Interaction of the concentration of S2–, SO4 2– and microbial community by controlling nitrogen stripping during sulfate-rich wastewater treatment

ABSTRACT Nitrogen stripping plays a vital role during the sulfate-rich wastewater treatment process by sulfate-reducing bacteria (SRB). However, the interaction between nitrogen stripping, sulfate reduction capacity, S2 – concentration, and microbial community structure is rarely studied. Here, two anaerobic systems were conducted to analyze these interactions simultaneously. Results showed that the toxic products (H2S) were stripped by nitrogen at first. Then, the increased relative abundance of fermentative bacteria accelerated the carbon sources conversion, and Desulfovibrio, Dethiosulfovibrio, and Sulfurospirillum facilitated the sulfate reduction after nitrogen stripping. Finally, the produced S2− concentration was increased by 12%, and the sulfate reduction rate was increased by 9.2% compared to that without nitrogen stripping.


Introduction
The annual production of mine wastewater in China is more than 1.1 × 10 10 m 3 , and the resource utilization rate is only 20%, with a large amount of mine wastewater is directly discharged [1,2]. Acid mine drainage (AMD) is continuously produced from abandoned mines and contains many high-concentration sulfate ions (SO 4 2-) and various heavy metals with low pH. It brings severe threats to mine equipment, the ecological environment, and human health [3]. There is an urgent need for an economical, practical, and resourcerecyclable method to protect the sustainability of ecosystems and reduce resource waste. There are many methods for treating AMD, among which biological technology by SRB has attracted much attention due to their low cost and environmental friendliness [4,5]. S 2plays an essential role in reducing sulfate by SRB [6]. On the one hand, S 2transformed by excessive sulfate can bind with metals to precipitate as metal sulfides under the metabolism of SRB [7]. On the other hand, S 2can be converted to elemental sulfur by sulfur-oxidizing bacteria. Unfortunately, current research predominantly focuses on growth characteristics, utilization of different substrates, and sulfate reduction ability of SRB [8][9][10][11]. However, little research has been done on S 2under the metabolic reactions of SRB. Based on previous studies, only a tiny S 2reacts with heavy metals, and a large amount of sulfur is not analyzed [12]. A study by Vasquez et al. reported that 833 mg/L of SO 4 2--S in the influent and a total of 473 mg L -1 of S 2in the effluent (including sulfate and generated metal sulfides), with approximately 43% of sulfur are not described in the effluent [13]. Therefore, it is necessary to pay attention to S 2when SRB treats sulfate-rich wastewater. The formation of S 2is affected by many factors. On the one hand, diverse microorganisms have different metabolic profiles because of their inherently different functions. As well, there is a competition between SRB and other bacteria. Additionally, the coexisting hydrolytic bacteria and fermentative bacteria may affect the activity of SRB. The above reason may affect the activity of SRB, which has the most considerable influence on S 2production [14]. Therefore, it is essential to understand the microbial community successions during the sulfate-rich wastewater reduction process. On the other hand, several types of sulfur-containing products (H 2 S, HS -, and S 2-) are found during the SRB metabolism process [15], and the undissociated H 2 S can penetrate the cell wall and inhibit the activity of SRB, causing the change in S 2production [16]. If toxic products (such as H 2 S) can be removed, the activity of SRB will not be inhibited, and thus the sulfate reduction reaction will be accelerated. Yao compared the effect of nitrogen stripping on sulfide production when using SRB to treat AMD and found that the sulfide content in the nitrogen stripping group is 2.09 times higher than that in the group without nitrogen stripping [17]. Thusly, nitrogen stripping is a feasible method for removing toxic products in the sulfate-rich wastewater treatment by SRB. Furthermore, the existence of toxic products can influence the microbial community composition. As reported earlier, the added H 2 S changed the microbial growth environment, which might cause the relative abundance of Hydrogenophaga sp, Clostridium populeti, Bacteroides sp, Pseudomonas sp, and Dysgonomonas wimpennyi to change [18]. Here, the formation of S 2is affected by the microbial community and nitrogen stripping. Meanwhile, the succession of microbial communities is affected by nitrogen stripping in the sulfate-rich wastewater treatment by SRB. However, it is still unclear about the relationship of microbial community succession, nitrogen stripping, and S 2formation during the sulfate-rich wastewater treatment by SRB.
Therefore, this study conducted two anaerobic systems to simultaneously analyze the interaction of microbial community succession, nitrogen stripping, S 2-, and sulfate reduction ability. The SRB inoculum was enriched, and the sulfate reduction ability, the concentration of S 2by controlling nitrogen stripping during sulfate-rich wastewater treatment by SRB were considered. Additionally, the response to microbial communities was also investigated.

SRB inoculum
The microorganisms were derived from anaerobic sludge at the Nibu Bay Wastewater Treatment Plant (Qingdao, Shandong Province, China). Postgate'S medium C was used for the culture of SRB [19]. This medium was composed of 0.65 g L -1 KH 2 PO 4 , 0.06 g L -1 MgCl 2 · 6H 2 O, 1.0 g L -1 Na 2 SO 4 , 0.04 g L -1 CaCl 2 · 2H 2 O, 1.0 g L -1 NH 4 Cl, 1.0 g L -1 yeast extract, and 2.0 mL L -1 sodium lactate. The pH of the media was adjusted to 7 and then sterilized at 121°C for 20 min. After cooling, 1.2 g L -1 of Fe(NH 4 ) 2 ·(SO 4 ) 2 · 6H 2 O and 0.5 g L -1 of L-cysteine sterilized by ultraviolet (UV) radiation for 30 min were added, and the samples were cultured at 28°C and 70 rpm min -1 in a shaker. After two days, when the solution turned ink-colored and the bottle smelled like rotten eggs (detected by using lead acetate test paper), 10% (v/v) of the above culture solution was transferred to fresh Postgate'S medium C, and the same culture conditions were repeated three times to enrich the SRB. Efficient SRBs were obtained and identified by Shanghai Majorbio Bio-pharm Technology Co., Ltd.
Subsequently, the morphology and the growth process of the SRB inoculum were evaluated. The pH and redox potential (ORP) during growth were tested together to investigate the growth characteristics of the SRB inoculum. SRB inoculum (10% [v/v]) were added to the fresh Postgate'S medium C (without the addition of Fe(NH 4 ) 2 ·(SO 4 ) 2 · 6H 2 O, the medium that was not inoculated was used as a blank control, and the OD 600 , pH, and ORP in the medium were measured every two hours. All experiments were performed three times.

Design, set up, and operation of anaerobic system SRB reaction device
Considering the influence of initial sulfate concentration and nitrogen stripping on sulfate reduction by SRB, two set anaerobic systems of experimental devices were designed ( Figure 1). Device (a) consisted of bottle A only, filled with sulfate-rich wastewater and SRB. Device (b) was composed of bottles A, B, and nitrogen bottles. Bottle B is filled with CuCl 2 solution, absorbing H 2 S by nitrogen stripped. The practical volumes of bottles A and B were 1 L, and rubber stoppers and polytetrafluoroethylene tapes sealed all the entrances of the bottles to ensure an anaerobic environment.
In each group of experiments, the initial sulfate concentrations of wastewater were 600 and 1000 mg L -1 , and 30% (v/v) SRB inoculum was added to sulfate-rich wastewater (Stripe nitrogen for 10 min before use to remove oxygen). The concentration of copper ions in bottle B used for absorption was 500 mg L -1 . The sulfate concentration, S 2concentration, pH, and ORP during the reaction were measured at 0, 3, 6, 9, 12, and 15 h. All experiments were conducted and analyzed three times.

Analytical procedures
The morphology of the SRB inoculum was tested through scanning electron microscopy (FEI Quanta 250 FEG), and sample processing steps were based on Shi's study [20]. A pH meter (PHS-3C) and automatic potentiometric titrator (ZD-2) measured pH and ORP. Sulfate concentration was tested following the 'Water Quality-Determination of sulfate-barium chromate spectrophotometry' (HJ/T 342-2007). The S 2concentration was determined using a sulfur ion concentration meter (Banter 931-S). The concentration of Cu 2+ remaining in bottle B was measured to account for the H 2 S produced in bottle A using a flame atomic absorption spectrometer (TAS-986 F).

Microbial community analysis
The samples (0 h, 15 h-without N 2 , and 15 h-with N 2 ) at 600 mg L -1 initial sulfate concentration were filtered by a 0.22 µm aqueous filter membrane were collected first for high-throughput sequencing analysis. DNA extraction and polymerase chain reaction (PCR) amplification were performed. According to the manufacturer's instructions, the samples were using a FastDNA® Spin Kit for Soil (MP Biomedicals, U.S.). The DNA concentration and purity were determined with a NanoDrop 2000 UV-vis spectrophotometer, and DNA extract was checked on 1% agarose gel. PCR amplification was run by two primers of 338 F (ACTCCTACGGGAGGCAGCAG) and 806 R (GGACTACHVGGGTWTCTAAT) for each DNA sample, and the region was V3-V4 hypervariable region. PCR reactions were performed in triplicate. According to the manufacturer's instructions, the PCR product was extracted from 2% agarose gel, purified using an AxyPrep DNA Gel Extraction Kit (Axygen Biosciences, USA), and quantified using a Quantus™ Fluorometer (Promega, USA). The third step was Illumina MiSeq sequencing. Purified amplicons were pooled in equimolar and paired-end sequenced on an Illumina MiSeq PE300 platform (Illumina, USA) according to the standard protocols by Majorbio Bio-Pharm Technology Co. Ltd. (Shanghai, China). The final step was the processing of sequencing data. Operational taxonomic units (OTUs) with 97% similarity cutoff were clustered, and chimeric sequences were identified and removed.

Characteristics of SRB inoculum
In order to further analyze the succession of microbial community structure during the sulfate reduction process by SRB, we first studied the characteristics of SRB inoculum. The morphology of microorganisms is shown in Figure 2(a). The SRB was arc and rodshaped, and their length was approximately 0.5 to 2 μm. As shown in Figures 2(b-c), the microorganisms were constituted by Escherichia-Shigella, Desulfovibrio, Citrobacter, Rhodobacteraceae, Dethiosulfovibrio, and others at the genus level. Since the first discovery of SRB by Beijerinck in 1895, more than 40 genera of SRB, such as Desulfovibrio, Desulfotomaculum, Desulfonema, Desulfomonas, Desulfobacter, and Desulfococcus, have been isolated successively [21].

Effect of different conditions on sulfate reduction ability by microorganisms
Different initial sulfate concentrations affect microorganisms' growth and reduction ability due to the different osmotic pressures. As observed in Figure 3, the sulfate reduction rate was 40.82% when the initial sulfate concentration was 600 mg L -1 , and it was 32.40% when the initial sulfate concentration was 1000 mg L -1 . Regardless of the initial sulfate concentration, the pH during the reactions was maintained in the range of 7.02-7.11, and the ORP values decreased more obviously at the initial sulfate concentration of 600 mg L -1 than that of the initial sulfate concentration was 1000 mg L -1 . With nitrogen stripping, the sulfate reduction rate was improved by 9.25% compared with that without nitrogen stripping at 600 mg L -1 initial sulfate concentration, and there was an approximately 7.48% increase at 1000 mg L -1 initial sulfate concentration. These demonstrated that nitrogen stripping is effective for sulfate-rich wastewater treatment by SRB. In addition, a visible elevation of alkalinity was observed in the SRB treatment of sulfate-rich wastewater after nitrogen stripping.

Effect of different conditions on S 2production during sulfate reduction process by microorganisms
S 2as the target product of SRB metabolism and can form stable metal sulfide precipitation with heavy metal ions in wastewater, the S 2concentration with different nitrogen stripping conditions was studied, and the results were shown in Figure 4. The S 2concentration increased from 97 mg L -1 to 103 mg L -1 with 600 mg L -1 initial sulfate concentration, and it grew from 98 mg L -1 to 106 mg L -1 with 1000 mg L -1 initial sulfate concentration. After nitrogen stripping, the concentration of S 2increased from 97 mg L -1 to 115 mg L -1 and 120 mg L -1 when the initial sulfate concentration increased from 600 mg L -1 to 1000 mg L -1 , which was approximately 12% higher than the final S 2concentrations in the system without nitrogen stripping. After 15 h bioreaction, the concentration of H 2 S (as sulfur) blown off was 85.14 and 109.99 mg L -1 with an initial sulfate concentration of 600 and 1000 mg L -1 , respectively.

Alpha diversity analysis of different samples
16S rRNA genes were amplified and sequenced through high-throughput sequencing, and all OTUs were clustered at a setting cut off 97% similarity. A total of 32,048 sequences per sample were collected after subsampling to determine microbial diversity. Community richness and diversity in each sample are estimated, and the results are shown in Figure 5. As shown in Figures 5(a-c), the microbial community richness has a slight downward trend after nitrogen stripping. As for microbial community diversity, an increasing tendency was found under the sulfate-rich wastewater treatment process of SRB ( Figures 5(d-f)). Moreover, the results indicated that nitrogen stripping negatively influenced microbial community diversity.

Microbial community composition analysis of different samples
The distribution of microorganisms was examined to reveal the functional microbial community among the three samples. The phylum with the highest representation was Proteobacteria representing 77.86%, 68.91%, and 59.96% among the samples of 0 h, 15 h-without N 2 , and 15 h-with N 2 , respectively, followed by Desulfobacterota (14.12%, 15.58%, and 16.63%), Bacteroidota, Synergistota, and Firmicutes ( Figure 6(a)).
As shown in Figure 6(c), the bacterial community of samples altered after sulfate reduction (only the top 10 genera). The distribution of some specific functional genera among the three samples is illustrated in Figure 6(d). The principal genera responsible for sulfate reduction were Desulfovibrio, accounting for 14.12%, 15.58%, and 16.63% among the samples of 0 h, 15 h-without N 2 , and 15 h-with N 2 , respectively, which performed an increasing tendency after the bioreaction. At the same time, Dethiosulfovibrio accounted for 4.13%, 2.06%, and 8.70%, respectively.

Discussion
This study comprehensively investigated the sulfate reduction rate, S 2concentration, and microbial community by controlling nitrogen stripping during sulfate-rich wastewater treatment by SRB. In the sulfate reduction process, SRB generated several reduction products, including H 2 S, HS -, and S 2-, especially when H 2 S caused the most toxicity to the growth and performance of SRB [23]. Consequently, nitrogen gas can strip the H 2 S generated during the sulfate reduction by SRB to decrease the toxicity. Other studies suggested that when the concentration of H 2 S reached 40-50 mg L -1 , SRB was inhibited, and SRB activity was irreversibly destroyed when the concentration of H 2 S exceeded the toxicity level for 3-6 h [24]. Different SRB possessed different tolerances to H 2 S, and the toxicity of H 2 S produced during sulfate reduction to SRB may be direct and reversible [25,26].
In this paper, the concentration of H 2 S (as sulfur) blown off by nitrogen was 85.14 mg L -1 after 15 h bioreaction with an initial sulfate concentration of 1000 mg L -1 , respectively. At the same time, the S 2− production concentration was increased from 103 to 120 mg L -1 compared to that without nitrogen stripping. Moreover, the sulfate reduction rate was increased by 9.2% to 50.08%. In other words, toxicity products were blown by nitrogen, and SRB activity was restored, thereby enhancing the sulfate reduction ability.
Furthermore, the H 2 S was blown off by nitrogen stripping, with the evolution of microbial community structure. Specifically, the relative abundance of Desulfovibrio, Macellibacteroides, Dethiosulfovibrio, and Sulfurospirillum increased. Desulfovibrio has an essential role in anaerobic systems [27], and it was a typical hydrogenotrophic SRB, indicating that these SRB can use hydrogen to reduce sulfate [28]. Moreover, Dethiosulfovibrio reduced thiosulfate and elemental sulfur but not sulfate to hydrogen sulfide [29]. Macellibacteroides, a type of fermentative bacteria, produced hydrogen for hydrogenotrophic and enhanced the activity of Desulfovibrio [30]. Sulfurospirillum, a type of sulfite-reducing bacteria [31] and Nogueira et al listed Sulfurospirillum as SRB genera [5]. In addition, the species of Bacteroides has a relatively high abundance in 15 h-with N 2 sample. Bacteroides was a hydrolytic bacterium that could participate in protein degradation and convert amino acids to acetate and had the function of being tolerant to high concentrations of metals and metalloids [32]. It is inferred from these results that the increased fermentative bacteria (Macellibacteroides) accelerated the conversion of carbon sources. The sulfate was transformed to sulfite by Desulfovibrio, and the bacteria of Dethiosulfovibrio and Sulfurospirillum converted sulfite to S 2-. Thus, the produced S 2− concentration increased, and sulfate reduction ability improved.

Conclusions
In this study, the contributions of nitrogen stripping to sulfate reduction by SRB were comprehensively investigated. Stripped H 2 S, favorable S 2production, and considerable sulfate reduction ability implied a vital contribution of nitrogen stripping in the sulfate-rich wastewater. The hydrolytic and fermentative microorganisms coexistent with the SRB, and the increased fermentative bacteria accelerated the conversion of carbon sources. The relative abundance of Desulfovibrio, Dethiosulfovibrio, and Sulfurospirillum increased after nitrogen stripping, thereby facilitating the sulfate reduction. These findings provide advanced incisive insights into the effect of nitrogen stripping on sulfate-rich wastewater treatment by SRB from S 2-, sulfate reduction ability, and microbial community successions.