLANOSOL, was air-dried and pounded to break up clods, then sieved through a 4-mm mesh and placed in plastic pots with capacity for 1 dm -3 . Samples were prepared and analyzed for physical and chemical characteristics according to the methodologies proposed by Embrapa (2017). Biological analyses were also conducted to quantify the number of viable rhizobium cells using the most probable number (MPN) technique according to Hungria and Araújo (1994) and arbuscular mycorrhizal fungi (AMF) spore density, by extracting 100 g of soil through wet sieving, following the procedures described by (Gerdemann & Nicolson, 1963) (Table  2).  The arboreal leguminous species, Sabiá (M. caesalpiniaefolia), with capacity to establish efficient symbiosis with NNFB, was sown using a total of three seeds per pot, maintaining one plant after thinning. For seed inoculation, an inoculant (MIX) containing the best three isolates of isolated and authenticated strains of each area, as described previously in section 2.2, was produced under orbital shaking of 105 rpm at 28 ºC until the cultures reached the log phase (10 9 cell mL -1 ). Initially, inoculation was performed by placing NNFB cells in peat using a sucrose solution (10%) as adhesive solution and, after seedling germination, a follow-up application was made using 2.0 mL pot -1 of culture in YMB (Vincent, 1970). The NNFB isolates of contaminated soil were selected based on the result of the phenotypical evaluation of nodulation (Table 3). Sabiá plants were harvested 60 days after sowing, when the following variables were analyzed: number of nodules (NN); nodule dry matter (NDM); shoot dry matter (SDM); nitrogen content in the shoots, determined by the micro-Kjeldahl method (Tedesco, Bohnem, Gianello, Bissani, & Volkweiss, 1995); manganese content in the shoots, in an extract obtained from nitric-perchloric digestion (Silva, 2009), determined by spectrophotometry of atomic absorption; and nitrogen accumulation in the shoots (NAS). N accumulation in the shoots was calculated by multiplying shoot dry matter by the N content.
The obtained data were subjected to analysis of variance. When significant difference was observed, qualitative data were compared by Tukey test (p < 0.05), while quantitative data (doses) were subjected to regression analysis, using the statistical program SISVAR (Ferreira, 2011).

Isolation, Authentication and Cultural Characterization of Isolates of Native Strains of NNFB From Manganese Mining Area
The leguminous species used as bait plant, sabiá, was able to nodulate with the strains native to the different soils collected in the Mn 2+ -mining area, which demonstrates compatibility between macro and microsymbionts in the three situations evaluated (native forest area, degraded area and tailings). After authentication of the selected isolates, 14 isolates of NNFB strains were obtained, 5 of which from the preserved forest area, 4 from the area degraded by mineral extraction and 5 from the tailings (Table 4).  In the authentication process, it was observed that out of all isolates of the selected strains, in a total of 15, only 1 was negative, while the remaining 14 (Table 4) were able to establish an efficient symbiotic relationship with the host, characterized by the intense green color of the leaves and intense red color of nodule center, indicating high concentration of leghemoglobin, and consequently good efficiency in N 2 fixation.
Soils contaminated by heavy metals usually have conditions of low fertility and become highly stressing environments for the biological community (Abdu et al., 2017). Thus, studies on the obtaining of NNFB adapted to such conditions become very important for successful programs of revegetation of soils contaminated by metals. It is worth pointing out that there is no vegetation in the area degraded by mining and in the generated tailings, so that the NNFB present in these environments survive without the direct influence of this environmental component, besides other conditions of high stress such as low availability of organic matter, P and K, slightly acid pH and high Mn concentration, which may indicate a high capacity of survival and tolerance of these bacteria, especially to the presence of high Mn concentrations. Some mechanisms used by these microorganisms in the tolerance to heavy metals have already been identified, particularly the production of exopolysaccharides (Nocelli, Bogino, Banchio, & Giordano, 2016), production of siderophores (Schalk, Hannauer, & Braud, 2011) and efflux of excess metals (Cubillas et al., 2014).
The cultural and morphological characterization of the isolates of strains authenticated and selected for the essay on the effect of Mn on the growth and nodulation of sabiá (Table 3) is presented in Table 5. It can be noted that all isolates showed Gram negative reaction, which is characteristic of the NNFB group belonging to proteobacteria (Brenner, Krieg, & Staley, 2005   Regarding the change in culture medium pH, the nine isolates selected were able to acidify the culture medium. For the growth time, all showed fast growth (3 days) in YMA culture medium mixed with Bromothymol blue. Fast growth capacity is already well documented and is typical of strains from arid and semi-arid climates, where the ability to fix N 2 is not the most important aspect for these microorganisms, but in fact their survival (Sprent, 1994). This characteristic also brings advantages in the competition with other microorganisms in the rhizosphere (Martins, Neves, & Rumjanek, 1997).
As observed, the isolates were similar, sharing the same characteristics with respect to several parameters, showing differences only in colony form. Possibly these isolates may belong to the same species, since they had similar characteristics, but it is not possible to state clearly because no molecular techniques were used for a definitive position in terms of taxonomy of these bacteria. In species belonging to the subfamily Mimosaceae (Mimosa), researchers have recently discovered that they can be nodulated mainly by species from the group of β-rizhobia (Burkholderia, Cupriavidus) (Bontemps et al., 2010;Chen et al., 2003), similarly to what occurs with sabiá (M. caesalpiniifolia) (Chen et al., 2008a;Martins et al., 2015).

Effect of Manganese on the Growth and Nodulation of Mimosa caesalpiniaefolia
Benth.
The analysis of variance for the studied variables of M. caesalpiniaefolia is presented in Table 6, corresponding to the values of probability and significance of F test for the main effects of treatments and their interactions. The interaction between Mn doses and NNFB inoculation treatments was significant for number of nodules (NN), nodule dry matter (NDM), shoot dry matter (SDM) and manganese content in the shoots (Mn), whereas the nitrogen content (N) and accumulation (NAS) were significantly affected only by the factor NNFB.  The variables number of nodule and nodule dry matter varied as a function of the interaction between Mn doses and NNFB, but no linear or quadratic fit could explain them, therefore, adjustments were made to the models most appropriate to the behavior of the data.
All inoculation treatments were able to establish effective symbiosis and form nodules with the microsymbiont, even in the presence of Mn. The strains of the control treatment had the lowest number of nodules compared to the isolates from mining area inoculated. The mix forest area showed higher number of nodules, with 39.00 and 31.75 nodules plant -1 at doses of 20 and 80 mg dm -3 , respectively. The mix degraded area, although there was no significant statistical difference between doses, reduced the number of nodules with increasing doses of Mn, demonstrating sensitivity to the presence of the metal. The mix tailings showed slight sensitivity to the presence of Mn only at dose of 20 mg dm -3 , statistically differing from the other treatments. The other doses did not cause effect on the isolates in terms of formation of nodules with the microsymbiont (Figure 1).

Figure
As observe between m number of small com lower num production Dobereine  Strains native to the soil were sensitive to the presence of Mn at the highest concentrations in the soil, evidenced by the negative effect on shoot dry matter production, whereas rhizobium isolates from Mn-mining area were tolerant to the presence of the metal at the concentrations tested in the present study. As mentioned by Ferreira et al. (2012), the tolerance of N 2 -fixing bacteria to heavy metals may be related to the origin of the isolates, even when they belong to the same genus or same species. As observed here, the isolates which composed the mixes were obtained from areas with high Mn concentrations, unlike the strains native to the soil used as substrate (control treatment), different from the area contaminated by Mn, which under this experimental condition are similar to what was mentioned by these authors, although one should not generalize such model of behavior.
It is known that one of the mechanisms through which rhizobia can improve the growth of leguminous species in stressing environments is the production of phytohormones such as auxins (indoleacetic acid-IAA). This hormone acts directly on cell division and elongation, thus improving plant growth (Brígido & Glick, 2015). Hence, it is possible to suppose that the rhizobium isolates from Mn-mining area had higher capacity to produce phytohormones which promote plant growth, compared to the strains native to the natural soil used in the experiment.
According to research studies, sabiá has specificity in the symbiosis with rhizobia, demonstrating to be preferentially nodulated by strains of the genus Burkholderia (Chen et al., 2008a). This group of bacteria is well characterized as plant-growth promoter through solubilization of phosphates, production of phytohormones (auxins, gibberellins and cytokinin) and through symbiotic N 2 fixation (Bontemps et al., 2010;Chen et al., 2003;Martins et al., 2015).
Regarding the nitrogen content in the shoots, the control treatment was statistically different from the others, showing the lowest N content. The isolates from mining area were statistically similar and did not differ from one another. N accumulation followed the same behavior of N content, with difference between the isolates from mining area and the control treatment, besides similarity between the different mixes (Table 7). All plants benefited from the symbiotic N 2 fixation by their microsymbionts because there were no symptoms of deficiency of this element until 60 days, and no other source was used. Rhizobium isolates from mining area were more efficient in terms of fixation and, consequently, supply of N to plants. By comparing the isolates from the mix tailings to the community native to the soil (control), it is possible to note that there was an increment of about 47.0% in the N content in the shoots of inoculated plants. Despite the inferiority in N content and accumulation in plants by the rhizobium community native to the soil, one should take into account that, even with lower production of nodules (at most 3 nodules per plant) (Figure 1), these native strains were able to supply a considerable quantity of N to plants. Malavolta, Vitti, and Oliveira (1997) suggest that adequate N contents for forest species should range from 16 to 35 g kg -1 , values found in all plants of the present experiment. Ferreira et al. (2013), evaluating the effect of heavy metals on the symbiosis of Mimosa caesalpiniaefolia and strains of Cupriavidus necator, detected N contents in the shoots from 32 to 39 g kg -1 . Similar results were also obtained by Ferreira et al. (2012), studying the symbiotic efficiency of the same association, with N contents in the plant ranging from 11.0 to 36.5 g kg -1 and maximum N accumulation of approximately 6 mg plant -1 .
Based on the analysis of the interaction between doses and NNFB for Mn content in the shoots, there was significant statistical difference for the treatments of inoculation with mix forest area, mix degraded area and mix tailings. The control treatment did not fit to any mathematical model ( Figure 4).
The data of the treatment mix forest area fitted to a quadratic regression model, in which the minimum Mn content (6.97 mg kg -1 ) was obtained at the dose 39.4 mg dm -3 . The data of the mix degraded area fitted to a linear regression model, and Mn content in the shoots decreased with the increasing levels of Mn in the soil. The data of the mix tailings fitted to a quadratic model in which the minimum Mn content (9.55 mg kg -1 ) was obtained at Mn dose of 61.20 mg dm -3 ( Figure 4). In general, the treatments of inoculation with the rhizobium isolates had influence on Mn content in the shoots, which decreased as the Mn levels in the soil increased. Chen et al. (2008b) observed an increase in the absorption of Pb, Cu and Cd in Mimosa pudica plants inoculated with Cupriavidus taiwanensis, compared to non-inoculated plants. Wani and Khan (2013), studying the association of the rhizobium strain RL9 with lentil plants, observed reduction in Ni absorption compared to non-inoculated plants.
According to Brígido and Glick (2015), one of the ways to reduce the absorption of metals by plants inoculated with rhizobia would be the accumulation of metals in root nodules by the strains which nodulate their hosts, while free strains that remain in the rhizosphere perform detoxification by different mechanisms, such as release of