Biodegradation of 17β-estradiol by Bacterial Co-culture Isolated from Manure

Animal wastes are potential sources of natural and steroidal estrogen hormones into the environment. These hormones can be removed by microorganisms with induced enzymes. Two strains of 17β-estradiol-degrading bacteria (LM1 and LY1) were isolated from animal wastes. Based on biochemical characteristics and 16 S rDNA gene sequences, we identified strains LM1 and LY1 as belonging to the genus of Acinetobacter and Pseudomonas, respectively. Bacterial co-culture containing LM1 and LY1 bacterial strains could rapidly remove approximately 98% of E2 (5 mg L−1) within 7 days. However, strains LM1 and LY1 degraded 77% and 68% of E2 when they were incubated alone, respectively. More than 90% of 17β-estradiol (E2, ≤ 20 mg L−1) could be removed by bacterial co-culture. Low C/N ratio (1:35) was more suitable for bacterial growth and E2 degradation. The optimal pH for bacterial co-culture to degrade E2 ranged from 7.00 to 9.00. Coexisting sodium acetate, glucose and sodium citrate decreased E2 degradation in the first 4 days, but more E2 was removed when they were depleted. The growth of the bacterial co-culture was not significantly decreased by Ni, Pb, Cd or Cu at or below 0.8, 1.2, 1.6 or 0.8 mg L−1, respectively. These data highlight the usefulness of bacterial co-culture in the bioremediation of estrogen-contaminated environments.

degrade 17β-estradiol by the synergistic effect are present in livestock and poultry wastes and that (2) the synergistic effect among bacteria enhances their adaptability to the environment. In the present study, two bacterial strains that exhibit a synergistic effect were isolated and identified. Furthermore, the effect of many factors such as nitrogen source, C/N ratio, pH, carbon coexistence, and heavy metals on the bacterial growth and 17β-estradiol degradation were also studied.

Results and Discussion
Isolation of 17β-estradiol-degrading bacterial co-culture. Two strains of bacteria with a strong ability to degrade E2 were successfully isolated from the composite samples after 50 days of enrichment cultivation. They were arbitrarily named as strains LM1 and LY1. Figure 1a shows that the bacterial co-culture containing strains LM1 and LY1 continually grew with an average maximum OD 600 of 0.352 during the 7 days of incubation. In comparison, after 2 days of incubation, the individual growth of strains LM1 and LY1 almost stopped and the average maximum OD 600 was 0.228 and 0.197, respectively. The bacterial co-culture containing LM1 and LY1 bacterial strains could rapidly remove approximately 98% E2 when 5 mg L −1 of E2 was used as a sole carbon source within 7 days of incubation. However, concentration of E2 decreased only 77% and 68% when strains LM1 and LY1 were incubated alone, respectively (Fig. 1b). Overall, the bacterial co-culture could grow on E2 as the sole carbon source and continually utilize the metabolite, which led to the continuous growth of bacteria and decrease of E2. In the case of individual growth, lethal intermediate metabolites may be produced by strain LM1 or LY1, which led to unchanged bacterial growth and E2 degradation after the logarithmic phase. Yu et al. divided the degradation of E2 by a single bacterial strain into three patterns according to the degree of estrogens being transformed and the estrogenicity of metabolites and/or the end products of estrogen degradation 28 . In pattern A, the concentration of estrone increased and accumulated over the experimental period due to the degradation of E2, and no degradation of estrone was observed. In pattern B, the transformation of E2 to estrone occurred much more slowly and only a small part of produced estrone was degraded. In pattern C, degradation of E2 and estrone were much improved with no estrogenic metabolites and/or end products from E2 degradation. In addition, several bacteria reversibly converted E2 to E1 but could not decrease estrogenic activity 26 . According to the E2 biodegradation trend, the bacterial co-culture containing strains LM1 and LY1 may degrade E2 in pattern C. By comparison, the individual degradation of E2 by strain LM1 or LY1 may be recognized as pattern B. So, we can infer that there are different metabolic mechanisms which prevent the further degradation of E2 by strain LM1 or LY1. When strains LM1 and LY1 are together, this inhibitory can be overcome through inter-organism interactions 29 . Nevertheless, the bacterial co-culture may play an important role in the degradation of estrogens from livestock and poultry manure. Therefore, strains LM1 and LY1 were identified, and their ability to degrade E2 under different conditions was tested.
Characterization of strains LM1 and LY1. The following is the characterization of physiological and biochemical reactions for strain LM1: Gram-negative, rod-shaped, catalase-positive, oxidase-negative and grew on glucose with acidification. In addition, Strain LM1 utilized glucose and lactose. And hydrogen sulfide test showed that there was no hydrogen sulfide produced. Strain LY1 was Gram-negative, short rod-shaped, had no spores, and grew on succinate with alkalization. In addition, strain LY1 utilized citrate and malonate.
The analysis of 16 S rDNA gene sequences of strains LM1 and LY1 revealed that they exhibited the highest similarity to the type strains of Acinetobacter and Pseudomonas, respectively (Fig. 2a,b). This analysis, along with the morphological and biochemical characteristics of the isolates, suggested that the bacterium LM1 belonged to the genus Acinetobacter and is most closely related to Acinetobacter calcoaceticus strain M10. Strain LY1 belonged to the genus Pseudomonas and is most closely related to Pseudomonas putida strain LB22. Isabelle et al. isolated one strain of Pseudomonas that could convert E2 to estrone from activated sludge bioreactor treating swine waste 20 . Degradation kinetics of E2 by bacterial co-culture. The degradation of E2 by bacterial co-culture at concentrations ranging from 1 to 20 ml L −1 was investigated. E2 degradation by bacterial combination was fitted by first-order kinetic equation (Fig. 3) as follows: − Where C 0 is initial concentration of E2 (mg L −1 ), C is E2 concentration at t time (mg L −1 ), t is time (d), K is the first order rate constant. The half-life of E2 biodegradation by bacterial combination was calculated according to the equation: The kinetic equations and the relative parameters for the biodegradation of E2 at different initial concentrations are listed in Table 1. The calculated results showed that E2 biodegradation reaction fitted well with first-order kinetics. When the initial concentration of E2 was increased from 1 to 20 mg L −1 , the first order rate constant and t 1/2 changed from 1.004 to 1.136 d and 1.298 to 1.947 d, respectively. The results showed that most E2 at the initial concentration of less than 10 mg L −1 was quickly degraded within 3 days of incubation. After 7 days of incubation, the degradation extent was 99%, 98% and 95% when the initial concentration of E2 was 1, 5 and 10 mg L −1 , respectively. More time was needed for bacterial co-culture to accommodate, grow and degrade E2 when E2 concentration was increased to 20 mg L −1 . However, the bacterial co-culture still degraded 91% of E2 after 7 days of incubation. These results show that E2 would be quickly removed by bacterial co-culture in application because   the concentration of estrogens in livestock and poultry wastes is very low 5,7 . The ability of bacterial co-culture to degrade E2 was not the strongest among the E2-degrading bacteria isolated from different sources 3,11,12 . Nevertheless, bacterial co-culture has significant practical value because the strains LM1 and LY1 were indigenous bacteria and their synergistic action may improve their survivability in actual microenvironment.
Effect of nitrogen source and C/N ratio on bacterial growth and E2 degradation. Two types of nitrogen sources (NH 4 + and NO 3 − ) were used to test the effect of nitrogen source on bacterial growth and E2 degradation. As shown in Fig. 4(a,b), the bacterial co-culture could use two kinds of nitrogen (NH 4 + and NO 3 − ) to grow and metabolize E2. However, at the same C/N ratio, the bacterial co-culture grew more efficiently when NO 3 − was used as the sole nitrogen source instead of NH 4 + . The reason may be different metabolic and growth pathways when the form of N was changed, for example the nitrification or denitrification. For example, the degradation extent of E2 was 98% and 72% when NO 3 − or NH 4 + was used as a sole nitrogen source at a C/N ratio of 1:35. It was found that most E2 degrading bacteria were isolated through a medium culture in which NH 4 + was used as the sole nitrogen source 12,13,26 . However, Yu et al. isolated 14 strains of E2 degrading bacteria from the activated sludge in a waste water treatment plant through a nitrate salt containing culture medium 28 . Kurisu et al. isolated two strains of E2-degrading bacteria from soil samples through an enrichment culture in a medium containing NH 4 Cl and NaNO 3 4 . So, nitrogen form used in the mineral salt medium maybe significantly affect bacterial growth and E2 degradation and further research is needed to investigate the underlying mechanisms. Generally, NH 4 + and NO 3 − simultaneously exist in poultry and livestock wastes and polluted environments. Therefore, bacteria that can utilize either NO 3 − or NH 4 + would have important practical application value. From Fig. 4b, it can be seen that degradation extent of E2 by bacterial co-culture decreased with increasing C/N ratio. For example, the degradation extent of E2 decreased from 98% to 75% when the C/N ratio increased from 1:35 to 1:10 in a culture medium with NO 3 − as the sole nitrogen source. The above results show that high nitrogen content in culture mediums promotes bacterial growth and consequently contributes to E2 degradation. So, the C/N ratio may be an important factor that affects the effectiveness of E2 degradation by bacteria because the C/N ratio of most of the wastes and environmental samples is greater than 5:1 30 .
Effect of pH on E2 degradation. The effects of the culture medium pH on bacterial growth and E2 degradation are shown in Fig. 5. The E2 degradation extent increased when pH increased from 5.15 to 7.06. The almost quantivalent E2 degradation extent was for pH 7.06, 8.12 and 9.14. Degradation of E2 by bacterial co-culture was almost inhibited in the relatively acidic (pH 4.02 and 5.15) condition. However, 98% of E2 was degraded even at  pH 9.14. As a whole, the pH at which bacterial co-culture satisfactorily degraded E2 ranged from 7.00 to 9.00, which is meaningful for their application in the bioremediation of estrogen-polluted environments.
Effect of coexisting carbon source on E2 degradation. As shown in Fig. 6(a,b), adding other carbon sources (sodium acetate, glucose and, sodium citrate) to the acetone-free MS medium with 5 mg L −1 of E2 promoted the growth of bacterial co-culture, thereby showing that bacterial co-culture can grow on the three carbon sources. E2 was degraded by 63.4% when E2 was used as the sole carbon source in the first 3 days of incubation, whereas the E2 degradation extent was only approximately 37.2%, 38.7% and 38.9%, respectively, when sodium acetate, glucose, sodium citrate were added to the culture medium for the first 3 days of incubation. However, E2 was rapidly degraded from 3 to 4 days in the carbon source coexistence cultural medium. Consequently, the E2 degradation extent was slightly higher in carbon coexistence culture medium than in the sole carbon culture medium after 4 days of incubation. These data showed that the addition of other carbon source affected the E2 degradation by bacterial co-culture, but the bacterial growth was promoted and thus increased the degradation extent of E2 after the other carbon source was depleted.
Effect of heavy metals on bacterial growth. Heavy metal Cu is frequently detected because it can promote animal growth and disease resistance and has been used as feed additives in China. Cd, Pb and Ni are usually found in environment due to the release of pollutants and the application of phosphate fertilizer 31 . Thus, the effect of Cu, Ni, Pb and Cd on bacterial growth was investigated (Fig. 7). The growth of the bacterial co-culture at the initial stage (after 3 days of incubation) was adversely affected when the concentrations of Ni, Pb, Cd and Cu were all 1.6 mg L −1 , respectively. The growth of the bacterial co-culture was not significantly decreased by Ni, Pb, Cd or Cu at or below 0.8 12, 1.6 or 0.8 mg L −1 . Notably, the bacterial co-culture was more resistant to Pb than other heavy metals. These results also show that application of a bacterial co-culture would not be affected by heavy metals in the environment because the available ions of Cu, Ni, Pb and Cd were lower than the highest concentration used in the experiments.

Conclusion
Two Gram-negative strains (LM1 and LY1) isolated from livestock and poultry wastes were identified as 17β-estradiol-degrading bacteria. An analysis of 16 S rDNA gene sequences identified them as belonging to the genus Acinetobacter (LM1) and Pseudomonas (LY1). The bacterial co-culture containing LM1 and LY1 bacterial strains could rapidly remove approximately 98% of E2 when 5 mg L-1 of E2 was used as a sole carbon source within 7 days of incubation. However, E2 was degraded by 77% and 68% when strains LM1 and LY1 were incubated alone, respectively. More than 90% of E2 could be removed by bacterial co-culture when E2 concentration was equal or less than 20 mg L −1 . The bacterial co-culture utilized two types of nitrogen sources (nitrate > ammonium) to support the growth and degrade E2. Low C/N ratio was more suitable for bacterial growth and E2 degradation. The optimal pH level for bacterial co-culture to degrade E2 ranged from 7 to 9. The coexisting sodium acetate, glucose, or sodium affected E2 degradation by bacterial co-culture in the first 4 days, but more E2 was removed by bacterial co-culture when they were depleted. The growth of the bacterial co-culture was not significantly decreased by Ni, Pb, Cd or Cu at or below 0.8, 12, 1.6 or 0.8 mg L −1 . In view of the results here obtained, the two strains possess potential for their application in the biodegradation of animal wastes. To reveal the mechanism of degradation, the metabolites produced by bacterial co-culture should be identified, and the enzymes involved in the estrogen biodegradation, such as dehydrogenase and carbonyl enzymes, should be further studied in detail.

Methods
Chemicals. E2 (>99% pure), purchased from Sigma-Aldrich and dissolved at a concentration of 1000 mg L −1 in acetone, were used as stock solutions. The organic solvents, methanol, ethyl acetate, acetonitrile, and HPLCgrade acetone were also obtained from Sigma-Aldrich. All other chemicals were of the highest commercially available purity. Enrichment and isolation of 17β-estradiol-degrading bacteria. Waste samples (including poultry, livestock wastes, manure compost, and manure-applied soil collected from Jilin Province, Northeast China) were used as inoculum source for obtaining the 17β-estradiol-degrading bacteria. To ensure the diversity of bacteria in the inocula, a composite sample was prepared by mixing 40 individual samples (1:1, mass ratio). Bacterial suspension was prepared by adding 5 g of the composite sample into a 100-ml flask containing 50 ml of sterilized water. The suspension was stirred by hand for 30 s and was left to stand for 20 min thereafter. A 10-ml aliquot of the resulting bacterial suspension was added to 100-ml acetone-free mineral salts (MS) medium with 5 mg L −1 of E2 as the sole carbon source. The suspension was then incubated at 150 rpm and 25 °C in the dark until the cultures became turbid and the E2 concentration significantly decreased (after 5 days of incubation). Then 1 ml of the enrichment culture was transferred into 100 ml of fresh acetone-free MS medium with 5 mg L −1 of E2 and re-incubated under the same condition for 5 days. This procedure was repeated five times. The acetone-free MS medium was prepared as follows: known amounts of estrogen stock solution were added to the MS medium, the medium was heated to 80 °C, and the medium was then purged with air at a flow rate of 120 ml min −1 for 30 min to remove the remaining acetone from the medium. After cooling to room temperature, the acetone-free MS medium was used for degradation experiments 28 . The MS medium comprised (per liter of distilled water) 1.0 g of K 2 HPO 4 , 0.6 g of NaH 2 PO 4 ·2H 2 O, 0.2 g of MgSO 4 ·7H 2 O, 0.2 g of KCl, 1 g of NaNO 3 , and 1 ml of trace element solution, as described by Park et al. 25 .
After 50 days of enrichment incubation, the last enrichment culture was used as a bacterial source for the isolation of 17β-estradiol-degrading bacteria. One milliliter of the culture was serially diluted and plated on R2A agar plates containing 5 mg L −1 of E2. The plates were incubated at 25 °C in the dark for 24-48 h. The morphologically distinct colonies were selected and purified by streaking on R2A agar plates. The 17β-estradiol-degrading experiment was performed by incubating strain LM1, LY1 or their co-culture in acetone-free MS medium with 5 mg L −1 of E2 as a sole carbon source. Finally, the bacterial co-culture consisting of strains LM1 and LY1 (arbitrarily named), which showed a superior degradation towards to E2, were used for further study.
Characterization of strains LY1 and LM1. Cells of strain LM1 or LY1 were suspended in sterilized water and boiled at 100 °C for 5 min. One milliliter of the above lysate as a template was used to amplify the 16 S rDNA gene sequences of strain LM1 or LY1 via polymerase chain reaction (PCR). The universal eubacterial primers Pf similarity search of the nucleotide sequence retrieved sequences from species within the genus Acinetobacter and Pseudomonas. A multiple alignment of all the sequences included in the analysis was produced by ClustalX 2.1. The region of this alignment, in which all the sequences overlap (with intermittent gaps and uncertain bases omitted) was used to construct a phylogenetic tree using the neighbor-joining method in MEGA 6.0 32 .
In addition, morphological appearance, Gram-staining results, and other physiological and biochemical characteristics were used to identify the isolated bacteria. The two strains were deposited in the China General Microbiological Culture Collection Center in May 18, 2015. The CGMCC numbers of strains LM1 and LY1 are 10814 and 10815, respectively.
Preparation of bacterial co-culture cell suspension. A single colony of strain LM1 or LY1 was grown in 10 ml of acetone-free MS medium containing 5 mg L −1 of E2 at 150 rpm and 25 °C. After 5 days of cultivation, the cultures were wholly transferred to 100 ml of the above fresh medium. They were then incubated under the same conditions for another 5 days. Then the culture was centrifuged at 8000 rpm for 5 min to collect the bacterial cells. The harvested cells of strains LM1 and LY1 were washed three times with MS medium and finally suspended in fresh MS medium at an OD 600 of 0.1. To obtain bacterial co-culture cell suspensions, the two cell suspensions were mixed at a volume ratio of 1:1 because the results from the viable cell count experiment showed that the cell quantity ratio of strain LM1 to LY1 in the cultures was approximately 1:1.
Degradation of E2 at different concentrations by bacterial co-culture. In this experiment, 250-ml Erlenmeyer flasks containing acetone-free MS medium with different initial concentrations of E2 and 2% of cell suspension were incubated in a rotary shaker (150 rpm) at 25 °C in the dark. The initial concentrations of E2 were set as 1, 5, 10 and 20 mg L −1 . At designated intervals, culture samples were collected for the determination of OD 600 and E2 residue.

Effect of factors on bacterial growth and E2 degradation.
The experiment on the effect of nitrogen source on bacterial growth and 17β-estradiol degradation was performed by using 1.0 g L −1 of NaNO 3 or 0.63 g L −1 of NH 4 Cl as a sole nitrogen source in the acetone-free MS medium. To investigate the effect of molar concentration ratio of carbon to nitrogen (C/N ratio) on bacterial growth and E2 degradation, the C/N ratios were set as 1:35, 1:20 and 1:10 by changing the amount of nitrogen source in the culture medium with 5 mg L −1 of E2, and we adjusted the pH by adding solution of HCl or NaOH to be 7.0. Regarding the pH effect, MS media with six different pH values (4.02, 5.15, 6.01, 7.06, 8.12 and 9.14) adjusted with 0.1 mol L −1 HCl and 0.1 mol L −1 NaOH were used. Experiments on the effect of carbon coexistence on bacterial growth and 17β-estradiol degradation were conducted by adding 22.5 mg L −1 of sodium acetate trihydrate, 10.9 mg L −1 of glucose and 14.2 mg L −1 of sodium citrate to the acetone-free MS medium with 5 mg L −1 of E2. The effect of heavy metals on bacterial growth were evaluated by adding Ni (0, 0.2, 0.4, 0.8, 1.6 and 2 mg L −1 ), Pb (0, 2, 4, 8, 12 and16 mg L −1 ), Cd (0, 0.2, 0.4, 0.8, 1.6 and 2 mg L −1 ), and Cu (0, 0.2, 0.4, 0.8 and 1.6 mg L −1 ) to the MS medium in form of metal nitrate. After 5 days of incubation, samples were collected to measure OD 600 .
The above E2 degradation experiments were conducted in a series of 250-ml flasks containing 100 ml of the specific cultural medium, 2% of cell suspension, and 5 mg L −1 of E2. The flasks were incubated in a shaker (150 rpm) at 25 °C in the dark. As a control without inoculation, 50 ml acetone-free MS medium with 5 mg L −1 of E2 was submitted to the same experimental conditions. All the above experiments were conducted in triplicate, and the data were expressed as means ± standard deviations. Estrogen analysis. Bacterial cells were separated from the cultures by centrifuging them at 10,000 rpm for 10 min at 4 °C. Estrogen in the supernatant was extracted with ethyl acetate and concentrated by evaporating ethyl acetate and re-dissolving the residual precipitates in methanol. The final solution was filtered through a 0.22-µm membrane and analyzed using a high-performance liquid chromatography (Agilent 1260) system equipped with a UV detector (λ = 205 nm), a 3-cm C-18 guard column, and a Lua end-capped (C18) reversed-phased column (4.6 × 150 mm, 5 µm particles; Phenomenex, Torrance, CA). 50-µl injection volumes were used; the mobile phase comprised acetonitrile and water at a volume ratio of 45:55 and had a flow rate of 1.0 ml min −1 14,33 . The column temperature was adjusted to 25 °C. Retention time for E2 was 7.28-7.45 min. External calibration curves were generated to estimate sample concentrations from peak areas and were linear between 10 µg L −1 and 10 mg L −1 . In addition, an internal standard was set to correct for losses during processing.