Bioimmobilization of lead by Bacillus subtilis X3 biomass isolated from lead mine soil under promotion of multiple adsorption mechanisms

In this study, a lead-resistant bacterium, Bacillus subtilis X3, was used to prepare a lead bioadsorbent for immobilization and removal of lead in lead solution. The lead shot precipitate was analysed by scanning electron microscopy combined with energy dispersive X-ray fluorescence microscopy, Fourier transform infrared spectroscopy, X-ray diffraction and X-ray photoelectron spectroscopy. The adsorbed lead was mainly mineralized to form Pb5(PO4)3OH, Pb10(PO4)6(OH)2 and Pb5(PO4)3Cl; however, other mechanisms that can also promote the mineralization of lead should not be ignored. For example, Na+ and Ca2+ on the cell wall surface were exchanged with Pb2+ in solution, which confirmed that the ion-exchange process occurred before mineralization. Moreover, adsorption bridging caused by extracellular polymeric substances also accelerated the further aggregation of lead, and the biomass was encapsulated by lead gradually. Hydroxyl, carbonyl, carboxyl and amine groups were not observed in lead mineral crystals, but the complexation between lead and these groups still benefited the mineralization of lead. The valence of Pb(II) was not changed after mineralization, which indicated that the biosorption process was not a redox reaction. Finally, biosorption occurred on the outer surface of the cell, but its specific surface area was relatively small, limiting the amount and efficiency of biosorption.


Isolation and identification of microorganism
The lead-resistant bacterial strain was isolated from the lead-contaminated soil of a lead mine in Nanjing, Jiangsu Province, China. To enrich the bacteria, 10 g of soil was added into 100 ml Luria-Bertani (LB) basal medium and then incubated at 150 rpm and 378C for 48 h. Next, 1 ml of the above culture was added into 100 ml basal medium containing 400 mg l 21 Pb(NO 3 ) 2 and cultured under the same conditions described above. The isolates were then purified by the dilution-plate method [27]. Individual bacterial colonies were isolated on LB medium containing 400 mg l 21 Pb(NO 3 ) 2 . The formation of black lead shot precipitation and lead reduced in the culture indicated that the bacterial colony could immobilize lead. The concentration of lead in solution was detected by flame atomic adsorption spectrometry (TAS-900, PGENERAL).
royalsocietypublishing.org/journal/rsos R. Soc. open sci. 6: 181701 Genomic DNA of strain X3 was extracted using a Bacterial Genomic DNA Extraction Kit (Sangon, China). The 16S rRNA gene was amplified using the primers f1 (5 0 -AGTTTGATCMTGGCTCAG-3 0 ) and r1 (5 0 -GGTTACCTTGTTACGACTT-3 0 ) by polymerase chain reaction, after which the fragments were purified using a gel extraction kit (Sangon, China), and the sequencing of the 16S rRNA gene was conducted by Sangon (Shanghai, China). The BLAST programme (http://blast.ncbi.nlm.nih.gov/Blast. cgi) was used for a sequence similarity search with the standard programme by default. Multiple sequence alignment and data analysis were conducted using the software package MEGA v. 5.1, and a phylogenetic tree was constructed using the neighbour-joining method.

Bacterial biomass adsorption preparation
The isolated and purified strain X3 was inoculated in LB medium and cultured at 378C and 150 rpm for 24 h. The cells were then harvested by centrifugation for 10 min at 5000g, after which the biomass was rinsed with sterile deionized water three times to remove the culture solution. Finally, the obtained cells were freeze-dried using a lyophilizer.

Lead adsorption experiments
The prepared biomass adsorbent was placed in 100 ml solution containing Pb(NO 3 ) 2 (200, 400, 600, 800, 1000, 1200, 1400 mg l 21 of Pb 2þ ). The pH of the solution was adjusted to 1 -5 by 0.1 M NaOH solution and 0.1 M HNO 3 solution. Next, each solution was amended with biomass (0.01, 0.02, 0.03, 0.04, 0.05, 0.06, 0.07 and 0.08 g). The reaction system was then placed in a shaker at 258C at 150 rpm for half an hour, after which it was allowed to stand for 1-30 min. Following centrifugation at 8000 rpm for 5 min, the concentration of lead in the supernatant was detected by flame atomic adsorption spectrometry. Each experiment was performed with three biological and technical replicates. The 95% confidence interval ( p , 0.05) was set as the significance threshold. The lead shot precipitate was collected by rinsing three times using sterile deionized water. Finally, the adsorption capacity of the biomass adsorbent was calculated by the following formula: where C 0 and C e were the initial and final concentrations of lead (mg l 21 ), respectively, and V and M were the volume of solution (l) and the weight of the biomass (g), respectively.

Characterization of lead shot
The collected lead shot was dried at 1058C for 2 h, then used for analysis. The surface morphological and elemental components of the lead shot could be analysed by scanning electron microscopy combined with energy dispersive X-ray fluorescence microscopy (SEM-EDS, Shimadzu, UK). The SEM spectra enabled the direct observation of changes in the surface structures of the bacteria [5]. Lyophilized samples of the bacteria before and after contact with Pb 2þ were used for X-ray diffraction (XRD) analysis (Rigaku, Japan). Diffraction patterns were collected at angles ranging from 158 to 658 with a 0.028 step-length. X-ray photoelectron spectroscopy (XPS, Shimadzu) was used to characterize the surface chemical compositions of the lyophilized samples [28], while the surface functional groups of the bacteria were analysed by Fourier infrared spectroscopy [5], after diluting the adsorbent to 5% in KBr and casting the samples in disks (Brooke, Germany).

Isolation and identification of bacteria
Seven different isolates exhibiting lead resistance were obtained from the lead-contaminated soil at a lead mine plant. One isolate, strain X3, was selected for further study because of its high lead tolerance and removal rate. The 16S rRNA sequence of strain X3 was amplified and sequenced, after which the sequence was submitted to GenBank under accession number KX966417.  figure S1). Finally, strain X3 was identified as B. subtilis X3 based on the results of the morphological observations and the 16S rRNA sequence analysis. The growth of B. subtilis X3 on plates with different lead concentrations showed that the maximum amount of lead B. subtilis X3 could tolerate was 2000 mg l 21 .

Different factors in biosorption
To investigate the effects of different pH on the adsorption of biomass, the pH of solution was adjusted from 7.0 to 1.0. When the pH was 6.0, some lead shot precipitate appeared. The lead removal at pH values of 1-5 are shown in figure 1a. The adsorption capacity increased with the increase in pH, eventually reaching the maximum value of 192.05 mg g 21 . Contact time is an important factor influencing the removal rate of lead biosorption. Batch experiments were conducted for different reaction times to determine the optimal contact time for the biomass of B. subtilis X3 (figure 1b). Briefly, 0.11 g of biomass was added into 50 ml of solution with 400 mg l 21 Pb(NO 3 ) 2 . The amount of bioadsorbed lead increased with increased contact time, and reached a maximum of 260 mg g 21 at 10 min.
The dosage of biomass was also an important factor influencing the adsorption of lead. To ensure its effects on lead adsorption, varying amounts were introduced into lead solution. As shown in figure 1c, the adsorption capacity of the biomass was 703.25, 321.875, 228.08, 188.62, 180.9, 144.83, 130.46 and 133.31 mg g 21 . These results showed that the adsorption capacity of the adsorbent decreased as the biomass increased.
To investigate the effects of initial lead concentration on biomass adsorption, the lead concentration after adsorption was determined under the initial concentration of 150-400 mg l 21 . As shown in figure 1d, the adsorption amount increased with the initial lead concentration, and the adsorption capacities were 129.8, 137.9, 163.55, 194.05, 218.95 and 283.75 mg g 21 . These results showed that the amount of lead adsorbed increased with the initial lead concentration. High concentrations can increase the mass transfer power of lead and overcome the obstacle of mass transfer between the solid and liquid phase of lead. When the concentration of lead in the solution was 350-400 mg l 21 , the effects of cell adsorption increased, indicating that there is a more rapid and effective mechanism of adsorption in solutions containing high concentrations of lead. The chemical components of the lead shots were analysed using EDS images. As shown in table 1, the percentage of phosphorus, calcium and lead increased, while that of sodium decreased from 3.16 to 1.66% when compared with the control. These results showed that ion exchange occurred between lead, calcium and sodium. Not only was lead-hydroxyapatite formed, but also Ca 3 (PO4) 2 . Moreover, the decrease in carbon and oxygen ion showed that the lead shot had covered the cell biomass.

XPS of lead shot
The chemical composition of the cell surface and identification of the oxidized state of the lead samples were evaluated by XPS analysis ( figure 3). The results of XPS indicated that the core level peaks were C-1s (284.6 eV), Pb-4f (138.40 eV) and Pb-4f5/2 (142.70 eV). The C-1s peak fluctuated around 284.1 eV, indicating that the C (C, H) or amino acid side chain functional groups were involved in the adsorption process and that carboxylate or ester C¼O functional groups were at 289.0 eV during this process. Pb-4f7/2 and Pb-4f5/2 peaks were observed at 138.40 eV and 142.70 eV, respectively. The corresponding spin orbital division of the peak was 5.01 eV, indicating that the lead adsorbed on the outer surface by the bacterial adsorbent was still in the divalent valence state.

XRD of lead shot
Compounds that undergo lead transition during adsorption were identified by XRD spectral analysis. As shown in figure 4, a new adsorption peak appeared at 308 after adsorption. Analysis using the MID Jade software analysis showed that Pb 2þ was transferred to Pb 5 (PO 4 ) 3 , Pb 10 (PO 4 ) 6 (OH) 2 and Pb 5 (PO 4 ) 2 Cl during the adsorption process.

FT-IR of lead shot
The results of the Fourier transform infrared spectroscopy (FT-IR) can be used to determine functional groups on the outer surface of the lead shot and their sources. Many functional groups on the outer surface of bacteria can chelate or complex lead, such as hydroxyl, carboxyl and phosphate groups. The general distribution of the various major groups on the lead shot in the infrared adsorption spectrum is shown in figure 5. The results showed that there was a broad and strong adsorption peak in the    region of 3500-3300 cm 21 , which was the stretching peak of O -H and N-H derived from carboxylic acid, amino acid or alcohol [29]. The adsorption peak at 2928 cm 21 can be assigned to C-H stretching.
Obvious changes between biomass and the lead shot at 2361 cm 21 reflecting the C;C stretching vibration peak were observed. The adsorption peak at 1650 cm 21 was indicative of C¼O group [30]. Additionally, a P¼O stretching vibration peak at 1400 cm 21 derived from the phosphoric acid group was observed, as was a C-N stretching vibration peak of the saccharide at 1236 cm 21 . Finally, a stretching vibration peak at 1050 cm 21 originated from the OH of alcohol or C-N groups [31]. These results showed that the adsorption process was mainly because of the adsorption of functional groups on the outer surface of the biomass, which was confirmed by the experimental results of FT-IR.

Discussion
In this study, lead-resistant B. subtilis X3 was isolated from the soil of a lead mine plant, formed into a biomass adsorbent and applied to the removal of lead in solution. When compared with similar studies (table 2), the biomass generated here has good potential for practical application.   Adsorption of lead in solution by biomass is influenced by many factors, with solution pH being one of the most important. In this study, the adsorption capacity of B. subtilis X3 biomass reached the maximum amount when the pH was 4, which was similar to the results of previous reports [42]. Ren et al. [5] pointed out that pH is one of the most important parameters affecting the solubility of metal ions and functional groups on the cell wall of biomass. A functional group such as a carboxylate on the bacterial cell wall is protonated below 4.0, and Pb(II) is weak in competition with hydrogen ions at the adsorption site on the surface of the cell. Therefore, the lower Pb(II) absorption capacity was caused by more hydrogen ions at a pH of 3.0-4.0 [5].
The initial lead concentration also affected the biosorption of lead. The amount of Bacillus thuringiensis 016 adsorbed increased with the increasing initial lead concentration [42], but in a study conducted by Ren et al. [5], the amount of lead biosorbed to Bacillus sp. PZ-1 decreased with increasing initial concentration. When compared with other microorganisms, the time required to reach biosorption equilibrium in the B. subtilis X3 biomass was obviously shorter.
Adsorption during biosorption occurs via ion exchange, complexation and biomineralization [43]. In previous studies, Pb 2þ was found to be transformed to PbHPO 4 , Pb 9 (PO 4 ) 6 or PbS [44,45]. In the present study, the products of Pb 2þ biomineralization were Pb 5 (PO 4 ) 3 OH, Pb 10 (PO 4 ) 6 (OH) 2 and Pb 5 (PO 4 ) 3 Cl. However, we found no evidence of PbS and PbSiO 3 formation [46]. We assumed that the formation of lead-phosphorus was determined by the species of microorganism, amount and kinds of functional groups on the cell surface and the solution composition. The immobilization of lead occurs via a complexation reaction between Pb 2þ in solution and functional groups on the cell surface that lead to formation of an insoluble substance. These functional groups include hydroxyl, carbonyl, carboxyl, amine and phosphoric acid groups [15]. In this study, we found the same groups on the lead shot upon FT-IR spectral analysis, but we did not find these functional groups upon XRD analysis of the lead mineral crystals. We assumed that complexation could help form the lead mineral, although the process did not play a major role. Furthermore, we found that some C;C bonds bound lead, which was different from the results of other studies. However, we did not conclude from which substances the C;C bond was derived.
Adsorption of Pb 2þ onto the biomass occurred via ion exchange with Na þ , K þ and Ca 2þ in solution. In this study, the concentration of Na þ decreased in solution, while it increased obviously in the lead precipitate (figure 2). The EPS secreted by microbes was an important factor influencing lead adsorption. Moreover, EPS was usually exported to solution to bind lead, although it could remain on the surface of microbes even though microbial biomass was obtained by centrifugation and freezedrying. We assumed that adsorption bridging action promoted the immobilization of lead, and that this was contributed to by EPS on the surface of the microbial biomass. Electron transfer will not occur during binding of Pb 2þ and functional groups [47]. The valence of lead did not change, indicating that there was no electron transfer and that the lead-phosphorus mineral was formed by electrostatic incorporation.

Conclusion
In this study, B. subtilis X3 isolated from the soil in a lead mine plant was able to adsorb lead efficiently in solution. To investigate its adsorption mechanism, SEM-EDS, XRD, XPS and FT-IR methods were applied to analyse the characteristics of lead shot precipitate formed during adsorption. The analysis showed that the lead minerals formed were composed of Pb 5 (PO 4 ) 3 OH, Pb 10 (PO 4 ) 6 (OH) 2 and Pb 5 (PO 4 ) 3 Cl, which are the major immobilization mechanisms. During the mineralization of lead, ion exchange and adsorption bridging played important roles, indicating that they can be used to promote mineralization. The biosorption of lead mainly occurred on the outer surface, where lead can bind with hydroxyl, carbonyl, carboxyl, amine and phosphoric acid functional groups. We also found that C;C bonds were involved in the binding of lead. These groups can accelerate the formation of lead mineral crystals, although they were not found in lead shots by XRD analysis.
Data accessibility. DNA sequences: Genbank accessions KX966417. Authors' contributions. Y.Z. carried out isolation and identification of the lead-resistant bacteria, participated in data analysis; H.X. carried out the adsorption experiments of bacteria biomass, and the design of the study and drafted the manuscript; H.X. and S.L. carried out the acquisition of data, analysis and interpretation of data; Y.L., S.W. and W.W. carried out determination of lead shot characteristics; W.Q. conceived of the study, designed the study, coordinated the study, helped draft the manuscript and revised it critically for important intellectual content, and final approval of the version to be published. All authors gave final approval for publication. All authors agreed to royalsocietypublishing.org/journal/rsos R. Soc. open sci. 6: 181701