Dimethyl phthalate (DMP), diethyl phthalate (DEP), di-n-butyl phthalate (DBP), butyl benzyl phthalate (BBP), bis (2-ethylhexyl) phthalate (DEHP) (purity >98%) was purchased from Aladdin Chemistry Co., Ltd. (Shanghai, China). Chromatographic grade methanol and dichloromethane were purchased from Sigma (Germany). All organic solvents used were of chromatographic grade, and other chemical reagents were of analytical grade. The basic mineral salts medium (MSM) contained KH₂PO₄ (1.0 g/L), K₂HPO₄ (1.0 g/L), (NH₄)₂SO₄ (1.0 g/L), MgSO₄·7H₂O (0.2 g/L), CaCl₂ (0.02 g/L), and 1.0 mL trace element solution per liter of medium (39.9 mg/L MnSO₄·H₂O, 42.8 mg/L ZnSO₄·H₂O, and 34.7 mg/L (NH₄)₆Mo₇O₂₄·4H₂O). The final pH of media was adjusted to 7.2 and then sterilized at 121°C in an autoclave for 20 min.

The sampling location is in Wujiaqu City, Xinjiang, China (44°18′55″-44°40′00″N, 87°25′05″-87°36′05″E). This site is a cotton field mulched by agricultural plastic film for a long time. Five typical sampling points were selected, and the snake shaped sampling method was used to collect samples from the surface soil (0–20 cm). After removing stones, dead branches, leaves, and other impurities, the samples were thoroughly mixed, and 1 kg was retained by quartering. The samples were transported to the laboratory in waterproof paper bags and stored at 4°C. The enrichment culture method was used to isolate bacteria that could grow on DBP as the sole carbon source (Shariati et al., 2022). Briefly, 10 g of soil was placed into a flask containing 90 mL of sterile water, and shaked at 180 rpm for 30 min. After standing for 5 min, 2 mL of the suspension was added to 100 mL of MSM containing 50 mg/L of DBP. The obtained suspension was incubated at 30°C and 150 rpm for 4 days, followed by four series of subcultures, gradually increasing the DBP concentration (50, 100, 200, and 500 mg/L). Finally, 100 μL of the enriched culture from the fourth subculture flask containing 500 mg/L DBP was spread onto MSM agar medium supplemented with 100 mg/L DBP and incubated at 30°C for 72 h. Subsequently, individual bacterial colonies were inoculated onto MSM agar plates containing 100 mg/L of DBP for further purification. The colonies grown on MSM agar plates were identified as potential DBP degrading strains.

The strain with high DBP degrading ability was selected for further identification. The morphology observation was carried out by optical microscope and scanning electron microscopy (SEM). The genomic DNA of the isolated strain was extracted using a DNA extraction kit (Omega BioTek, United States). The bacterial 16S rRNA gene was amplified using universal PCR primers 27F (5′-AGAGTTTGATCCTGGCTCAG-3′) and 1492R (5′-GGTTACCT TGTTACGACTT-3′). The PCR products were purified on 1% agarose gel, and sequenced by Sangon Biotech Co., Ltd. (Shanghai, China). The sequence of the isolate was compared with known sequences in the NCBI GenBank database to search for relevant reference strains with high sequence similarity. A phylogenetic tree was constructed using MEGA software (version 11.0).

The substrate utilization efficiency for PAEs by isolated strain was determined by testing the growth ability in liquid MSM supplemented with one of the following substrates: DMP, DEP, DBP, BBP, and DEHP. The detail experimental procedure was as follows. The purified strains were inoculated into R2A medium and cultured for 24 h at 30°C and 150 rpm. Subsequently, the cultured bacterial solution (2%, v/v) was inoculated into fresh MSM containing different PAEs (100 mg/L) as sole source of carbon and energy. After 3 days of cultivation, residual PAEs were extracted and purified from the culture medium according to the method reported in previous literature (Feng N. et al., 2018). The extracts were analyzed using gas chromatography–mass spectrometry (GC–MS, Aglient 7893-5975) with a DB-5 MS capillary column (length 30 m, i.d. 0.25 mm, film 0.25 μm, Agilent). The carrier gas was helium (99.999% purity) with a constant flow of 1 mL/min. The carrier gas was helium (99.999% purity) with a constant flow of 1 mL/min. The temperatures for the transfer line and ion source were 280°C and 230°C, respectively. Analyte ionization was performed using electron ionization (70 eV), and signal acquisition was performed in selected ion-monitoring (SIM) mode. The PAEs recoveries in samples ranged from 82.40 to 103.50%. The detection limit of PAEs in samples was 0.05–0.15 μg/L. The degradation rate of the strain was calculated according to the following formula: Degradation rate (%) = (1-C/C₀) × 100%, where C and C₀ represent the concentration of PAEs in inoculated and non-inoculated media, respectively. All experiments were repeated three times.

The optimal parameters for the maximal degradation rate of DBP by strain P-7 were initially determined through single-factor experimental design. These parameters included pH (5 ~ 9), inoculation amount (1 ~ 5%), metal ions (Fe²⁺, Ca²⁺, Mn²⁺, Mg²⁺, and Cu²⁺), carbon sources (dextrin, corn, starch, sucrose, glucose, xylose), and nitrogen sources (yeast extract, tryptone, urea, ammonium sulfate, beef extract). Subsequently, based on the single-factor experimental results, principal component analysis (PCA) and response surface methodology (RSM) were used to optimize the culture conditions for efficient DBP degradation by strain P-7, which was used to select the three key factors involved in the DBP degradation. By comprehensive score of above five growth factors (Supplementary Table S1), three important factors (metal ions, carbon source, and nitrogen source) were identified from PCA results as the independent variables for RSM analysis. Based on the Box–Behnken design, the range of independent-variables and levels are shown in Supplementary Table S2. The RSM was further applied to optimize the degradation conditions (Alghamdi et al., 2023), resulting in 15 experiments with each independent variable at three different levels (−1, 0, and +1) using Box–Behnken (Table 1). A Box–Behnken matrix and the response of dependent variable for DBP degradation were generated using Design Expert. All experiments were conducted in 100 mL Erlenmeyer flasks containing 20 mL of trypticase soy broth (TSB) medium and 100 mg/mL of DBP. After incubation for 8 h, the residual DBP in medium was measured to calculate the degradation rate. A group without strain P-7 was used as control (CK), and all treatments were performed in triplicates.

The obtained data were finally analyzed using the response surface regression procedure to fit the following second-order polynomial equation (Du et al., 2024):

Where Y is the response variable (DBP degradation), X_i and X_j are the independent variables (metal ions, carbon source, and nitrogen source), β₀ is the constant coefficient, β_i is the line coefficient, β_ij is the interaction coefficient, and β_ii is the quadratic coefficient. ANOVA and F-value were performed to evaluate the statistical significance and efficiency of the model. The multiple determination coefficient (R²) was calculated to indicate the suitability of model. To demonstrate the individual and interactive effects of the independent variables on DBP degradation, three-dimensional response surface plots and contour plots were constructed to intuitively predict the impact of three factors on degradation interactions.

The strain P-7 cells were harvested for further DNA extraction using Wizard^® Genomic DNA Purification Kit (Promega, United States). The size, quantity and quality of purified DNA were checked by TBS-380 fluorometer (Turner BioSystem Inc., United States). The whole genome was sequenced using Illumina HiSeq platforms and PacBio RS II Single Molecule Real Time (SMRT) system at Shanghai Majorbio Bio-pharm Technology CO., Ltd. (Majorbio, China). The raw reads obtained were trimmed into clean reads, and then assembled into a contig by Unicycle v0.4.8. The bioinformatic analysis was performed using Majorbio Cloud Platform.¹ The protein-coding sequences (CDSs) were predicted using Glimmer.² The following databases were used for gene functional annotation: Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG), Cluster of Orthologous Groups (COG), Swiss-Prot, and non-redundant protein.

To analyze the metabolites involved in DBP degradation, bacterial cells were inoculated at a 2% inoculation volume into MSM liquid medium containing 100 mg/L DBP. Then, all cultures were incubated at 30°C with shaking at 180 rpm for 24 h. Samples were collected at 0, 6, 8, and 18 h. Metabolic intermediates were extracted using an equal volume of n-hexane, dried under nitrogen gas (purity > 99.99%), and redissolved in methanol. The solution was filtered through a 0.22 μm membrane, and the filtrate was transferred into 2 mL glass vials for identification. Finally, the metabolic intermediates of DBP were determined by Panomix CO., Ltd. (Suzhou, China) using ultra-high performance liquid chromatography-mass spectrometry (UHPLC–MS/MS) (Thermo Fisher Scientific, United States).

All statistical analysis were conducted using SPSS 22.0 software. Statistical significance was set at p < 0.05. The data were fitted and plotted using Origin Pro. 2021. Other sequencing results were directly generated from the Majorbio Cloud Platform (see Footnote 1). All experiments were carried out in triplicate.

Using DBP as the sole carbon and energy source, 8 bacterial strains capable of utilizing DBP were isolated from film mulched cotton field soil through enrichment and adaptation (Supplementary Table S3). The isolated strains, designated as L-2, L-7, L-15, P-7, P-9, P-14, P-16, and P-21, showed varied efficiencies in DBP biodegradation (Figure 1). Overall, all 8 strains exhibited the ability to degrade DBP, though with significant differences in degradation efficiencies. Strains L-2, L-7, and P-21 demonstrated degradation rates of less than 40% within 3 days, while strains P-9 and L − 15 showed degradation rates of less than 20%, indicating relatively low degradation capability. In contrast, strains P-7 and P-14 achieved higher degradation efficiency, with strain P-7 performing best, reaching over 93%. These results suggest that strain P-7 holds great potential as an efficient bacterial strain for DBP degradation in the environment. Therefore, it was selected as a candidate for further cultivation optimization and DBP biodegradation pathway study.

Next, the isolated strain P-7 was identified according to the biochemical identification, morphological observation, and 16S rRNA gene sequencing. As shown in Supplementary Table S4, the physiological and biochemical assays indicated that strain P-7 was a Gram-positive strain with strong tolerance to salt, temperature, alkali, and phenol. Additionally, it was capable of producing catalase and protease, which would help to catalyze DBP and other organic pollutants. After 24 h culturation on R2A agar medium, the colonies of strain P-7 were faint yellow, with a smooth, moist surface, regular edges, and an opaque appearance (Figure 2A). SEM observation revealed that strain P-7 cells were short and rod-shaped with a width of 1.5–2.0 μm and length of 4.2–7.8 μm (Figure 2B). Further comparison of the 16S rRNA gene sequence of strain P-7 with NCBI database and phylogenetic tree construction showed that strain P-7 belonged to the genus Priestia and clustered with Priestia megaterium NBRC 15308 (JMH01000057) with a homology of over 99% (Figure 2C). In conclusion, strain P-7 was ultimately identified as Priestia megaterium and designated as P. megaterium P-7.

To obtain the optimal conditions for P. megaterium P-7 to degrade DBP, single-factor experiment was conducted to explore the effects of five different growth factors (pH, inoculation amount, metal ions, carbon and nitrogen source) on DBP degradation. As shown in Figure 3A, a low inoculum amount (1%) of P. megaterium P-7 suspension was able to achieve a DBP degradation efficiency of nearly 60%. Further increasing the inoculum amount, the degradation rate of DBP remained stable at over 80%, indicating the enormous potential of P. megaterium P-7 for DBP degradation. In addition, P. megaterium P-7 was sensitive to pH, especially under acidic condition (Figure 3B). The DBP degradation rate increased rapidly when the pH was adjusted to neutral (pH 7.0) and weakly alkaline conditions (pH 9.0). Notably, it was observed that the significant enhancement in DBP degradation was effectively supported at low concentrations of metal ions except for Cu²⁺, in the following order: Fe²⁺ > Mg²⁺ > Ca²⁺ > Mn²⁺ > Cu²⁺ (Figure 3C), which was mainly achieved by affecting enzyme activity. Moreover, the supplementation of carbon and nitrogen sources also significantly increased the DBP degradation efficiency of P. megaterium P-7. Among the tested carbon sources, sucrose, glucose, corn starch, dextrin, and xylose proved to be suitable options, demonstrating a broad adaptability (Figure 3D). In contrast, the effective nitrogen sources were relatively limited, mainly including beef extract, yeast extract, and tryptone (Figure 3E). Based on the above results, we determined that the optimal metal ions and carbon/nitrogen sources for DBP degradation by P. megaterium P-7 were Fe²⁺, sucrose and beef extract.

To optimize the conditions for DBP biodegradation, single-factor experiments were first conducted, followed by PCA to identify the key factors influencing the degradation rate. The relationship between the DBP degradation rate and the three main factors was further explored using RSM design matrix, along with a polynomial regression model, to determine the optimal combination of variables. As shown in Table 2, the ANOVA for the quadratic response surface model fit (R² = 0.9246) indicates that the model can explain more than 92% of the experimental and predicted data. Additionally, the adjusted R² (0.7888) is also acceptable, confirming the model’s accuracy and reliability. According to the results of the F-test and t-test, the factors affecting DBP degradation rate are ranked as follows: Fe²⁺ > beef extract > sucrose. Only the linear term (AC) and the quadratic terms (A² and B²) significantly affect DBP biodegradation (p < 0.05). Therefore, the predictive equation for DBP degradation rate (%) with significant model terms is as follows:

Where Y is the DBP degradation rate (%), A is the beef extract content (g/L), B is the sucrose content (g/L), and C is the Fe²⁺ content (g/L).

The experimental results were analyzed using Design-Expert 13.0 software, and the response surface was plotted. The interaction between the factors can be observed in Figure 4. When Fe²⁺ is at the zero level, the optimal concentrations of beef extract and sucrose are 6.44 g/L and 0.16 g/L, respectively. When sucrose is at the zero level, the optimal concentrations of beef extract and Fe²⁺ are 7.06 g/L and 0.18 g/L, respectively. The optimal conditions obtained from the mathematical model are a beef extract concentration of 9.92 g/L, sucrose concentration of 0.19 g/L, and Fe²⁺ concentration of 0.12 g/L. Under these conditions, the model predicts that P. megaterium P-7 will achieve the highest DBP degradation rate of 74.31%. A verification experiment was carried out under the optimal conditions, and the actual DBP degradation rate was found to be 78.5%. The difference between the predicted and experimental values is only 4.19%, confirming the validity and effectiveness of the optimized model.

Under these optimal degradation conditions, the relationship between the growth and DBP degradation of P. megaterium P-7 was achieved. As shown in Supplementary Figure S1, P. megaterium P-7 was able to rapidly degrade DBP and utilized it as a growth substrate, with DBP degradation positively correlated with bacterial growth. Moreover, the growth of P. megaterium P-7 exhibited no significant lag phase in the presence of DBP, but quickly entered the exponential growth phase, with 0–12 h being the exponential phase. During this time, the degradation rate of DBP increased rapidly and synchronously. As the growth of P. megaterium P-7 reached a stationary phase (12–20 h), the degradation curve also tended to stabilize. Ultimately, DBP in the culture medium was almost undetectable and the strain began to enter a decline phase, indicating the complete consumption of the growth substrate. The comparison of DBP biodegradation among different strains further demonstrates the superior degradation capability of P. megaterium P-7 (Supplementary Table S5), providing important theoretical support and potential applications for the bioremediation of DBP-induced soil pollution.

To evaluate substrate utilization, P. megaterium P-7 was cultured in liquid medium supplemented with 100 mg/L of each PAE substrates, including DMP, DEP, DBP, BBP, and DEHP. As shown in Supplementary Figure S2, the results indicated that P. megaterium P-7 was capable of utilizing all five PAEs as sole carbon sources, demonstrating the effective degradation ability. Notably, the strain P. megaterium P-7 exhibited a significantly higher degradation rate (over 60%) for PAEs with shorter side chains (DMP, DEP, DBP, and BBP) compare to DEHP, which has longer side chains. These findings highlight the strong potential of P. megaterium P-7 for efficient PAEs degradation and its promise as a candidate for bioremediation of PAE-contaminated environments.

To further decipher the genetic information of P. megaterium P-7 and explore its functional genes, whole-genome sequencing and genomic analysis of this strain were conducted. The genomic characteristics of P. megaterium P-7 were summarized in Supplementary Table S6 and Figure 5. The genome of P. megaterium P-7 consisted of a single circular chromosome and was 5,567,352 bp in length with a GC content of 37.61%, of which 4,660,332 bp encodes genes. It contained a total of 5,653 predicted protein-coding sequences (CDSs), accounting for 83.71% of the total genome length. At the same time, there were 103 tRNA genes and 17 rRNA genes (1 of 16S rRNA, 1 of 23S rRNA, and 15 of 5S rRNA). These protein-coding genes were subjected to functional annotation using multiple databases, including GO, NR, COG, KEGG, Swiss-Prot, and Pfam databases. GO functional annotation results in Figure 6A showed that a total of 261 CDSs were mainly related to “regulation of DNA-templated transcription” (1.59%), “proteolysis” (1.47%) and “phosphorylation” (1.46%) in terms of biological processes. For cell composition, the CDSs were mainly related to “integral component of membrane” (17.13%), plasma membrane (6.57%), and cytoplasm (6.15%), while the genes were mainly related to ATP binding, DNA binding based on molecular function. Based on COG functional annotation, 4,669 CDSs were classified into 23 COG function classes, with the three most abundant categories being amino acid transport and metabolism, transcription, and general function prediction only (Figure 6B). According to KEGG pathway annotation, it was found that a total of 3,332 CDSs could be assigned to 40 metabolic pathways, which were involved in 6 biological pathways: cellular processes, environmental information processing, genetic information processing, human diseases, metabolism, organismal systems (Figure 6C). Among these genes, the majority were related to metabolic pathways, with 286 genes annotated for amino acid metabolism, 332 genes annotated for carbohydrate metabolism, and 214 genes annotated for metabolism of cofactors and vitamins. In addition, 56 genes associated with 14 pathways of xenobiotics biodegradation and metabolism was found. Especially, 33, 7, 8, and 9 genes were annotated to benzoate degradation, dioxin degradation, aminobenzoate degradation and xylene degradation, which exhibited the potential to degrade and metabolize organic pollutants in P. megaterium P-7.

Based on the genome annotation of P. megaterium P-7, multiple genes and gene clusters potentially involved in the degradation of DBP were identified, including those encoding esterases, hydrolases, decarboxylases, and dioxygenases (Table 3). The positions of genes and gene clusters implicated in DBP degradation were delineated within the chromosomal genome (Supplementary Figure S3A), and their structural organization was characterized (Supplementary Figure S3B). Additionally, five promoter sequences were predicted among the 28 key genes (Supplementary Table S7). Key genes widely associated with ester bond hydrolysis and transesterification were annotated, such as lip (triacylglycerol lipase), aes (acetyl esterase), ybfF (esterase), estA (putative tributyrin esterase), and yvaK (carboxylesterase). These enzymes played a critical role in converting DBP to MBP and PA (Nahurira et al., 2019; Yoo et al., 2020; Wang et al., 2024). Subsequently, the genes encoding decarboxylases and dioxygenases facilitated the conversion of above intermediates into protocatechuate (PCA), which was a key intermediate in the aerobic degradation pathway of aromatic compounds (Geiger et al., 2019). According to the known metabolic pathways of PAEs, it was found that pdc (phenolic acid decarboxylase), bsdCD (4-hydroxybenzoate decarboxylase subunit), mdcACDH (malonate decarboxylase alpha subunit), and lysA (diaminopimelate decarboxylase) genes could catalyze the decarboxylation of MBP and PA to produce benzoic acid. Further steps are accomplished by the hydroxylation of the aromatic ring by dioxygenases to form the common intermediates such as catechol and PCA. Although only 4 genes encoding dioxygenases (K07104, K08967, K00452, K00453) were annotated using KEGG analysis, no previously reported phthalate dioxygenase genes (pht gene clusters) were identified in this study (Feng L. et al., 2018). However, Swiss-Prot annotation revealed two dioxygenase genes (gene1285 and gene1474) involved in aromatic ring hydroxylation. Meanwhile, the ring of aromatic compounds was further opened by ring-cleaving dioxygenase (gene0497, gene3652, and gene5204) and catechol 2,3-dioxygenase (catE) to produce muconate derivative. These intermediates subsequently enter the catabolic pathway via the β-ketoadipate oxidized to acetyl-CoA and are ultimately entered the tricarboxylic acid (TCA) cycle, leading to the production of H₂O and CO₂ (Al-Khalid and Ei-Naas, 2012).

To further elucidate the metabolic mechanism of DBP in P. megaterium P-7, the metabolic intermediates were preliminary measured based on UHPLC–MS/MS (Supplementary Figure S4). The metabolomics results indicated that a total of 2,457 metabolites were determined, among which 8 key intermediate products were closely related to DBP degradation (Figure 7; Supplementary Table S8). According to the identification of metabolic intermediates, an alkyl chain (-C₄H₉) in DBP was hydrolyzed by esterases to form the primary metabolites MBP, followed by the hydrolysis of another alky chain to form PA. Further, PA was converted to catechol via PCA under the reaction of dioxygenases. Finally, the benzene ring of catechol was opened to form cis, cis-muconate, which was further used for cell growth through the β-ketoadipate pathway (Chen et al., 2021). It has been reported that benzoic acid metabolic pathway is a key pathway for bacterial degradation of DBP, which is generated through the hydrolysis of MBP and the decarboxylation of PA (Feng et al., 2024). However, we did not identify benzoic acid among the metabolites, but instead identified its derivatives, such as salicylic acid and 4-hydroxybenzoic acid. Therefore, we hypothesized another metabolic pathway in which the primary metabolite MBP was transformed to BB under the catalysis of decarboxylase, followed by ester hydrolysis to form BA. Additionally, diethyl phthalate (DEP) was also detected and identified in the metabolites of the fermentation broth, which was generally formed by β-oxidation of DBP (Feng et al., 2024; Ren et al., 2024). This observation indicated that DBP degradation occurred via β-oxidation, resulting in the formation of DEP, which was further transformed to PA through de-esterification and subsequently entered PA metabolic pathway. Based on whole genome analysis and identification of metabolites, we ultimately proposed the detailed metabolic pathway of DBP in P. megaterium P-7, which was presented in Figure 7.