Introduction

Microbial remediation is a promising nature-based, sustainable, and green remediation strategy [1, 2]. Nitrogen availability is vital for maintaining bioremediation processes [3, 4]. In hydrocarbon-contaminated soils (e.g., petroleum-contaminated soils), carbon availability increases, but nitrogen availability decreases dramatically, which restricts pollutant degradation by bacteria [4, 5]. Biological nitrogen fixation by diazotrophs, which convert atmospheric nitrogen to ammonia, is a major source of bioavailable nitrogen and plays a key role in maintaining elemental biogeochemical cycles and global ecosystem productivity, particularly in nitrogen-limited environments [6,7,8]. Hydrocarbon-degrading bacteria and diazotrophs often share the same microhabitat and coexist as a community, and these communities may be the best bioresources for remediating hydrocarbon-contaminated natural environments [9]. However, the interactions between diazotrophs and hydrocarbon-degrading bacteria in nitrogen-deficient contaminated environments have not been characterized.

The members of microbial communities may interact bidirectionally or unidirectionally in location and nutrient utilization, resulting in cooperation or competition [10, 11]. Under nutrient-limited conditions, auxotrophic bacteria rely on other members of the community to provide essential nutrients, prompting greater local exchange among cooperative bacteria and potentially increasing reciprocity [12, 13]. Consequently, in nitrogen-deficient environments, diazotrophs and degraders may interact synergistically in the degradation of organic pollutants. For instance, introducing diazotrophs into a consortium of bacterial degraders increased the degradation rate of diesel by 3.4–6.9% comparable to the increases achieved by supplementing the consortium with inorganic nitrogen [14]. The same stimulation effect was found in one study with the dechlorination rate of bacterial consortia increasing by 16% through introducing diazotrophs [15]. Another study found that stimulating diazotrophs by adding molybdenum and tungsten increased the removal rate of polycyclic aromatic hydrocarbons (PAHs) in farmland soil by 7.5% [16]. Hence, diazotrophs may indirectly contribute to the degradation of organic pollutants by providing nitrogen to degraders. However, whether diazotrophs provide nitrogen to degraders during organic pollutant degradation and the mechanism of any synergistic interactions remain unclear.

Interactions between members in different communities are frequently driven by metabolism. Resource sharing via cross-feeding of metabolites can underlie synergistic interactions between species [17, 18]. Under nutrient-limited conditions, oligotrophic strains rely on small peptides or vitamins produced by other members of the consortium for better growth and maintenance of function [19]. Bidirectional cross-feeding studies have confirmed that the division of labor between species in metabolic pathways such as amino acid synthesis, sugar synthesis, and organic pollutant degradation reduces energy costs and improves metabolic efficiency [20,21,22]. Some diazotrophs possess degradation function genes and can reduce environmental toxicity through aerobic degradation and anaerobic dechlorination [23, 24]. Thus, in addition to supplying nitrogen to their neighbors in contaminated nitrogen-free microhabitats, diazotrophs may also be directly involved in the degradation and bidirectional feeding of organic pollutant degradation intermediates with other degraders. Synthetic communities (SynCom) are advantageous for studying species interactions and their contributions to community function [25, 26]. The molecular and metabolic mechanisms underlying interspecies interactions in SynCom can be explored by biofilm imaging and omics analyses, such as transcriptomics and metabolomics [27,28,29]. Applying these methods to elucidate the molecular and metabolic mechanisms of interactions between diazotrophs and degraders in nitrogen-deficient environments contaminated with organic pollutants could enhance the efficiency of bioremediation in natural environments.

In the present study, a stable SynCom was constructed using Paracoccus aminovorans HPD-2, a degrader of high-molecular-weight-polycyclic aromatic hydrocarbons (HMW-PAHs) that was isolated from historically PAH-contaminated soil, and the diazotroph Azotobacter chroococcum HN. HMW-PAHs are a huge international environmental problem due to their persistence and recalcitrance to degradation [30, 31]. Using the HMW-PAH pyrene as the target contaminant, we determined whether A. chroococcum HN provides nitrogen to P. aminovorans HPD-2 in a contaminated nitrogen-free microhabitat using surface-enhanced Raman spectroscopy coupled with 15N stable isotope probing (SERS-15N SIP) experiments. We performed transcriptomic and metabolomic analyses of the two strains to compare their gene expression and metabolite in co-culture and monoculture in mineral salt medium (MSM). We also analyzed the external morphology and spatial three-dimensional structure of the cells in co-cultured biofilms by electron microscopy scanning and fluorescence in situ hybridization to assess cell-cell interactions.

Materials and methods

Strains and media

P. aminovorans HPD-2 was obtained by isolation and screening from historically PAH-contaminated soil collected from Wuxi, Jiangsu Province, China. P. aminovorans HPD-2 is an oligotrophic-nitrogen bacterium that grows slowly in nitrogen-free media and cannot fix nitrogen [32] (Fig. S1 and Supplementary Table 1). A. chroococcum ATCC 9043 (HN) was purchased from China General Microbiological Culture Collection Center (CGMCC). Nitrogen-free modified MSM was used for pyrene-degradation experiments [23, 33]. The medium contained the following components (L−1): 2 g of mannitol, 0.15 g of anhydrous CaCl2, 0.2 g of MgSO4 7H2O, 0.04 g of FeSO4 7H2O, 0.005 g of NaMoO4, 0.8 g of K2HPO4, pH 6.8–7.0, and 10 mL of trace metal salt solution. The trace element solution contained (L−1) 10 mg of ZnSO4 7H2O, 5 mg of MnSO4 H2O, 330 mg of H3BO, 24 mg of CoSO4 7H2O, and 5 mg of CuSO4 5H2O, pH 7.2. P. aminovorans HPD-2 was cultivated in Luria-Bertani (LB) medium containing (L–1) 10 g of NaCl, 5 g of yeast extract, and 10 g of tryptone. A. chroococcum HN was cultivated in Brown’s medium [33] containing (L−1) 5 g of glucose, 0.15 g of anhydrous CaCl2, 0.2 g of MgSO4 7H2O, 0.04 g of FeSO4 7H2O, 0.005 g of NaMoO4, and 0.8 g of K2HPO4, pH 6.8–7.0.

Monoculture and interspecies inhibition test

Colonies of P. aminovorans HPD-2 and A. chroococcum HN were inoculated into 250 mL of LB or Brown’s medium, respectively, and incubated in an orbital shaker at 150 rpm and 30 °C for 24 h. The cells were separated from the medium by centrifugation, washed three times with MSM and re-suspended in MSM to obtain a bacterial suspension. The optical density at a wavelength of 600 nm (OD600) of the bacterial suspension was measured (Bio-TekμQuant, Winooski, VT) and adjusted to OD600 value of 1.0 (2.8 ×108 CFU mL–1).

Antagonism experiments between P. aminovorans HPD-2 and A. chroococcum HN were performed in MSM agar plates by the punch hole method [34]. Briefly, MSM plates supplemented with pyrene (1 mg L−1) were perforated and inoculated with strains in two ways. On one side of the plate, equal volumes of monoculture of P. aminovorans HPD-2 and A. chroococcum HN were inoculated in separate holes and in the same hole. On the other side of the plate, pairs of holes with different distances between holes were inoculated separately with equal volumes of monoculture of P. aminovorans HPD-2 and A. chroococcum HN. All plates were incubated at 30 °C for 72 h.

Growth and pyrene degradation during bacterial cultivation

A volume of pyrene stock solution was added to a 50-mL conical flask containing 10 mL of MSM to obtain a final pyrene concentration of 1 mg L−1. Next, suspensions of P. aminovorans HPD-2 and/or A. chroococcum HN at OD600 1.0 were added to the conical flask as a monoculture or co-culture (volume, 1:1) as a 5% inoculum. The non-inoculated treatment group was a blank control. All cultures were set up in three replicates. The conical flasks were placed in a shaker (30 °C, 150 rpm) for 10 days. On days 0, 1, 3, 5, 7, and 10, samples were removed for destructive sampling to extract pyrene and pyrene degradation intermediates. The OD600 of the sample was measured to characterize the bacterial growth. The growth rate of live bacteria (colony-forming units, CFU) was determined by dilution plating (LB medium and Brown’s medium) and colony counting.

Detection of pH and ammonium ions (NH4 +) during bacterial cultivation

The pH of the cultures was determined using a pH meter (TOA, Co. Ltd.). The concentration of NH4+ released during pyrene degradation was determined by ion chromatography [35]. Briefly, the culture supernatant was filtered through a 0.22-μm aqueous-phase microporous membrane to remove impurities and injected into a Dionex ICS-1100 ion chromatograph (Dionex Corp., Sunnyvale, CA, USA). The concentration of NH4+ was calculated using the external standard method.

Detection of pyrene and its degradation intermediates during bacterial cultivation

Pyrene was extracted from bacterial cultures by adding 2 times volumes of ethyl acetate and 0.4 g of (NH4)2SO4 (emulsion breaker) to the conical flask [36, 37]. The extraction was vortexed for 10 min and left to stand for 1 h. One milliliter of the upper organic phase was dried under a stream of N2 and re-dissolved in 1 mL of acetonitrile. The concentration of pyrene was measured by high-performance liquid chromatography (Shimadzu, Kyoto, Japan) according to a previous study [38].

Pyrene degradation intermediates were extracted and measured as described previously [36, 37]. In brief, the culture supernatant was acidified to pH 2.0 using 1 M HCl. A volume of ethyl acetate equal to the volume of the culture supernatant was added, and the organic-phase extracts were transferred to tubes and dried under a stream of N2. Finally, 0.1 mL of ethyl acetate and 0.1 mL of the derivatization reagent BSTFA-TMCS (99:1, v/v) were added to the tubes and reacted at 60 °C for 30 min, and the volume was adjusted to 1 mL with ethyl acetate. The solution was analyzed using Agilent gas chromatography-mass spectrometer (GC − MS, 6890A-5973N, Agilent Technologies, Wilmington, DE) equipped with an HP-5 MS capillary column (30 m × 0.25 mm × 0.25 μm). For qualitative determination, the MS system was operated in full-scan mode at m/z 50–450. The mode was electron impact energy mode, and the electron impact energy was 70 eV. The carrier gas was high-purity nitrogen, and the inlet was operated at a constant flow rate of 1.0 mL min−1. The injection port temperature was 280 °C. The sample volume was 1 μL without splitting, and the sample was purged after 0.75 min.

Cross-feeding experiments

A stock solution of phthalic acid or 2,5-dihydroxybenzoic acid was added to a 50-mL conical flask containing 10 mL of MSM to achieve a final concentration of 600 μg L−1. A suspension of P. aminovorans HPD-2 or A. chroococcum HN at OD600 1.0 was added to the conical flask as a 5% inoculum. The non-inoculated treatment group was a blank control. All cultures were set up in three replicates. The conical flasks were placed in a shaker (30 °C, 150 rpm) for 5 days, and samples were removed on days 0, 1, 3, and 5 for destructive analysis of phthalic acid or 2,5-dihydroxybenzoic acid. In addition, the OD600 of the culture was measured to characterize the bacterial growth potential.

Phthalic acid and 2,5-dihydroxybenzoic acid were extracted and measured according to the literature with some modifications [39, 40]. Culture samples (1 mL) were removed from the conical flasks and placed in 5-mL centrifuge tubes, and 1 mL of methanol (to precipitate the salt component) was added and mixed well with a vortexer. The tubes were centrifuged for 5 min at 5000 rpm. The supernatant was filtered through a 0.22-μm sterile organic membrane and transferred to an injection vial. The concentration of phthalic acid or 2,5-dihydroxybenzoic acid was determined by high-performance liquid chromatography-mass spectrometry (HPLC-MS, Tq8050, Shimadzu, Kyoto, Japan) on an Agilent SB-Aq column (3 mm × 150 mm, 2.7 μm, Waters) at a flow rate of 0.3 mL min−1 at 30 °C. The mobile phase consisted of (A) 0.08% formic acid and 1.2 mM ammonium acetate in water and (B) methanol. The sample injection volume was 5 μL, and the sample injector was maintained at 15 °C throughout the analysis. Mass spectrometry analysis was performed on an ABI/SCIEX 4500 Triple Quad™ with a Turbo V ion source. Phthalic acid and 2,5-dihydroxybenzoic acid were detected in negative ion mode. The nebulizer and heater gas pressures were 50 and 60 psi, respectively. The electrospray voltage was 5000 V, and the turbo ion spray source temperature was 500 °C.

Bacterial cultivation and single-cell SERS experiment

Sterile MSM (5 mL) was added to a 30-mL crimp-top vial and spiked with pyrene stock solution to achieve an initial pyrene concentration of 1 mg L–1. Bacterial suspensions of P. aminovorans HPD-2 and A. chroococcum HN at OD600 1.0 were added to the vial either alone or simultaneously (volume, 1:1) as a 5% inoculum, and the vial was sealed. The air in the headspace was replaced with 15N2 (99 atom %, purity >98.5%, Aladdin, China) and oxygen in a volume ratio of 4:1. The control treatments were supplied with lab air. All vials were placed in a shaker at 150 rpm and 30 °C in the dark for 5 days. The bacteria were harvested and washed twice with ultrapure water by centrifugation at 5000 rpm for 5 min. To prepare single-cell surface-enhanced Raman spectroscopy (SERS) samples, the cell pellet was mixed with 10 μL of concentrated Ag (∼7000 mg L−1) [41], and 2 μL of the mixture was dropped on an aluminum slide and air-dried prior to spectroscopy.

A scanning electron microscope (SEM) coupled with a confocal Raman microscope (TESCAN-MAIA 3 GMU, WITec, Czechia) was used for Raman spectroscopy of single cells in co-culture. First, a back-scattered electron (BSE) detector was employed to distinguish and obtain single-cell images of the two species. In order to distinguish between the two types of spherical cells in the Raman spectroscopy, then, the specimen stage from the SEM was moved to the Raman microscope inside the chamber to find the view and map it with the SEM single-cell images to eventually achieve overlap. Eventually, Raman spectra were acquired for two types of cells based on their morphology.

Raman analysis was conducted as follows. Single cells of P. aminovorans HPD-2 or A. chroococcum HN were excited by a 532-nm laser via a 100×/0.75 objective lens at a laser power of 20 μW with a grating of 600 g mm−1 [41]. The signal was acquired for 3 s per cell.

Scanning electron microscopy and fluorescence in situ hybridization

Biofilm contents were measured in 96-well plates using the crystal violet staining method [42]. Biofilms for SEM and FISH experiments were prepared as follows. A sterile 18-mm circular coverslip was placed on the bottom of a 35-mm glass culture plate, and 3 mL of MSM and 6 μL of pyrene solution (final concentration of 1 mg L−1) were added. Next, suspensions of P. aminovorans HPD-2 and A. chroococcum at OD600 1.0 were added to the culture plate alone or simultaneously (volume, 1:1) as a 5% inoculum. The biofilms on the coverslips were harvested after 7 days of incubation at 30 °C and 150 rpm in a constant-temperature shaker.

The biofilms of monocultures and co-culture of P. aminovorans HPD-2 and A. chroococcum were processed by referring to a previous publication [43]. Finally, the biofilms were freeze dried and gold sputter coated, and field emission SEM combined with energy-dispersive X-ray spectroscopy (FE-SEM-EDS, S-4800, Hitachi, Tokyo, Japan) was performed. FISH was performed as described previously [42]. In brief, biofilm samples were fixed with 4% paraformaldehyde for 6 h at 4 °C, washed three times with PBS, air dried, and dehydrated in 50%, 80%, and 100% ethanol for 3 min each at room temperature. Dehydrated slides can be stored at −20 °C for several months. Probes for P. aminovorans HPD-2 and A. chroococcum HN were commercially synthesized and labeled with 6-carboxyfluorescein (FAM, green) and the sulfoindocyanine dye indocarbocyanine (Cy3, red), respectively. The specimens were incubated with the fluorochrome-labeled oligonucleotide probes at a concentration of 25 ng μL−1 of hybridization buffer [0.9 M NaCl, 20 mM Tris/HCl (pH 7.2), 30% formamide, 0.01% (w/v) SDS] at 46 °C for 4 h. Following hybridization, the samples were quickly placed in a wash buffer [20 mM Tris/HCl (pH 7.5), 5 mM EDTA, 108 mM NaCl, and 0.01% (w/v) SDS] and incubated for 20 min at 48 °C. After washing, the samples were submerged in cold ddH2O for a couple of seconds and air dried completely at ambient temperature. Next, 10 μL of DAPI (diluted to 1 μg μL−1 with methanol) was added to the coverslips dropwise, and the samples were stained on ice for 10 min. After washing several times with water and air drying completely, the coverslips were fixed on slides and sealed around the perimeter with colorless nail polish. Images were acquired by confocal laser scanning microscopy (CLSM, LSM710, Carl Zeiss Light Microscopy, Germany) with a 40 × oil-immersion objective using the Quick-LUT (Look Up Table) function to set the pixel saturation limits.

Transcriptomic analysis

A total of three treatment groups were subjected to transcriptomic analysis: the monocultures of P. aminovorans HPD-2 and A. chroococcum HN and the co-culture. All treatments were centrifuged at 5000 rpm for 15 min to collect cells. Total cellular RNA was extracted using TRIzol reagent according to the manufacturer’s instructions (Invitrogen, Carlsbad, CA), and RNA was quantified using a NanoDrop ND-1000 UV–Vis spectrophotometer (NanoDrop Technologies, Wilmington, DE, USA). RNA-Seq transcriptome libraries were constructed using an Illumina TruSeq RNA Sample Preparation Kit (San Diego, CA) and sequenced using the Illumina NovaSeq 6000 instrument by Majorbio Biopharm Technology Co., Ltd. (Shanghai, China). Raw RNA-Seq data (raw reads) were processed using in-house Perl scripts with FastQC software. Adapter and low-quality sequences were removed from the raw data to obtain clean RNA-Seq data. The Q20 and Q30 were counted and were more than 96.15% and 91.29%, respectively, indicating a good-quality assembly. The clean reads of each sample were sequenced against the assigned reference genome separately, and the matching rate ranged from 62.29% to 90.43%. Transcriptome sequencing results are shown in Supplementary Table 2. The read count data were normalized, and unigenes (p-adjust < 0.05 and |log2 fold change| ≥ 1) were identified as differentially expressed genes (DEGs) between two treatments using the DESeq2 (1.30.1). The sequences were annotated according to Gene Ontology (GO) enrichment analysis and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis for functional and metabolic pathway analyses.

Metabolomic analysis

Supernatants from 50 mL of the monocultures (control) or co-culture of A. chroococcum HN and P. aminovorans HPD-2 were collected by centrifugation at 5000 rpm for 15 min and freeze dried for 72 h. The solid sample (50 mg) was transferred to a 2-mL centrifuge tube, and a 6-mm-diameter grinding bead was added. Metabolites were extracted by adding 400 μL of extraction solution (methanol: water = 4:1 (v:v)) containing 0.02 mg mL−1 internal standard (L-2-chlorophenylalanine). The samples were ground with a Wonbio-96c (Shanghai Wanbo Biotechnology Co., Ltd) frozen tissue grinder for 6 min (−10 °C, 50 Hz), followed by low-temperature ultrasonic extraction for 30 min (5 °C, 40 kHz). The samples were stored at −20 °C for 30 min and centrifuged for 15 min (4 °C, 13000 g), and the supernatant was transferred to an injection vial for LC-MS/MS analysis. LC-MS/MS analysis was conducted on a Thermo UHPLC-Q Exactive HF-X system equipped with an ACQUITY HSS T3 column (100 mm × 2.1 mm i.d., 1.8 μm; Waters, USA) at Majorbio Biopharm Technology Co., Ltd. The mobile phase consisted of 0.1% formic acid in water: acetonitrile (95:5, v/v) (solvent A) and 0.1% formic acid in acetonitrile: isopropanol: water (47.5:47.5:5, v/v/v) (solvent B). Mass spectrometry data were collected on a Thermo UHPLC-Q Exactive HF-X Mass Spectrometer equipped with an electrospray ionization (ESI) source operating in positive mode and negative mode. Raw data were corrected and mapped using Progenesis QI (Waters Corporation, Milford, USA). Identified metabolites were annotated using KEGG for metabolic pathway analysis and compound classification. Metabolite differences between two groups were considered significant at p < 0.05 and VIP > 1.

Statistical analysis

One-way analysis of variance (ANOVA) and Tukey’s HSD tests were performed using the “multcomp” package (4.0.4) to evaluate the significance of differences between treatments.

Results

Co-culturing A. chroococcum HN and P. aminovorans HPD-2 promotes mutual growth and pyrene degradation

The growth characteristics and pyrene degradation of co-cultured A. chroococcum HN and P. aminovorans HPD-2 were compared with the corresponding monocultures. The co-culture and monocultures both grew rapidly after 1 day of incubation, implying that both bacteria used mannitol as the main carbon source for growth during the initial stage of culture (Fig. 1a). After 10 days of incubation, the pyrene concentration in the co-culture decreased from 985.99 μg kg−1 to 461.02 μg kg−1, resulting in a degradation rate of 32.39% after excluding the degradation in the abiotic control (20.91%) (Fig. 1b and Supplementary Table 3). By contrast, the pyrene concentration in the A. chroococcum HN and P. aminovorans HPD-2 monocultures decreased from 996.36 μg kg−1 and 1006.83 μg kg−1 to 633.95 μg kg−1 and 554.24 μg kg−1, respectively, corresponding to degradation rates of 15.45% and 24.05%. Moreover, dilution plating colony counting showed that both A. chroococcum HN and P. aminovorans HPD-2 grew better in co-culture than in monoculture (Fig. 1c). Interestingly, co-culture in MSM strongly favored the growth of A. chroococcum HN during the lag phase, potentially due to greater utilization of mannitol compared with P. aminovorans HPD-2 in the nitrogen-free environment (Fig. 1d). There was a shift toward increased P. aminovorans HPD-2 in the exponential phase, possibly due to the consumption of nitrogen released by A. chroococcum HN. Thus, co-culture not only promoted the growth of A. chroococcum HN and P. aminovorans HPD-2 but also significantly accelerated the degradation of pyrene. The antagonism experiments also visually demonstrated that A. chroococcum HN and P. aminovorans HPD-2 were better able to promote each other’s growth when co-cultured in direct contact (Fig. 1e, f).

Fig. 1: Growth and pyrene degradation curves of A. chroococcum HN and P. aminovorans HPD-2 monocultures and co-culture in nitrogen-free MSM supplemented with mannitol (0.2 g L−1) and pyrene (1 mg L−1) as carbon sources.
figure 1

a Growth curves of A. chroococcum HN and P. aminovorans HPD-2 monocultures and co-culture. b Pyrene degradation curves of A. chroococcum HN and P. aminovorans HPD-2 monocultures and co-culture. c Growth curves of A. chroococcum HN and P. aminovorans HPD-2 in monocultures and co-culture. d Relative proportions of A. chroococcum HN and P. aminovorans HPD-2 under co-culture conditions. Antagonism analysis between A. chroococcum HN and P. aminovorans HPD-2 after 1 day (e) and 3 days (f) of incubation in nitrogen-free MSM agar plates with mannitol (0.2 g L−1) and pyrene (1 mg L−1) as carbon sources. MIX—co-culture of A. chroococcum HN and P. aminovorans HPD-2, HN—monoculture of A. chroococcum HN, HPD-2—monoculture of P. aminovorans HPD-2. Error bars represent standard deviations (N = 3). Letters indicate the significance of the differences between the values of the different treatments (p < 0.05).

Full size image

A. chroococcum HN provides nitrogen to P. aminovorans HPD-2 to mitigate nutrient-poor environments

To investigate whether A. chroococcum HN provided nitrogen to P. aminovorans HPD-2 during pyrene degradation in co-culture, single-cell SERS − 15N SIP experiments were performed. The Raman spectra of monocultured A. chroococcum showed bands at 730 or 717 cm−1 due to N assimilation when incubated with 15N2 or 14N2 (Fig. 2a and Supplementary Fig. S2). However, the Raman spectra of monocultured P. aminovorans HPD-2 did not include these bands due to the inability of this bacterium to fix nitrogen. Scanning electron microscopy observations combined with Raman imaging revealed that the Raman spectra of both A. chroococcum HN and P. aminovorans HPD-2 in co-culture included bands at 730 or 717 cm−1 when incubated with 15N2 or 14N2, respectively (Fig. 2a, b), implying that A. chroococcum HN provided nitrogen for P. aminovorans HPD-2. Furthermore, A. chroococcum HN released significant amounts of NH4+ during pyrene degradation in monoculture, whereas the monoculture of P. aminovorans HPD-2 did not (Fig. 2c). It is worth noting that no NH4+ was detected in the co-culture of A. chroococcum HN and P. aminovorans HPD-2, suggesting that A. chroococcum HN provided nitrogen to P. aminovorans HPD-2 by releasing NH4+.

Fig. 2: Characteristic of NH4+ concentrations and nitrogen-absorbing cell in studied cultures.
figure 2

a SERS spectra of A. chroococcum HN and P. aminovorans HPD-2 incubated with 15N2 and 14N2 in monoculture and co-culture. Spectra a, b, d, and e correspond to A. chroococcum HN incubated with 15N2 in co-culture, 15N2 in monoculture, 14N2 in co-culture, and 14N2 in monoculture, respectively; spectra c and f belong to P. aminovorans HPD-2 incubated with 15N2 or 14N2 in co-culture, respectively; and spectra g and h correspond to P. aminovorans HPD-2 incubated with 15N2 or 14N2 in monoculture, respectively. b Scanning electron microscopy mapping images combined with Raman imaging of A. chroococcum HN and P. aminovorans HPD-2 in co-culture. The red and green circles indicate cells of A. chroococcum HN and P. aminovorans HPD-2 in co-culture, respectively. c Changes in NH4+ concentrations in the monocultures and co-culture of A. chroococcum HN and P. aminovorans HPD-2. MIX—co-culture of A. chroococcum HN and P. aminovorans HPD-2, HN—monoculture of A. chroococcum HN, HPD-2—monoculture of P. aminovorans HPD-2. Error bars represent standard deviations (N = 3).

Full size image

Cross-feeding between A. chroococcum HN and P. aminovorans HPD-2 in pyrene degradation

Since A. chroococcum HN degraded pyrene, we speculated that cross-feeding of pyrene intermediate metabolites between A. chroococcum HN and the degrader P. aminovorans HPD-2 may occur during pyrene degradation. To evaluate this possibility, changes in pH and pyrene metabolites were examined in the monocultures, co-culture, and cell-free controls. In the monoculture of A. chroococcum HN, the pH was 7.0 on day 0 and gradually increased to approximately 7.4 as the cells grew and pyrene was degraded; the pH of the monoculture of A. chroococcum HN was not significantly different from that of the cell-free control (Supplementary Fig. S3). After day 7 of culture, three pyrene degradation intermediates were detected: phenanthrene, 2-naphthol, and salicylic acid (Supplementary Fig. S4 and Supplementary Table 4). The pH of the P. aminovorans HPD-2 monoculture was 7.3 on day 0 but rapidly decreased to 6.7 on day 1 and then gradually increased to approximately 7.0 as pyrene was degraded. The pH of the P. aminovorans HPD-2 monoculture was significantly lower than that of the cell-free control. Six pyrene degradation intermediates were detected in the monoculture of P. aminovorans HPD-2: phenanthrene, (2-oxo-2H-chromen-3-yl)-acetic acid, 2,5-dihydroxybenzoic acid, 2-hydroxyphenylacetic acid, phenylacetic acid, and salicylic acid. In the co-culture of A. chroococcum HN and P. aminovorans HPD-2, the pH increased gradually with the degradation of pyrene and was between the pH values of the two monocultures (7.0–7.2). Four pyrene degradation intermediates were produced in the co-culture: phenanthrene, 2-naphthol, 2-hydroxyphenylacetic acid, and phthalic acid (Supplementary Fig. S4 and Supplementary Table 4). Because phenanthrene was detected in all cultures, we deduced that the pathway of pyrene degradation by the bacteria was oxidative ring opening and decarboxylation of pyrene to phenanthrene by mono- or dioxygenase, followed by continued degradation via the phthalic or salicylic acid metabolic pathway (Fig. 3a).

Fig. 3: Possible pyrene degradation pathways in studied cultures.
figure 3

a Possible pyrene degradation pathways inferred from intermediate metabolites of pyrene degradation in the monocultures and co-culture of A. chroococcum HN and P. aminovorans HPD-2. The solid and dotted lines represent the paths between the lines where the two substances can or cannot be degraded in one step, respectively. MIX—co-culture of A. chroococcum HN and P. aminovorans HPD-2, HN—monoculture of A. chroococcum HN, HPD-2—monoculture of P. aminovorans HPD-2. b Changes in the concentrations of phthalic acid and growth curves of A. chroococcum HN and P. aminovorans HPD-2 in nitrogen-free MSM supplemented with mannitol (0.2 g L−1) and phthalic acid (600 μg L−1) as carbon sources. c Changes in the concentrations of 2, 5-dihydroxybenzoic acid and growth curves of A. chroococcum HN and P. aminovorans HPD-2 in nitrogen-free MSM supplemented with mannitol (0.2 g L−1) and 2, 5-dihydroxybenzoic acid (600 μg L−1) as carbon sources. Error bars represent standard deviations (N = 3). Letters indicate the significance of the differences between the values of the different treatments (p < 0.05).

Full size image

Based on the functional genes associated with PAH metabolism in the whole genomes of the two species and the differences in pyrene degradation intermediates (Fig. 3a and Supplementary Fig. S5), 2,5-dihydroxybenzoic acid and phthalic acid feeding experiments were performed in monocultures of P. aminovorans HPD-2 and A. chroococcum HN. Both A. chroococcum HN and P. aminovorans HPD-2 grew well in the experimental system supplemented with phthalic acid. Compared with the sterile control treatment, both A. chroococcum HN and P. aminovorans HPD-2 degraded approximately 31.22% of the phthalic acid after 5 days of culture (Fig. 3b). In the experimental system with 2,5-dihydroxybenzoic acid, both A. chroococcum HN and P. aminovorans HPD-2 grew well. However, compared to the sterile control treatment, A. chroococcum HN reduced the concentration of 2,5-dihydroxybenzoic acid from 624.25 μg kg−1 to approximately 92.21 μg kg−1 after 1 day of incubation, corresponding to a degradation rate of 85.23% (Fig. 3c). For P. aminovorans HPD-2, the degradation rate was only 19.56% after 5 days of incubation. These results suggest that cross-feeding of 2,5-dihydroxybenzoic acid or phthalic acid occurred between P. aminovorans HPD-2 and A. chroococcum during pyrene degradation.

Transcriptomics and metabolomics of A. chroococcum HN and P. aminovorans HPD-2 in co-culture

To uncover the molecular mechanisms of the interactions between A. chroococcum HN and P. aminovorans HPD-2 for pyrene degradation in co-culture, we performed transcriptomics and extracellular untargeted metabolomics on the monocultures and co-culture. After standardizing RNA read counts, 1020 genes were significantly upregulated and 996 genes were significantly downregulated in A. chroococcum HN in the co-culture compared with the monoculture (adj. p ≤ 0.05 and |log2FC | ≥ 1) (Fig. 4a). The differentially expressed genes (DEGs) were analyzed using KEGG ontology and GO functional category enrichment (Fig. 4b and Supplementary Fig. S6). The significantly upregulated DEGs were distributed among nitrogen metabolism, cellular metabolism, degradation of aromatic compounds, TCA cycle, amino acid metabolism, oxidative phosphorylation, biofilm formation, and protein export (Fig. 4b). In P. aminovorans HPD-2, 270 genes were significantly upregulated and 314 genes were significantly downregulated in the co-culture compared with the monoculture (Fig. 4c). The upregulated DEGs were mainly involved in quorum sensing, degradation of aromatic compounds, amino acid metabolism, oxidoreductase activity, cellular metabolism, ABC transporters, and iron transport (Fig. 4d).

Fig. 4: Volcano plots of the differentially expressed genes and enrichment analysis of KEGG and GO functional categories in A. chroococcum HN and P. aminovorans HPD-2 in the co-culture compared with the corresponding monocultures.
figure 4

a Volcano plots of differentially expressed genes in A. chroococcum HN in co-culture versus monoculture. b Enrichment analysis of KEGG and GO functional categories in A. chroococcum HN in co-culture versus monoculture. c Volcano plots of the differentially expressed genes in P. aminovorans HPD-2 in co-culture versus monoculture. d Enrichment analysis of KEGG and GO functional categories in P. aminovorans HPD-2 in co-culture versus monoculture. MIX—co-culture of A. chroococcum HN and P. aminovorans HPD-2, HN—monoculture of A. chroococcum HN, HPD-2—monoculture of P. aminovorans HPD-2. Three replicates (N = 3) were used for transcriptomics analyses.

Full size image

Based on the KEGG results, the analysis further focused on upregulated DEGs associated with major metabolic pathways such as nitrogen metabolism, amino acid metabolism, membrane transport, degradation of aromatic compounds, cellular metabolism, TCA cycle, and biofilm formation in A. chroococcum HN and P. aminovorans HPD-2. Among nitrogen metabolic pathways, many upregulated DEGs in A. chroococcum HN belonged to nitrogen fixation pathways, including nifH/D/K (nitrogenase iron protein; nitrogenase molybdenum-iron protein alpha chain/subunit beta) and vnfD/K/G (vanadium-dependent nitrogenase alpha chain/beta chain; vanadium nitrogenase delta subunit). By contrast, few of the upregulated genes in P. aminovorans HPD-2 were related to nitrogen metabolism (Fig. 5a and Supplementary Table 5). In addition, a large number of genes related to amino acid synthesis and metabolism were upregulated in A. chroococcum HN and P. aminovorans HPD-2, particularly those related to amino acids such as glutamate, alanine, arginine, aspartate, glycine, serine, threonine, tryptophan, valine, leucine, isoleucine, cysteine, methionine, proline, lysine, and tyrosine (Fig. 5b and Supplementary Table 6). In the nitrogen-free co-culture environment, nitrogen fixed by A. chroococcum HN was the only source of nitrogen for cell growth. Consequently, the upregulation of the nitrogen fixation pathway in A. chroococcum HN and the amino acid synthesis pathways in both species indicate that P. aminovorans HPD-2 utilized nitrogen fixed by A. chroococcum HN, such as NH4+. The protein and peptide metabolic process and protein export pathways of A. chroococcum HN were also significantly upregulated (Fig. 4b), and analysis of the membrane transport pathway revealed that the main upregulated genes were livH/M/G/F (branched-chain amino acid transport system permease protein/ATP-binding protein) and dppF/C/B/A (dipeptide transport system ATP-binding protein/permease protein/substrate-binding protein) (Fig. 5b and Supplementary Table 7). The same genes were upregulated in P. aminovorans HPD-2. In addition, the ammonium transporter in the P. aminovorans HPD-2 membrane transport pathway was upregulated. We performed metabolomics analysis to compare amino acid and peptide metabolites between the co-culture and monocultures. The abundances of most dipeptides in the co-culture were significantly lower than those in the A. chroococcum HN monoculture but significantly higher than those in the P. aminovorans HPD-2 monoculture, including aspartylglutamate, prolylasparagine, serinyltyrosine, threoninylphenylalanine, and valyl-hydroxyproline (Fig. 5c and Supplementary Table 9). These differences suggest that P. aminovorans HPD-2 used amino acids and dipeptides secreted by A. chroococcum HN as a source of nitrogen and energy for growth, consistent with the transcriptomic results.

Fig. 5: Distribution of upregulated genes and metabolites across major pathways in the co-culture compared with the monoculture of each species and identification of the main differential metabolites.
figure 5

a Genes related to nitrogen metabolism, aromatic compound degradation, cell growth and death and the TCA cycle were upregulated in A. chroococcum HN and P. aminovorans HPD-2 in co-culture. b Genes related to amino acid metabolism, membrane transport and biofilm formation were upregulated in A. chroococcum HN and P. aminovorans HPD-2 in co-culture. Upregulated gene expression across major pathways was measured by the log 2-fold change (Log2FC) of gene expression in each species in the co-culture compared with the corresponding monoculture. c Metabolite abundances of amino acids and peptides in the monocultures and co-culture of A. chroococcum HN and P. aminovorans HPD-2. d Metabolite abundances of aromatic compounds in the monocultures and co-culture of A. chroococcum HN and P. aminovorans HPD-2. MIX—co-culture of A. chroococcum HN and P. aminovorans HPD-2, HN—monoculture of A. chroococcum HN, HPD-2—monoculture of P. aminovorans HPD-2. Four replicates (N = 4) were used for metabolomics analyses.

Full size image

Among aromatic degradation pathway genes, co-culture significantly upregulated genes encoding key enzymes for ring-opening of PAHs in A. chroococcum HN, such as catE (catechol 2, 3-dioxygenase), pcaH (protocatechuate 3, 4-dioxygenase), and a series of related dehydrogenases (Fig. 5a and Supplementary Table 5). However, co-culture did not upregulate genes encoding key enzymes associated with PAH degradation in P. aminovorans HPD-2. A series of downstream metabolites of aromatic compounds, including salicylic acid, gentisic acid, toluene, phenol, and 3-hydroxyphenylacetic acid, were identified by metabolomics analysis of the monocultures and co-culture (Fig. 5d and Supplementary Table 10). The abundances of salicylic acid, gentisic acid, and phenol metabolites in the co-culture were significantly increased compared to those in the A. chroococcum HN monoculture but significantly decreased compared to those in the P. aminovorans HPD-2 monoculture, implying that downstream metabolites from pyrene degradation by P. aminovorans HPD-2 were used as carbon sources by A. chroococcum HN. This result was consistent with the results of the feeding experiment.

Cell growth and death genes and TCA cycle-related genes were significantly upregulated in P. aminovorans HPD-2 and A. chroococcum HN when grown in co-culture (Fig. 5a), further indicating that these species contributed significantly to each other’s growth and carbohydrate recycling. Biofilm-related genes were significantly upregulated in both species in the co-culture (Fig. 5b and Supplementary Table 8), as was the quorum-sensing pathway in P. aminovorans HPD-2 (Fig. 4d). These results suggest that biofilm formation as well as intercellular signaling may play an important role in interspecies interactions.

External morphology and three-dimensional structure of cells in biofilms from A. chroococcum HN and P. aminovorans HPD-2 co-culture

To further study the interactions and interdependence of P. aminovorans HPD-2 and A. chroococcum HN in pyrene degradation, the external morphology, and three-dimensional structure of cells in biofilms from the co-culture and monocultures and the distribution of pyrene on the bacterial biofilms were examined. A. chroococcum HN produced a large number of biofilms along with pyrene degradation in monoculture, and electron microscopy revealed that the A. chroococcum HN cells were spherical and surrounded by a thick extracellular matrix (ECM), with a small amount of granular pyrene distributed on the surface (Fig. 6a and Supplementary Fig. S7). By contrast, P. aminovorans HPD-2 produced a small amount of biofilms. Electron microscopy of these biofilms showed that the cells were also spherical but were covered only by a thin ECM, and a small amount of granular pyrene was detected on the surface. The co-culture produced a moderate number of biofilms, and electron microscopy showed that the cells were embedded in the ECM and that there was a bridge-like structure between the cells of A. chroococcum HN and P. aminovorans HPD-2 (Fig. 6a and Supplementary Fig. S7). This structure, which EDS indicated was mainly composed of carbon, nitrogen, and oxygen, may contribute to the transportation of substances between cells. In addition, a large distribution of pyrene was found on the surface of the biofilms of both species. Spatial localization analysis by FISH further confirmed these observations. A. chroococcum HN and P. aminovorans HPD-2 maintained a stable relative biovolume in the co-culture biofilm layer (Fig. 6b). The analysis of the three-dimensional structure of the cells in the biofilm showed that A. chroococcum HN predominantly occupied the bottom layer and had extremely high biomass (Fig. 6c, d). By contrast, P. aminovorans HPD-2 occupied the top layer and had a relatively low biovolume in the biofilm layer; however, colonization was greatly increased compared to the monoculture biofilm.

Fig. 6: Microscopic analysis of biofilms from the A. chroococcum HN and P. aminovorans HPD-2 monocultures and co-culture.
figure 6

MIX—co-culture of A. chroococcum HN and P. aminovorans HPD-2, HN—monoculture of A. chroococcum HN, HPD-2—monoculture of P. aminovorans HPD-2. a Scanning electron micrographs of biofilms from the monocultures and co-culture and EDS analysis of the bridge-like structure observed in the co-culture biofilm. b Confocal laser scanning image of FISH-stained biofilm planes of A. chroococcum HN and P. aminovorans HPD-2 from the monocultures and co-culture. Images from left to right include the monoculture of A. chroococcum HN, monoculture of P. aminovorans HPD-2, and co-cultures of A. chroococcum HN and P. aminovorans HPD-2. Red represents A. chroococcum HN, green represents P. aminovorans HPD-2, and blue represents the nucleus. c, d Confocal laser scanning image of FISH-stained biofilm 3D structure from monocultures and co-culture of A. chroococcum HN and P. aminovorans HPD-2 with or without the blue of the nucleus.

Full size image

Discussion

Nitrogen is a key nutrient limiting degraders function, and biological nitrogen fixation by diazotrophs may synergistically contribute to hydrocarbon degradation in nitrogen-free contaminated environments by providing nitrogen to degraders [4, 9]. The mechanisms of synergistic interactions that enhance function in nutrient-poor environments include cross-feeding, communication through chemical signals, and direct contact through biofilms [44,45,46]. In this study, we verified that nitrogen transfer, cross-feeding, and biofilm production were mechanisms of synergy between A. chroococcum and P. aminovorans for promoting growth and degradation (Fig. 7). These synergistic interactions indicate that consortia of diazotrophs and degraders would be the best bioresources for the remediation of PAH-contaminated natural environments.

Fig. 7: Schematic illustration of the interactions between A. chroococcum HN and P. aminovorans HPD-2 during pyrene degradation in a nitrogen-free environment.
figure 7

A. chroococcum HN produces NH4+, amino acids and peptides through biological nitrogen fixation and provides nutrients to P. aminovorans HPD-2 through ammonium ion and amino acid transporters. Cross-feeding of downstream intermediates of pyrene, such as 2, 5-dihydroxybenzoic acid and phthalic acid, between P. aminovorans HPD-2 and A. chroococcum HN synergistically accelerated pyrene degradation.

Full size image

Although it has been proposed that diazotrophs provide nitrogen to degraders and promote hydrocarbon degradation in different environments [5, 7], direct evidence of this relationship is limited. A study reported significant shifts in the 730 cm–1 SERS-15N SIP band for different bacterial species due to the substitution of “light” 14N by “heavy” 15N; thus, SERS-15N SIP is useful for identifying active nitrogen-absorbing cells [41]. In the present study, 15N2/14N2 culture experiments combined with SERS-15N SIP revealed that P. aminovorans HPD-2 was able to take up and utilize nitrogen released by A. chroococcum HN to mitigate nutrient-deficient environments. This study is the first to demonstrate that diazotrophs are capable of supplying nitrogen to degraders in a nitrogen-free contaminated environment.

Under nutrient-limited conditions, oligotrophic strains must rely on metabolites produced by their neighbors, such as vitamins, amino acids, and NH4+, for growth and function [20, 44, 47]. In this work, we found that A. chroococcum HN released large amounts of NH4+ during growth, either by secretion or by release after cell metabolic death. In general, Azotobacter only fixes sufficient nitrogen for its own use and cannot secrete NH4+ to other co-cultured microbes [48]. No NH4+ was detected in the co-culture and the transcriptomic data indicated that the ammonium transporter in the P. aminovorans HPD-2 membrane transport pathway was upregulated. This implies that the NH4+ released by A. chroococcum HN was utilized by P. aminovorans HPD-2 as a nitrogen source. The transcriptomic and metabolomic data further revealed that P. aminovorans HPD-2 significantly promoted the nitrogen fixation pathway of A. chroococcum HN in co-culture, suggesting that the nitrogen metabolites secreted by A. chroococcum HN cross-fed P. aminovorans HPD-2 and accelerated the nitrogen fixation pathway via positive feedback [49]. The up-regulation of genes related to branched-chain amino acid and dipeptide protein transport in co-cultures of A. chroococcum HN and P. aminovorans HPD-2, as well as the related metabolites identified, suggests that A. chroococcum HN also fed amino acids and small peptides to P. aminovorans HPD-2. Overall, these results further emphasize the role of synergistic interactions between microorganisms in a nutrient-deficient environment.

Co-metabolic degradation provides the benefit of removing contaminants as the degraders are not dependent on the carbon or energy source of the contaminants [50, 51]. In this study, A. chroococcum HN did not grow in the absence of an additional carbon source (data not shown), and 2% mannitol was added to the nitrogen-free MSM as a carbon source to promote bacterial growth. A. chroococcum HN degraded pyrene in monoculture, indicating its reliance on additional carbon sources for pyrene degradation. Microorganisms have diverse PAH-degrading characteristics [31]. One of the main forms of interaction between degraders in a microbial consortium during the degradation of organic pollutants such as PAHs and crude oil is the cross-feeding of metabolites [52, 53]. The degradation rate of pyrene was 1.3-2.1 times higher in the co-culture than in either monoculture, implying a synergistic effect. Metabolite analysis revealed differences in the abundances of metabolites such as 2-naphthol, phthalic acid, salicylic acid, and 2,5-dihydroxybenzoic acid between the monocultures and co-culture, further suggesting cross-feeding of pyrene metabolites between A. chroococcum HN and P. aminovorans HPD-2 in co-culture.

Genomic and metabolic data can be used to define possible interactions between partners, as has been done in recent models (constraint-based metabolic modeling) [54]. Pyrene degradation intermediates such as phthalic acid and 2,5-dihydroxybenzoic acid are the most common downstream metabolites in the degradation of PAHs and were detected in this study. Feeding experiments showed that both A. chroococcum HN and P. aminovorans HPD-2 utilized phthalic acid and 2,5-dihydroxybenzoic acid. In particular, A. chroococcum HN efficiently degraded 2,5-dihydroxybenzoic acid, which is produced during pyrene metabolism by P. aminovorans HPD-2. These results imply an effective division of labor between species with complementary genes that facilitates pollutant degradation under stress conditions through positive feedback. These positive interactions in natural soils are essential for microbial growth and survival in communities [55].

The results of transcriptome analysis showed that co-culture significantly promoted PAH degradation pathways, naphthalene degradation pathways, and xenobiotic metabolism by the cytochrome P450 pathway in A. chroococcum HN and P. aminovorans HPD-2. In particular, genes encoding key enzymes of the PAH ring-opening step, such as catechol 2, 3-dioxygenase and protocatechuate 3, 4-dioxygenase, were upregulated in A. chroococcum HN. In most microorganisms, the degradation of aromatic compounds begins with oxidation by dioxygenase to yield a product with two hydroxyl groups [31]. Although co-culture upregulated naphthalene degradation and xenobiotic metabolism by the cytochrome P450 pathway in P. aminovorans HPD-2, it did not promote the expression of key enzymes involved in PAH degradation in P. aminovorans HPD-2. We speculated that this occurred because as a degrader, P. aminovorans HPD-2 is insensitive to pyrene stress; in addition, pyrene was not the only carbon source for growth in the culture environment. By contrast, A. chroococcum HN relies on external carbon sources for degradation and is sensitive to pyrene stress (Fig. S8). Consequently, key PAH-degradation genes were significantly upregulated in A. chroococcum HN but not in P. aminovorans HPD-2. Furthermore, the TCA cycle, oxidative phosphorylation, and ABC transporters were all significantly upregulated in co-cultured A. chroococcum HN and P. aminovorans HPD-2, implying that these pathways are associated with pyrene degradation. The TCA cycle is a major pathway for the further degradation of PAH metabolites and for providing energy to the bacterium. Previous studies have found that the TCA cycle, ATP synthesis pathway, and ABC transporters are all upregulated when PAHs are the sole carbon source [56].

Biofilms are formed by cells adhering to each other on a substrate through the production of extracellular polymeric substance (EPS) and can act as a medium for the exchange of metabolites through cross-feeding [43, 57]. In this study, electron microscopy and FISH experiments showed that A. chroococcum HN produced a large number of biofilms and was found in the bottom layer of co-culture biofilms. P. aminovorans HPD-2 formed fewer biofilms and was tightly attached to A. chroococcum HN. Importantly, biofilm formation by P. aminovorans HPD-2 was significantly higher in the co-culture than in the monoculture, suggesting that co-culture with A. chroococcum HN significantly promoted P. aminovorans HPD-2 growth and biofilm formation. Transcriptomic data showed that co-culture promoted the biofilm formation pathways of A. chroococcum HN and P. aminovorans HPD-2, particularly the ferric ion transmembrane transporter in P. aminovorans HPD-2. Biofilm formation involves the regulatory function of the siderophore system, which enables iron to regulate several functional metabolites [43]. In addition, we found that co-culture significantly promoted quorum sensing in P. aminovorans HPD-2. Microorganisms engage in quorum sensing at the population level to enable microbial cell-to-cell communication. Quorum sensing plays a significant role in coordinating biofilm organization and regulating genes through the use of specific chemical signals known as autoinducers (AI) [58]. The control of biofilm substrate components by intracellular and extracellular signals is achieved through transcriptional and post-transcriptional regulation and environmental factors [59]. Thus, co-culture with A. chroococcum HN in a nitrogen-deficient environment promoted communication between P. aminovorans HPD-2 and A. chroococcum HN cells, resulting in sufficient nitrogen and division of labor to co-metabolize pyrene. Although a bridge-like structure was found between the cells of P. aminovorans HPD-2 and A. chroococcum HN, which may act as a channel for material transfer between cells [60], further in-depth investigation of biofilms is needed to determine how the cells communicate and cooperate with each other.

Conclusions

In this study, we observed synergistic interactions between A. chroococcum and P. aminovorans during the degradation of pyrene in a nitrogen-free contaminated environment. This study is the first to show that the degrader P. aminovorans can receive nitrogen for growth from a syntrophic partner, A. chroococcum. A. chroococcum HN fed P. aminovorans HPD-2 mainly by secreting NH4+, amino acids, and peptides, thus generating a positive feedback mechanism that enhanced nitrogen fixation by A. chroococcum HN. Furthermore, cross-feeding of downstream metabolites of pyrene, such as 2, 5-dihydroxybenzoic acid and phthalic acid, between P. aminovorans HPD-2 and A. chroococcum HN synergistically accelerated pyrene degradation. Enhanced biofilm formation and quorum sensing by P. aminovorans HPD-2 may provide chemical signals and mediators for nitrogen supplementation and metabolite exchange in a nitrogen-deficient environment. In conclusion, this study provides new insights into the mechanisms of synergistic interactions between P. aminovorans HPD-2 and A. chroococcum HN in a nitrogen-deficient contaminated environment that may be useful for improving soil bioremediation.