Molecular Mechanism and Genetic Determinants of Buprofezin Degradation

ABSTRACT

Buprofezin is a widely used insect growth regulator whose residue has been frequently detected in the environment, posing a threat to aquatic organisms and nontarget insects. Microorganisms play an important role in the degradation of buprofezin in the natural environment. However, the relevant catabolic pathway has not been fully characterized, and the molecular mechanism of catabolism is still completely unknown. Rhodococcus qingshengii YL-1 can utilize buprofezin as a sole source of carbon and energy for growth. In this study, the upstream catabolic pathway in strain YL-1 was identified using tandem mass spectrometry. Buprofezin is composed of a benzene ring and a heterocyclic ring. The degradation is initiated by the dihydroxylation of the benzene ring and continues via dehydrogenation, aromatic ring cleavage, breaking of an amide bond, and the release of the heterocyclic ring 2-tert-butylimino-3-isopropyl-1,3,5-thiadiazinan-4-one (2-BI). A buprofezin degradation-deficient mutant strain YL-0 was isolated. A comparative genomic analysis combined with gene deletion and complementation experiments revealed that the gene cluster bfzBA3A4A1A2C is responsible for the upstream catabolic pathway of buprofezin. The bfzA3A4A1A2 cluster encodes a novel Rieske nonheme iron oxygenase (RHO) system that is responsible for the dihydroxylation of buprofezin at the benzene ring; bfzB is involved in dehydrogenation, and bfzC is in charge of benzene ring cleavage. Furthermore, the products of bfzBA3A4A1A2C can also catalyze dihydroxylation, dehydrogenation, and aromatic ring cleavage of biphenyl, flavanone, flavone, and bifenthrin. In addition, a transcriptional study revealed that bfzBA3A4A1A2C is organized in one transcriptional unit that is constitutively expressed in strain YL-1.

IMPORTANCE There is an increasing concern about the residue and environmental fate of buprofezin. Microbial metabolism is an important mechanism responsible for the buprofezin degradation in the natural environment. However, the molecular mechanism and genetic determinants of microbial degradation of buprofezin have not been well identified. This work revealed that gene cluster bfzBA3A4A1A2C is responsible for the upstream catabolic pathway of buprofezin in Rhodococcus qingshengii YL-1. The products of bfzBA3A4A1A2C could also degrade bifenthrin, a widely used pyrethroid insecticide. These findings enhance our understanding of the microbial degradation mechanism of buprofezin and benefit the application of strain YL-1 and bfzBA3A4A1A2C in the bioremediation of buprofezin contamination.

INTRODUCTION

Buprofezin (2-tert-butylimino-3-isopropyl-5-phenyl-1,3,5-thiadiazinan-4-one), which was first synthesized and marketed by Nihon Nohyaku Co., Ltd. (Tokyo, Japan), in the 1980s, is an insect growth regulator that inhibits chitin biosynthesis and integumentary cuticle deposition in the immature developmental stage of homopteran pests (1–3). It has been applied extensively on rice, tea, potato, fruit, cotton, and vegetables and exhibits a persistent larvicidal action against the brown rice planthopper (Nilaparvata lugens) and the greenhouse whitefly (Trialeurodes vaporariorum) (4–6). Due to its widespread use, buprofezin residue has been frequently detected in food, soil, and drinking water (7–10). Toxicity studies have shown that buprofezin is toxic to some aquatic organisms and nontarget insects (11–15). Buprofezin is considered to have low acute toxicity to humans and other mammals (acute rat oral 50% lethal dose [LD₅₀] is 1,635 mg · kg⁻¹ in males). However, recent research indicated that buprofezin could perturb the energy metabolism of mouse liver (16). Buprofezin residue can be easily absorbed by the human body via the oral cavity, the skin, and the respiratory tract (17) and therefore has potentially adverse effects on human health.

Microbial degradation plays an important role in the elimination of xenobiotics (18). To date, several pure bacterial cultures that are capable of degrading buprofezin have been isolated. Paracoccus sp. strain BF3 (19) and Pseudomonas sp. strain DFS35-4 (20) were able to cometabolize buprofezin, while Rhodococcus sp. strain YL-1 (21) and Bacillus sp. strain BF5 (22) could use buprofezin as their sole sources of carbon and energy for growth. Strain YL-1 was isolated from a buprofezin-contaminated paddy field, and strain BF-5 was isolated from the activated sludge of a buprofezin-producing plant. Strains YL-1 and BF-5 catabolize buprofezin in a similar manner (Fig. 1). The upstream catabolic pathway results in the release of the heterocyclic ring (2-tert-butylimino-3-isopropyl-1,3,5-thiadiazinan-4-one [2-BI]), which is further degraded in the downstream catabolic pathway. The upstream catabolic pathway is critical for the degradation of buprofezin. However, the steps before the production of 2-BI have not been identified, and moreover, the genetic basis of the entire catabolic pathway is completely unknown.

FIG 1.

Proposed buprofezin catabolic pathway in R. qingshengii YL-1 and organization of the genes involved in the upstream catabolic pathway. (A) Proposed buprofezin catabolic pathway of strain YL-1. The intermediate metabolites that were previously identified are shown in green, and the intermediate metabolites and the involved enzymes that are identified in this study are shown in blue. BF-DD, buprofezindihydrodiol; DHBF, dihydroxybuprofezin; RCDB, aromatic ring-cleaved dihydroxybuprofezin; 2-BI, 2-tert-butylimino-3-isopropyl-1,3,5-thiadiazinan-4-one; N-BA, N-tert-butyl-thioformimidic acid formylaminomethyl ester; 2-IM, 2-isothiocyanato-2-methyl-propane; 2-IP, 2-isothiocyanato-propane. (B) Organization of the missing genes involved in buprofezin catabolism in strain YL-1.

Previous work showed that the inoculation of soil with strain YL-1 significantly accelerated the degradation of buprofezin in the soil (21), implying the potential use of this strain in bioremediation. However, the buprofezin-degrading phenotype of strain YL-1 is unstable when it grows in nutrient-rich medium. To improve the stability of the buprofezin-degrading ability of strain YL-1 through genetic engineering, this study aimed to elucidate the molecular mechanism and genetic determinants of buprofezin degradation in this strain.

RESULTS AND DISCUSSION

Identification of the upstream catabolic pathway of buprofezin.

To identify the upstream catabolic pathway of buprofezin in strain YL-1, the intermediate metabolites that are made before the production of 2-BI were analyzed using ultrahigh-performance liquid chromatography-tandem mass spectrometry (UHPLC-MS/MS). As shown in Fig. 2, three new compounds (A, B, and C) (Fig. 2A to C) were detected in addition to 2-BI (Fig. 2D). The molecular ion mass of compound A was m/z 340.1679 [M+H]⁺, which was in line with the protonated derivative of buprofezindihydrodiol (C₁₆H₂₆N₃O₃S⁺, m/z 340.1689) with a −2.9-ppm error. The molecular ion mass of compound B was m/z 338.1535 [M+H]⁺, which was in good agreement with the protonated derivative of dihydroxybuprofezin (C₁₆H₂₄N₃O₃S⁺, m/z 338.1533) with a 0.59-ppm error. The molecular ion mass of compound C was m/z 370.1432 [M+H]⁺, which was consistent with the protonated derivative of the aromatic ring-cleaved dihydroxybuprofezin (C₁₆H₂₄O₅N₃S⁺, m/z 370.1431) with a mass error of 0.27 ppm. Generally, a mass error between −5 ppm and 5 ppm is acceptable (23), so the following upstream buprofezin catabolic pathway was proposed (Fig. 1A): the benzene ring of buprofezin was initially dihydroxylated into buprofezindihydrodiol, which was subsequently dehydrogenated to generate dihydroxybuprofezin. Dihydroxybuprofezin was then subjected to aromatic ring cleavage, and finally, 2-BI was released through the breaking of the amide bond.

FIG 2.

UHPLC-MS/MS analysis of the intermediate metabolites in the upstream buprofezin catabolic pathway in R. qingshengii YL-1. The data analysis of the tandem mass spectrometry was in positive mode. (A) Tandem mass spectrometry of compound A. (B) Tandem mass spectrometry of compound B. (C) Tandem mass spectrometry of compound C. (D) Tandem mass spectrometry of 2-tert-butylimino-3-isopropyl-1,3,5-thiadiazinan-4-one (2-BI).

Isolation of the buprofezin degradation-deficient mutant strain YL-0.

When grown on 0.2× LB plates supplemented with 50 mg · liter⁻¹ buprofezin, colonies of strain YL-1 can form a transparent halo, probably due to the conversion of buprofezin into some relatively high-solubility intermediate products. However, this phenotype was unstable; some colonies lost the ability to produce the transparent halo after continuous transfer in LB broth, with one such mutant strain designated YL-0 (Fig. 3A). The mutant strain YL-0 showed no detectable degrading activity toward buprofezin (see Fig. S1 in the supplemental material), indicating that at least the genes responsible for the initial degradation step of buprofezin were lost. In addition, mutant strain YL-0 exhibited an obviously shorter lag phase than the wild type when grown in LB broth (Fig. 3B), implying that the expression of the lost genes is a metabolic burden to strain YL-1 under this culture condition.

FIG 3.

Differences between strain YL-1 and mutant strain YL-0. (A) Transparent rings formed by tested strains on 0.2× LB agar supplemented with 50 mg · liter⁻¹ buprofezin. (B) Growth curves of strain YL-1 and the mutant strain YL-0 in LB broth.

Prediction of buprofezin-catabolic genes via comparative genomic analysis.

To find the genes missing in the buprofezin degradation-deficient mutant through comparative genomic analysis, the complete genome of strain YL-1 and the draft genome of mutant strain YL-0 were sequenced. The complete genome of strain YL-1 revealed six replicons, consisting of one circular chromosome (6,367,154 bp, 62.45% G+C, 6,231 open reading frames [ORFs]), three circular plasmids (pYLC1 [423,076 bp, 62.46% G+C, 524 ORFs], pYLC2 [64,544 bp, 62.19% G+C, 77 ORFs], and pYLC3 [7,616 bp, 68.99% G+C, 524 ORFs]), and two putative linear plasmids (pYLL1 [393,079 bp, 61.95% G+C, 468 ORFs] and pYLL2 [339,489 bp, 61.15% G+C, 398 ORFs]). Collectively, the whole genome of strain YL-1 is 7,594,958 bp in length and has an average G+C content of 62.32%, with a total of 7,710 protein-coding genes being predicted. The draft genome sequence of strain YL-0 is 7,249,781 bp in length (62.37% G+C, 6,965 ORFs) and consists of 159 scaffolds. The genome sequence of strain YL-1 shares 83.2% symmetric identity with that of Rhodococcus qingshengii JCM 15477^T. The 16S rRNA gene sequence of strain YL-1 is identical to that of R. qingshengii JCM 15477^T (see Fig. S2). Furthermore, a phylogenetic tree based on an alignment of concatenated amino acid sequences from 30 proteins with housekeeping function was constructed to analyze the taxonomic position of strain YL-1 among the related species (Fig. S3). The result showed strain YL-1 is closest to R. qingshengii JCM 15477^T. Based on these data, strain YL-1 was classified as Rhodococcus qingshengii.

The results of the sequence comparison and the PCR amplification showed that, compared with strain YL-1, mutant strain YL-0 lost a fragment of approximately 200 kb (Fig. 1B) that is located on plasmid pYLC1 of strain YL-1 ranging from 159,918 to 365,552 bp. A total of 245 ORFs were predicted to be in this missing fragment. The candidate genes that are involved in the degradation of buprofezin were then predicted from these ORFs. Based on the upstream catabolic pathway proposed above, a gene cluster, bfzBA3A4A1A2C, was targeted. As shown in Fig. 1B and in Table 1, the genes in bfzA3A4A1A2 were predicted to encode a multicomponent Rieske nonheme iron dioxygenase (RHO). RHOs are typically involved in hydroxylating aromatic ring substrates (24) and are composed of a reductase that obtains electrons from NADPH, generally a Rieske ferredoxin (in a three-component system), which shuttles the electrons, and an oxygenase that performs the catalysis. BfzA1 and BfzA2 exhibit similarities to the corresponding α and β subunits of some aromatic ring-hydroxylating dioxygenases. BfzA1 shares 38% and 35% identity with HcaE from Escherichia coli HS (25) and BphA1 from Pseudomonas sp. strain KKS102 (26), respectively. BfzA2 shows 47% and 45% identity to BphA2 from Pseudomonas sp. strain KKS102 and BnzB from Pseudomonas putida F1 (27), respectively. BfzA3 and BfzA4 show similarities to the corresponding ferredoxin and ferredoxin reductase components of the electron transport chain (ETC) of RHOs. BfzA3 shares 50% and 48% identity with BphA3 from Rhodococcus jostii RHA1 (28) and TodB from P. putida F1 (27), respectively. BfzA4 shares 34% and 32% identity with HcaD from E. coli SMS-3-5 (29) and BedA from P. putida ML2 (30), respectively. All of these analyses indicated that bfzA3A4A1A2 is most likely responsible for the dioxygenation of buprofezin to generate buprofezindihydrodiol.

TABLE 1.

Deduced function of partial ORFs within the missing fragment of mutant strain YL-0

Gene	Position in pYLC1, product size (no. of amino acids)	Homologous protein (GenBank accession no.), source	Identity (%)	Proposed function
istB	160709–159918, 263	Helper protein for insertion element IS1415 (AF002247.1), Rhodococcus erythropolis NI86/21a	100	Helper protein for insertion element IS1415
istA	162250–160709, 513	Transposase for insertion element IS1415 (AF002247.1), Rhodococcus erythropolis NI86/21a	99	Transposase for insertion element IS1415
orf1	240243–241067, 274	Transcriptional repressor IclR (P16528.1), Escherichia coli K-12b	31	Hypothetical protein
bfzB	241178–242155, 325	Aryl-alcohol dehydrogenase (Q01752.1), Phanerochaete chrysosporiumb	24	Dehydrogenase
bfzA3	242279–242503, 74	Biphenyl 2,3-dioxygenase, ferredoxin component (Q53124.1), Rhodococcus jostii RHA1b	50	Buprofezin dioxygenase ferredoxin component
orf2	242624–243043, 139	HTH-type transcriptional regulator BetI (Q8YFY3.1), Brucella melitensisb	41	Hypothetical protein
orf3	243125–243250, 41	Mitochondrial Rho GTPase 3 (Q9MA88.1), Arabidopsis thalianab	34	Hypothetical protein
bfzA4	243409–244788, 459	Benzene 1,2-dioxygenase ferredoxin reductase (Q07946.1), Pseudomonas putida ML2b	32	Buprofezin dioxygenase reductase component
bfzA1	244896–246248, 450	3-Phenylpropionate/cinnamic acid dioxygenase α subunit (A8A344.1), Escherichia coli HSb	38	Buprofezin dioxygenase α subunit
bfzA2	246275–246814, 179	Biphenyl dioxygenase β subunit (Q52439.1), Pseudomonas sp. strain KKS102b	47	Buprofezin dioxygenase β subunit
bfzC	246862–247758, 298	Biphenyl-2,3-diol 1,2-dioxygenase (P08695.2), Pseudomonas pseudoalcaligenes KF707b	41	Aromatic ring-cleaving dioxygenase
tnpX	248831–247941, 296	Insertion element IS630 uncharacterized 39-kDa protein (P16943.1), Shigella sonneib	23	Transposase
istB	364011–363220, 263	Helper protein for insertion element IS1415 (AF002247.1), Rhodococcus erythropolis NI86/21a	100	Helper protein for insertion element IS1415
istA	365552–364011, 513	Transposase for insertion element IS1415 (AF002247.1), Rhodococcus erythropolis NI86/21b	99	Transposase for insertion element IS1415

^aThe top BLASTp hit whose function has been characterized was selected from the NCBI nonredundant protein sequences database.

^bThe top BLASTP hit whose function has been characterized was selected from the NCBI Swiss-Prot protein sequences database.

The product of bfzB consists of 325 amino acids and exhibits 24% identity to the aryl-alcohol dehydrogenase from Phanerochaete chrysosporium (31), which indicates that BfzB is probably involved in the dehydrogenation of buprofezindihydrodiol. BfzC shows 45% identity to BphC, the 2,3-dihydroxybiphenyl dioxygenase from Pseudomonas pseudoalcaligenes KF707 (32), implying that BfzC is likely responsible for the aromatic ring cleavage of dihydroxybuprofezin. Additionally, two genes, istA and istB, are located at both ends of the missing fragment. IstA and IstB share 99 and 100% identities with the transposase and helper protein, respectively, of insertion sequences of the IS21-related transposable element IS1415 in Rhodococcus erythropolis NI86/21 (33).

Gene cluster bfzBA3A4A1A2C is involved in the upstream catabolic pathway of buprofezin.

To confirm the bfzBA3A4A1A2C cluster is involved in the upstream catabolic pathway of buprofezin in strain YL-1, the plasmid pREBAC, which contains bfzBA3A4A1A2C, was introduced into mutant strain YL-0. The plasmid pREBAC restored the abilities of mutant strain YL-0 to form a transparent halo around the colony on 0.2× LB plates supplemented with 50 mg · liter⁻¹ buprofezin and to utilize buprofezin as the sole carbon and nitrogen sources (data not shown). High-performance liquid chromatography (HPLC) analysis showed that strain YL-0(pREBAC) was able to degrade buprofezin as the wild-type strain did (data not shown). These results indicated that gene cluster bfzBA3A4A1A2C is responsible for the dihydroxylation, dehydrogenation, and aromatic ring cleavage of buprofezin; in addition, the genes responsible for breaking the amide bond of aromatic ring-cleaved dihydroxybuprofezin and for the downstream catabolic pathway have not been lost in the mutant strain YL-0.

Confirmation of the roles of each gene in bfzBA4A3A1A2C.

To further validate the roles of each gene in bfzBA4A3A1A2C, unmarked deletion strains of these genes were individually constructed. BfzA1A2 is a putative oxygenase component that catalyzes the hydroxylation of buprofezin. Knockouts of bfzA1 or bfzA2 resulted in the inability of mutant strains YL-1 ΔbfzA1 and YL-1 ΔbfzA2 to degrade buprofezin. The complemented strains YL-1(pREA1) ΔbfzA1 and YL-1(pREA2) ΔbfzA2 regained their abilities to degrade buprofezin. The product of bfzA3 is a putative ferredoxin that transfers electrons that originated from reducing equivalents directly to BfzA1A2. Mutant strain YL-1 ΔbfzA3 lost the ability to degrade buprofezin, but this ability was restored by the introduction of the complementation plasmid pREA3, indicating that the dioxygenase component (BfzA1A2) has a high specificity for the ferredoxin component (BfzA3). BfzA4 is predicted to be a ferredoxin reductase that is responsible for obtaining electrons from NADPH and transferring them to BfzA3. Mutant strain YL-1 ΔbfzA4 showed a decreased ability to degrade buprofezin compared with that of its wild type. This result implied that another ferredoxin reductase in the strain can partially replace BfzA4. BfzB is deduced to catalyze the dehydrogenation of buprofezindihydrodiol. Deletion of bfzB led to the accumulation of buprofezindihydrodiol as well as trace amounts of dihydroxybuprofezin, indicating that another dehydrogenase in the host participates, to some extent, in the dehydrogenation of buprofezindihydrodiol. The accumulation of dihydroxybuprofezin was observed in the bfzC knockout mutant YL-1 ΔbfzC. The plasmid pREC bearing the entire bfzC complemented the mutation of strain YL-1 ΔbfzC, indicating that the product of bfzC is responsible for the aromatic ring cleavage of dihydroxybuprofezin (see Fig. S4).

Products of bfzBA4A3A1A2C could also attack biphenyl and partially degrade the structurally related compounds flavone, flavanone, and bifenthrin.

Since the sequence of bfzBA4A3A1A2C exhibited homology to the genes involved in the upstream catabolic pathway of biphenyl, we investigated whether the products of bfzBA4A3A1A2C could degrade biphenyl and its analogs, such as flavone, flavanone, and bifenthrin. As shown in Fig. 4, the strain YL-1 could metabolize the four tested substrates, and yellow-colored (for biphenyl, flavanone, and bifenthrin) or orange-colored (for flavone) intermediate metabolites appeared during metabolism. The aromatic ring-cleaved products of the four substrates were detected using UHPLC-MS/MS analysis, with mass errors between −5 ppm and 5 ppm (see Fig. S5). By contrast, the mutant strains YL-0 and YL-1 ΔbfzA1 showed no detectable catalytic activity toward the four substrates. No significant color change was observed in the reactions of mutant strain YL-1 ΔbfzC. All these results indicated that the products of bfzBA4A3A1A2C could also catalyze the dihydroxylation, dehydrogenation, and aromatic ring cleavage of biphenyl, flavone, flavanone, and bifenthrin.

FIG 4.

Degradation of biphenyl, bifenthrin, flavanone, and flavone by R. qingshengii YL-1 and its mutant strains. The intermediate metabolites were analyzed by UHPLC-MS/MS (see Fig. S5 in the supplemental material).

Bifenthrin is an effective pyrethroid insecticide that is widely used in both agricultural and urban environments (34–36). However, due to its high toxicity against aquatic organisms and its potential risk to human health, bifenthrin residue in the environment has drawn increasing public concern (37–39). In 2009, bifenthrin was listed in toxicity class II (moderately hazardous) by the World Health Organization (WHO). The previously reported biodegradation pathway of bifenthrin is initiated by the breaking of the carboxylesterase linkage (40), which is catalyzed by a pyrethroid-hydrolyzing carboxylesterase (41, 42). By contrast, we showed that strain YL-1 degraded bifenthrin in a different way, implying the potential use of strain YL-1 in the dissipation of bifenthrin residue in the environment.

Transcriptional analysis of the bfz cluster was carried out to investigate whether the enzyme-encoding genes were highly transcribed in response to the substrates buprofezin and biphenyl. The bfzBA4A3A1A2C genes were proven to be organized in one transcriptional unit via real-time PCR (RT-PCR) (see Fig. S6). Therefore, bfzA1 and bfzB were used as two representative genes to study the transcriptional levels of this transcriptional unit. Real-time quantitative PCR (RT-qPCR) showed that bfzBA4A3A1A2C genes were constitutively expressed and not enhanced by buprofezin or biphenyl.

Constitutive expression of bfzBA3A4A1A2C poses a metabolic burden to the growth of strain YL-1 in LB medium, which may be one of the reasons that mutant strain YL-0 exhibited an obviously shorter lag phase in contrast to that of its wild type. Furthermore, the 205,635-bp missing fragment is surrounded by two sets of transposase genes (istA and istB) (33). When grown in LB medium, strain YL-1 tends to lose the 205,635-bp fragment to alleviate the metabolic burden. On the other hand, strain YL-1 was isolated from a paddy field. It is not known whether or not the expression of bfzBA3A4A1A2C is still constitutive under those field conditions. If this cluster is still constitutively expressed, there should be a certain substrate that persistently exists in the paddy field that acts as a selective pressure to maintain the phenotype stability of the degrader YL-1. The metabolism of such a substrate facilitates the survival of strain YL-1.

Phylogenetic analysis of BfzA1.

It was proposed that RHOs could be divided into two major groups, one for catalyzing aromatic ring hydroxylation and one for C—O or C—N bond-cleaving reactions (43). The phylogenetic analysis of BfzA1, comparing it with the large subunits of 60 characterized RHOs, indicated that BfzA1 belongs to the group involved in aromatic ring hydroxylation. However, BfzA1 formed a new branch within this group. In another classification system suggested by Kweon et al., the RHOs are classified into five groups based on the analysis of the oxygenase and ETC components of the RHOs (44). According to this classification system, BfzA1A2A3A4 falls into the type IV RHOs, whose members generally contain a hetero-oligomer-type oxygenase (α_nβ_n), a [2Fe-2S]-type ferredoxin, and a glutathione reductase (GR)-type reductase. BfzA1 also forms a new branch within this group, indicating it is a new member of the type IV RHOs (Fig. 5A). Since bfzBA3A4A1A2C was homologous with the genes responsible for the upstream catabolic pathway of biphenyl, the sequence of BfzA1 was compared with those of the large subunits of the characterized biphenyl dioxygenases. As shown in Fig. 5B, the large subunits of biphenyl dioxygenases form a clade, while BfzA1 occupied a separate clade.

FIG 5.

Phylogenetic comparison of BfzA1 with the α subunits of characterized RHOs. The multiple alignment analysis of the amino acid sequences was performed with Clustal X 2.1. The neighbor-joining method was used to construct the phylogenetic unrooted tree with MEGA 5.0. Bootstrap percentages (based on 1,000 replications) >70% are shown on some branches. Each protein is shown as “protein name. strain name (GenBank accession no. of the protein).” (A) Phylogenetic tree constructed on the basis of the comparison of BfzA1 with 60 characterized RHOs. A total of 559 positions were used in the analysis. The clustering of RHOs according to their respective native substrate is clearly displayed in different colors. The magenta, green, blue, and purple colors correspond to the RHOs that catalyze the respective reactions: biphenyl, benzene, and substituted benzene hydroxylation; polyaromatic hydrocarbon and nitroarene hydroxylation; benzoate and substituted benzoate hydroxylation; and C—O/C—N bond cleavage. The RHOs are further classified into five types according to the classification system suggested by Kweon et al. (44), and these classifications are indicated by the outer ring label. Type I RHOs consist of an oxygenase and an ferredoxin-NADP+ reductase (FNR)_C-type reductase, type II RHOs consist of an oxygenase and an FNR_N-type reductase, type III RHOs consist of an oxygenase, a [2Fe-2S]-type ferredoxin, and an FNR_N-type reductase, type IV RHOs consist of a hetero-oligomeric oxygenase, a [2Fe-2S]-type ferredoxin, and a GR-type reductase, and type V RHOs consist of a hetero-oligomeric oxygenase, a [3Fe-4S]-type ferredoxin, and a GR-type reductase. The hosts of the proteins are as follows: BfzA1. YL-1, R. qingshengii YL-1; BphA1. TK102, Comamonas testosteroni TK102; BphA1. KKS102, Pseudomonas sp. strain KKS102; BphA. B-356, C. testosteroni B-356; IpBAa. RE204, P. putida RE204; BphA. LB400, Burkholderia xenovorans LB400; BphA1. KF707, P. pseudoalcaligenes KF707; BpdC1. M5, Rhodococcus sp. strain M5; BphA1. RHA1, R. jostii RHA1; BedC1. ML2, P. putida ML2; TodC1. F1, P. putida F1; TecA1. PS12, Burkholderia sp. strain PS12; TcbAa. P51, Pseudomonas sp. strain P51; HcaE. K-12, Escherichia coli K-12; BphA1. TA421, R. erythropolis TA421; DxnA1. RW1, Sphingomonas wittichii RW1; DfdA1. YK3, Terrabacter sp. strain YK3; PhnA1a. CHY-1, Sphingomonas sp. strain CHY-1; PhnA1. A5, Cycloclasticus sp. strain A5; ArhA1. A4, Sphingomonas sp. strain A4; PahAc. OUS82, P. putida OUS82; NahAc. NCIB 9816-4, Pseudomonas sp. strain NCIB 9816-4; NagAc. U2, Ralstonia sp. strain U2; DntAc. R34, B. cepacia R34; NtdAc. JS42, Pseudomonas sp. strain JS42; NbzAc. JS765, Comamonas sp. strain JS765; PhtA1. DBF63, Terrabacter sp. strain DBF63; PhtAa. 12B, Arthrobacter keyseri 12B; PhtAa. PYR-1, Mycobacterium vanbaalenii PYR-1; PdoA2. 6PY1, Mycobacterium sp. strain 6PY1; PhdA. KP7, Nocardioides sp. strain KP7; PdoA1. 6PY1, Mycobacterium sp. strain 6PY1; NidA3. PYR-1, M. vanbaalenii PYR-1; NidA. PYR-1, M. vanbaalenii PYR-1; TdnA1. UCC22, P. putida UCC22; ORF7Nc. 7N, Delftia acidovorans 7N; AtdA. YAA, Acinetobacter sp. strain YAA; AntA. ADP1, Acinetobacter sp. strain ADP1; AntA. CA10, Pseudomonas sp. strain CA10; CbdA. 2CBS, B. cepacia 2CBS; XylX. mt2, P. putida mt2; BenA. ADP1, Acinetobacter sp. strain ADP1; CmtAb. F1, P. putida F1; PsbAb. No.7, Rhodopseudomonas palustris No.7; BphA1e. B1, Sphingobium yanoikuyae B1; AhdA1e. P2, Sphingomonas sp. strain P2; NagG. U2, Ralstonia sp. strain U2; HybB. JB2, P. aeruginosa JB2; AhdA1d. P2, Sphingomonas sp. strain P2; PhnA1b. CHY-1, Sphingomonas sp. strain CHY-1; CarAa. CA10, P. resinovorans CA10; CarAa. J3, Janthinobacterium sp. strain J3; CarAa1. KA1, Sphingomonas sp. strain KA1; NdmA. BB5, P. putida CBB5; NdmB. CBB5, P. putida CBB5; DdmC. DI-6, P. maltophilia DI-6; TsaM. T-2, C. testosteroni T-2; VanA. HR199, Pseudomonas sp. strain HR199; VanA. ATCC 19151, Pseudomonas sp. strain ATCC 19151; LigX. SYK-6, S. paucimobilis SYK-6; PobA. POB310, P. pseudoalcaligenes POB310; BphA1. P6, R. globerulus P6; BphA1. R04, Rhodococcus sp. strain R04. (B) The phylogenetic tree constructed on the basis of comparing BfzA1 with the α subunits of the characterized biphenyl dioxygenases. A total of 487 positions were used in the analysis. The host of each protein is as follows: BphA. B-356, C. testosteroni B-356; BphA1. KKS102, Pseudomonas sp. strain KKS102; BphA. LB400, B. xenovorans LB400; BphA1. KF707, P. pseudoalcaligenes KF707; BphA1. RHA1, R. jostii RHA1; BphA1. P6, R. globerulus P6; BpdC1. M5, Rhodococcus sp. strain M5; BphA1. R04, Rhodococcus sp. strain R04.

Conclusion.

Taken together, the results from this work demonstrated that a novel bph-like gene cluster, bfzBA3A4A1A2C, is responsible for the upstream catabolic pathway of buprofezin in R. qingshengii YL-1. The product of bfzA3A4A1A2 initiated the catabolic pathway by incorporating two hydroxyl groups into the benzene ring to generate buprofezindihydrodiol, which was dehydrogenated by the product of bfzB to produce dihydroxybuprofezin. The product of bfzC catalyzed the aromatic ring cleavage of dihydroxybuprofezin, which was followed by the release of 2-BI through the breakage of the amide bond via an unidentified amidase. Additionally, BfzBA3A4A1A2C could also attack biphenyl and partially degrade the structurally related compounds, such as flavone, flavanone, and bifenthrin. These results lay a foundation for a better use of microbes and enzymes to dissipate buprofezin contamination in the environment.

MATERIALS AND METHODS

Chemicals.

Buprofezin (98% purity) was kindly provided by Zhenbo Agrochemical Co., Ltd., Jiangsu Province, China. Biphenyl (99% purity), flavone (99% purity), flavanone (99% purity), and bifenthrin (99% purity) were purchased from Dr. Ehrenstorfer-Schafers Chemical Co. (Augsburg, Germany).

Bacterial strains, plasmids, culture conditions, and oligonucleotides.

The bacterial strains and plasmids used in this study are listed in Table 2. The primers used for PCR are listed in Table 3. The Rhodococcus strains were cultivated aerobically at 30°C either in Luria-Bertani (LB) medium or in minimal salt medium (MSM) (45) supplemented with various carbon sources. The E. coli strains were grown at 37°C in LB medium. When required, the antibiotic kanamycin (Km) was added at 50 mg · liter⁻¹ for E. coli and 200 mg · liter⁻¹ for Rhodococcus.

TABLE 2.

Strains and plasmids used in this study^a

Strain or plasmid	Description	Source or reference
R. qingshengii strains
YL-1	Wild type, able to degrade buprofezin	21, CCTCC AB 2017132
YL-0	Mutant of strain YL-1, unable to degrade buprofezin	This study, CCTCC AB 2017133
YL-1 ΔbfzA1	R. qingshengii YL-1 ΔbfzA1	This study
YL-1 ΔbfzA2	R. qingshengii YL-1 ΔbfzA2	This study
YL-1 ΔbfzA3	R. qingshengii YL-1 ΔbfzA3	This study
YL-1 ΔbfzA4	R. qingshengii YL-1 ΔbfzA4	This study
YL-1 ΔbfzB	R. qingshengii YL-1 ΔbfzB	This study
YL-1 ΔbfzC	R. qingshengii YL-1 ΔbfzC	This study
E. coli DH5α	supE44 lacU169 (ϕ80lacZΔM15) recA1 endA1 hsdR17 thi-1 gyrA96 relA1	TaKaRa
Plasmids
pK18mobsacB	A sacB-based vector for unmarked allelic exchange, Kmr	55
pK18mobsacB-ΔbfzA1	pK18mobsacB containing the deletion construction of bfzA1	This study
pK18mobsacB-ΔbfzA2	pK18mobsacB containing the deletion construction of bfzA2	This study
pK18mobsacB-ΔbfzA3	pK18mobsacB containing the deletion construction of bfzA3	This study
pK18mobsacB-ΔbfzA4	pK18mobsacB containing the deletion construction of bfzA4	This study
pK18mobsacB-ΔbfzB	pK18mobsacB containing the deletion construction of bfzB	This study
pK18mobsacB-ΔbfzC	pK18mobsacB containing the deletion construction of bfzC	This study
pRESQ	Rhodococcus-E. coli shuttle vector, Kmr	56
pREBAC	pRESQ containing bfzBA3A4A1A2C	This study
pREA1	pRESQ containing bfzA1 for gene complementation analysis	This study
pREA2	pRESQ containing bfzA2 for gene complementation analysis	This study
pREA3	pRESQ containing bfzA3 for gene complementation analysis	This study
pREC	pRESQ containing bfzC for gene complementation analysis	This study

^aKm^r, kanamycin resistance; CCTCC, China Center for Type Culture Collection.

TABLE 3.

Primers used in the study

Primer	Sequence (5′ to 3′)	Purpose
pREbfzABC-F	ACCGAGCTCAGATCTACTAGTGTGTCCGGCACGAAAGTGTC	Amplification of the fragment containing bfzBA3A4A1A2C for functional study
pREbfzABC-R	ACACTGGCGGCCGTTACTAGTTCAGCTCGGCTGAGACGC
KO-bfzA1u-F	TCCTCTAGAGTCGACCTGCAGAGGATCGAGCATTGGGACA	Amplification of upstream sequence of bfzA1 for gene knockout
KO-bfzA1u-R	ATGCTCGTTGGTTGCGTCTCAATCTCGCCTGTCTCCA
KO-bfzA1d-F	TGGAGACAGGCGAGATTGAGACGCAACCAACGAGCAT	Amplification of downstream sequence of bfzA1 for gene knockout
KO-bfzA1d-R	GCCAAGCTTGCATGCCTGCAGTCAAGTTACGGGACAGAAGC
KO-bfzA2u-F	TCCTCTAGAGTCGACCTGCAGAGTGGCGAAACCGACCTG	Amplification of upstream sequence of bfzA2 for gene knockout
KO-bfzA2u-R	TAGCGGTGGTAGCGATGAGACGAAATCTCCTGGTTTGGT
KO-bfzA2d-F	ACCAAACCAGGAGATTTCGTCTCATCGCTACCACCGCTA	Amplification of downstream sequence of bfzA2 for gene knockout
KO-bfzA2d-R	GCCAAGCTTGCATGCCTGCAGGACGAAAGTGGAATCACTGGAT
KO-bfzA3u-F	TCCTCTAGAGTCGACCTGCAGCGTGCGGCAAGGTGTTAT	Amplification of upstream sequence of bfzA3 for gene knockout
KO-bfzA3u-R	TCAAAACAAGCCGGGCGTGGCGTGGTTTTGCGGATTGGAT
KO-bfzA3d-F	ATCCAATCCGCAAAACCACGCCACGCCCGGCTTGTTTTGA	Amplification of downstream sequence of bfzA3 for gene knockout
KO-bfzA3d-R	GCCAAGCTTGCATGCCTGCAGGGCTTCCTCACCGTAGATG
KO-bfzA4u-F	TCCTCTAGAGTCGACCTGCAGGAGTTGGCGCGGGAAATCGTGTCGT	Amplification of upstream sequence of bfzA4 for gene knockout
KO-bfzA4u-R	GCCTATCGACATGTTCCACGTGATGATCGGCTGGCAGTCAACGAGTCAGC
KO-bfzA4d-F	GCTGACTCGTTGACTGCCAGCCGATCATCACGTGGAACATGTCGATAGGC	Amplification of downstream sequence of bfzA4 for gene knockout
KO-bfzA4d-R	GCCAAGCTTGCATGCCTGCAGCGAATGTTGCAAACACGAATCCCTT
KO-bfzBu-F	TCCTCTAGAGTCGACCTGCAGGGGATCGAGCAGATTGGG	Amplification of upstream sequence of bfzB for gene knockout
KO-bfzBu-R	TGGGTGTAGAGGTTCGCCGCCACCGTCTGGCGAAGTA
KO-bfzBd-F	TACTTCGCCAGACGGTGGCGGCGAACCTCTACACCCA	Amplification of downstream sequence of bfzB for gene knockout
KO-bfzBd-R	GCCAAGCTTGCATGCCTGCAGTTCCCAACAGGACACGAGTCAG
KO-bfzCu-F	TCCTCTAGAGTCGACCTGCAGTGACCCTCAGCACTGAAATCCGCAC	Amplification of upstream sequence of bfzC for gene knockout
KO-bfzCu-R	ACTCCCCGGCAAGCTACCCAGTCCGCTGCTGATGTAG
KO-bfzCd-F	CTACATCAGCAGCGGACTGGGTAGCTTGCCGGGGAGT	Amplification of downstream sequence of bfzC for gene knockout
KO-bfzCd-R	GCCAAGCTTGCATGCCTGCAGCCGCCACCAGGAGTTTCT
GC-bfzA1-F	ACCGAGCTCAGATCTACTAGTATGCCAATCGATCAAGACGA	Amplification of bfzA1 for gene complementation
GC-bfzA1-R	ACACTGGCGGCCGTTACTAGTTCAGAGAGTGTCGCCGGGTG
GC-bfzA2-F	ACCGAGCTCAGATCTACTAGTATGACCCTCAGCACTGAAATC	Amplification of bfzA2 for gene complementation
GC-bfzA2-R	ACACTGGCGGCCGTTACTAGTTCAGAAGAAGTTGCTCAAGT
GC-bfzA3-F	ACCGAGCTCAGATCTACTAGTATGCCCCTCATACGAGTATGC	Amplification of bfzA3 for gene complementation
GC-bfzA3-R	ACACTGGCGGCCGTTACTAGTCTAGTCCACCTCGACGTAGA
GC-bfzC-F	ACCGAGCTCAGATCTACTAGTATGGATTACGTCAGTTCAGT	Amplification of bfzC for gene complementation
GC-bfzC-R	ACACTGGCGGCCGTTACTAGTTCAGCTCGGCTGAGACGCCGA
RT41-F	CACGTTCACCCACCGAGAC	Amplification of 340 bp of orf4-orf1-spanning region
RT41-R	ACAGGCGTGCAACGGATT
RT1B-F	GACCGATGCAGTCCGCAAGCTCT	Amplification of 326 bp of orf1-bfzB-spanning region
RT1B-R	GCCGTGTCGATCACATTGATGCC
RTBA3-F	CGACGAACTTACCGAACTTCTACC	Amplification of 290 bp of bfzB-bfzA3-spanning region
RTBA3-R	AGTCGTCATGGCTGCAAGTGTCC
RTA32-F	GCATGGCACTTTGCGAAGTTCTG	Amplification of the 342-bp region spanning bfzA3 and orf2
RTA32-R	GAAGTTGCGCCCAGCAGCAGTCA
RT23-F	CGGGAAATCGTGTCGTACTGGAA	Amplification of 339 bp of orf2-orf3-spanning region
RT23-R	TCAGCTGTCGAAAATTCCCAACA
RT3A4-F	CGCCACTCTTCGACGACCAGCAG	Amplification of 396 bp of orf3-bfzA4-spanning region
RT3A4-R	GCTCCAACGATTGCGATTTTCCT
RTA4A1-F	GCGTTACAGGACCCACAAACCGAT	Amplification of 324 bp of bfzA4-bfzA1-spanning region
RTA4A1-R	GGAAAAGCCATGCACGACCGAAT
RTA1A2-F	TATCGAGGACGCAACCAACGAGC	Amplification of 317 bp of bfzA1-bfzA2-spanning region
RTA1A2-R	GTCCGCCGAATGGGCATAAAGTA
RTA2C-F	CGAGGAGGACCAATGGATCGGAT	Amplification of 303 bp of bfzA2-bfzC-spanning region
RTA2C-R	CCACTGGTGGTTCGGGTCGTCAC
RTq-16S-F	TCCTGGTGTAGCGGTGAA	Amplification of 88-bp fragment of 16S rDNA for RT-qPCR
RTq-16S-R	CGTTCCTCAGCGTCAGTTAC
RTq-A1-F	TCGTCGTTCGTCAGAAGG	Amplification of 130-bp fragment of bfzA1 for RT-qPCR
RTq-A1-R	AAGGTCCAGCCGTGATAG
RTq-B-F	CTCTCGTTCACTTCGCAATC	Amplification of 181-bp fragment of bfzB for RT-qPCR
RTq-B-R	ACCTTAGTAGCCAGAATGATGT

Biotransformation and identification of intermediates.

Biotransformation was performed as described previously, with minor modifications (45). LB-grown cells were harvested, washed twice, and diluted to an optical density at 600 nm (OD₆₀₀) of 2.0 with MSM prior to the addition of 30 mg · liter⁻¹ substrate (buprofezin, biphenyl, flavone, flavone, or bifenthrin). Samples from buprofezin reactions were extracted with equal volumes of dichloromethane, dried over anhydrous Na₂SO₄ and evaporated to dryness with nitrogen gas. Samples from biphenyl, flavone, flavanone, and bifenthrin reactions were acidified to pH 3.0 with 1 M HCl, were extracted twice with equal volumes of ethyl acetate, and were evaporated to dryness with a rotary evaporator. The dried samples were dissolved in 500 μl of methanol and filtered through a 0.22-μm-pore Millipore membrane for HPLC and UHPLC-MS/MS analyses.

Analytical methods.

For HPLC analysis, a Kromasil 100-5-C₁₈ separation column (internal diameter, 4.6 mm; length, 250 mm) was used. The mobile phase was methanol-water (80:20 [vol/vol]), and the flow rate was 0.8 ml · min⁻¹. The detection wavelength was 230 nm, and the injection volume was 20 μl. For UHPLC-MS/MS analysis, the samples were detected using a UHPLC system (Dionex, Thermo, USA) that was connected to an LTQ-Orbitrap XL hybrid mass spectrometer (Thermo Fisher Scientific), which is equipped with an electrospray ionization (ESI) probe. The UHPLC column was a Hypersil Gold C₁₈ (100 mm by 2.10 mm, 3-μm particle size; Thermo Fisher Scientific). The mobile phase consisted of solvents A (ultrapure water) and B (acetonitrile) with a gradient program that started with maintaining 30% B for 4 min, followed by increasing to 95% B from 4 to 21 min, holding at 95% B from 21 to 25 min, and then returning to 5% B within 5 min. The flow rate was 0.2 ml · min⁻¹. The injection volume was 5 μl. For samples from buprofezin, the data analysis was in positive mode. For samples from biphenyl, flavone, flavanone, and bifenthrin, the data analysis was in negative mode.

Sequencing, assembly, annotation, and genome comparison.

The total genomic DNA (gDNA) was extracted according to the method described by Pitcher et al. (46). The genome sequencing was performed by the Shanghai Majorbio Bio-pharm Technology Co., Ltd. (Shanghai, China). The complete genome sequence of strain YL-1 was assembled using a method combining Illumina MiSeq sequencing technology, the Pacific Biosciences platform, and PCR validation (47–49). The draft genome sequence of mutant strain YL-0 was generated using the Illumina HiSeq 2000 system (50). Annotation was performed using Glimmer 3.02, tRNAscan-SE version 1.3.1, and Barrnap 0.4.2. An all-versus-all genome alignment between strain YL-1 and mutant strain YL-0 was executed to identify the missing DNA fragment in mutant strain YL-0 using Mauve version 1.2.3 (51). BLASTn and BLASTp were employed to search for the related nucleotide sequences and to deduce the amino acid identities, respectively.

To analyze phylogenetic relationships, sequences were first aligned by Clustal X version 2.1 (52); subsequently, the phylogenetic tree was generated by the neighbor-joining method using MEGA version 5.0 (53). The evolutionary distances between branches were calculated using the Kimura two-parameter distance model, with bootstrap analysis of 1,000 resamplings to evaluate the tree topology.

Complementation of bfzBA3A4A1A2C in mutant strain YL-0.

The bfzBA3A4A1A2C cluster was amplified by PCR from genomic DNA of strain YL-1 with the primer pair listed in Table 3 and was cloned into the Rhodococcus-E. coli shuttle vector pRESQ at the SpeI site using a ClonExpress II One Step cloning kit (Vazyme, Nanjing, China). The fragment that was inserted into the plasmid was verified by sequencing. The resulting construct, pREBAC, was then introduced into strain YL-0 by electrotransformation (54). The ability of the complemented strain to degrade buprofezin was determined as described previously.

Deletion and complementation of bfzA1, bfzA2, bfzA3, bfzA4, bfzB, and bfzC.

The deletions of bfzA1, bfzA2, bfzA3, bfzA4, bfzB, and bfzC from strain YL-1 were performed using the suicide plasmid pK18mobsacB (55, 56). Briefly, plasmid pK18mobsacB-ΔbfzA1 (for the bfzA1 gene knockout) was constructed by fusing the upstream and downstream fragments of the target gene (bfzA1) to PstI-digested pK18mobsacB with a ClonExpress II One Step cloning kit (the primers used are listed in Table 3). The corresponding plasmid was then introduced into strain YL-1 by electrotransformation. Single-crossover clones were selected on LB plates containing Km, while double-crossover mutant strains were screened on LB plates containing 20% (wt/vol) sucrose. Mutant strains with the gene disruption were confirmed by colony PCR and DNA sequencing. The knockout of the bfzA2, bfzA3, bfzA4, bfzB, and bfzC genes was carried out in the same way as that for bfzA1. The plasmid pREA1, for gene complementation, was constructed by fusing the entire bfzA1 into SpeI-digested pRESQ (56). The plasmid was then transformed into the mutant strain YL-1 ΔbfzA1 by electrotransformation to generate the bfzA1-complemented strain YL-1(pREA1) ΔbfzA1. Strains YL-1(pREA2) ΔbfzA2, YL-1(pREA3) ΔbfzA3, and YL-1(pREC) ΔbfzC were generated in the same way as described above.

Transcriptional analysis.

The experimental design, procedures, and data analysis were executed based on the minimum information for publication of quantitative real-time PCR experiments (MIQE) guidelines (57).

(i) Experimental design, sample, and RNA extraction.

For the preparation of total RNA, strain YL-1 was grown in R2A medium in triplicates. Inducer (biphenyl or buprofezin) was added into two tubes individually at a final concentration of 30 mg · liter⁻¹ when the culture was grown to an OD₆₀₀ of 0.5, and the cultures were incubated for another 5 h before harvest (approximately half of the inducer was degraded). Afterwards, the pellets were washed twice with RNase-free H₂O and immediately subjected to RNA extraction.

The total RNA was extracted using an RNA isolation kit (Sangon Biotech, Shanghai, China) according to the manufacturer's instructions with slight modifications. The lysis of the cells of strain YL-1 was achieved by grinding the cells in liquid nitrogen. The concentration and purity (A₂₆₀/A₂₈₀) of RNA samples were determined using a NanoDrop (Thermo Scientific), while the quality of RNA samples was estimated visually by 2.0% agarose gel electrophoresis.

(ii) Reverse transcription.

The reverse transcription of RNA samples was performed with the PrimeScript RT reagent kit with a gDNA Eraser (TaKaRa, Dalian, China) to obtain the cDNA. RNA (750 ng) was first treated with gDNA Eraser for the removal of genomic DNA (gDNA) in a 10-μl-volume reaction system according to the manufacturer's instructions. The cDNA was synthesized in a 20-μl-volume reaction system containing 10 μl of the gDNA-removed reaction solution, 1 μl of PrimeScript RT enzyme mix I, 1 μl of RT primer mix, 4 μl of 5× PrimeScript buffer 2 (for real-time PCR), and 4 μl of RNase-free H₂O. The reaction mixtures were incubated at 37°C for 15 min, followed by 85°C for 5 s for inactivation of the reverse transcriptase enzyme. Finally, the cDNA samples were stored at −20°C until use.

(iii) Primer design, qPCR, and data analysis.

Primers for target gene amplification were designed using Beacon Designer version 7.5 (58). The sequences and amplicon lengths of each primer pair are listed in Table 3. Reaction mixtures (20 μl each) were set up in a 96-well half-skirt automation-compatible PCR microplate (Axygen, Hangzhou, China), and each mixture contained 10 μl of SYBR Premix Ex Taq II (Tli RNaseH Plus) (2×; TaKaRa, Dalian, China), 2 μl of the diluted cDNA (50 ng · μl⁻¹), 0.4 μl of each forward and reverse primer (10 μM), and 7.2 μl of RNase-free water. Real-time quantitative PCR (RT-qPCR) was performed using the CFX96 real-time PCR detection system (Bio-Rad, Hercules, CA). The qPCRs for each target gene were performed in technical triplicates, with a no-template control (NTC). The qPCR program was as follows: 95°C for 30 s and 40 cycles of 95°C for 5 s and 60°C for 34 s. Melting curve analysis was performed by increasing the temperature from 60°C to 95°C. The amplification efficiency of each primer was estimated with standard curves using serial dilutions of synthesized cDNA as the template and calculated using the slope of a linear regression model (10^−1/slope − 1) (59). Three independent biological replicates were assayed for each strain under each experimental condition. The expression levels of all the genes tested were normalized to 16S rRNA gene expression as the internal standard and quantified using the cycle threshold (2^-ΔΔCT) method (60). Results were expressed as mean values and standard deviations from three independent experiments. Tukey trend test analyses were performed using Statistical Package for the Social Sciences (SPSS) version 19.0.

Accession number(s).

The complete genome sequence of R. qingshengii YL-1 has been deposited in the GenBank database under the following accession numbers: CP017299 (chromosome), CP017300 (pYLC1), CP017301 (pYLC2), CP017302 (pYLC3), CP017303 (pYLL1), and CP017304 (pYLL2). The draft genome sequence of R. qingshengii YL-0 has been submitted to the GenBank database under accession number MSQF00000000. The GenBank accession number of the 6,581-bp fragment containing bfzBA3A4A1A2C is KY785168.