Characterization of Microbial Communities, Identification of Cr(VI) Reducing Bacteria in Constructed Wetland and Cr(VI) Removal Ability of Bacillus cereus

In this study, the contribution of substrates microorganisms in three different constructed wetlands (CWs) to Cr(VI) purification was discussed. In addition, the microbial communities in the substrate of different CWs were characterized, and rhizosphere Cr(VI) reducing bacteria was also identified. The results showed that microorganisms could improved Cr(VI) removal to 76.5%, and result in that more Cr(VI) was reduced to Cr(III). The dominant strains in the substrates of different CWs were Sphingomonas sp., Cystobacter sp., Acidobacteria bacterium, Sporotrichum and Pellicularia species. The Cr(VI) reducing bacteria from Leersia hexandra Swartz rhizosphere was identified as Bacillus cereus. Furthermore, under suitable conditions, the removal rate of Cr(VI) by Bacillus cereus was close to 100%.

A suitable substrate not only could remove Cr(VI) by adsorption or ion exchange 11,12 , but also could increase the permeability and hydraulic load of the CWs 13 . At present, biochar used as substrate has attracted extensive attention because of its enhancement of soil fertility, attraction of beneficial microorganisms, and improvement of pollutants removal 14,15 . For example, the addition of biochar to the substrate can significantly increase the abundance of Proteobacteria, Acidobacteria, Firmicutes and Actinobacteria, and promote the transformation of the dominant species from Rudeea to Bacillus which has strong degradation ability to chlorpyrifos 16 . The contribution of microorganisms to Cr(VI) removal is achieved through biosorption and bioreduction 17,18 . Singh et al. 19 confirmed that inoculation of Cr-reducing bacteria at the root of plants can increase the removal of Cr(VI) by constructed wetlands. Moreover, microorganisms can also form symbionts with plant roots, thereby increasing plant uptake of heavy metals 20 . Unfortunately, most researches pay attention on the effect of plants and matrix on Cr(VI) removal 4,21,22 , while there are a few reports about the effect of microorganisms on Cr(VI) removal 9,23 .
In this study, three different CWs were used to treat Cr(VI) contaminated water for three special purposes. First, the contribution and role of microorganisms in CWs for the Cr(VI) removal were determined. Second, the characteristics of microbial communities in substrate were investigated based on the results of polymerase chain reaction-denaturing gradient gel electrophoresis (PCR-DGGE). Last, the Cr(VI)-reducing bacteria in the plants rhizosphere were isolated and identified, and its ability of reducing Cr(VI) was also discussed.

Materials and Methods
Materials. The soil used as the substrate of the constructed wetland was collected from the surface of the beach (0-20 cm) at the junction of the Lijiang River and the Liangfeng River. After the soil is naturally air-dried, it passes through a 60 mesh sieve. The bagasse biochar, also used as a substrate, was prepared by a two-step pyrolysis process at 200 °C and 500 °C, respectively. The bagasse biochar also passed through a 60 mesh sieve. The physical and chemical properties of soil and bagasse biocahr are shown in Table 1.
The Cr(VI)-containing wastewater was prepared from potassium dichromate (analytical grade) and tap water. The Cr(VI) concentrations were 5.00 mg/L and 25 mg/L, respectively.
Leersia hexandra Swartz was collected from the non-polluted area of the Liangfeng River beach in Yanshan District, Guilin, China. It was trimmed to 15 cm and then washed with pure water. The trimmed Leersia hexandra Swartz was incubated in Hoagland nutrient solution for 5 d to restore the damaged roots during transport.
Design and operation of constructed wetlands. In this study, a three-stage wavy subsurface flow constructed wetland was designed, the details of which are shown in Fig. 1. On this basis, according to the experimental purpose, three different forms of wetlands, 1# CWs, 2# CWs and 3# CWs, were designed and operated. The details of these three wetlands are shown in Table 2. After 156 d of operation, the substrates of the three CWs were sampled using a quincunx method. The samples were freed of impurities and stored aseptically sealed at 4 °C.
Purification of Cr(VI) by microorganisms in substrates. The substrate sample collected from 1# CWs was divided into three groups for different treatments. The first group of samples did not take any treatment, while the second group and the third group of samples were sterilized by high-temperature autoclaving and chloroform sterilization, respectively.
Before usage, the potassium dichromate solution with Cr(VI) concentration of 25 mg/L was first filtered with a 0.22 micron membrane to remove bacteria. The differently treated substrates with the quality of 10 g were separately added to three 150 mL Erlenmeyer flasks, and then 50 mL of the filtered solution was added to  www.nature.com/scientificreports www.nature.com/scientificreports/ the Erlenmeyer flasks. The Erlenmeyer flasks were shaken at 150 rpm at 30 °C. After 72 h, the mixtures in the Erlenmeyer flasks were transferred to three 10 mL centrifuge tubes, respectively, and centrifuged at 4000 rpm for 10 min. The Cr(VI) and total Cr concentrations in the supernatant samples were determined by diphenylcarbazide spectrophotometry and potassium permanganate oxidation-diphenylcarbazide spectrophotometry, respectively. The Cr(VI) and total Cr in the solid phase were extracted by basic digestion method (EPA3060A) and acidic digestion method, respectively, and their concentrations were also determined by diphenylcarbazide spectrophotometry and potassium permanganate oxidation-diphenylcarbazide spectrophotometry, respectively. The concentration of Cr(III) is the difference between total Cr and Cr(VI). The PCR products were separated by DGGE on 8% polyacrylamide gels in 1 × TAE buffer, and detected by the D-Code Mutation Detection System (Bio-Rad, US) 19 . The linear denaturing gradients for bacteria and fungi were 35~65% and 30~50%, respectively. Electrophoreses were performed at 70 V and 60 °C for 14 h. Then, the gels were rinsed with ultrapure water and stained with a dye bath containing 1% Goldview. The Quantity One software was used to identify the lanes and bands of the DGGE profiles. Basing on the results of Quantity One, the coefficient of similarity (Cs) and diversity index (Di) were calculated.

Analysis of microbial communities in substrates.
The representative bands were excised and crushed. After elution with 50 μL sterile distillation-distillation H 2 O, they were stored overnight at 4 °C. The 1 μL soaking solution was selected as a template and amplified using primers 357F/518R. The PCR products were sequenced with primer F357 by Invitrogen Biotechnology Co. Ltd. (Shanghai, China). And the sequence was submitted to the NCBI database for BLAST alignment.
Isolation and identification of rhizosphere Cr(VI) reducing bacteria. Liquid Luria-Bertani culture and solid Luria-Bertani culture were used to separate and culture Cr(VI) reducing bacteria. Their components were as follows: peptone 10 g/L, yeast extract powder 5 g/L, NaCl 5 g/L (liquid Luria-Bertani culture); peptone 10 g/L, yeast extract powder 5 g/L, NaCl 5 g/L, 1.5% agar (solid Luria-Bertani culture). They were sterilized at 121 °C for 20 min.
The soil collected from the rhizosphere of Leersia hexandra Swartz was cultured in liquid Luria-Bertani culture for 12 h at 37 °C. Then, the supernatant was transferred to a liquid Luria-Bertani culture containing 5 mmol of Cr(VI), and also cultured at 37 °C for 12 h. In order to screen out the Cr(VI) reduced strain, the bacteria enriched in liquid Luria-Bertani culture were transferred to solid Luria-Bertani culture and cultured for 48 h at 37 °C 28 .
The isolated strain was first rejuvenated and cultured for 35 h. After this, the DNA of the strain was extracted and subjected to 16S rDNA PCR amplification. The primers of PCR amplification were 5′-AGA GTT TGA TCC TGG CTC AG-3′ (16 s rDNA forward primer 27F), 5′-ACG GTT ACC TTG TTA CGA CTT-3′ (16 s rDNA reverse primer 1492R) 29 , 5′-AAAgTTTAAAgAAgTACAAgAAgC-3′ (dnaJ gene forward primer) and 5′-CTTTACCATgAACAgTAggAAC-3′ (dnaJ gene reverse primer Test of reducing ability of Cr(VI) reducing bacteria. The effects of Cr(VI) initial concentration, initial pH value, reaction temperature and inoculation amount on the reducing ability of Cr(VI) reducing bacteria in rhizosphere were investigated. A certain amount of K 2 Cr 2 O 7 was added to the liquid Luria-Bertani cultures in which the Cr(VI) reducing bacteria inoculations were 1%, 5%, 10%, 15%, and 20%, respectively, so that the concentration of Cr(VI) was in the range of 20 to 100 mg/L. The pH of the liquid medium was adjusted in the range of

Results and Discussion
Contribution of microorganisms to Cr(VI) purification. The contribution of substrates microorganisms to Cr(VI) purification is shown in Fig. 2. Form Fig. 2(a), it can be seen that the contribution of microorganisms to Cr(VI) purification is great, especially in the early stages of Cr(VI) removal. When the time was 12 h, the removal rate of Cr(VI) was 76.5% in untreated substrate, which was 2.36 and 3.18 times that of high-temperature autoclaving and chloroform sterilization, respectively. According to the chromium mass balance (Fig. 2(b)), there is more Cr(VI) in substrate of no treatment, indicating microorganisms immobilize more of the Cr(VI) in the substrate by biosorption 17 . At the same time, the total content of Cr(III) in the substrate and supernatant is the highest in the untreated substrate test, confirming that the microorganism can achieve detoxification of Cr(VI) through biological reduction 18 .
Phylotypes and diversity of microbial community. Figure 3 shows the DGGE profiles of bacteria and fungi in substrates of different CWs. As shown in Fig. 3(a), there are significant differences in the number and population of bacteria in the three constructed wetland substrates. Firstly, the obvious bands number of the 2# CWs DGGE profile is the highest, indicating that under the stress of Cr(VI), the microbial community in the substrate is inhibited and the biodiversity is reduced. which may cause the substrate ecosystem to become fragile 32 Table 5. They are directed to Sphingomonas sp., Cystobacter sp., and Acidobacteria bacterium, respectively. From Fig. 3(b), it can be seen that the number of bands in the three DEEG profiles are basically the same, while the positions of the bands are very different. This indicates a significant change in the population of the fungus. The diversity index (Table 3) and similarity coefficients (Table 4) Table 5. They mainly belong to Sporotrichum, and Pellicularia species.
Identification of rhizosphere Cr(VI) reducing bacteria. The Cr(VI) reducing bacteria isolated from the rhizosphere of Leersia hexandra Swartzt are labeled as GG. Figure 4 and Table 6 show its cell morphology and physiological and biochemical characteristics, respectively. Compared with other studies 33,34 , the Cr(VI) reducing bacteria may be Bacillus cereus. Moreover, the phylogenetic tree of GG (Fig. 5) shows a significant genetic relationship with Bacillus cereus. Therefore, GG was identified as Bacillus cereus. It should be noted that DGGE www.nature.com/scientificreports www.nature.com/scientificreports/   www.nature.com/scientificreports www.nature.com/scientificreports/ profiles did not recover Bacillus cereus in the previous microbial community analysis. The reason may be that although Bacillus cereus is highly resistant to Cr(VI), it does not form a dominant flora in wetlands. Fig. 6(a) shown, Bacillus cereus has higher Cr(VI) removal rates under neutral and alkaline conditions, and the maximum removals were nearly 100% after 48 h. The reasons may be as follow: (1) under neutral and alkaline conditions, Bacillus cereus has stronger biological activity and cell metabolism which are benefit for Cr(VI) reduction 35 ; (2) at low pH level, more hydrated ions are formed, causing protonation of negative sites on bacterial cells, which will reduce the amount of Cr(VI) biosorption 36 .

Effect of Cr(VI) concentration on Cr(VI) removal.
It is obvious that with the increase of Cr(VI) concentration, the reduction effect of Cr(VI) by Bacillus cereus is decreasing (Fig. 6(b)). However, when the contact time is longer than 24 h, the influence of the Cr(VI) concentration becomes insignificant. The reasons may be as follows: (1) In the early stage, high concentration of Cr(VI) has a large negative effect on Bacillus cereus, such as oxidative stress, DNA and protein damage 37 ; (2) After a period of time, Bacillus cereus is resistant to Cr(VI) stress and restores activity. This also indicates that Bacillus cereus has good adaptability to high concentrations of Cr(VI).
Effect of temperature on Cr(VI) removal. The effect of temperature on Cr(VI) removal is shown in Fig. 6(c). It is clear that when the temperature is 25 °C and 45 °C, Cr(VI) removals are lower. This may be attributed to the following reasons: (1) the Cr(VI) reductase activity of Bacillus cereus is lower when the temperature is higher or lower 38 ; (2) When the temperature is in the range of 30-40 °C, the affinity of the Cr(VI) site or the binding site on the Bacillus cereus is higher 39 . In addition, the curves of 35 °C and 40 °C are almost coincident, indicating that the suitable temperature range for the reduction of Cr(VI) by Bacillus cereus is 35-40 °C.    Fig. 6(d) depict Cr(VI) removal during 48 h, when Bacillus cereus inoculation is increased from1.0 to 20%. When the time is less than 36 h, the removal rate of Cr(VI) increases with the increase of inoculation, which may be due to the increase of cell density