Synergistic PAHs biodegradation by a mixed bacterial community: Based on a multisubstrate enrichment approach

Polycyclic aromatic hydrocarbons (PAHs) are highly hard-biodegradable compounds. Therefore, in this work, a multisubstrate enrichment approach was proposed to develop a bacterial community named MBF from activated sludge of coking wastewater plant capable of degrading mixed-PAHs consisting of phenanthrene and pyrene (50 mg/L of each) by 98.8% and 73.3% within 5 days, respectively. The bacterial community could maintain its degradation ability to mixed PAHs relatively under temperatures (20°C–35°C), pH values (5.0–9.0), and salinities (0–10 g/L NaCl). Additionally, the bacterial community MBF degraded 58.9%, 79.9%, and 80.7% of mixed PAHs in the presence of catechol, salicylic acid, and phthalic acid, respectively within 5 days. High-throughput sequencing of 16S rRNA gene amplicon analysis showed that the bacterial community MBF was dominated by Pseudomonas in most treatments, and Burkholderia was predominant under both acidic condition and high salt concentrations. Furthermore, the composition of microbial communities of the bacterial community was signicantly different with/without addition of pathway intermediate metabolites after biodegradation of mixed PAHs, revealing the metabolic burden may be distributed between members of this bacterial community. Those results demonstrate that the biodegradation ability of MBF could be maintained with the bacterial community structure altering when facing environmental variations or changes in composition of target contaminants.


Introduction
Polycyclic aromatic hydrocarbons (PAHs) are organic compounds composed of two or more aromatic rings with different structural con guration (Nzila, 2018). As a kind of typical persistent organic pollutants, PAHs are widely present in environmental media such as soil, sediments and water bodies owing to some natural and human activities, including wild res, volcanic explosion, oil re ning and petrochemical industries (Haritash and Kaushik, 2009). PAHs contamination has raised signi cant concerns for environmental and human health due to their high toxicity, reluctance to biodegradation and high bioaccumulation (Blumer et al., 1977;Downward et al., 2014). Due to all the above characteristics, 16 kinds of PAHs have been listed as priority pollutans by the USEPA.
Compared to conventional physicochemical approaches, bioremediation referring to the technique that employs microorganism to eliminate pollutants has been proven environmentally sustainable, ecofriendly and economical for PAHs degradation (Gillespie and Philp, 2013;Kumari et al., 2018;Li et al., 2009;Ma et al., 2021). The contaminated environment by PAHs usually contains multiple PAHs and various intermediates produced during the microbial metabolic process of PAHs. Considering many different interactions (e.g., co-metabolism, inhibition, and cross induction) have been identi ed among PAHs and/or PAHs metabolites, thus it is essential to evaluate the degradation of PAHs in their metabolite mixture for successful remediation (Dean-Ross et al., 2002;Gupta et al., 2015). Furthermore, the overall degradation and removal rate of PAHs are closely related to environmental factors including pH value, temperature, and salinity. These factors may have an impact on microbial growth, activity and pollutant bioavailability during PAH biodegradation (Patel et al., 2018;Yu et al., 2018). Consequently, it is signi cant to develop a microbial community that can exhibit stable and effective degradability for a mixture of PAHs with their metabolites and under various environmental conditions. Several studies have shown that mixed bacterial consortia exhibits more effectively than pure bacterial cultures, because there may be multiple degradation pathways, e cient and synergistic metabolic activities (Kumari et al., 2018;Kuppusamy et al., 2016b;Wanapaisan et al., 2018). The de ned microbial consortia was constructed for bioprocessing using 'top-down' or 'bottom-up' methods. The 'bottom-up' approach starts to mix two or more isolated and characterized strains as the inoculums to degrade pollutants, and the interaction among these strains may be metabolically engineered. However, microbial consortia constructed by this approach lack stability and metabolic diversity due to inherent limitation by knowledge of characterized strains (Iwabuchi et al., 2002). The 'top-down' approach down-selects stable microbial communities sourced from the natural environment to build consortia which possess remarkable stability and resilience (Gilmore and O'Malley, 2016). The controlled initial microbiome is exposed to multiple selection cycles and/or regimes, which can lead to the microbiome acquiring or optimizing speci c functions through adaptation or evolution (Lawson et al., 2019). Several studies reported that an initial microbiome collected from the contaminated environment by PAHs was successively transferred with gradually increase in concentration of target pollutants to obtain an e cient and stable consortium (Lu et al., 2019;Patel et al., 2018Patel et al., , 2012. Despite the high degradability to PAHs, the consortium may lack the ability to e ciently degrade the metabolites produced during the biodegradation of PAHs, and the accumulation of metabolites can inhibit biodegradation process (Meng et al., 2014;Wen et al., 2011). Moreover, it takes a long time to complete the whole enrichment procedure for adaptation of microbiome to target pollutants. To solve these problems, we propose a multisubstrate enrichment approach to obtain a bacterial community with e cient degradation of target pollutants and intermediate metabolites in a short period of time. The opening of target pollutant is the hardest link in the biodegradation process, and it is also a key factor limiting the mineralization effect of pollutants. As the product of forthcoming ring opening, the intermediate metabolites are used as the primary energy and carbon source for the initial microbiota rstly. The intermediate metabolites could induce some enzyme activities (e.g., C 12 O, C 23 O) of microbiota, therefore reducing the adaptation time of microbiota to the provided carbon source and accelerating their growth (Sei et al., 2004). In the later stage of enrichment, the microbiota was directly exposed to the target pollutant to acclimate the microbiota that can effectively degrade the target pollutant. Preliminary laboratory studies evaluating the potential of a PAHdegrading consortium are indispensable for the successful implementation of biodegradation of PAHs as a remediation technology (Bacosa and Inoue, 2015).
In this current study, the mixed bacterial community MBF was developed from activated sludge of coking wastewater treatment plant (CWWTP) of Huayu Gas (Xuzhou, China) through a multisubstrate enrichment approach, in which catechol, salicylic acid and phthalic acid were selected as pathway intermediate metabolites of PAHs, and phenanthrene and pyrene were target contaminant. Furthermore, the mixture of phenanthrene and pyrene degradation by MBF and its microbial community structure under environmental conditions (pH, temperature and salinity) and in the presence of pathway intermediate metabolites (catechol, salicylic acid and phthalic acid) after biodegradation were also investigated.

Sampling and multisubstrate enrichment process
Activated sludge, as the microbial source, was collected from CWWTP of Huayu Gas, located in Xuzhou, Jiangsu Province of China. The wastewater contained a variety of PAHs, thus the microbiota that could degrade PAHs de nitely existed in the activated sludge.
The acclimation process of mixed PAH-degrading bacterial ccommunity was based on a multisubstrate enrichment approach, as illustrated in Fig. 1. The speci c operation was as follows: activated sludge suspension was diluted with saline, and then aerated in a plastic bucket for 7 d at 25°C, and then began the acclimation with addition of pathway intermediate metabolites as primary carbon source. The catechol, salicylic acid and phthalic acid were selected as common pathway intermediate metabolites of PAHs (Patel et al., 2019). The supernatant (5 mL) was picked out to Erlenmeyer ask (250 mL) provided with 45 mL sterilized MSM, 200 mg/L catechol, and 50 mg/L of phenanthrene and pyrene (each 25 mg/L), and incubated for 7 d (25°C, 150 rpm). Successive transfers were performed with gradually increase in concentration (200 mg/L with each transfer after 7 d incubation) of catechol from 200 mg/L to 1000 mg/L. Similar acclimation with addition of salicylic acid or phthalic acid were carried out synchronously. After the acclimation with addition of pathway intermediate metabolites, 5 mL bacterial uid (15 mL in total) were transferred to 45 mL sterilized MSM with 100 mg/L phenanthrene and pyrene (each 50 mg/L) as the solo carbon and energy source. With successive transferring that the concentration of phenanthrene and pyrene increased from 100 mg/L to 400 mg/L (50 mg/L with each transfer after 7 d incubation), a mixed PAH-degrading bacterial community was obtained (designated as 'MBF'). The consortium was stored at 4°C for further experiments.

Effects of initial concentration on PAH degradation
The degradation experiments were carried out in 50 mL shaking ask containing 9 mL MSM and 1 mL inoculum provided with a different initial concentration (25 mg/L, 50 mg/L, 100 mg/L and 200 mg/L) of phenanthrene and pyrene separately at 25°C under shaking condition (150 rpm). For inoculum preparation, pre-grown MBF was centrifuged at 6000 rpm for 6 min at 4°C, washed with fresh MSM and then re-suspended in fresh MSM. Three replicate asks were collected at 1, 3 and 5 d for degradation of phenanthrene and pyrene. Uninoculated asks containing only MSM supplemented with PAH was used as a control. The biodegradation pro les of phenanthrene and pyrene as a single compound and as a mixture of PAHs were also investigated.

Effects of environmental factors on PAHs degradation
To determine the effects of environmental factors on the degradation of 100 mg/L of phenanthrene and pyrene (each 50 mg/L) by MBF, experiments were performed at different temperature (20°C, 25°C, 28°C, 30°C, and 35°C), pH value (5, 6, 7, 8, and 9) and salinity (0 g/L, 1 g/L, 3 g/L, 5 g/L, and 10 g/L NaCl). The inoculum process was similar as described in Section 2.3. The phenanthrene and pyrene degradation were monitored on 1st, 3rd and 5th day of incubation. Another three replicate asks were collected for DNA extraction. Uninoculated asks containing only MSM supplemented with PAH was used as a control.

Effects of pathway intermediate metabolites on PAH degradation
To investigate the effects of pathway intermediate metabolites on the degradation of 100 mg/L of phenanthrene and pyrene (each 50 mg/L) by MBF, experiments were performed with the supplementation of 0.05% (w/v) catechol, phthalic acid or salicylic acid. The inoculum process and incubation conditions were similar as described in Section 2.3. In addition, three replicate asks were collected for DNA extraction. The asks without addition of pathway intermediates were kept under similar conditions and served as a control.
2.6. DNA extraction and bacterial community structure analysis Total genomic DNA was extracted from original activated sludge and the consortium using the Mobio PowerSoil ® DNA isolation kit (Qiagen, Shanghai, China) according to manufacturer's instruction. DNA samples in all treatments were extracted in triplicate and combined for further analysis. The V3-V4 variable region of the 16S rRNA genes was ampli ed using forward primer 338F and reverse primer 806R (Liu et al., 2021). The PCR were performed using the following protocol: 3 min of denaturation at 95°C, 30 s at 95°C, 30 s at 55°C, 45 s at 72°C, and 27 cycles, a nal extension at 72°C for 10 min. The PCR productions were further puri ed using 2% agarose gel and AxyPrep DNA Gel Extraction Kit (Axygen Biosciences, Union City, CA, USA). Puri ed Amplicons were sequenced in a paired end format using the Illumine MiSeq PE300 platform by Majorbio BioPharm Technology Co, Ltd (Shanghai, China).
Raw sequences were demultiplexed and quality-ltered using Quantitative Insights Into Microbial Ecology (QIIME, version 1.9.1) with the following instructions: (1) the reads with average quality score < 20 over a matching, reads containing ambiguous characters were removed; and (3) only sequences that overlapped longer than 10 bp were assembled according to the overlap sequences. After trimming, sequences with at least 97% identi ed were de ned as an operational taxonomic unit (OTU) using UPARSE (Edgar, 2013). The RDP Classi er was used to assign the taxonomy of each OUT against the Silva 16S rRNA gene database (Quast et al., 2012). The raw data of fungal sequences was deposited in the Sequence Read Archive of NCBI database (PRJNA746429).

Determination of PAHs
The entire content of ash (10 mL) was added with 20 mL of methanol using the ultrasonic dissolution promotion to extract the residual phenanthrene and pyrene (Gu et al., 2015). After ltration through 0.22 µm polytetra uoroethylene membrane, the residual phenanthrene and pyrene in the mixed solution were determined in Flexar Quaternary LC Pump Platform (Flexar LC, PerkinElmer, Singapore) tted with a C18 column (Brownlee C18, 5 µm, 150 × 4.6, PerkinElmer) and a UV/Vis detector. Phenanthrene and pyrene were eluted with a mobile phase (methanol: water, 80:20) at a ow rate of 1 mL/min and monitored at 254 nm. The column temperature was 30°C, and the injection volume was 50 µL. The chromatograms were recorded and integrated with system software (Chromera 2.1, PerkinElmer).

Statistical analysis
All experiments were carried out in triplicate, and the values of degradation ratio were performed as the mean ± standard deviation (SD) which was represented in the error bar to show variation within the same experiments. The statistical analyses were carried out on IBM SPSS Statistics 25. The differences between different samples were analyzed using one-way analysis of variance (ANOVA) test followed by Duncan's multiple range test. P-values ≤ 0.05 were considered to be signi cant.

Results And Discussion
3.1. Dynamics of the bacterial community during the multisubstrate enrichment process The bacterial communities in activated sludge and consortium during the multisubstrate enrichment process were characterized using high-throughput sequencing of 16S rRNA gene amplicons, seen in Fig. 2. Species richness and diversity index were obtained (Table S1). It can be concluded from the Shannon and Simpson indices that the bacterial community of each samples varied and the bacterial diversity became lower as the multisubstrate enrichment proceeded. For the complete dataset, the most abundant phylum was Proteobacteria (91.6-98.8%) in all samples followed by Actinobacteria (0.9-5.6%). Various studies have implicated that higher relative abundance of Proteobacteria and Actinobacteria and their positive correlation in PAHs degradation in PAH-contaminated sites (Chen et al., 2017;Lee et al., 2018;Sharma et al., 2016).
In addition, the difference and species richness of top 15 dominant genera in all samples during the multisubstrate enrichment process was re ected using a heat map (Fig. S1). The dominant genus was Pseudomonas in original sample AS, but became Ralstonia, Delftia, and Achromobacter in sample CA, SA, and PA, respectively. Ralstonia (Di Gioia et al., 2002) and Achromobacter (Pradeep et al., 2015) have been reported as showing the ability of degrading catechol and phthalic acid, respectively, and Delftia have not previously been reported with biodegradation of salicylic acid. The contribution of the other genera with increasing percentages associated with PAHs biodegradation are supported by previous studies for Stenotrophomonas (Zang et al., 2020), Pseudacidovorax (Dealtry et al., 2018), Novosphingobium (Fida et al., 2017), Enterobacter (Lors et al., 2012), Comamonas (Qin et al., 2019), Chryseobacterium (Xiao et al., 2019), and Burkholderia (Morya et al., 2020). After the acclimation that phenanthrene and pyrene were added as carbon source, the dominant genus was Pseudomonas again in sample MBF. This may due to that Pseudomonas has been considered to be one of the most frequently isolated genera in PAHs biodegradation (Nie et al., 2010). Figure 3 illustrates phenanthrene and pyrene degradation by MBF when they were provided as individual PAH and as mixed-PAHs under different initial concentration (25 mg/L, 50 mg/L, 100 mg/L, and 200 mg/L). When the initial concentration of phenanthrene and pyrene was 25 mg/L or 50 mg/L, the degradation ratio of phenanthrene was almost 100.0%, both in its presence alone and as a mixture with pyrene. Degradation of pyrene were 99.0% and 73.3%, respectively, when provided in the mixture, as compared to 73.1% and 66.3%, respectively, when supplemented as individual PAH. When the initial concentration was 100 mg/L or 200 mg/L, degradation ratio of individual PAH was rapid as compared to the degradation of individual PAH in the mixture. Reason of these results was that, the cross acclimation and co-metabolism may enhance the biodegradation of PAHs under lower initial concentration of mixed PAHs, and the competitive inhibition and re ection of increased levels of toxicity were revealed under higher initial concentration of mixed PAHs (Couling et al., 2010;Yuan et al., 2002). Besides, phenanthrene was quickly degraded followed by pyrene in above four initial concentrations. It was consistent with previous studies in which the degradation ratios generally increased with decreasing molecular weight (Wu et al., 2013).

Effect of environmental factors on PAHs degradation
The PAHs biodegradation processes are affected by environmental factors such as temperature, pH, and salinity, and these factors may determine the effectiveness of bioremediation (Ghosal et al., 2016). To evaluate the potential degradation capacity of MBF for mixed-PAH, a range of degradation tests were conducted at various temperatures, pH values, and salinities.
Temperature, as an important factor during the bioremediation of PAHs, has an impact on the microbial metabolism as well as solubility of PAHs (Bandowe et al., 2014;Vidonish et al., 2016). Fig. 4a demonstrated that MBF exhibited good biodegradation e ciency between 25°C and 35°C. One-way ANOVA analyses implicated that there had no signi cant difference for phenanthrene degradation under various temperatures except 20°C, and for pyrene degradation except 20°C and 25°C within 5 d. Generally, microbial activities increase with increasing temperature in the appropriate range, due to its enhancement to enzymatic activity and microbial metabolism, which will facilitate the degradation e ciency of PAHs (Liu et al., 2017). On the other hand, with temperature increasing, the dissolved oxygen becomes lower that may reduce the metabolic activity of aerobic microorganisms (Ghosal et al., 2016). Though the enrichment of MBF was carried out at 25°C, the results of this study showed it had good adaptability to temperature changes.
Generally, most heterotrophic bacteria prefer a neutral to alkaline pH for their metabolic activity (Al-Hawash et al., 2018). In this study, no signi cant difference for phenanthrene and pyrene degradation was observed under pH values ranging from 5.0 to 8.0 at day 5, and the degradation rate of phenanthrene still remained > 70% at pH value of 9.0, suggesting the practical implementation of MBF at large scale ( Fig. 4b). In fact, the cell permeability barrier may be disrupted under acidic conditions, resulting in a increase of the substrate concentration in the cytosol (Y. H. . The remarkable decrease in PAH-biodegradation at higher pH value could be attributed to the inhibition to bacterial growth under alkaline environment (Tao et al., 2007). According to Sihag Pathak and Jaroli (2014), the PAH biodegradation e ciency could be maximized under the optimal pH conditions of 6.5-8, and this is because a neutral to nearly alkaline environment is suitable for the carboxylation reaction.
Effect of salinity conditions on PAH-biodegradation was presented in Fig. 4c. After 5 days of incubation, MBF were able to degrade phenanthrene > 90% and pyrene > 60% under NaCl concentrations of 0 g/L to 3 g/L, indicating its tolerance to a degree of salinity conditions. When the NaCl concentration increased to 10 g/L, the degradation rates signi cantly decreased to < 40% and < 20%, corresponding to phenanthrene and pyrene, respectively. These results agreed with previous study in which higher salinity level could causes higher osmotic pressure and lower dissolved oxygen in the medium, thus retarding the enzyme synthesis, bacterial growth, and PAHs biodegradation (Imron and Titah, 2018).

Effect of environmental factors on community structure of the consortium MBF during mixed-PAH degradation
Under different temperatures, the dominated genus of bacterial communities in all treatment after mixed-PAH degradation was Pseudomonas, and no obvious effect of temperature on bacterial communities in consortium MBF was observed (Fig. 5a). Genus Pseudomonas are well known degraders of PAHs and has been widely applied to PAHs remediation (Khan et al., 2014;Rabodonirina et al., 2019;Wang et al., 2020). Ma et al. (2012) isolated Pseudomonas sp. JM2 from active sewage sludge of chemical plant and found that it could degrade > 40% of the mixture of uorene and phenanthrene (50 mg/L) in 4 days at 37°C, and also could up to 24% of uorene and 12% phenanthrene when temperature was 4°C.
Under different incubation pH, the bacterial compositions in all treatments after mixed-PAH degradation were dominated by Pseudomonas, Burkholderia, Chryseobacterium, and Stenotrophomonas (Fig. 5b). Kuppusamy et al. (2016a) isolated two strain Pseudomonas (MTS-1) and Stenotrophomonas (MTS-2) from contaminated site soils and reported that the Pseudomonas (MTS-1) was able to degrade more than 85% of pyrene (150 mg/L) at pH 5-8 after 30 days, and the Stenotrophomonas (MTS-2) could remove more than 90% of pyrene (150 mg/L) at pH 5-7 after 30 days. The proportions of Burkholderia remarkably increased at pH 5, 6 and 9 which may due to its adaptability to a wide range of acidic and alkaline conditions (Morya et al., 2020;Somtrakoon et al., 2008). Report has been implied that Chryseobacterium could degrade all kinds of PAHs such as naphthalene, acenaphthene, phenanthrene (Oberoi et al., 2015).
The bacterial communities in MBF after mixed-PAH biodegradation were affected with addition of NaCl, as shown in Fig. 5c. The proportion of Burkholderia and Novosphingobium signi cantly increased with the concentration of NaCl increasing. There have been reports that 0.5-3% NaCl was found to be optimal for cell growth of some Burkholderia strains and phenanthrene removal Lu et al., 2012). Yuan et al. (2015) analyzed the diversity of bacteria associated with degradation of PAHs in sea water, and the results indicated that bacteria intimately accessorial with Novosphingobium existed in different bacterial consortia.
Additionally, a principal coordinate analysis (PCoA) based on weighted UniFrac dissimilarity was used to compare the bacterial diversity among all samples (Fig. 5d). About 87.2% of bacterial community variance can be explained from the rst two principal components. The initial communities (MBF) and the communities in T treatments, pH treatments (pH = 7 and pH = 6), and salinity treatments (NaCl_5 and NaCl_10) clustered in the left bottom quadrant. In addition, the communities in T treatments showed least variation over temperature. Among pH treatments, the communities in acid group (pH = 5 and pH = 6) were clearly distinguished from the initial communities (MBF) along the rst principal coordinate, while the separation between alkaline group (pH = 8 and pH = 9) and MBF was seen along the second principal coordinate. The NaCl_10 treatment exhibited the most variation in community composition among salinity treatments. The results above demonstrated that MBF showed a response in community composition to environment condition changes. It is worthy noticed that more than 90% of phenanthrene and 65% of pyrene were removal at pH 5, indicating that MBF was able to obtain good biodegradation of PAHs under a wide range of environmental conditions by changing the interaction among members of MBF. Figure 6 depicts the effect of chose pathway intermediate metabolites on phenanthrene and pyrene degradation by MBF. After incubation of 5 d, degradation of phenanthrene and pyrene was decreased from 93.8-64.6%, 91.7%, and 89.6%, and from 72.2-53.5%, 68.1%, and 71.8%, respectively, in presence of CA, SA, and PA. It was worth noting that the initial rate of phenanthrene and pyrene were slowed down signi cantly, and, nevertheless, it became less distinct with 5 d. This may be due to that the pathway intermediate metabolites were preferentially utilized as carbon source because of their simpler structure, resulting in the reduction of phenanthrene and pyrene biodegradation (Huang et al., 2011). In addition, the structural similarity between pathway intermediate metabolites and target contaminant and broad substrate speci city of enzymes lead to a wide range of catabolic activity of PAHs degraders (J. D. . Under multi-component contaminated substrate conditions, some key enzymes induced by easily degradable contaminants would promote the simultaneous metabolism of the refractory contaminants by microorganisms (Gupta et al., 2015).

Effect of pathway intermediate metabolites on mixed-PAH degradation by consortium MBF
3.6. Effect of pathway intermediates on community structure of the consortium MBF after mixed-PAH degradation At genus level, obvious changes in composition of microbial communities were observed between adding and non-adding pathway intermediate metabolites samples and demonstrated in Fig. 7. The relative abundance of Pseudomonas in CK sample increased after PAHs biodegradation, while reduced in SA, PA, and CA sample. Additionally, the composition of microbial communities was signi cantly different among the treatments with pathway intermediate metabolites adding. The dominant bacterial genera in SA sample were Delftia (37.4%), Curvibacter (14.7%), Chryseobacterium (13.6%), and Comamonas (12.7%), respectively. In PA sample, the dominant bacterial genera were Achromobacter (27.2%), Pseudacidovorax (26.6%), Novosphingobium (17.3%), Delftia (12.3%), and Burkholderia (9.1%). In CA sample, Ralstonia (60.1%), Comamonas (14.5%), and Burkholderia (7.3%) were dominant. The cluster analysis was further conducted to better understand the microbial response in PAHs biodegradation with addition of pathway intermediate metabolites. The cluster plot showed that the control sample without adding pathway intermediate metabolites was clustered with MBF. The bacterial communities of the CA and PA samples were more closely linked and were clustered into one group, and the SA sample was not closely linked to either group.
By analyzing the changes in microbial community composition in the multisubstrate enrichment process ( Fig. S1) and the experiment of biodegradation of phenanthrene and pyrene as a mixture in the presence of pathway intermediate metabolites (Fig. 7), we could nd that the dominant bacterial genera were similar. This indicated that the addition of pathway intermediate metabolites changed microbial community composition, which may due to the multisubstrate enrichment approach used in our study. In the enrichment, the percentage of dominant genera (Delftia, Achromobacter, Ralstonia) in the process of pathway intermediate metabolites adding were nally less than 10% after the whole complete enrichment (Fig. S1). However, the proportion of the three dominant genera increased after biodegradation of phenanthrene and pyrene as a mixture in the presence of pathway intermediate metabolites (Fig. 7). These results revealed that the metabolic burden could be distributed between members of MBF. At rst glance, during this multisubstrate enrichment process, the members behaved signi cantly different under stresses of different pathway intermediate metabolites. Besides, different member dominated in differing degrading process. While our study did not provide direct evidence of potential functions for different bacteria taxa, nevertheless it could provide useful information in developing strategies to boost bioremediation performance of PAHs.

Conclusion
In this study, we put forward an approach of multisubstrate enrichment to develop a bacterial community named MBF from activated sludge of coking wastewater plant. This bacterial community exhibited a relatively good capacity to degrade the mixture of phenanthrene and pyrene under temperatures (20°C-35°C), pH values (5.0-9.0), and salinities (0-10 g/L NaCl), indicating its excellent resistibility to different environmental stress. The consortium MBF was dominated by Pseudomonas in all environmental conditions after biodegradation of mixed PAHs. Burkholderia was found to be the major genus under both acidic environment and high salt concentrations. Additionally, phenanthrene and pyrene were degraded along with pathway intermediate metabolites like catechol, salicylic acid, and phthalic acid demonstrating the high competency of MBF. The community structure in MBF obviously changed with the addition of pathway intermediate metabolites, implying the metabolic burden could be distributed between members of MBF. Thus, the study provides valuable information that can lead to designing an effective strategy for bioremediation of PAHs contaminated environments.

Declarations
Availability of data and material Original data are available from the corresponding author on reasonable request.

Funding
This research was nancially supported by the National Natural Science Foundation of China (No. 51974314, No. 51778612

Contributions
All the authors contributed to the study conception and design. Liping Wang was responsible for this conceptualization. The experimental equipment and scheme were designed by Li et al. The experiment    Biodegradation of phenanthrene and pyrene as a mixture by consortium MBF after incubation of 1, 3, and 5 days under different temperatures (a), pH values (b), and salinities (c). The initial concentration of phenanthrene and pyrene was 50 mg/L. The different letters represented signi cant differences at Pvalues ≤ 0.05 based on one-way ANOVA analyses (Regular letter and Italic letter correspond to phenanthrene and pyrene biodegradation, respectively).