The engineering of TBBPA-degrading synthetic microbiomes with integrated strategies

Abstract

The capability to understand and construct synthetic microbiomes is crucial in biotechnological innovation and application. Tetrabromobisphenol A (TBBPA) is an emerging pollutant, and the understanding of its biodegradation is very limited. Here, a top-down approach was applied for the enrichment of TBBPA-degrading microbiomes from natural microbiomes. Ten keystone taxa correlated to TBBPA degradation and their co-occurrence interactions were identified by the dissection of the degrading microbiomes. Those keystone taxa were targeted and cultivated, and the genomic information was obtained by genome sequencing of strains and metagenomic binning. The keystone bacterial strains showed efficient degradation of TBBPA, and L-amino acids were important co-metabolic substrates to promote the degradation. Guided by this knowledge, a bottom-up approach was applied to design and construct a simplified synthetic consortium SynCon2, that consisted of four strains. The SynCon2 demonstrated efficient TBBPA degradation activity and soil bioremediation. Our study demonstrates the importance of the application of multiple tools in understanding the functions of microbiomes and provides an integrated top-down and bottom-up strategy for the construction of synthetic microbiomes with various applications.

Introduction

Microbiomes are key components across diverse ecosystems and drive global biogeochemical cycles^1,2. Pioneering studies have exploited microbiomes to synthesize valuable products³, protect plant hosts^4,5, remove pollutants^6,7, and control human diseases^8,9. Natural microbiomes are often complex and difficult to apply in practice, and their functions are limited by natural evolution. Design and synthesis of microbiomes hold promise for addressing challenges in agriculture, human health, and environmental sustainability^10,11. Recently, synthetic and engineered microbiomes based on top-down or bottom-up approaches have provided novel solutions to address these challenges^11,12. The top-down approach starts with a natural seed microbiome and drives the seed microbiome to evolve for highly optimized functions^13,14. This approach is easy to implement, but it is less controllable and difficult to scale-up. In contrast, the bottom-up approach starts with a number of defined bacterial strains, of which physiological features have been well characterized^15,16. This approach is limited by the shortage of cultivated microbial strain resources and the empirical selection of microbial strains. The top-down and bottom-up approaches, respectively, have been applied in the construction of functional microbiomes^{17, 18–19}. It is conceivable that the integration of the top-down and bottom-up approaches will overcome their limitations: The naturally evolved microbiome with the top-down approach could be used for understanding the interactions and guiding bacterial cultivation for strain resources, and the bottom-up approach could apply this knowledge and strains for rationale design and construct synthetic microbiomes. Currently, this integrated engineering strategy is still limited to microbiome construction. Organic pollutant degradation relies on the collective activities of diverse microbes^20,21. Recreating microbial cooperation and robust activities in a synthetic microbiome is a promising revenue for the bioremediation of refractory pollutants²². Synthetic microbiomes have been obtained for pollutants with known metabolic interactions and cultivated bacterial strains^23,24, but not for emerging pollutants such as tetrabrombisphenol A (TBBPA), of which the knowledge of biodegradation and bacterial resources is very limited.

TBBPA, the most common brominated flame retardant, has been widely used in plastics, furniture, textiles, and electronic equipment in recent years²⁵. TBBPA persists in environments, threatening natural ecosystems and human health^26,27. TBBPA is reductively debrominated by organohalide-respiring bacteria under anaerobic conditions and eventually produces another organic pollutant, bisphenol A (BPA)^28,29. In recent studies, aerobic settings have been more conducive to the biodegradation of TBBPA³⁰. Bacterial strains that degraded TBBPA under aerobic conditions were obtained in the presence of additional carbon sources^{31, 32, 33, 34–35}, and microbial interactions among the TBBPA-degrading microorganisms remain unknown. Moreover, the establishment of complex functions by combining multiple strains will help overcome the limitations of the metabolic capacity of a single strain and share the unwanted metabolic burden³⁶. Biodegradation of contaminants with complex metabolic processes can be carried out more efficiently compared to a single strain³⁷. The construction of synthetic microbiomes that degrade TBBPA and downstream metabolites is an essential part and challenging task for bioremediation³⁸.

Here, an integrated strategy to construct synthetic consortia for TBBPA degradation is reported for the first time. Firstly, a top-down approach was applied to obtain TBBPA-degrading microbiomes (naturally evolved, TBBPA-degrading microbiomes). Secondly, the microbiomes were dissected with multiple tools to understand keystone taxa and bacterial interactions. Thirdly, bacterial strains representing keystone taxa were cultivated and characterized. Finally, guided by the emerging knowledge, bacterial strains were rationally selected to construct synthetic consortia for efficient TBBPA degradation. Our study provides a general strategy for engineering synthetic microbiomes without prior knowledge of bacterial interactions and strain resources, and this strategy could be extended to wider applications for synthetic microbiomes than for pollutant biodegradation.

Results

TBBPA-degrading microbiomes are successfully enriched and have multiple pathways of TBBPA debromination

An overview of this study was framed in Fig. 1a. Previous studies showed that the sediments of the Beijing Olympic Park dragon-shape aquatic system contained diverse and interactive microbial communities^39,40. Sediments were collected from the dragon-shape aquatic system and were used as seeds for the enrichment of TBBPA-degrading microbiomes. After 21 days of inoculation, the initial TBBPA-degrading microbiome (M1) was obtained (Fig. 1a, b). From this M1 microcosms, we obtained 11 TBBPA-degrading and 6 BPA-degrading microbiomes, according to transfer times (every 14 days or 30 days), media (MSM or GSM), and treatments (TBBPA or BPA) (Fig. 1a). The 11 TBBPA-degrading microbiomes (groups G1, M2, and M3; Fig. 1b) showed different activities of TBBPA degradation: The TBBPA degradation percentages was decreased in the microbiomes of group M2 (M2-1, M2-2, and M2-3), while group M3 (M3-1, M3-2, M3-3 and M3-4) increased TBBPA degradation percentages (with the highest 91.6%), suggesting that long transfer time (every 30 days) was essential for TBBPA-degrading microbiomes. The degradation activities of group G1 were relatively stable, with degradation percentages ranging 28.6–40.5%. Debromination is a key step in the biodegradation of TBBPA^30,41. We tested debromination during the TBBPA degradation of the Group M3 microbiomes (Fig. 1c). The results showed that the group M3 microbiomes were efficiently debrominating TBBPA, and the debromination varied with their TBBPA degradation percentages (Fig. 1b, c). Different from the TBBPA-degrading microbiomes, all 6 BPA-degrading microbiomes (BG-1, BG-2, BG-3, BM-1, BM-2, and BM-3) had high BPA degradation percentages (87.7–92.9%; Fig. 1e).

[See PDF for image]

Fig. 1

TBBPA and BPA-degrading microbiomes are obtained by a top-down strategy.

a Overview of experimental design. Some of the elements were created in BioRender (https://BioRender.com/k0m85rp). Degradation (b) and debromination (c) of TBBPA in GSM medium by TBBPA-degrading microbiomes. M3-1: TBBPA degraded 7.67 μM and Br⁻ released 5.55 μM; M3-2: TBBPA degraded 6.03 μM and Br^- released 4.59 μM; M3-3: TBBPA degraded 16.83 μM and Br⁻ released 24.78 μM; M3-4: TBBPA degraded 12.99 μM and Br⁻ released 13.73 μM. d Scanning electron microscopy of TBBPA-degrading microbiomes. e Degradation of BPA in GSM medium by BPA-degrading microbiomes. The group containing G represents transfer in GSM medium, and the group containing M represents transfer in MSM medium. The error bar represents the Standard Error of the Mean.

HPLC-MS was used to identify the intermediates of TBBPA degradation (Fig. S1). DP1 and DP2 are typical debromination products of TBBPA, which can usually be further debrominated to BPA^28,35. DP3, DP4, and DP5 are oxidation products via beta-scission of TBBPA, and have been previously found in TBBPA biodegradation^33,35. Interestingly, DP6, DP7, and DP8 are nitrated products, and similar metabolites were rarely reported in the biotransformation of TBBPA. Based on the intermediates of this study, three initial reactions of TBBPA degradation were proposed, i.e., direct debromination³³, beta-scission of oxidation³⁵, and beta-scission of nitration⁴².

The microbial diversity, keystone taxa and co-occurrence network of TBBPA-degrading microbiomes

To explore the microbial dynamics driven by TBBPA contamination, we performed amplicon sequencing of the V4–V5 variable regions of the 16S rRNA genes of the TBBPA-degrading microbiome. Compared with the initial microbiome M1, the alpha diversities (Chao1 index) of continuously enriched microbiomes (groups G1, M2, and M3) decreased, and group M3 decreased most significantly (Fig. S2a). The principal coordinates analysis (PCoA) of the Bray-Curtis distance clearly separated the groups G1, M2, and M3 from each other, and showed that the microbiomes within each group clustered (Fig. S2b). We observed that Gammaproteobacteria, Alphaproteobacteria and Betaproteobacteria were the top 3 in abundance at the class level of all groups of TBBPA-degrading microbiomes (Fig. S2c), but were different at the family level. Groups G1 and M3 were dominated by Pseudomonadaceae, and families Xanthomonadaceae, Alcaligenaceae and Phyllobacteriaceae also accounted for a large proportion (Fig. S2d). The poorly TBBPA-degrading microbiomes of group M2 were dominated by Enterobacteriaceae. The dominant taxa of the BPA-degrading microbiomes overlapped with those of TBBPA-degrading microbiomes. Proteobacteria was also the dominant taxon in all BPA-degrading microbiomes and Pseudomonadaceae, Alcaligenaceae and Xanthomonadaceae accounted for a high proportion at the family level (Fig. S3a, b).

Keystone taxa dominate the functions of microbiomes. We used Spearman’s correlation, Random Forest regression, and LEfSe tools to explore the keystone taxa for TBBPA degradation. With Spearman analysis, we calculated the correlation between the microbial abundance at the genus level and TBBPA degradation. The results showed that a total of 16 genera were positively correlated with TBBPA degradation, of which 7 genera showed a statistically significant correlation (p < 0.05) (Fig. 2a). The correlation is shown in Fig. S4. Seven significantly correlated genera were considered as potential keystone taxa and might be closely involved in TBBPA degradation. Among them, the keystone taxa of Pseudomonas, Achromobacter, and Pseudoxanthomonas had higher average relative abundances in the TBBPA-degrading microbiomes.

[See PDF for image]

Fig. 2

Identification of keystone taxa and co-occurrence networks.

a Heatmap showing correlation between the microbial abundance at genus level and TBBPA degradation. The average relative abundance of each genus was greater than 0.1%. * p < 0.05, ** p < 0.01. b Bacterial taxonomic biomarkers of TBBPA degradation in TBBPA-degrading microbiomes. The biomarkers were identified by Random Forests regression. c Heatmap showing the abundances of biomarkers against TBBPA degradation. The insert represents 10-fold cross-validation error as a function of the number of input genera. The dashed gray line marks the optimal cut-off for biomarker selection. d TBBPA co-occurrence network at the genus level (left), and modularized network (right). Red and green links represent positive and negative interactions, respectively. The color changes depending on the analysis.

Further, a degradation regression model was constructed based on the Random Forests machine learning algorithm, and the genera whose abundances differentiated along with TBBPA degradation activity of microbiomes were screened. To reveal important genera as biomarkers correlated with TBBPA degradation, we performed 10-fold cross-validation with five repeats to evaluate the importance of bacterial genera. As shown in Fig. 2b, when 9 important taxa were used, the minimum cross-validation error was obtained. The relative abundances of the 9 genera against TBBPA degradation are shown in Fig. 2c. The genera of Stenotrophomonas and Enterobacteriaceae Unassigned accumulated in the communities with low degradation efficiency and were biomarkers of low-degradation microbiomes. With the increase of the relative abundance of Rhodanobacter, Achromobacter, and Brucella, the degradation activity of microbiomes on TBBPA was improved, and they could be used as biomarkers for the microbiomes to develop toward moderate degradation activity. Ancylobacter, Pseudomonas, Herbaspirillum, and Azospirillum were biomarkers of high-degradation microbiomes, which accumulated in microbiomes with high degradation activity. We also included biomarkers representing moderate and high degradation activities as potential keystone taxa. In addition to the keystone taxa already identified by Spearman correlation analysis, Herbaspirillum, Brucella, and Rhodanobacter were added to the list of keystone taxa. Besides, the results of LEfSe (LDA > 3.0) analysis further overlap and corroborate with the above two analysis methods (Fig. S5). In conclusion, a total of 10 genera were defined as keystone taxa in our analysis (Table 1). Microorganisms from these taxa should be of interest during the biodegradation of TBBPA.

Table 1. The keystone taxa identified from the TBBPA-degrading microbiomes

Keystone taxa (genera)	Identification tools	Network modules	Representative strains
Herbaspirillum	Random Forest regression, LEfSe	Module 1	Herbaspirillum sp. WTB6
Pseudomonas	Spearman’s correlation, Random Forest regression, LEfSe	Module 1	Pseudomonas asiatica WTB3
Azospirillum	Spearman’s correlation, Random Forest regression	Module 1	Azospirillum oryzae WTB4
Rhodopseudomonas	Spearman’s correlation, LEfSe	Module 1	Rhodopseudomonas faecalis WTB18
Kaistia	Spearman’s correlation, LEfSe	Module 2	Kaistia hirudinis WTB13
Ancylobacter	Spearman’s correlation, Random Forest regression, LEfSe	Module 2	Ancylobacter rudongensis WTB12
Achromobacter	Spearman’s correlation, Random Forest regression, LEfSe	Module 3	Achromobacter anxifer WTB1
Brucella	Random Forest regression, LEfSe	Module 3	Brucella cytisi WTB2
Pseudoxanthomonas	Spearman’s correlation, LEfSe	Module 3	Pseudoxanthomonas humi WTB8
Rhodanobacter	Random Forest regression, LEfSe	Module 5	Not acquired

To understand the microbial correlation and potential interactions of keystone taxa with others, we constructed co-occurrence networks of TBBPA- and BPA-degrading microbiomes (Figs. 2d and S6). Their network indices were summarized in Table S1. There were 37 nodes and 103 edges in the TBBPA-degrading microbiome network, showing more extensive and complex relationships than the BPA network. Nodes in the TBBPA network were grouped into six modules (Fig. 2d). Interestingly, the keystone taxa related to TBBPA degradation were mostly found in modules 1–3, and showed extensive correlations. These suggest that the taxa and interactions in modules 1–3 may be more important for the degradation of TBBPA. In the TBBPA network, the keystone taxa of Pseudomonas, Rhodopseudomonas, Herbaspirillum, and Azospirillum were in module 1; Kaistia and Ancylobacter were in module 2; Achromobacter, Pseudoxanthomonas, and Brucella were in module 3 (Fig. 2d, Table 1). Based on degree and betweenness centrality (BC), Achromobacter (degree:9, BC: 154.88) was the most important node in the modules (modules 1–3) (Table S2) and had a positive correlation with different taxa, including keystone taxa Brucella and Pseudoxanthomonas. The BPA network contains three different modules (Fig. S6). Substrate changing drove the reassembly of microbial communities during enrichment. Herbaspirillum had the largest degree and betweenness centrality in the BPA network (Table S2). As a keystone taxon of TBBPA degradation, Herbaspirillum might play an important role in the degradation of both TBBPA and BPA.

Targeted bacterial cultivation for keystone taxa and characterization of TBBPA degradation

A total of 594 bacterial isolates were obtained from the TBBPA-degrading microbiomes by R2A, GSM and MSM medium, which belonged to 35 genera and 47 species (Fig. 3a). Members of Pseudomonas (26.9%), Achromobacter (18.9%) and Brucella (15.9%) were the most frequent isolates (Fig. S7). As shown in Table 1, we obtained strains that corresponded to all keystone taxa except Rhodanobacter. We first tested TBBPA degradation in the MSM and GSM media that were used for enriching TBBPA-degrading microbiomes. According to their relatedness to keystone taxa, growth in GSM and taxonomy, 25 bacterial strains from the 594 isolates that overlapped 9 keystone taxa and represented 25 genera were selected for detailed degradation testing (Table S3). The results showed that only 7 strains marginally degraded TBBPA (degradation percentage < 10%) (Fig. 3b). We tried to co-culture different strains in GSM medium to degrade TBBPA. Although the combinations of multiple strains achieved higher degradation activity compared to single strains (Fig. S8), the degradation percentages of TBBPA were still weak compared with the naturally evolved microbiome, such as M3-3.

[See PDF for image]

Fig. 3

Bacterial strains and genomic resources for TBBPA-degradation.

a Neighbor-joining tree based on the 16S rRNA gene sequence of isolates. b Degradations of TBBPA and BPA with various representative strains in different media. The strains marked in red are keystone taxa. c Phylogenomic tree and COG characteristics of MAGs and complete genomes of strains. The genomes marked in red are keystone taxa. Abbreviations in Metabolism categories, E: Amino acid transport and metabolism; C: Energy production and conversion; P: Inorganic ion transport and metabolism; I: Lipid transport and metabolism; Q: Secondary metabolites biosynthesis, transport and catabolism; H: Coenzyme transport and metabolism; G: Carbohydrate transport and metabolism; F: Nucleotide transport and metabolism. d Degradation activity on TBBPA with three keystone strains in the presence of different L-amino acids. See the Methods for specific L-amino acid groups. e Degradation and growth of TBBPA or BPA by strain WTB6 in the presence of L-alanine (2 g/L), L-valine (1 g/L), and L-leucine (1 g/L).

We then tested the 25 strains with LB (a peptide-rich medium) and found that 8 strains (including 5 keystone taxa strains, WTB1, WTB2, WTB3, WTB4, and WTB8) significantly degraded TBBPA (Fig. 3b). The keystone taxa of strains WTB1, WTB2, and WTB3 were more efficient among the 8 TBBPA degraders (Fig. S9). After degradation optimization, these keystone taxa strains could achieve nearly 60% TBBPA degradation in Luria-Broth (Fig. S10, Tables S4 and S5). However, some keystone taxa strains (Herbaspirillum sp. WTB6, Rhodopseudomonas harwoodiae WTB18, and Ancylobacter rudongensis WTB12) did not grow in LB medium and had no degradation activities.

To find suitable carbon sources to promote TBBPA degradation from the genetic foundation, we sequenced the genomes of 6 selected strains and the metagenomes of the TBBPA-degrading microbiomes. The most abundant functional pathway in the tested TBBPA-degrading microbiome was amino acid metabolism (5.03%) based on KEGG pathways annotation (Fig. S11). We also found high contents of genes for COG categories of amino acid transport and metabolism in the genomes of TBBPA-degrading strains such as WTB1, WTB2, WTB6, and the metagenome assembled genomes (MAGs) (Fig. 3c). Also inspired by other studies^{43, 44–45}, amino acids were tested for promoting TBBPA degradation. We supplemented L-amino acids (ASM medium) to test the degradation of all 25 strains (Fig. 3b). The results showed that the keystone taxa strains WTB1, WTB2, WTB3, WTB4, WTB6, WTB8, WTB12, and WTB18 degraded TBBPA in ASM medium. Except for strain WTB13, all keystone taxa strains showed the ability to degrade TBBPA. The strains WTB1, WTB2, and WTB6 had stronger degradation abilities. Two strains of non-keystone taxa, WTB19 (Module 3) and WTB7 (Module 1), also showed TBBPA degradation ability in LB and ASM media. As shown in Fig. 3b, all the TBBPA-degrading strains, but strain WTB8, also degraded BPA in ASM media. Therefore, we deduced that the amino acid-related metabolic functions in the microbiomes were attributed to TBBPA and BPA degradation.

We were interested in which amino acid(s) stimulated TBBPA degradation. The three keystone taxa strains with high degradation activity, WTB1, WTB2, and WTB6, were selected for further degradation in the presence of L-amino acid groups with different properties (group 1–group 4). Compared to the complete L-amino acids mixture, the degradation of strains WTB1 and WTB2 significantly decreased or even disappeared in those amino acid groups (Fig. 3d). Interestingly, strain WTB6 only required nonpolar aliphatic amino acids of group 1 to obtain high degradation activity. L-alanine was the key to the growth and degradation of strain WTB6 (Fig. S12a, b). When only L-alanine was added, a high concentration of L-alanine (2 g/L) was required to produce efficient degradation of TBBPA (Fig. S12c). The addition of branched-chain amino acids (L-valine, L-leucine) on the basis of L-alanine could promote the degradation of WTB6, which could completely degrade 10 mg/L TBBPA within 42 h (Fig. 3e). The nitration products (DP6, DP7, and DP8) were detected in the degradation of TBBPA by the strain WTB6, indicating that strain WTB6 could mediate the beta-scission and nitration transformation of TBBPA. The nitrogen metabolism of strain WTB6 in the presence of L-amino acid may be involved in the degradation of TBBPA.

The key genes of the keystone taxa strains were analyzed to further confirm their biological functions. The genomes of these degrading strains were rich in genes putatively associated with TBBPA degradation (Fig. S13): Glutathione S-transferase (gst), cytochrome P450 (cyp450), ferredoxin-NADP reductase (fpr) have been reported to be required for the debromination^35,46,47, which is considered to be a key step in TBBPA biodegradation and detoxification⁴¹. These genes were repeatedly found in the genomes of strains WTB1, WTB2, and WTB6. Furthermore, these strains harbored haloacid dehalogenase (HADH) or haloalkane dehalogenases (HLDs). In particular, strain WTB2 had homologous proteins of aerobic bromophenol dehalogenase (ARV76518, Brucella intermedia) on its plasmid, which has been believed to be involved in the aerobic debromination of TBBPA⁴⁸. Strain WTB1 had the most genes in the benzoate degradation pathway (67), followed by WTB6 (28) and WTB2 (19). A variety of aromatic ring dioxygenases and monooxygenases could participate in the degradation of aromatic compounds, and methyltransferase genes (bioC, pcm, metE, ubiG, and ubiE) could regulate the transfer of methyl groups on aromatic rings or O-H bonds. It is worth noting that these keystone taxa strains contained abundant genes involved in the nitrogen metabolism pathway, especially strains WTB2 and WTB6 have a relatively complete denitrification pathway, which could be related to the nitration transformation of TBBPA in the degrading microbiomes (Fig. S14). These strains contain most of the genes related to the TBBPA metabolism process, which might be the functional basis of the co-metabolic degradation activity.

Construction, characterization, and application of synthetic consortia in TBBPA-polluted soil

In order to obtain a cost-effective and degrading-efficient synthetic consortia, the strain WTB6 (Herbaspirillum) was selected from keystone strains as the basis for the synthetic consortia, which could efficiently degrade TBBPA with fewer resources. The construction process is shown in Fig. 4a. Firstly, we selected taxa located in the same module (module 1) with the Herbaspirillum taxon in the TBBPA network for consortia construction. Among them, Acinetobacter taxa had a significant positive correlation with Herbaspirillum taxon, and the synthetic consortium SynCon1 composed of strains WTB5 (Acinetobacter sp.) and WTB6, could accelerate the degradation of TBBPA (Fig. S15). This stimulated us to test combinations of more keystone strains of adjacent modules for synthetic consortia to promote TBBPA degradation and reduce the demand for amino acids in the degradation.

[See PDF for image]

Fig. 4

Construction of synthetic consortia and proposed degradation of TBBPA.

a The construction of the synthetic consortium SynCon2. b Upstream metabolites of TBBPA degradation and degradation of aromatic organic compounds by the synthetic consortium SynCon2. The circles under the arrows indicate that different strains possess metabolic genes for this step. Gene names are summarized in Table S6. c Degradation of TBBPA by synthetic consortia at low L-alanine concentration (1 g/L). The degradation percentages were determined after incubation at 30 °C for 36 h. d Degradation of TBBPA in soil microcosms with and without SynCon2. Residual TBBPA was extracted from soils at day 7 of treatment. Error bars are the Standard Error of the Mean of triplicate experiments. Significance was analyzed by unpaired, two-tailed Student’s t-test. *p < 0.05, ** p < 0.01, *** p < 0.001.

The strains WTB1 (Achromobacter, the most important node) and WTB2 (Brucella) were keystone taxa and tightly associated in the TBBPA network, which showed stronger TBBPA degradation potential among the keystone strains. Genome-based metabolic analysis suggested that the association of strains WTB1 and WTB2 could enhance the degradation of related aromatic compounds via cross-feeding into the metabolic pathways of 4-hydroxybenzoate and 4-hydroxyphenyl acetate (Fig. 4b, Table S6), which are the downstream pathways involved in the degradation of TBBPA and BPA in the environment⁴⁹. In addition, the genomes of WTB1 and WTB2 contained a large number of genes encoding for amino acid transport and metabolism, and amino acids stimulated TBBPA degradation. Therefore, we considered strains WTB1 and WTB2 to have great potential for synthetic consortia. Lastly and finally, we formulated a consortium SynCon2 of four strains, WTB6, WTB5, WTB1, and WTB2. Compared with strain WTB6 and SynCon1, which degraded TBBPA at a concentration of 2 g/L alanine, the SynCon2 could achieve efficient degradation of TBBPA at 1 g/L alanine concentration, and combining strains of all keystone taxa was not the most optimal option in the construction of synthetic consortium (Fig. 4c). We also observed that SynCon2 could degrade all 4-hydroxybenzoate and 4-hydroxyphenyl acetate with initial addition of 10 mg/L, indicating that the synthetic consortium indeed extended the degradation of related aromatic compounds.

We determined 9 TBBPA metabolites from the SynCon2 (Figs. 4b and S16). Among them, products SDP3, SDP7, SDP8, and SDP9 were identical to the TBBPA degradation products of the enriched microbiomes (DP5, DP6, DP7, and DP8), indicating that the synthetic consortium could carry out degradation pathways with beta-scission of oxidation and nitration, just as the enriched microbiomes. Furthermore, we found more oxidation (SDP4) and nitro-debromination (SDP1, SDP2, SDP5, SDP6) products in the metabolites. The diverse metabolites showed that the TBBPA degradation processes of debromination, hydroxylation, methylation, nitration, and isopropyl cleavage occurred in the SynCon2.

To evaluate the effectiveness of bioremediation, we applied the synthetic consortium to TBBPA-polluted, sterilized, and non-sterilized soil microcosms (Fig. 4d). The degradation of TBBPA in sterilized soil was about 6%, and inoculation of the SynCon2 significantly improved TBBPA degradation. Data showed that more than 20% of TBBPA was degraded in 7 days. In the non-sterilized soil microcosm, the degradation of TBBPA could be further promoted by the SynCon2, achieving nearly 40% TBBPA degradation in 7 days. This result suggested that the SynCon2 could adapt to the soil environment and might further invoke and collaborate with indigenous microbiomes for the degradation of TBBPA.

Discussion

In recent years, synthetic and engineered microbiomes are increasingly being employed as a solution to challenges in health, agriculture, and environmental remediation⁵⁰. Pioneering studies have attempted to exploit synthetic consortia for the bioremediation of organic pollutants with known microbial degradation pathways and bacterial resources²³. However, for emerging pollutants with a limited understanding of microbial degradation^51,52, the construction of synthetic microbiomes is a difficult and urgent issue. In this study, the organic pollutant TBBPA was taken as a case study to find effective solutions to this issue. We used a top-down strategy to reassemble the natural microbiome into degrading microbiomes. Compared to BPA degradation, the enrichment of TBBPA-degrading microbiomes was more difficult. Consistent with previous studies, TBBPA is difficult to be directly utilized by microbes^33,53,54 and longer transfer time during the enrichment was conducive to the evolution of the degradation function. Under different enrichment processes, we obtained a series of microbiomes with different degradation and debromination activities.

The degrading microbiomes obtained by top-down enrichment processes are usually difficult to apply in practice due to the lack of controllability and indefinity⁵⁵. Traditionally, the study is subsequently carried out with strain isolation from the degrading microbiomes and trial-and-error experiments to find degrading strains and to construct combinatorial degrading consortia¹¹. This study strategy is time-consuming and difficult to conduct for the refractory pollutants³¹. Especially for pollutants that rely on microbial co-metabolism degradation, it is difficult to select suitable nutritional conditions for screening degrading strains. In addition, existing studies have ignored microbial interactions, and some potential degradation-promoting microbes have been missed. We attempted to gain insights from top-down microbiomes to address these issues in this current study, including: (1) identifying keystone taxa for TBBPA degradation through the correlation between community dynamics and degradation activities; (2) revealing microbial interactions by the co-occurrence network analysis; (3) analyzing metabolic characteristics of microbes to find suitable co-metabolism carbon sources to promote degradation.

Based on our framework, Spearman’s correlation analysis, random forest regression analysis, and LEfSe analysis were integrated to explore microbial taxa associated with TBBPA degradation. A TBBPA-degrading regression model was constructed with the degradation percentages of different TBBPA-degrading microbiomes as parameters. Finally, ten genera were identified as keystone taxa for TBBPA degradation without prior knowledge of TBBPA degradation. Co-occurrence networks are considered an effective means to identify microbial interactions in microbiomes⁵⁶. In the TBBPA network, keystone taxa were concentrated in modules 1–3 with a close correlation. From their network indices, we could get a more useful understanding. For example, Achromobacter was the most important node in the TBBPA network, which might cooperate with different microbial taxa to perform functions. Herbaspirillum might be involved in the degradation of both TBBPA and BPA. So far, microbes from Achromobacter and Herbaspirillum have not been reported to have the degradability of TBBPA. Especially for Herbaspirillum, as a genus with active nitrogen metabolism, there are few studies related to the degradation of aromatic compounds⁵⁷.

Through targeted isolation and cultivation, we obtained representative strains of different keystone taxa. However, compared with the TBBPA-degrading microbiomes, single and combined strains had difficulty degrading in GSM medium, indicating that glucose was not a direct co-metabolism substrate for the microbes in the TBBPA-degrading microbiomes. This stimulated us to search for more suitable co-metabolic substrates to promote and simplify the degradation of TBBPA by strains and synthetic consortia. Based on metagenomic and strain genome data, we found that most microbial taxa in the TBBPA-degrading microbiome had a high abundance of genes related to amino acid transport and metabolism functions. Recently, amino acids have been found to play an important role in microbial positive interactions⁴³ and promote the degradation of organic pollutants by microorganisms as co-metabolic substrates⁴⁵. Combined with this information, we speculated that L-amino acids are important substrates produced and transported in TBBPA-degrading microbiomes and promote the degradation of TBBPA by keystone taxa. As expected, the keystone taxa strains showed efficient degradation of TBBPA when L-amino acids were added as nutrients, and most of the degrading strains had better degradation activity than those in LB medium. The degradation and growth curves of strain WTB6 in the presence of amino acids were determined to characterize the degradation rate and biomass changes, and strain WTB6 had increased biomass in the presence of amino acids and rapidly degraded TBBPA during the plateau of bacterial growth. In addition, we tried the degradation of TBBPA by sodium pyruvate and vitamins for strain WTB6, which also promoted the bacteria's growth to a high biomass but did not produce significant TBBPA degradation. Therefore, the effects of amino acids on TBBPA degradation might rely on co-metabolism and increased bacterial biomass, which requires further exploration. We revealed for the first time that L-amino acids could be used as co-metabolic substrates to promote the degradation of TBBPA by different microbial taxa. The identification of key degrading strains and suitable nutrient substrates helped us to construct simplified synthetic consortia from the bottom-up approach.

The combination of all keystone strains is not the most cost-effective and degradation-efficient option²³. Especially for pollutants that require co-metabolism degradation, each strain in the combination might cause unnecessary resource consumption. Our strategy was to select a keystone strain as the basis, and then screen other strains to construct consortia based on co-occurrence interactions and metabolic interactions with this base strain. When the keystone strain WTB6 was used as the basis, we obtained a synthetic consortium SynCon2 consisting of four strains. The artificially combined SynCon2 consortium had a stable degradation effect on TBBPA. Moreover, SynCon2 achieved efficient degradation of TBBPA at a low L-amino acids dosage and improved the overall bioremediation function of related aromatic compounds. In addition, the SynCon2 showed similar TBBPA metabolic processes as the enriched TBBPA-degrading microbiomes, suggesting that the simplified consortium dominated by keystone taxa strains could achieve the main reaction of TBBPA degradation. Synthetic consortia generally show better environmental adaptability than engineered strains⁵⁸, and SynCon2 showed enhanced bioremediation of TBBPA in soil environments. Our results demonstrate the potential of an integrated engineering strategy for the bioremediation of emerging organic pollutants. Moreover, the omics data-driven design of synthetic microbiomes is flexible, which allows for the continued construction of more diverse degradation systems based on different functional strains. For example, if strains targeting intermediate products are identified, they could be supplemented into synthetic consortia to promote the complete mineralization of TBBPA. The strains and genomic resources obtained in this strategy can be further applied to the construction of consortia in the future as the relevant metabolic mechanisms are elucidated.

In conclusion, this study provides an integrated engineering strategy for the biodegradation of emerging environmental pollutants. Keystone taxa and co-occurrence interactions were identified, and the study showed that L-amino acids were important for co-metabolism and promoting TBBPA degradation. This information was applied to design a simplified synthetic consortium SynCon2 of four strains with bottom-up approach, and SynCon2 had strong TBBPA-degradation activity and was efficient in soil bioremediation. This strategy does not need a prerequisite understanding of the metabolism and interaction mechanisms of microbes, paving the way for bioremediation of emerging pollutants with a limited understanding of biodegradation. These findings provide a new insight into the biotransformation and artificial bioremediation of TBBPA and other organic pollutants in natural environments.

Methods

Chemicals and culture media

TBBPA (98%) and BPA (99%) analytical standard used throughout the experiments was acquired from Aladdin (Shanghai, China). 4-hydroxybenzoate and 4-Hydroxyphenyl acetate were purchased from OKA and Mreda (Beijing, China). The minimal salt medium (MSM) and glucose source medium (GSM, 0.1% glucose (w/v)) were as in our previous study⁵⁹. Amino acid medium (ASM) was GSM medium supplemented with 20 essential L-amino acids. The 20 L-amino acids were divided into four groups according to experimental needs, Group 1 (simple non-polar aliphatic): L-glycine, L-alanine, L-valine, L-leucine, L-isoleucine; Group 2 (heterocyclic and aromatic): L-proline, L-phenylalanine, L-tyrosine, L-tryptophan; Group 3 (uncharged): L-methionine, L-serine, L-threonine, L-cysteine, L-aspartate, L-glutamine; Group 4 (positively and negatively charged): L-aspartate, L-glutamic acid, L-lysine, L-arginine, L-histidine. High L-alanine concentration (2 g/L) medium was GSM medium supplemented with 2 g/L L-alanine, 1 g/L valine and 1 g/L leucine. Low L-alanine concentration (1 g/L) medium was GSM medium supplemented with 1 g/L L-alanine, 1 g/L L-valine and 1 g/L L-leucine. The composition of Reasoner’s 2 medium (R2A) was as previously reported⁶⁰. Agar (1.5%) was added to obtain a solid medium if necessary.

Enrichment and establishment of microbiomes

Sample collection was performed according to our previous study³⁹. The sediments of dragon-shaped aquatic system (Beijing, China) were used as a source of microorganisms. The enrichment process was shown in Fig. 1a. Briefly, 5 g sediment samples were inoculated into a 250 mL flask containing 100 ml MSM medium that was amended with 10 mg/L TBBPA. After treatment at 30 °C and 150 rpm for 21 days, the enrichment culture was the initial microbiome M1. The enrichment culture (10% v/v) was transferred to a new medium with 20 mg/L TBBPA. In the M2 and M3 transfer groups, the culture was transferred to MSM medium and transferred every 14 days (M2-1, M2-2, M2-3) or 30 days (M3-1, M3-2, M3-3, M3-4). G1 transfer group (G1-1, G1-2, G1-3, G1-4) was transferred to the GSM medium and transferred every 14 days. In a similar enrichment process, the enrichment culture of initial microbiome M1 (10% v/v) was transferred to a new medium with 8.4 mg/L BPA (same molarity as TBBPA). In the BM transfer group, the culture was transferred to the MSM medium and transferred every 30 days (BM-1, BM-2, BM-3). The BG transfer group (BG-1, BG-2, BG-3) was transferred to the GSM medium and transferred every 14 days.

Bacterial isolation, cultivation, and identification

The morphology of the cells in the microbiomes was observed by scanning electron microscope (Quanta 200; FEI). 5 mL of enrichment culture was evenly dispersed in 45 mL of sterile 0.85% NaCl (w/v) solution, then serially diluted tenfold. Aliquots (100 μL) of the diluted samples were then spread onto R2A, GSM, and MSM plates containing 10 mg/L TBBPA and cultivated at 30 °C for 48–72 h. Colonies were then transferred, and bacterial strains were preserved at −80 °C.

Bacterial purity was confirmed by colony morphology observation and 16S rRNA gene sequencing. Additionally, the 16S rRNA gene was amplified with the universal bacterial primers 27F (AGAGTTTGATCATGGCTCAG) and 1492R (TACGGTTACCTTGTTACGACTT). The obtained sequences were then compared with 16S rRNA gene sequences by BLAST searches of the GenBank database (https://blast.ncbi.nlm.nih.gov/Blast.cgi) and EZBioCloud (https://www.ezbiocloud.net/identify).

The 25 strains used in the detailed experiments were publicly deposited in the China General Microbiological Culture Collection Center (CGMCC, Beijing, China). See Table S3 for details.

Batch biodegradation experiments

Batch degradation experiments amended with TBBPA or BPA were constructed to examine the degradation activity of the enrichment microbiomes, strains, and co-culture of strains. Unless specified, biodegradation experiments were conducted with 10 mg/L TBBPA or 5 mg/L BPA under aerobic conditions at 30 °C and 150 rpm. For the test of enrichment microbiomes, enrichment culture was added to fresh MSM or GSM medium at a volume of 1% (v/v) for 10 days. For the test of single or co-culture strains, strains were revitalized in R2A medium, after which 1% (v/v) of the culture was transferred to a fresh R2A medium and cultivated to the end of the exponential growth phase. Cells of strains were washed three times with MSM (5000 rpm, 10 min) and precipitated to OD₆₀₀ = 1. The bacteria were subsequently added to prepare the media at a volume of 0.5% (v/v) and cultured for the corresponding time for detection.

DNA extraction and sequencing of microbiome samples

DNA extraction was performed on biomass pellet after centrifuging at 13,000×g for 10 min at 4 °C using FastDNA^TM Spin Kit (MP Biomedicals, Solon, OH, USA) following the manufacturer’s instructions. The 60 samples of TBBPA microbiomes and 30 samples of BPA microbiomes were subjected to 16S rRNA gene amplicon sequencing. The V4–V5 region of the bacteria 16S rRNA gene was amplified by PCR using the 341F/806R primer set (515F: 5’-GTGYCAGCMGCCGCGGTAA-3’, 926R: 5’-CCGYCAATTYMTTTRAGTTT-3’). The amplicon library was sequenced on an Illumina Miseq PE250 platform (Guangzhou Pluslife Biotech Co., Ltd, Beijing, China) according to the standard protocols. Since metagenomic analysis requires high DNA content, we selected 6 samples from the G1 transfer group with high biomass for metagenomic sequencing. After DNA extraction, the Illumina HiSeq 2500 PE150 platform (Guangzhou Pluslife Biotech Co., Ltd, Beijing, China) was used for sequencing.

16S rRNA gene amplicon sequencing and statistical analyses

Amplicon sequencing bioinformatics was performed according to the EasyAmplicon v1.09 analysis process⁶¹. We used the --derep_fulllength subcommand in vsearch v2.15 to remove sequence redundancy⁶². The non-redundancy sequences were denoised into amplicon sequence variants (ASVs) with USEARCH v10.0 2⁶³. ASVs were based on the RDP training set v16 and used the USEARCH sintax algorithm for taxonomic annotation⁶⁴. The important ASVs sequences with low confidence in the genus-level annotation were individually compared with the GenBank database for verification. Diversity analysis was used QIIME 2 and vegan v2.5-6 package in R v4.1.

In the statistical process, for data that conforms to the normal distribution and has homogeneous variances, one-way analysis of variance (ANOVA) and Tukey’s honest significant difference (HSD) were used for the significant differences test. The Wilcox rank sum test was used for the data that did not conform to the normal distribution. PCoA analysis was performed using the permutation-based hypothesis tests (PREMANOVA). The degradation percentage and response surface analysis methodology were described in our previous study⁵⁹. The Data on degradation were analyzed using GraphPad Prism version 8.0 (GraphPad Software, Inc., La Jolla, CA, USA) with an unpaired, two-tailed Student’s t-test.

Data visualization was mainly performed using the ggplot2 v3.3.2 package. Corrplot and ComplexHeatmap packages were used to calculate correlation and plotted heatmaps⁶⁵. Random forest regression analysis (randomForest package) was used to classify genera whose abundances differentiated along with the TBBPA degradation activity of microbiomes. Linear discriminant analysis (LDA) was used to perform LEfSe differential comparison⁶⁶ and the microbiomeViz package was used to plot the evolutionary branch diagram with ggtree⁶⁷. In the co-occurrence network analysis, genera with average relative abundance >0.01% were selected for network construction. Each node of the co-occurrence network represented one genus, and each edge denoted a significant correlation. When the Spearman correlation coefficient between two nodes was p > 0.2 and p < 0.01, they were retained in the network. The igraph package was used to construct and analyze the network, and the visualization and topology calculations were completed in the Gephi software.

Metagenome sequencing and metagenome‑assembled genome (MAG)

Metagenomics sequencing data were assembled using MEGAHIT⁶⁸. Prodigal was used to predict open reading frames (ORFs) and to assess the quality of the ORFs⁶⁹. The default parameters of Mmseqs software were used for gene clustering and redundancy removal to obtain non-redundant gene catalogs⁷⁰. High-quality reads were aligned to these non-redundant gene catalogs to calculate gene abundance using bbmap. Diamond (Version 0.9.30)⁷¹ was used to align the protein sequences of the non-redundant gene catalogs with databases such as the Non-Redundant Protein Database (NR), Kyoto Encyclopedia of Genes and Genomes (KEGG), and Clusters of Orthologous Groups (COG) to obtain functional annotation information. Combined with the gene abundance table, the functional composition of each classification level was obtained.

Metagenome binning used the MetaWRAP v1.2 integrated analysis process⁷², in which the binning commands were based on MetaBAT v2.12.1⁷³, MaxBin v2.2.5⁷⁴, and CONCOCT v1.0.0 software⁷⁵. The taxonomic assignments of metagenome-assembled genomes (MAGs) were obtained based on the classify_bins command invoking taxator-tk v1.3.3e⁷⁶. Prokka v1.13.3 was used to predict and annotate the functions of genes in the MAGs⁷⁷. High-quality MAGs were distinguished by completeness >90% and contamination <5%.

Whole-genome sequencing and functional annotation

Genomic DNAs of strains were extracted with a commercial TIANamp Bacteria DNA kit according to the manufacturer’s instructions (TIANGEN, Beijing, China). Whole-genome sequencing was performed using the Illumina HiSeq 2500 system and the PacBio Sequel platform. Genome sequences were assembled with SMRT Link v5.1.0 using clean data⁷⁸. Coding genes and prophage were predicted using Glimmer3⁷⁹ and PHAST⁸⁰. Genome functional annotation was mainly accomplished through homology alignment with reference databases, including the NR, KEGG, and COG. Gene cluster visualization was conducted using the ChiPlot online tool (https://www.chiplot.online/).

Soil bioremediation experiments

The soil samples were dried at 30 °C sieved to 5 mm, and mixed thoroughly. Part of the soil samples were sterilized at 115 °C for 30 min. Fresh and sterile soil samples were added with TBBPA (dissolved in methanol, 1 g/L) to maintain their TBBPA concentrations at 20 mg/kg. The soil samples were then stirred and dried at 30 °C. Cells of four strains were washed three times with MSM (5000 rpm, 10 min) and precipitated to OD₆₀₀ = 1. The equal volume combination of four strains was the initial inoculation SynCon2 consortium, and inoculated at a volume of 400 μl per gram of soil. Then, soil samples were incubated with ddH₂O and synthetic consortia, respectively. Experiments were conducted at 30 °C for 7 days. The TBBPA in soil was fully extracted by methanol and quantified by sampling.

Phylogenetic analysis

The 16S rRNA gene phylogenetic trees were established using the neighbor-joining algorithm. The sequences were aligned using CLUSTAL W. Phylogenetic trees were constructed and bootstrapped with 1,000 replicates of each sequence using MEGA version 7.0. OrthoFinder v2.3.3 software was used to analyze and extract the universal single-copy genes in the MAGs and strain genomes⁸¹. After multiple sequence alignment in MUSCLE, the phylogenomic trees based on bacterial genomes were constructed using IQ-TREE v1.6.11⁸². The phylogenetic trees and phylogenomic trees were further modified using iTOL (https://itol.embl.de/).

Analytical method

TBBPA and related compounds were analyzed by high-performance liquid chromatography (HPLC, Agilent 1260 Infinity II, USA) using an HPLC system with a variable wavelength detector (VWD, G7114A, Germany) and an extend-C18 column (4.6 × 250 mm, 5 μm). The mobile phase was 0.1% (v/v) formic acid in ultrapure water (solvent A) and 80% (v/v) methanol in ultrapure water (solvent B), which was applied at a flow rate of 1 mL/min for 20 min. The injection volume was 10 μL. The UV signals were recorded at 220 and 280 nm, corresponding to the maximum absorbance of TBBPA and BPA, respectively. The concentration of free bromine ions in the culture system was measured by an ion chromatograph DX600. The intermediate products of TBBPA biodegradation were extracted by ethyl acetate and determined by liquid chromatography/high-resolution electrospray ionization mass spectrometry (LC/ESI-MS) analyses using an Agilent Accurate-Mass-Q-TOF MS 6520B system. The fragmentor and capillary voltages were set at 130 and 3500 V, respectively. Full-scan spectra were acquired over a scan range of m/z 80–2000 at 1.03 spectra s⁻¹ in positive ionization mode.

Acknowledgements

This work was funded by the National Natural Science Foundation of China (Grants Nos. 41991333 and 32270084) and was supported by the Ministry of Science and Technology of the People’s Republic of China (No. 2019YFA0905500). We thank Wen-Zhao Wang (Institute of Microbiology, Chinese Academy of Sciences, China) for his help with the operation of high-resolution mass spectrometry.

Author contributions

T.W.: original draft, methodology, investigation, data analysis; S.Z.G., Y.Z., X.Z.Z., C.G.R., F.L.L., and R.H.W.: investigation, data analysis; D.F.L. and H.Z.Z.: supervision and review; C.Y.J., X.H.S., and S.J.L.: conceptualization, supervision, editing, and finalizing the paper, project administration, and funding acquisition.

Data availability

The 16S rRNA gene amplicons data and metagenomes data have been deposited in the National Center for Biotechnology Information GenBank repository (accession numbers: SRR31160755-SRR31160844; SRR31326938-SRR31326943). The strains used in the experiments are publicly available in the China General Microbiological Culture Collection Center (CGMCC).

Code availability

The source codes for the amplicon analyses and metagenomic analyses are available on GitHub (https://github.com/YongxinLiu/EasyAmplicon and https://github.com/YongxinLiu/EasyMetagenome).

Competing interests

The authors declare no competing interests.

Supplementary information

The online version contains supplementary material available at https://doi.org/10.1038/s41522-025-00777-9.

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

References

1. Thompson, LR et al. A communal catalogue reveals Earth’s multiscale microbial diversity. Nature; 2017; 551, pp. 457-463.1:CAS:528:DC%2BC2sXhslehurzL [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/29088705][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6192678]

2. Philippot, L; Raaijmakers, JM; Lemanceau, P; van der Putten, WH. Going back to the roots: the microbial ecology of the rhizosphere. Nat. Rev. Microbiol.; 2013; 11, pp. 789-799.1:CAS:528:DC%2BC3sXhsVyqt7nI [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/24056930]

3. Zhou, K; Qiao, K; Edgar, S; Stephanopoulos, G. Distributing a metabolic pathway among a microbial consortium enhances production of natural products. Nat. Biotechnol.; 2015; 33, pp. 377-383.1:CAS:528:DC%2BC2MXislOltA%3D%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/25558867][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4867547]

4. Compant, S; Samad, A; Faist, H; Sessitsch, A. A review on the plant microbiome: ecology, functions, and emerging trends in microbial application. J. Adv. Res.; 2019; 19, pp. 29-37.1:CAS:528:DC%2BC1MXitFKktrvP [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/31341667][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6630030]

5. Liu, Y; Zhu, A; Tan, H; Cao, L; Zhang, R. Engineering banana endosphere microbiome to improve Fusarium wilt resistance in banana. Microbiome; 2019; 7, [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/31092296][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6521393]74.

6. Hu, S et al. A synergistic consortium involved in rac-dichlorprop degradation as revealed by DNA stable isotope probing and metagenomic analysis. Appl. Environ. Microbiol.; 2021; 87, pp. e01562-01521.1:CAS:528:DC%2BB3MXisVOjtrzE [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/34524896][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8552887]

7. Meyer-Cifuentes, IE et al. Synergistic biodegradation of aromatic-aliphatic copolyester plastic by a marine microbial consortium. Nat. Commun.; 2020; 11, 1:CAS:528:DC%2BB3cXitlKltrjE [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/33188179][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7666164]5790.

8. Khanna, S et al. A novel microbiome therapeutic increases gut microbial diversity and prevents recurrent Clostridiumdifficile infection. J. Infect. Dis.; 2016; 214, pp. 173-181. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/26908752]

9. Van der Lelie, D et al. Rationally designed bacterial consortia to treat chronic immune-mediated colitis and restore intestinal homeostasis. Nat. Commun.; 2021; 12, [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/34050144][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8163890]3105.

10. Toju, H et al. Core microbiomes for sustainable agroecosystems. Nat. Plants; 2018; 4, pp. 247-257. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/29725101]

11. Lawson, CE et al. Common principles and best practices for engineering microbiomes. Nat. Rev. Microbiol.; 2019; 17, pp. 725-741.1:CAS:528:DC%2BC1MXhvVarsb7M [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/31548653][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8323346]

12. Hu, H et al. Guided by the principles of microbiome engineering: accomplishments and perspectives for environmental use. mLife; 2022; 1, pp. 382-398. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/38818482][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10989833]

13. Vega, NM; Gore, J. Simple organizing principles in microbial communities. Curr. Opin. Microbiol.; 2018; 45, pp. 195-202. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/30503875]

14. Chang, C-Y et al. Engineering complex communities by directed evolution. Nat. Ecol. Evol.; 2021; 5, pp. 1011-1023. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/33986540][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8263491]

15. De Roy, K; Marzorati, M; Van den Abbeele, P; Van de Wiele, T; Boon, N. Synthetic microbial ecosystems: an exciting tool to understand and apply microbial communities. Environ. Microbiol.; 2014; 16, pp. 1472-1481. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/24274586]

16. Jia, X et al. Design, analysis and application of synthetic microbial consortia. Synth. Syst. Biotechnol.; 2016; 1, pp. 109-117. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/29062933][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5640696]

17. Xie, X; Müller, N. Enhanced aniline degradation by Desulfatiglans anilini in a synthetic microbial community with the phototrophic purple sulfur bacterium Thiocapsa roseopersicina. Syst. Appl. Microbiol.; 2019; 42, 125998.1:CAS:528:DC%2BC1MXhsVals7nJ [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/31345671]

18. Díaz-García, L et al. Dilution-to-stimulation/extinction method: a combination enrichment strategy to develop a minimal and versatile lignocellulolytic bacterial consortium. Appl. Environ. Microbiol.; 2021; 87, pp. e02427-02420. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/33127812][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7783344]

19. Wang, X et al. Nitrogen transfer and cross-feeding between Azotobacter chroococcum and Paracoccus aminovorans promotes pyrene degradation. ISME J.; 2023; 17, pp. 2169-2181.1:CAS:528:DC%2BB3sXitVelt7jI [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/37775536][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10689768]

20. Borchert, E; Hammerschmidt, K; Hentschel, U; Deines, P. Enhancing microbial pollutant degradation by integrating eco-evolutionary principles with environmental biotechnology. Trends Microbiol.; 2021; 29, pp. 908-918.1:CAS:528:DC%2BB3MXntFWhsbs%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/33812769]

21. Zelezniak, A et al. Metabolic dependencies drive species co-occurrence in diverse microbial communities. Proc. Natl Acad. Sci. USA; 2015; 112, pp. 6449-6454.1:CAS:528:DC%2BC2MXns1Oks78%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/25941371][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4443341]

22. Xu, X et al. Modeling microbial communities from atrazine contaminated soils promotes the development of biostimulation solutions. ISME J.; 2019; 13, pp. 494-508.1:CAS:528:DC%2BC1cXhvVyht7fI [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/30291327]

23. Ruan, Z et al. Engineering natural microbiomes toward enhanced bioremediation by microbiome modeling. Nat. Commun.; 2024; 15, 1:CAS:528:DC%2BB2cXhtl2msbrJ [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/38824157][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11144243]4694.

24. Yu, K et al. An integrated meta-omics approach reveals substrates involved in synergistic interactions in a bisphenol A (BPA)-degrading microbial community. Microbiome; 2019; 7, [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/30728080][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6366072]16.

25. Zhou, H; Yin, N; Faiola, F. Tetrabromobisphenol A (TBBPA): a controversial environmental pollutant. Environ. Sci. (China); 2020; 97, pp. 54-66.1:CAS:528:DC%2BB38XislKmurzN

26. Parsons, A et al. Molecular mechanisms and tissue targets of brominated flame retardants, BDE-47 and TBBPA, in embryo-larval life stages of zebrafish (Danio rerio). Aquat. Toxicol.; 2019; 209, pp. 99-112.1:CAS:528:DC%2BC1MXivVShs7w%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/30763833]

27. Wang, W et al. A comparative assessment of human exposure to tetrabromobisphenol A and eight bisphenols including bisphenol A via indoor dust ingestion in twelve countries. Environ. Int.; 2015; 83, pp. 183-191.1:CAS:528:DC%2BC2MXht1Wrt7vO [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/26177148]

28. Arbeli, Z; Ronen, Z. Enrichment of a microbial culture capable of reductive debromination of the flame retardant Tetrabromobisphenol-A, and identification of the intermediate metabolites produced in the process. Biodegradation; 2003; 14, pp. 385-395.1:CAS:528:DC%2BD3sXoslWkur0%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/14669869]

29. Zhang, C et al. A humin-dependent Dehalobacter species is involved in reductive debromination of tetrabromobisphenol A. Chemosphere; 2013; 92, pp. 1343-1348.1:CAS:528:DC%2BC3sXptlGgsbs%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/23769323]

30. Macêdo, WV; Sánchez, FE; Zaiat, M. What drives Tetrabromobisphenol A degradation in biotreatment systems?. Rev. Environ. Sci. Bio.; 2021; 20, pp. 729-750.

31. Ma, Y et al. Effects of Cu²⁺ and humic acids on degradation and fate of TBBPA in pure culture of Pseudomonas sp. strain CDT. J. Environ. Sci. (China); 2017; 62, pp. 60-67.1:CAS:528:DC%2BB3cXis1Slsb3I [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/29289293]

32. Xu, S; Wang, Y-F; Yang, L-Y; Ji, R; Miao, A-J. Transformation of tetrabromobisphenol A by Rhodococcus jostii RHA1: effects of heavy metals. Chemosphere; 2018; 196, pp. 206-213.1:CAS:528:DC%2BC1cXjt1CgsQ%3D%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/29304458]

33. Gu, C et al. Extracellular degradation of tetrabromobisphenol A via biogenic reactive oxygen species by a marine Pseudoalteromonas sp. Water Res.; 2018; 142, pp. 354-362.1:CAS:528:DC%2BC1cXhtFaktr%2FP [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/29908463]

34. Gu, C et al. Biogenic fenton-like reaction involvement in cometabolic degradation of Tetrabromobisphenol A by Pseudomonas sp. fz. Environ. Sci. Technol.; 2016; 50, pp. 9981-9989.1:CAS:528:DC%2BC28Xhtl2ksL3I [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/27556415]

35. Peng, X et al. Efficient biodegradation of tetrabromobisphenol A by the novel strain Enterobacter sp. T2 with good environmental adaptation: kinetics, pathways and genomic characteristics. J. Hazard. Mater.; 2022; 429, 128335.1:CAS:528:DC%2BB38Xis1Wrs74%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/35121290]

36. Mee, M et al. Syntrophic exchange in synthetic microbial communities. Proc. Natl Acad. Sci. USA; 2014; 111, pp. E2149-E2156.1:CAS:528:DC%2BC2cXntFGis7s%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/24778240][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4034247]

37. Deng, Y; Li, B; Zhang, T. Bacteria that make a meal of sulfonamide antibiotics: blind spots and emerging opportunities. Environ. Sci. Technol.; 2018; 52, pp. 3854-3868.1:CAS:528:DC%2BC1cXjs1Kjtb0%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/29498514]

38. Ghosh, S; Chowdhury, R; Bhattacharya, P. Mixed consortia in bioprocesses: role of microbial interactions. Appl. Microbiol. Biotechnol.; 2016; 100, pp. 4283-4295.1:CAS:528:DC%2BC28XlsVWnur8%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/27037693]

39. Zhu, H-Z; Jiang, M-Z; Zhou, N; Jiang, C-Y; Liu, S-J. Submerged macrophytes recruit unique microbial communities and drive functional zonation in an aquatic system. Appl. Microbiol. Biotechnol.; 2021; 105, pp. 7517-7528.1:CAS:528:DC%2BB3MXitVegsbzP [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/34519857]

40. Jiang, M-Z et al. Droplet microfluidics-based high-throughput bacterial cultivation for validation of taxon pairs in microbial co-occurrence networks. Sci. Rep.; 2022; 12, 1:CAS:528:DC%2BB38Xisl2qt73F [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/36307549][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9616874]18145.

41. Liu, A et al. Transformation/degradation of tetrabromobisphenol A and its derivatives: a review of the metabolism and metabolites. Environ. Pollut.; 2018; 243, pp. 1141-1153. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/30261454]

42. Li, F et al. Enhanced transformation of tetrabromobisphenol A by nitrifiers in nitrifying activated sludge. Environ. Sci. Technol.; 2015; 49, pp. 4283-4292.1:CAS:528:DC%2BC2MXjvFynu78%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/25754048]

43. Jiang, M-Z et al. Gut microbial interactions based on network construction and bacterial pairwise cultivation. Sci. China Life Sci.; 2024; 67, pp. 1751-1762.1:CAS:528:DC%2BB2cXpslylsLg%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/38600293]

44. Xu, C et al. Christensenella minuta interacts with multiple gut bacteria. Front. Microbiol.; 2024; 15, [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/38440147][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10910051]1301073.

45. Li, T; Zhao, Z; Wang, Q; Xie, P; Ma, J. Strongly enhanced fenton degradation of organic pollutants by cysteine: an aliphatic amino acid accelerator outweighs hydroquinone analogues. Water Res.; 2016; 105, pp. 479-486.1:CAS:528:DC%2BC28XhsFGqtr7O [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/27668992]

46. Li, Y-J et al. The degradation mechanisms of Rhodopseudomonas palustris toward hexabromocyclododecane by time-course transcriptome analysis. Chem. Eng. J.; 2021; 425, 130489.1:CAS:528:DC%2BB3MXht1Oksb%2FJ

47. Sasaki, M; Maki, J-i; Oshiman, K-i; Matsumura, Y; Tsuchido, T. Biodegradation of bisphenol A by cells and cell lysate from Sphingomonas sp. strain AO1. Biodegradation; 2005; 16, pp. 449-459.1:CAS:528:DC%2BD2MXhsF2mu7g%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/15865158]

48. Liang, Z; Li, G; Mai, B; Ma, H; An, T. Application of a novel gene encoding bromophenol dehalogenase from Ochrobactrum sp. T in TBBPA degradation. Chemosphere; 2019; 217, pp. 507-515.1:CAS:528:DC%2BC1cXit1Snt7nE [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/30445395]

49. Ronen, Z; Abeliovich, A. Anaerobic-aerobic process for microbial degradation of Tetrabromobisphenol A. Appl. Environ. Microbiol.; 2000; 66, pp. 2372-2377.1:CAS:528:DC%2BD3cXjvFWgu7g%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/10831413][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC110535]

50. Albright, MBN et al. Solutions in microbiome engineering: prioritizing barriers to organism establishment. ISME J.; 2022; 16, pp. 331-338. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/34420034]

51. Ahmad, HA et al. The environmental distribution and removal of emerging pollutants, highlighting the importance of using microbes as a potential degrader: a review. Sci. Total Environ.; 2022; 809, 151926.1:CAS:528:DC%2BB3MXisleltLzL [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/34838908]

52. Okeke, ES et al. Association of tetrabromobisphenol A (TBBPA) with micro/nano-plastics: a review of recent findings on ecotoxicological and health impacts. Sci. Total Environ.; 2024; 927, 172308.1:CAS:528:DC%2BB2cXoslGjsbc%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/38599396]

53. Gu, C et al. Aerobic cometabolism of tetrabromobisphenol A by marine bacterial consortia. Environ. Sci. Pollut. Res. Int.; 2019; 26, pp. 23832-23841.1:CAS:528:DC%2BC1MXhtFymtLnK [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/31209756]

54. Huang, W et al. Co-metabolic degradation of tetrabromobisphenol A by Pseudomonas aeruginosa and its auto-poisoning effect caused during degradation process. Ecotoxicol. Environ. Saf.; 2020; 202, 110919.1:CAS:528:DC%2BB3cXhtlCltbnI [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/32800254]

55. Liu, X; Chen, K; Chuang, S; Xu, X; Jiang, J. Shift in bacterial community structure drives different atrazine-degrading efficiencies. Front. Microbiol.; 2019; 10, 88. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/30761118][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6363660]

56. Lin, X et al. Characterization of two keystone taxa, sulfur-oxidizing, and nitrate-reducing bacteria, by tracking their role transitions in the benzo[a]pyrene degradative microbiome. Microbiome; 2023; 11, 1:CAS:528:DC%2BB3sXhtlKrsbnE [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/37355612][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10290299]139.

57. Baldani, JI et al. Emended description of Herbaspirillum; inccusion of [Pseudomonas] rubrisubalbicans, a mild plant pathogen, as Herbaspirillum rubrisubalbicans comb. nov.; and classification of a group of clinical isolates (EF Group 1) as Herbaspirillum species 3. Int. J. Syst. Bacteriol.; 1996; 46, pp. 802-810.1:STN:280:DyaK28zns1ygsw%3D%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/8782693]

58. Liang, Y; Ma, A; Zhuang, G. Construction of environmental synthetic microbial consortia: based on engineering and ecological principles. Front. Microbiol.; 2022; 13, 829717. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/35283862][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8905317]

59. Wu, T et al. The sulfonamide-resistance dihydropteroate synthase gene is crucial for efficient biodegradation of sulfamethoxazole by Paenarthrobacter species. Appl. Microbiol. Biotechnol.; 2023; 107, pp. 5813-5827.1:CAS:528:DC%2BB3sXhsVOmurbO [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/37439835]

60. Reasoner, DJ; Geldreich, EE. A new medium for the enumeration and subculture of bacteria from potable water. Appl. Environ. Microbiol.; 1985; 49, pp. 1-7.1:CAS:528:DyaL2MXhsVGltr8%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/3883894][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC238333]

61. Liu, Y-X et al. A practical guide to amplicon and metagenomic analysis of microbiome data. Protein Cell; 2021; 12, pp. 315-330. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/32394199]

62. Rognes, T; Flouri, T; Nichols, B; Quince, C; Mahé, F. VSEARCH: a versatile open source tool for metagenomics. PeerJ; 2016; 4, e2584. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/27781170][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5075697]

63. Edgar, RC. Search and clustering orders of magnitude faster than BLAST. Bioinformatics; 2010; 26, pp. 2460-2461.1:CAS:528:DC%2BC3cXht1WhtbzM [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/20709691]

64. Cole, JR et al. Ribosomal Database Project: data and tools for high throughput rRNA analysis. Nucleic Acids Res.; 2014; 42, pp. D633-D642.1:CAS:528:DC%2BC2cXoslGk [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/24288368]

65. Gu, Z. Complex heatmap visualization. iMeta; 2022; 1, 1:CAS:528:DC%2BB2MXosFKrtrk%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/38868715][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10989952]e43.

66. Segata, N et al. Metagenomic biomarker discovery and explanation. Genome Biol.; 2011; 12, [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/21702898][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3218848]R60.

67. Yu, G; Smith, DK; Zhu, H; Guan, Y; Lam, TT-Y. ggtree: an r package for visualization and annotation of phylogenetic trees with their covariates and other associated data. Methods Ecol. Evol.; 2017; 8, pp. 28-36.

68. Li, D; Liu, C-M; Luo, R; Sadakane, K; Lam, T-W. MEGAHIT: an ultra-fast single-node solution for large and complex metagenomics assembly via succinct de Bruijn graph. Bioinformatics; 2015; 31, pp. 1674-1676.1:CAS:528:DC%2BC28XhtFyltL3N [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/25609793]

69. Hyatt, D et al. Prodigal: prokaryotic gene recognition and translation initiation site identification. BMC Bioinform.; 2010; 11, 119.

70. Steinegger, M; Söding, J. MMseqs2 enables sensitive protein sequence searching for the analysis of massive data sets. Nat. Biotechnol.; 2017; 35, pp. 1026-1028.1:CAS:528:DC%2BC2sXhs1GqsLzE [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/29035372]

71. Buchfink, B; Xie, C; Huson, DH. Fast and sensitive protein alignment using DIAMOND. Nat. Methods; 2015; 12, pp. 59-60.1:CAS:528:DC%2BC2cXhvFKlsrzN [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/25402007]

72. Uritskiy, GV; DiRuggiero, J; Taylor, J. MetaWRAP—a flexible pipeline for genome-resolved metagenomic data analysis. Microbiome; 2018; 6, pp. 1-13.

73. Kang, DD et al. MetaBAT 2: an adaptive binning algorithm for robust and efficient genome reconstruction from metagenome assemblies. PeerJ; 2019; 7, [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/31388474][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6662567]e7359.

74. Wu, Y-W; Simmons, BA; Singer, SW. MaxBin 2.0: an automated binning algorithm to recover genomes from multiple metagenomic datasets. Bioinformatics; 2016; 32, pp. 605-607.1:CAS:528:DC%2BC28XhsVWhur3F [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/26515820]

75. Alneberg, J et al. Binning metagenomic contigs by coverage and composition. Nat. Methods; 2014; 11, pp. 1144-1146.1:CAS:528:DC%2BC2cXhsFOksrbF [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/25218180]

76. Dröge, J; Gregor, I; McHardy, AC. Taxator-tk: precise taxonomic assignment of metagenomes by fast approximation of evolutionary neighborhoods. Bioinformatics; 2015; 31, pp. 817-824. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/25388150]

77. Seemann, T. Prokka: rapid prokaryotic genome annotation. Bioinformatics; 2014; 30, pp. 2068-2069.1:CAS:528:DC%2BC2cXhtFCrtLjI [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/24642063]

78. Chin, CS et al. Nonhybrid, finished microbial genome assemblies from long-read SMRT sequencing data. Nat. Methods; 2013; 10, 563.1:CAS:528:DC%2BC3sXntVaks74%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/23644548]

79. Delcher, AL; Bratke, KA; Powers, EC; Salzberg, SL. Identifying bacterial genes and endosymbiont DNA with Glimmer. Bioinformatics; 2007; 23, pp. 673-679.1:CAS:528:DC%2BD2sXkt1GhtL8%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/17237039]

80. Zhou, Y; Liang, YJ; Lynch, KH; Dennis, JJ; Wishart, DS. PHAST: a fast phage search tool. Nucleic Acids Res.; 2011; 39, pp. W347-W352.1:CAS:528:DC%2BC3MXosVOmtrg%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/21672955][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3125810]

81. Emms, DM; Kelly, S. OrthoFinder: solving fundamental biases in whole genome comparisons dramatically improves orthogroup inference accuracy. Genome Biol.; 2015; 16, [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/26243257][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4531804]157.

82. Nguyen, L-T; Schmidt, HA; von Haeseler, A; Minh, BQ. IQ-TREE: a fast and effective stochastic algorithm for estimating maximum-likelihood phylogenies. Mol. Biol. Evol.; 2015; 32, pp. 268-274.1:CAS:528:DC%2BC2MXivFGltrs%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/25371430]