Natural Attenuation of Petroleum Hydrocarbons: Distinct Microbial Mechanisms in Soil Versus Groundwater

Abstract

Natural attenuation is a potential way to reduce total petroleum hydrocarbons (TPH) contamination, but the microbial mechanisms that explain differences in attenuation performance between soil and groundwater remain unclear. In this study, field investigation and metagenomic analysis were conducted at a decommissioned refinery site with more than 20 years of operation. Over a four-year period, the average TPH degradation rate in the soil attenuation zone reached 307.7 ± 135.2 mg kg⁻¹ year⁻¹, whereas the groundwater attenuation group showed an average degradation rate of 5.2 ± 3.6 mg L⁻¹ year⁻¹. Metagenomic results showed that TPH attenuation in soil and groundwater was associated with two different microbial consortia adapted to local conditions. In soil, the attenuation zone was characterized by a possibly sessile and cooperative consortium dominated by Pseudomonadota and Actinomycetota, with Sphingomonas and Nocardioides as representative genera. The consortium showed broader amino acid metabolic potential (e.g., ko00250, ko00260, and ko00310) and a higher abundance of functions related to biofilm formation and quorum sensing, which may promote stable and surface-attached growth. In groundwater, the attenuation zone was characterized by a possibly motile and more specialized consortium dominated by Pseudomonadota, including Novosphingobium, Sphingorhabdus, and Tabrizicola. The consortium possessed a less complex catabolic network for TPHs and intermediates (e.g., ko01220/ko00621/ko00624; nahAc/catE/fadA/pcaD/atoB), coupled with stronger potential for motility and secretion. In both soil and groundwater, attenuation was associated with lower eukaryotic activity and enrichment of prokaryotic functions related to oxidative stress defenses and high-yield respiration. These results showed that natural attenuation of TPHs in soil and groundwater involved different microbial features, which could improve the evaluation of natural attenuation in heterogeneous environments.

1. Introduction

Petroleum hydrocarbon contamination of soil and groundwater remains a serious global environmental challenge, especially in regions with intensive industrial activity [1]. As a remediation approach, natural attenuation is a passive, low-intervention approach that relies on intrinsic biogeochemical processes to degrade total petroleum hydrocarbon (TPH) and mitigate associated risks over time [2]. Most field investigations have predominantly targeted highly soluble and mobile TPH constituents such as benzene, toluene, ethylbenzene and xylenes (BTEX) [3,4,5,6,7,8], whereas long-term monitoring evidence for heavier, less mobile C10–C40 TPH remains limited [9,10,11,12].

In the limited literature, a 3.5-year field study at a petrochemical site in Liaoning Province (Northeast China) reported continuous decrease in soil C10–C40, with average degradation rates ranging from 4.0 to 70.1 mg kg⁻¹ year⁻¹ at initial concentration of 30~270 mg kg⁻¹ [11]. 16S rRNA analysis further suggested that C10–C40 attenuation was likely linked to bacterial phyla (e.g., Actinobacteria, Chloroflexi, Bacteroidoetes, and Firmicute). Independently, a 4-year field monitoring study at an oil-polluted site in Jilin Province (Northeast China) observed sustained decline in groundwater TPH concentration, with average degradation rates of 0.01–0.5 mg L⁻¹ year⁻¹ at initial concentration of 0.05~2.3 mg L⁻¹ [10]. The main TPH degradation mechanisms were likely related to sulfate reduction and methanogenesis.

Microorganisms play a critical role in the natural attenuation of petroleum hydrocarbons via aerobic or anaerobic pathways [13,14]. Using isolated and cultivated hydrocarbon-degrading microbes, the degradation mechanisms of TPHs have been investigated, including metabolic pathways, key enzymes, and functional genes. Under aerobic conditions, the microbial degradation of petroleum hydrocarbons follows established biochemical pathways. The catabolism of aromatic hydrocarbons typically involves ring-hydroxylating dioxygenases (e.g., nahAc, benA-xylX) and ring-cleavage enzymes (e.g., catE) to transform substrates into central metabolic intermediates [15,16]. For alkanes, the initial oxidation to primary alcohols is catalyzed by hydroxylases such as alkB and cytochrome P450 enzymes [17,18], with subsequent conversion to fatty acids that enter β-oxidation and the TCA cycle. Nevertheless, these isolate-based characterizations alone are insufficient to capture in situ attenuation, which is typically governed by complex microbial communities and metabolic interactions. Thus, metagenomics is essential for revealing TPH degradation potential and pathways at the community level directly in actual environmental samples [19].

Metagenomic studies have shown that microbial communities in petrochemical-contaminated brownfields exhibit different functional adaptations in soil and groundwater under varying contamination levels [20]. Other studies have reported vertical stratification of microbial hydrocarbon degradation potential across soil layers [21]. However, the microbial mechanisms governing multi-year natural attenuation at the same field locations remain unclear.

To delineate the microbial mechanisms governing the natural attenuation of TPHs, a comprehensive four-year investigation was undertaken at a decommissioned refinery site in East China. The site had a two-decade operational history prior to closure, providing an ideal site of long-term hydrocarbon influence. Field investigations revealed significant differences in the degradation rates of TPHs at different locations in the soil and groundwater. Therefore, this study aimed to: (i) employ metagenomic sequencing to decipher the key taxonomic and functional drivers responsible for the differential attenuation efficiencies; and (ii) reveal matrix-dependent mediation by analyzing the divergence in microbial communities and functional potentials between soil and groundwater. Collectively, this mechanistic, multi-scale understanding will provide a scientific basis for developing effective, site-specific remediation strategies, such as targeted biostimulation and bioaugmentation.

2. Materials and Methods

2.1. Selection of TPH-Contaminated Site

This study was conducted at a decommissioned oil refinery site in East China, covering an area of approximately 0.2 km². The site was selected based on two principal considerations: (i) This site was engaged in crude oil refining for over two decades, with documented oil leaks occurring since 1995, resulting in long-term and heterogeneous contamination of both soil and groundwater; (ii) Preliminary on-site investigations revealed spatially heterogeneous exceedances of TPH (C10–C40) concentrations relative to environmental standards in both soil and groundwater (Table S1). Notably, a clear spatial differentiation was observed between zones of TPH attenuation and persistent contamination, which provided a valuable opportunity to comparably analyze the underlying microbial ecological processes governing these divergent fate behaviors.

2.2. Sample Collection and TPH Analysis

Based on preliminary investigations, a total of 18 soil and 12 groundwater samples were collected during two sampling campaigns in October 2020 and July 2024. Preliminary vertical profiling showed C10–C40 hydrocarbons predominantly concentrated in the 0–1 m layer (Table S1); thus, subsequent sampling and metagenomic analysis focused on this surface soil layer. Surface soil samples (0–1 m) were collected, placed in sterile resealable bags, and transported to the laboratory. The distribution of the 9 sampling points, categorized as background (S1–S3), attenuation (S4–S6), and no-attenuation (S7–S9), was presented in Figure 1a. Groundwater samples (W1–W3: Attenuation; W4–W6: No-attenuation) were obtained from monitoring wells, immediately transferred to sterile centrifuge tubes, and transported under cold conditions to the laboratory. Subsamples were either processed for pollutant concentration quantification or stored at −80 °C for high-throughput sequencing. Here, the attenuation group refers to sampling points with a marked decline in TPH concentrations, whereas the no-attenuation group refers to points showing negligible decreases or relatively stable concentrations.

Petroleum hydrocarbons (C10–C40) in soil were quantified using the gas chromatography method specified in HJ 1021–2019. Briefly, 10 g soil sample was homogenized and dehydrated with anhydrous sodium sulfate, followed by extraction using Soxhlet extraction with n-hexane/acetone (1:1, v/v) for 16–18 h. The extract was concentrated to 1.0 mL, cleaned up using a magnesium silicate column, re-concentrated, and adjusted to 1.0 mL for instrumental analysis. Analysis was performed on a gas chromatograph (Agilent 7890B, Agilent Technologies, Santa Clara, CA, USA) with the method detection limit of 6 mg kg⁻¹. Extractable petroleum hydrocarbons (C10–C40) in groundwater were determined according to HJ 894–2017. A total of 1 L groundwater sample was transferred to a separatory funnel and extracted twice with dichloromethane. The combined extract was dried over anhydrous sodium sulfate, concentrated to 1 mL, solvent-exchanged into n-hexane, and then purified on a magnesium silicate column. The analyte was eluted with dichloromethane/n-hexane (1:4, v/v) and finally made up to 1.0 mL with n-hexane for GC analysis (Agilent 7890B) with the method detection limit of 0.01 mg L⁻¹. All samples were collected, transported, stored, and analyzed in total of 15 days. In addition, petroleum hydrocarbons in the C10–C40 range were further resolved into fractions of C10–C12, C13–C16, C17–C21, and C22–C40. These fractions were determined based on retention time windows in gas chromatography, with identification by retention time and quantification by peak area (Table S2).

2.3. DNA Extraction

DNA was extracted from the 9 soil and 6 groundwater samples collected in July 2024 for subsequent metagenomic analysis to characterize the microbial communities. Before DNA extraction, groundwater samples were filtered through 0.22 µm membranes (Millipore, Burlington, MA, USA). All the filters were stored at −80 °C until processing. Genomic DNA was extracted from the 15 samples using the A.E.Z.N.A.TM Soil DNA Kit (Omega Biotech, Inc., Norcross, GA, USA) following the manufacturer’s protocol with triplicate extractions performed for each sample [22]. The concentration and purity of obtained DNA were determined using SynergyHTX microplate reader (BioTek Instruments, Winooski, VT, USA) and a NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA), respectively. DNA quality was further checked on 1% agarose gel electrophoresis.

2.4. Metagenomic Sequencing and Annotation

For paired-end library construction, the extracted DNA was fragmented to an average size of approximately 350 bp using Covaris M220 ultrasonicator (Covaris LLC., Woburn, MA, USA). Sequencing libraries were then prepared from the fragmented DNA using NEXTFLEX Rapid DNA-Seq Kit (Bioo Scientific, Austin, TX, USA). Finally, Paired-end sequencing was performed on an Illumina NovaSeqTM X Plus platform (Illumina Inc., San Diego, CA, USA) at Majorbio Bio-Pharm Technology Co., Ltd. (Shanghai, China), using NovaSeq X Series 25B Reagent Kit (Illumina, San Diego, CA, USA) in accordance with the manufacturer’s instructions (www.illumina.com, accessed on 20 May 2026). In total, 792,539,268 raw reads and 118,880,890,200 raw bases data were generated (Table S3). The data were processed and a non-redundant (NR) gene catalog was constructed as described in Refs. [22,23]. Concurrently, the gene catalogue was aligned against the Kyoto Encyclopaedia of Genes and Genomes database (KEGG) for functional annotation (Detailed analysis results were provided in Text S1 in Supplementary Materials).

2.5. Statistical Analysis

For metagenomic data, to identify differences among groups, Kruskal–Wallis tests were applied on Majorbio Cloud Platform at the taxonomic, functional, and individual gene levels, using the relevant annotation and abundance data. Statistical analyses were also conducted using SPSS (IBM SPSS Statistics 26.0). One-way analysis of variance (ANOVA) with LSD and Tukey’s test was performed to evaluate differences in TPH degradation rates and microbial diversity among different samples. p < 0.05 was considered to indicate a statistically significant difference.

3. Results and Discussion

3.1. Spatial Heterogeneity of TPH Attenuation at the Site

Previous field studies have shown that natural attenuation is a feasible approach for managing TPH contamination. McHugh [24] conducted a large-scale analysis of more than one thousand sites in California and found that TPH contaminant plumes typically stabilized or contracted over time. Another multi-site study based on over 15 years of field monitoring data also confirmed that natural source zone depletion occurred at measurable rates and was likely driven by microbial activity [13].

However, these studies mainly assessed natural attenuation of TPH at the site scale and left the microbial mechanisms underlying spatial heterogeneity unresolved. Understanding the composition and functions of indigenous microbial communities is important for explaining differences in remediation outcomes and evaluating long-term stability. Although recent work by Yang [25] and Martirosyan [26] has linked microbial communities to natural attenuation, the mechanisms that explain differences in attenuation performance between soil and groundwater remain unclear.

We conducted a long-term field study at a decommissioned oil refinery site in East China (Figure 1a). The TPH concentration and fraction distribution at nine soil sampling locations (S1–S9) and six groundwater monitoring wells (W1–W6) in 2020 and 2024 are presented in Table S2. Preliminary investigations showed that the extent and concentration of TPH contamination at the whole site decreased over four years, which suggested natural attenuation (Table S1). Notably, the TPH concentration at this site exhibited long-term heterogeneity, with significant differences between areas of active degradation and persistent contamination. In soil, the average TPH degradation rate in the attenuation zone (S4–S6) was markedly high, at 307.7 ± 135.2 mg kg⁻¹ year⁻¹, approximately 146.5 times that of the background soil (2.1 ± 0.7 mg kg⁻¹ year⁻¹) (Figure 1b). In contrast, TPH degradation in the non-attenuation group (S7–S9) was negative (−4.3 ± 33.4 mg kg⁻¹ year⁻¹), which may be attributed to residual sources or plume migration. In groundwater, the mean TPH degradation rate in the attenuation group (W1–W3, 5.2 ± 3.6 mg L⁻¹ year⁻¹) was significantly higher than that in the non-attenuation group (W4–W6, 0.05 ± 0.13 mg L⁻¹ year⁻¹). This provided a basis for investigating microbial processes associated with TPH attenuation in soil and groundwater.

3.2. Microbial Community Diversity of Different Samples

Microorganisms play an important role in natural TPH attenuation [27]. Mass balance analysis showed that biodegradation accounted for most (~98%) of TPH removal in soils [28]. To identify microbial factors associated with the observed attenuation patterns, metagenomic sequencing was performed to characterize the microbial community and function in different zones. Metagenomic sequencing generated high-quality data for all 15 samples (nine soil and six groundwater), with DNA insert sizes of 434–466 bp (Table S3). After quality control, 99.99% of reads and 99.88–99.93% of bases were retained. Base composition was stable (Figures S1 and S2), and sequencing quality was high (Figures S3 and S4; >Q30). The clean data ranged from 6.1 to 9.2 billion base pairs per sample, supporting subsequent analysis of microbial community structure and function.

The alpha diversity analysis showed different responses of microbial communities in soil and groundwater to TPH contamination and natural attenuation (Table S4). In soil samples, no significant differences were observed in the Chao, Shannon, or Simpson indices among groups. This may be related to physicochemical partitioning and adsorption in the soil, which reduced TPH bioavailability and mitigated ecological stress on microbial diversity [29]. Natural attenuation may involve changes at the functional level including shifts in key degraders and functional genes, even when alpha diversity remained stable [27,30]. This interpretation was supported by observed differences in beta diversity (Figure 2a). In contrast, groundwater microbial communities were more sensitive to TPH concentration and redox conditions [31]. In this study, the Chao index was higher in the attenuation group, indicating higher species richness (Table S4).

3.3. Biomarkers Associated with TPH Attenuation

Previous studies have shown that TPH contamination could enrich specific functional populations such as hydrocarbon degraders and sulfate reducers, without changing overall microbial richness or evenness [32,33,34]. This highlighted the role of specific taxa in TPH attenuation. Therefore, identifying biomarkers in soil and groundwater may help understand microbial mechanisms of natural attenuation in these two matrices.

3.3.1. Diversification of Dominant Taxa in Attenuation Soil

A shift in soil community composition was observed under natural attenuation (Figure 3a). In the non-attenuation group, Pseudomonadota (formerly Proteobacteria) was the dominant phylum (36.7%), consistent with previous reports from oil-contaminated soils [26]. It demonstrated its potential for TPH tolerance. In contrast, the attenuation group showed a more diverse community structure. The relative abundance of Actinomycetota (formerly Actinobacteria) increased from 11.7% to 27.2%, forming a co-dominant pattern with Pseudomonadota (35.9%). The dual-dominance structure is noteworthy, because Proteobacteria and Actinobacteria could serve as key hubs for microbial interactions [35]. Furthermore, these two phyla have been reported to contribute the majority of identified hydrocarbon degradation genes [36].

At the genus level, the significant enrichment of Sphingomonas (Attenuation: 10.3% vs. No-attenuation: 0.9%) and Nocardioides (Attenuation: 5.8% vs. No-attenuation: 0.8%) in the attenuation group (Figure 4a and Figure 5a,b) may support the proposed hypothesis. Specifically, genomic and field studies have identified Sphingomonas as a key taxon in PAH-contaminated soils because of its diverse catabolic pathways and mobile genetic elements for degrading multi-ring aromatics [37,38]. Furthermore, isolates such as Sphingomonas paucimobilis demonstrated a strong capacity to form stable biofilms and survived under low-nutrient or stressed conditions [39]. In addition, Nocardioides (Actinomycetota) is also known to survive under nutrient-limited conditions and to use a wide range of carbon- and nitrogen-containing organic compounds [40]. The co-enrichment of these genera in attenuation soil suggested that they may contribute to TPH degradation together, although their direct interactions were not examined in this study.

3.3.2. Dominance of Key Taxa in Attenuation Groundwater

Unlike the possible multi-phylum collaboration in soil, the groundwater attenuation group was dominated by a single phylum. Pseudomonadota increased in relative abundance from 43.1% in the no-attenuation group to 62.6% in the attenuation group (Figure 3c). Correspondingly, Novosphingobium, Sphingorhabdus, and Tabrizicola were identified as key biomarkers in the groundwater attenuation group (Figure 4c and Figure 5c,d), all belonging to the phylum Pseudomonadota. These genera are commonly reported in aquatic and oligotrophic environments, indicating that they are well adapted to groundwater conditions [41,42,43]. The enrichment of these key taxa suggested that they may collectively contribute to TPH degradation in groundwater. For example, Novosphingobium could serve as the core aromatic degrader, equipped with aromatic ring-hydroxylating dioxygenase genes for attacking persistent pollutants [41,44]. Sphingorhabdus may expand the consortium’s metabolic range by degrading aliphatic and aromatic hydrocarbons [42]. Tabrizicola is possibly involved in the utilization of dissolved organic carbon by-products to help maintain energy flow within the community [43].

3.4. Microbially Mediated TPH Degradation Mechanism

3.4.1. Shared Features of Soil and Groundwater Attenuation Systems

(i)

Reduced eukaryotic functions and enhanced oxidative stress response

A common feature of both soil and groundwater attenuation systems was the marked downregulation of eukaryotic functions (No-attenuation vs. Attenuation), as shown in pathways governing ribosome biogenesis in eukaryotes (ko03008; Reporterscore: −3.35 in soil/−4.04 in groundwater), RNA polymerase (ko03020; −2.35 in soil), and the cell cycle (ko04110; −1.81/−3.79; ko04111; −3.49 in groundwater) (Figure 2b,e and Figure 6a,c). The pathway changes suggested reduced growth-related activity in fungi and protists and were accompanied by downregulation of the central mTOR signaling pathway (ko04150; −3.56 in groundwater) [45]. The reduction in eukaryotic activity was also accompanied by the enhanced oxidative stress response, evidenced by the upregulation of glutathione metabolism (ko00480; 2.76/3.63) in both matrices. The environment-specific induction of the peroxisome pathway (ko04146; 3.66) in groundwater and the chemical carcinogenesis-reactive oxygen species (ROS) pathway (ko05208; 3.76) in soil might suggest the intense oxidative pressure during TPH degradation.

Taken together, the suppressed eukaryotic growth coupled with intense oxidative stress responses suggested that contaminated soil and groundwater may favor metabolically versatile and stress-tolerant prokaryotes during TPH degradation (Figure S9). This interpretation was consistent with previous studies showing that bacterial communities were major contributors to hydrocarbon degradation because of their high taxonomic and functional diversity [46,47]. In a long-term TPH-contaminated soil bioremediation study, Liu [48] reported that bacteria dominated early-stage degradation of saturated and aromatic hydrocarbons while fungi contributed later to transforming intermediate metabolites. Our findings offered a molecular rationale for the niche’s selective advantage: Prokaryotes sustained catabolic activity under oxidative stress, whereas eukaryotes were constrained to survival mode.

(ii)

Convergence on energy-generating metabolic pathways

Microbial communities in both matrices appeared to shift from low-energy-yield fermentative/methanogenic metabolism to high-energy-yield respiratory and oxidative pathways. This shift was evidenced by the upregulation of oxidative phosphorylation (ko00190; 5.18/4.81) and the tricarboxylic acid (TCA) cycle (ko00020; 2.47/1.90) (Figure 2b,e and Figure 6a,c), which was consistent with the substantial energy demand of aerobic and facultatively anaerobic TPH degradation. Intermediates produced from initial hydrocarbon oxidation could enter the TCA cycle and drive ATP synthesis via oxidative phosphorylation, thus supporting the degradation process as reported previously [49,50]. The co-enrichment of these energy-related pathways and aromatic TPH-degrading microorganisms demonstrated a close relationship between energy metabolism and contaminant degradation. In contrast, methane metabolism (ko00680; −9.18/−4.23) was strongly suppressed, suggesting that low-energy-yield terminal-electron-accepting processes were less important in these attenuation systems. This was also consistent with previous studies that respiratory metabolism was thermodynamically and kinetically favored with high energy yields [51,52].

Carbon utilization differed between soil and groundwater, reflecting adaptive metabolic specialization. In soil, the upregulation of pyruvate metabolism (ko00620, 3.66), glycolysis/gluconeogenesis (ko00010, 1.84), starch and sucrose metabolism (ko00500, 3.44), and pentose/glucuronate interconversions (ko00040, 4.50) indicated that the microbial community may actively and concurrently utilize readily degradable carbohydrates. These substrates could serve as co-metabolic or biosynthetic precursors, supporting the rapid establishment and functional redundancy of degraders [53,54].

In contrast, groundwater systems showed downregulation of glycolysis/gluconeogenesis (ko00010, −1.84), starch and sucrose metabolism (ko00500, −2.74), and the pentose phosphate pathway (ko00030, −1.73), coupled with significant upregulation of glyoxylate and dicarboxylate metabolism (ko00630, 4.74). This distinct regulatory pattern revealed a more focused strategy: The aquatic community shifted away from conventional carbohydrate metabolism toward efficient assimilation of C2 intermediates derived directly from TPH degradation, possibly via the glyoxylate cycle [55], representing an adaptation to the more oligotrophic aquatic environment.

(iii)

Synergy of secondary metabolites and lipid catabolism

The biosynthesis of secondary metabolites (ko01110; Reporterscore: 7.29/3.81; KO_number: 1208/1174) and type I polyketide structures (ko01052; Reporterscore: 1.91/2.44) were markedly enriched in both soil and groundwater attenuation systems (Figure 2d and Figure 6a,c). This widespread enrichment suggested that microbial communities could produce secondary metabolites, which may contribute to TPH attenuation and the transformation of degradation intermediates, thereby alleviating their toxic effects [50,56]. Notably, the biosynthesis of enediyne antibiotics (ko01059; 2.16) and staurosporine biosynthesis (ko00404; 3.11) were specifically enriched in groundwater environment. By suppressing competing microbes, these antimicrobial compounds possibly give TPH-degrading taxa a competitive advantage, which helps to consolidate their ecological dominance within the contaminant plume [57].

In addition, the significant upregulation of key fatty acid metabolism pathways, including fatty acid metabolism (ko01212; 3.25/3.77), fatty acid biosynthesis (ko00061; 1.84/2.88), and fatty acid degradation (ko00071; 3.32/2.24), suggested that TPHs might be extensively converted into fatty acids following initial oxidation. These fatty acids then entered the acetyl-CoA/TCA cycle via efficient β-oxidation, thus providing crucial carbon flux and energy support for the natural attenuation process [32,58]. This was consistent with the previously observed upregulation of the TCA cycle (ko00020; Figure 6a,c).

3.4.2. A Sessile and Cooperative Consortium in Soil

In attenuation soil, several amino acid metabolism pathways were significantly enriched. These included valine, leucine and isoleucine degradation (ko00280; 5.56), alanine, aspartate and glutamate metabolism (ko00250; 4.65), glycine, serine and threonine metabolism (ko00260; 3.72), lysine degradation (ko00310; 3.89), tyrosine metabolism (ko00350; 3.04), and biosynthesis of amino acids (ko01230; 2.71) (Figure 6b). The concurrent upregulation of both catabolic and anabolic pathways indicated the presence of an intracellular amino acid metabolic network, reflecting a metabolic pattern that was more complex than a simple linear degradation route. The amino acid network might serve a dual role: Degradative pathways produced alternative carbon skeletons and energy for central metabolism (e.g., acetyl-CoA); while anabolic hubs, like glutamate metabolism (ko00250), drove nitrogen assimilation and de novo protein synthesis to support microbial growth and stress tolerance [59,60,61]. This may be an adaptation pattern for microbiota to the stress of carbon fluctuation and nitrogen limitation in soils.

Beyond intracellular metabolism, the attenuation soil also exhibited upregulation of pathways governing structured multicellular behavior: Biofilm formation-Pseudomonas aeruginosa (ko02025; 1.90), two-component system (ko02020; 3.02), quorum sensing (ko02024; 1.85), bacterial chemotaxis (ko02030; 2.77) and flagellar assembly (ko02040; 2.56), and ABC transporters (ko02010; 8.23) (Figure 6b). This co-enrichment of signaling, motility, and transport modules demonstrated the dominance of densely interacting, surface-attached consortia within the attenuation zone. This was further corroborated by correlation network analysis, which revealed a microbial interaction network of higher complexity and connectivity in soil compared to that in groundwater (Figure 4b,d).

This organized community potentially had an enhanced capacity to sense hydrocarbon/nutrient gradients (via two-component systems), migrate toward favorable micro-environments (via chemotaxis and flagella), and establish resilient biofilms at solid–liquid interfaces [62,63]. Within the biofilm matrix, quorum sensing and two-component systems could coordinate the expression of catabolic genes, stress responses, and the production of extracellular polymeric substances (EPS) [64,65]. In addition, the upregulated ABC transporters may sustain this dense lifestyle by mediating high-affinity uptake of essential nutrients (e.g., amino acids and peptides), thereby maintaining cellular homeostasis [66,67].

3.4.3. A Motile and TPH-Targeting Consortium in Groundwater

In the groundwater natural attenuation zone, a suite of xenobiotic degradation pathways was markedly enriched (Figure 6d). These include degradation of aromatic compounds (ko01220; 6.14), dioxin degradation (ko00621; 3.98), polycyclic aromatic hydrocarbon degradation (ko00624; 2.95), benzoate degradation (ko00362; 3.00), toluene degradation (ko00623; 2.56), xylene degradation (ko00622; 1.98), and chloroalkane and chloroalkene degradation (ko00625; 1.80). This co-enrichment pattern indicated the establishment of a microbial consortium with broad but highly specific catabolic functions related to the degradation of TPHs. These pathways were consistent with sequential transformation of complex hydrocarbons into smaller aromatic intermediates, such as benzoate and alkylbenzenes, which may subsequently enter central metabolism through ring-cleavage pathways [49,68].

The underlying genetic mechanism for this efficient breakdown was revealed by the upregulated TPH degradation pathways and genes (Table S5). This genetic profile showed a coherent aerobic catabolism: Initial oxidation of hydrocarbons by genes like benA-xylX and nahAc, followed by aromatic ring cleavage mediated by enzymes such as catE [15]. Notably, the most pronounced genetic signature was the exceptional abundance of genes governing downstream mineralization, particularly those for β-oxidation (e.g., fadA, fadN) and lactone hydrolysis (pcaD) [69]. This constituted a highly efficient terminal metabolic engine, ensuring the complete breakdown of aromatic intermediates into acetyl-CoA (atoB) for final mineralization via the TCA cycle [70]. Consequently, the microbial communities in this zone provided a strong molecular foundation for natural attenuation and engineered bioremediation strategies.

In addition, the metabolic specialization is coupled with distinct phenotypic adaptations to the aqueous environment. A pronounced upregulation of bacterial secretion system (ko03070; 2.89) and flagellar assembly (ko02040; 5.69) and downregulation of the biofilm formation-Pseudomonas aeruginosa (ko02025; −3.33) was also observed. This indicated that microbes are more dependent on a lifestyle characterized by motility, secretion, and potentially low aggregation. In the flowing, heterogeneous groundwater plume, enhanced flagellar motility enables cells to actively perform chemotaxis along hydrocarbon and oxygen gradients [71], allowing them to repeatedly colonize transient substrate hotspots instead of committing to fixed surfaces. The upregulated secretion system may further facilitate the extracellular release of enzymes and biosurfactants, which increase TPH bioavailability and mediate ecological interactions [72,73].

4. Conclusions

This study demonstrated that the natural attenuation of TPHs at the site constituted an ecological filter. This inhibited eukaryotes while selecting stress-tolerant prokaryotes that exhibited upregulated energy utilization mechanisms, including oxidative phosphorylation, the TCA cycle, and fatty acid metabolism (e.g., ko01212/ko00061/ko00071), as well as biosynthesis of secondary metabolites.

In soil, a possibly sessile and cooperative microbial community formed, characterized by taxonomic diversification (Pseudomonadota and Actinomycetota; Sphingomonas and Nocardioides), upregulation of biofilm formation and quorum-sensing pathways, and the establishment of a dynamic amino acid pool (e.g., ko00250/ko00260/ko00310) to overcome nutrient limitations. These adaptations collectively foster a stable, complexly interactive network. In groundwater, a possibly motile and specialized microbial community dominated (Pseudomonadota; Novosphingobium, Sphingorhabdus, and Tabrizicola). It was defined by abundant catabolic pathways (e.g., ko01220/ko00621/ko00624) and genes (e.g., nahAc/catE/fadA/pcaD/atoB) for TPHs and intermediates, significant upregulation of flagellar assembly and secretion systems, and downregulation of biofilm formation. This trait facilitated tracking of transient contaminant hotspots within the contaminant plume.

Environmental Implication

Findings on microbial consortia in attenuation zones guide the management of TPH-contaminated sites. The stark contrast between the “sessile, cooperative” soil consortium and the “motile, specialized” groundwater consortium invalidates a universal remediation approach. For soil, strategies should foster stable biofilms through biofilm-promoting amendments or engineered syntrophic consortia. For groundwater, interventions must enhance natural plume-tracking, such as by injecting motile specialist degraders or deploying slow-release substrates to guide motile communities to hotspots. The prokaryote-optimized niche also validates leveraging engineered bacterial chassis that incorporate both catabolic genes and the robust stress-response systems of native consortia for efficacy and resilience.