LDPE microplastic particles (MPPs) were prepared by grinding LDPE beads (CSS Industries, Inc., China) using an electric blender (Hanil SHMFP-40000, South Korea) and then passing the particles through a 120-mesh sieve to obtain a size of less than 125 μm. LDPE films were prepared by manually cutting LDPE plastic film (Hanjung Chemical, Geumsan, South Korea) into pieces smaller than 5 mm × 5 mm. Both LDPE MPPs and films were sterilized by washing with 70% ethanol and then dried in an oven at 50°C overnight prior to use [20].

Soil samples for the experiment were collected from agricultural soil exposed to polyethylene (PE) mulching films in G City, South Korea, and transported to the laboratory under cold conditions. Upon arrival, the soil samples were stored at 4°C until use.

Soil bacteria capable of degrading LDPE MPPs were isolated using the enrichment culture technique [21]. Approximately 10 g of soil was added to 250 mL Erlenmeyer flasks containing 100 mL of synthetic mineral medium (SMM) with the following composition (g/L): NH₄NO₃, 1.0; MgSO₄·7H₂O, 0.2; K₂HPO₄, 1.0; CaCl₂·2H₂O, 0.1; KCl, 0.15, supplemented with 0.5% (w/v) LDPE MPPs as the sole carbon source. The cultures were maintained on a rotary shaker at 120 rpm and 28°C for 8 weeks. After the enrichment period, bacterial isolates were obtained by plating on Tryptic Soy Agar (TSA) and further purified by repeated subculturing on TSA plates.

All bacterial isolates were screened for their LDPE MPP degradation capability. Screening was performed by culturing the isolates on Bushnell-Haas (BH) agar medium with the following composition (g/L): MgSO₄, 0.2; CaCl₂, 0.02; K₂HPO₄, 1.0; KH₂PO₄, 1.0; NH₄NO₃, 1.0; and FeCl₃, 0.05; and supplemented with 0.5% (w/v) LDPE MPPs as the sole carbon source [22, 23]. The plates were incubated at 28°C for 7 days. To obtain a uniform dispersion, LDPE MPPs (0.5 g) were first dispersed in 10 mL of warm BH broth; the resulting suspension was then gently incorporated into 100 mL of molten BH agar prior to pouring [23]. Control plates containing BH medium without LDPE MPPs were also prepared. Growth on the plates was observed, and isolates that showed notable growth on LDPE MPP-supplemented BH medium were selected for further investigation.

Molecular identification of bacterial isolates was carried out by Macrogen Inc. (Seoul, South Korea). Genomic DNA was extracted, and the 16S rRNA was amplified through PCR, followed by sequencing with the BigDye™ Terminator v3.1 Cycle Sequencing Kit (Applied Biosystems, USA). The obtained sequences were compared against the National Center for Biotechnology Information (NCBI) GenBank database to determine taxonomic identity based on sequence homology (see supplementary material Text S1 for details).

The selected bacterial strains were cultured in fresh Tryptic Soy Broth (TSB) and incubated at 30°C with shaking (150 rpm) until reaching the logarithmic growth phase, indicated by an optical density (OD₆₀₀) of 1.00. A 5% (v/v) aliquot of the bacterial culture was subsequently inoculated into 250 mL Erlenmeyer flasks containing 100 mL of sterile BH medium supplemented with pre-weighed LDPE MPPs as the sole carbon source. No additional carbon sources were provided to ensure that bacterial growth solely depended on LDPE MPPs. Before inoculation, LDPE MPPs were sterilized by washing with 70% (v/v) ethanol, followed by oven drying at 50 °C overnight. Uninoculated BH medium supplemented with LDPE MPPs was used as the control. At specified intervals (10, 20, and 30 days), residual LDPE MPPs were recovered from the culture medium by filtration through 47 mm glass fiber filters with a pore size of 0.7 μm (Whatman, UK). The particles were then incubated in 2% (v/v) sodium dodecyl sulfate (SDS) for 4 h to detach any adhering cells, rinsed thoroughly with sterile distilled water, and dried overnight at 50°C [20]. The initial and final dry weight of LDPE MPPs in each flask was measured using an analytical balance (Adventurer^TM, AX224KR, Ohaus Cooperation, New Jersey, USA) with a precision of 0.0001 g. The percentage weight loss of the LDPE MPPs was calculated using Eq. (1).

The data were further analyzed to calculate the rate constant for LDPE MPP degradation, applying the first-order kinetic model. This model was selected on the assumption that the degradation rate is proportional to the remaining polymer weight, experimental conditions remained constant for all treatments, and abiotic losses were negligible. Under these conditions, a first-order model provides a description of the weight loss and allows direct comparison of rate constants across strains.

This calculation was based on the initial and final weights of LDPE MPPs over defined intervals of 10 days using Eq. (2).

where Wt is the LDPE MPP weight at time t, W0 is the initial LDPE MPP weight, and K is the rate constant for LDPE MPPs per day.

The half-life (t_1/2) was calculated using Eq. (3).

The surface morphology of LDPE MPPs was analyzed using Field emission-SEM (FE-SEM; MERLIN Compact, ZEISS, Germany), while alterations in the chemical structure of LDPE MPPs were characterized using FTIR spectroscopy (Spectrum 3, PerkinElmer, USA).

CSH of LDPE-degrading bacterial strains was assessed using the bacterial adhesion to hydrocarbons (BATH) assay [24]. Briefly, bacterial cultures in the logarithmic growth phase were harvested by centrifugation at 10,000 rpm and washed twice with phosphate urea magnesium (PUM) buffer with the following composition (g/L): K₂HPO₄, 17; KH₂PO₄, 7.26; urea, 1.8; and MgSO₄, 0.2. Bacterial pellets were resuspended in PUM buffer to obtain an optical density (OD₆₀₀) of 1.0. Aliquots (1.2 mL) of this suspension were transferred into test tubes, and hexadecane (0.2 mL) was added as the hydrophobic phase. The mixtures were vigorously shaken for 10 min and subsequently allowed to stand for 2 min to enable phase separation. The OD₆₀₀ of the aqueous phase was measured using a UV spectrophotometer (Optizen POP QX, Mecasys Co., Republic of Korea), with cell-free PUM buffer serving as a blank control [25]. The CSH of bacterial isolates was calculated using Eq. (4).

Bacterial colonization on LDPE film surfaces was assessed in BH medium. At intervals (0, 3, 7, 10, 20, and 30 days), LDPE films were aseptically recovered from the culture medium. To ensure the removal of loosely adhered bacterial cells and debris, the films were gently rinsed with sterile distilled water. Subsequently, LDPE films were placed in a sterile tube containing 1 mL of saline solution (0.85% NaCl). To detach bacterial biofilms, LDPE films were subjected to ultrasonic treatment using a sonicator (POWER SONIC 605 6C5W160359, Hwashin Technology, South Korea) [26]. Aliquots of the resultant suspension were serially diluted, plated on TSA, and incubated at 30°C for 48 h to quantify viable bacteria as colony-forming units (CFU). To further visualize bacterial colonization on LDPE films, samples were examined using FE-SEM (MERLIN Compact, ZEISS, Germany).

Given the role of laccase in initiating the oxidative depolymerization of PE chains and generating oxidized fragments for downstream metabolism [12, 27], the laccase activity assay in cell-free supernatants was conducted. Laccase activity was assessed at 10, 20, and 30 days of incubation following the modified method of Khruengsai et al. [28]. Culture supernatants were collected by centrifugation at 10,000 × g for 10 min and filtered through 0.7 μm glass fiber filters (Whatman, UK). Laccase activity was determined by the oxidation of 2, 2′-azino-bis (3-ethylbenzothiazoline-6-sulfonic acid) (ABTS). In a 200 μL reaction mixture, 50 μL of culture filtrate was mixed with 0.1 mM ABTS and 100 mM sodium acetate buffer (pH 4.5). Oxidation of ABTS was monitored by recording absorbance changes at 420 nm (ɛ = 36,000 M⁻¹ cm⁻¹) using a UV spectrophotometer (Optizen POP QX, Mecasys Co., Republic of Korea). Laccase activity (μmol min⁻¹) was calculated using Beer-Lambert’s law. One unit of laccase activity was defined as the amount of enzyme catalyzing the oxidation of 1 μmol ABTS per min.

Bacterial protein-coding sequences (RefSeq GCF_017068375.1 and RefSeq GCF_029639525.1) were retrieved in FASTA format from the NCBI RefSeq database and submitted to GhostKOALA for KO assignment [29]. KO annotations were obtained, and key KOs involved in PE and alkane catabolism were extracted, including alkane monooxygenases, alcohol and aldehyde dehydrogenases, acyl-CoA synthetase, β-oxidation enzymes, and multicopper oxidases (MCO). This list was then mapped in reference mode using KEGG Mapper – Reconstruct Pathway (KEGG Mapper Reconstruct tool, v5) to visualize the metabolic pathway of interest. The resulting pathway overlays guided construction of a detailed LDPE biodegradation schematic, annotating each substrate conversion with enzyme names, EC numbers, and KO identifiers from polymer oxidation through β-oxidation to acetyl-CoA formation.

All experiments were conducted in triplicate, and data were statistically analyzed using GraphPad Prism (v8.0.1, GraphPad Software, Boston, USA). Data were first assessed for normality using the Shapiro–Wilk normality test, ensuring that the assumption of Gaussian distribution was met. One-way analysis of variance (ANOVA) was performed to assess differences between groups, followed by Tukey’s post hoc test for multiple comparisons. To explore potential relationships, correlation analysis was performed using Pearson’s correlation coefficient (two-tailed test). Statistical significance was set at p < 0.05, denoting a confidence level of 95%.

In this study, bacterial strains capable of degrading LDPE MPPs were isolated from soil samples collected from plastic-contaminated agricultural fields. After enrichment in SMM medium supplemented with LDPE MPPs as the sole carbon source, five distinct colonies were obtained and designated MFB1, MFB2, MFB3, MFB4, and MFB5 (Table S1). Of these, only MFB1 and MFB5 demonstrated rapid and reproducible growth on BH agar supplemented with LDPE MPPs and were therefore selected for further studies (Table S1). Molecular identification through 16S rRNA gene sequencing revealed that these strains correspond to Rhodococcus koreensis (MFB1) and Gordonia hongkongensis (MFB5), with their nucleotide sequences deposited in GenBank under accession numbers OR084265 and OR084252, respectively (Fig. S1). Notably, this represents the first report of LDPE biodegradation by these specific strains and highlights the adaptive response of indigenous soil bacteria to plastic pollution. Previous studies have demonstrated PE degradation by other species within the Rhodococcus [9, 30, 31] and Gordonia genera [32, 33]; however, this study expands the current understanding of LDPE biodegradation in these genera by demonstrating the degradation ability of R. koreensis and G. hongkongensis.

To determine the biodegradation ability of the isolated strains, LDPE MPP weight loss was monitored over a 30-day incubation period (Fig. 1(a)). Results indicate that R. koreensis MFB1 achieved a 13.34% reduction in LDPE MPP weight, which was slightly higher than the 12.21% reduction observed for G. hongkongensis MFB5 (Fig. 1(a)). The difference in the observed degradation performance of the strains may be attributed to strain-specific characteristics such as extracellular oxidative and hydrolytic enzyme activity, biofilm formation, or production of biosurfactants that increase polymer accessibility [9, 34]. Both strains exhibited a statistically significant reduction in LDPE MPP weight compared to the uninoculated control (Fig. 1(b)), substantiating their biodegradation capabilities.

A comparison of PE weight reduction (Table 1) reveals that the degradation efficiencies of R. koreensis MFB1 and G. hongkongensis MFB5 surpass those reported for previously studied Rhodococcus and Gordonia strains. For example, R. koreensis MFB1 achieved a 13.34% reduction in LDPE within 30 days, notably outperforming Rhodococcus sp., which degraded only 1% of PE over the same period [30]. Similarly, G. hongkongensis MFB5 demonstrated superior degradation efficiency, exceeding the 10.27% reduction reported for Gordonia sp. over the same period [32]. The enhanced degradation observed in this study may be attributed to the prolonged exposure of these strains to plastic-contaminated agricultural soils, which likely facilitated adaptation and the evolution of more efficient metabolic pathways for plastic breakdown [23].

Using a first-order kinetic model, with the assumption that the degradation rate is proportional to the remaining polymer weight and experimental conditions remained constant for all treatments, LDPE biodegradation rates for R. koreensis MFB1 and G. hongkongensis MFB5 were analyzed (Table S2). The degradation rate constant (K) for R. koreensis MFB1 was 0.00445 day⁻¹, corresponding to a half-life (t½) of 155.69 days and tau (τ) of 224.61 days. G. hongkongensis MFB5 exhibited a slightly lower degradation rate constant (K = 0.00402 day⁻¹), resulting in a longer half-life of 172.42 days and tau (τ) of 248.75 days. These results indicate that R. koreensis MFB1 degraded LDPE MPPs at a slightly faster rate than G. hongkongensis MFB5, as evidenced by its shorter half-life (Table S2). However, plateau values for both strains remained low (2.48 × 10⁻¹² g for R. koreensis MFB1 and 8.95 × 10⁻¹³ g for G. hongkongensis MFB5), suggesting that the complete mineralization of LDPE MPPs was not achieved within the experimental period. This aligns with previous studies that reported the slow biodegradation of PE MPs, likely due to the inherent structural resilience and recalcitrance of this polymer [4]. For instance, Bacillus cereus degraded PE with a degradation rate constant of 0.0004 day⁻¹, corresponding to a half-life of 1725 days [36]. These results underscore the challenges associated with achieving efficient LDPE biodegradation under natural environmental conditions.

SEM analysis further demonstrated LDPE MPP degradation after 30 days of exposure to the bacterial strains. SEM images revealed distinct morphological changes on LDPE MPP surfaces, characterized by pit formation, surface cracks, and pronounced surface erosion (Fig. 2). These changes were more evident in LDPE MPP samples treated with R. koreensis MFB1, which exhibited deep cracks and significant surface roughness compared to those treated with G. hongkongensis MFB5. This finding is consistent with the higher LDPE MPP weight loss recorded for R. koreensis MFB1 relative to G. hongkongensis MFB5 (Fig. 1). The observed relationship between the extent of surface degradation and the degree of weight loss underscores the role of microbial activity in facilitating LDPE MPP degradation. Previous studies indicate that biofilm formation enhances enzymatic access to hydrophobic polymer surfaces, facilitating surface degradation [9].

FTIR spectroscopy revealed chemical changes in LDPE MPPs after 30-day exposure to R. koreensis MFB1 and G. hongkongensis MFB5 (Fig. 3). The FTIR analysis of LDPE MPPs revealed characteristic absorption bands consistent with its hydrocarbon structure, with notable peaks at 2916 cm⁻¹ and 2849 cm⁻¹, corresponding to the asymmetric and symmetric C–H stretching vibrations of methylene (−CH₂) groups, respectively. Additionally, peaks at 1472 cm⁻¹ and 718 cm⁻¹ were associated with the bending and rocking vibrations of methylene (−CH₂) groups, consistent with the chemically inert and hydrophobic nature of LDPE [37]. After 30 days of treatment with R. koreensis MFB1 and G. hongkongensis MFB5, significant spectral changes were observed, including the emergence of a new peak at 1760 cm⁻¹, attributed to the carbonyl (C=O) stretching vibration (Table S3). This peak is indicative of oxidation products, such as ketones, aldehydes, or carboxylic acids, resulting from the oxidative degradation of the polymer [9, 32]. The increasing intensity of the carbonyl peak and reduction in methylene band intensities suggest progressive oxidative degradation through chain scission, facilitated by bacterial activity (Fig. 3, Table S3).

These findings corroborate earlier reports indicating that Rhodococcus and Gordonia strains facilitated LDPE oxidation by incorporating oxygen-rich functional groups, which enhance LDPE bioavailability for microbial enzymatic action [9, 32]. While SEM and FTIR changes show surface erosion and oxidation of LDPE, these changes do not confirm complete mineralization or identify low-molecular-weight degradation products; targeted chemical analyses (e.g., GC-MS, pyrolysis-GC-MS, LC-MS/MS, CO₂ evolution assays, and ₁₄C-labelled polymer) are therefore required to characterize degradation products and confirm mineralization.

CSH plays a pivotal role in microbial adhesion to hydrophobic polymers such as LDPE, initiating the biodegradation process [17]. The BATH assay revealed that R. koreensis MFB1 exhibited higher hydrophobicity (70.58%) compared to G. hongkongensis MFB5 (61.15%) (Fig. 4). This enhanced hydrophobicity likely facilitates stronger bacterial adhesion to LDPE surfaces, promoting biofilm formation and extracellular enzyme activity [8]. Consistent with this observation, R. koreensis MFB1 demonstrated higher LDPE MPP degradation efficiency, achieving 13.34% weight loss compared to 12.21% for G. hongkongensis MFB5 (Fig. 1). This observed relationship between CSH and biodegradation efficiency is consistent with previous reports on the role of hydrophobicity in microbial plastic degradation [8, 16, 18]. For example, Adithama et al. [8] observed that hydrophobic bacteria, such as Stenotrophomonas maltophilia, have a greater ability to degrade LDPE. Similarly, Tribedi and Sil [15] reported a correlation between CSH and polyethylene succinate (PES) biodegradation by Pseudomonas sp. These findings emphasize that CSH is a key determinant of the biodegradation potential of bacterial strains.

Fig. 5 illustrates the colonization pattern of R. koreensis MFB1 and G. hongkongensis MFB5 on LDPE film surface over 30 days, determined by viable plate count (CFU mL⁻¹) and SEM. Both strains demonstrated increasing colony-forming units (CFU) counts over time, with R. koreensis MFB1 exhibiting higher colonization. By day 3, R. koreensis MFB1 reached 16.0 × 10⁵ CFU/mL compared to 5.5 × 10⁵ CFU/mL for G. hongkongensis MFB5. Peak colonization for R. koreensis MFB1 was observed at 34.5 × 10⁵ CFU/mL on day 20, whereas G. hongkongensis MFB5 peaked at 27.0 × 10⁵ CFU/mL on day 10, followed by a decline by day 30 (Fig. 5(a)). The decline in colonization could reflect limitations in biofilm sustainability, likely due to competition for resources or spatial constraints on the LDPE surface [38]. This trend parallels other studies that reported a plateau in biofilm formation on LDPE in similar patterns [17, 38].

SEM images provided visual confirmation of bacterial attachment and biofilm formation on LDPE surfaces (Fig. 5(b)). Samples treated with R. koreensis MFB1 exhibited dense biofilm structures and extensive surface roughness, indicative of active colonization and degradation. In contrast, biofilms formed by G. hongkongensis MFB5 were less dense and showed fewer surface alterations (Fig. 5(b)). The observed differences in colonization patterns may stem from the higher CSH of R. koreensis MFB1, which facilitates stronger adhesion to LDPE surfaces, thereby enhancing biodegradation [17]. These findings align with previous studies that highlight microbial adhesion and biofilm formation as critical factors for the biodegradation of hydrophobic polymers [16, 18].

Extracellular laccase is known to catalyze the initial oxidation of PE, introducing oxygen-containing groups (e.g., carbonyls, carboxyls) that facilitate subsequent depolymerization [39, 40]. In this study, both R. koreensis MFB1 and G. hongkongensis MFB5 exhibited measurable laccase activity when cultivated on LDPE MPPs, with activity increasing over the 30-day incubation. R. koreensis MFB1 exhibited the highest laccase activity at Day 30, reaching 5.420±0.174 μmol min⁻¹ (Fig. 6). Moreover, laccase activity in R. koreensis MFB1 correlated strongly with LDPE weight loss (r = 0.953, p = 0.047; Table S4), supporting a key mechanistic role of laccase in initiating LDPE oxidation. Although G. hongkongensis MFB5 also demonstrated a positive trend (r = 0.898), the correlation did not achieve statistical significance (p = 0.103) (Table S4).

These findings further highlight the role of laccase in catalyzing the initial oxidation of LDPE, introducing oxygen-containing functional groups during degradation [39, 23]. In vitro studies demonstrate that laccase-mediator systems acting on UV-aged PE generate new carbonyl and hydroxyl groups, and produce low-molecular-weight compounds such as aldehydes and ketones [40]. Similarly, laccase-like oxidases from R. opacus R7 directly depolymerized commercial LDPE without UV pretreatment to generate oxygenated compounds [39]. Collectively, these results suggest that laccase-mediated surface oxidation generates the initial oxidative fragments that act as biochemical triggers for subsequent PE catabolism in bacteria.

GhostKOALA annotation of 9,598 and 4,467 protein sequences from R. koreensis (RefSeq GCF_017068375.1) and G. hongkongensis (RefSeq GCF_029639525.1), respectively, yielded KO assignments for 3,732 (38.9 %) and 1,951 (43.7 %) protein sequences. A curated set of unique KOs linked to PE and hydrocarbon catabolism enabled the first genome-scale reconstruction of the LDPE biodegradation pathway in these strains (Fig. 7). As shown in Table S5, both strains encode AlkB monooxygenases (K00496, EC 1.14.15.3) that hydroxylate LDPE fragments into primary alcohols. Subsequent oxidation by alcohol dehydrogenases (K00121; EC 1.1.1.1) and aldehyde dehydrogenases (K00128; EC 1.2.1.3) converts alcohol to fatty acids. These fatty acids are activated by acyl-CoA synthetase (K01897; EC 6.2.1.3) and funneled through β-oxidation via acyl-CoA dehydrogenase (K00249; EC 1.3.8.7), enoyl-CoA hydratase (K01692; EC 4.2.1.17), and acetyl-CoA acetyltransferase (K00626; EC 2.3.1.9), feeding acetyl-CoA into the TCA cycle. Both genomes also encode ferredoxin-cytochrome b₅ reductase (K00529; EC 1.18.1.–) to support electron transfer to oxygenases, and lipase (K19311; EC 3.1.1.-), suggesting a shared auxiliary mechanism for liberating fatty acids from oxidized LDPE fragments (Fig. 7, Table S5).

Comparative genomic analysis reveals a significant expansion in gene-copy number of core catabolic enzymes in R. koreensis compared to G. hongkongensis (Fig. S2). For example, R. koreensis encodes 34 aldehyde dehydrogenase homologs (K00128) compared with 6 in G. hongkongensis, and 53 acyl-CoA synthetase homologs (K01897) versus 12, suggesting greater metabolic redundancy in oxidative metabolism. Despite sharing a conserved suite of PE-degrading enzymes, each strain employs a unique extracellular MCO that distinguishes its initial oxidation strategies. R. koreensis encodes a CupA-family MCO (K22552; EC 1.16.3.1), which localizes oxidative activity at the cell surface to target high-molecular-weight polymer chains. In contrast, G. hongkongensis secretes a canonical laccase-type MCO (K00421; EC 1.10.3.2), enabling diffusible oxidation across the polymer surface (Fig. 7, Table S5). Such differences in enzyme localization and secretion are likely to translate into distinct polymer access and depolymerization kinetics. This is consistent with reports of MCO (laccase)-mediated PE oxidation in Rhodococcus opacus [39]. The complementary presence of AlkB, MCOs, lipase, and β-oxidation modules reiterates a multi-enzyme biodegradation model consistent with previous Rhodococcus studies [9]. While genome annotations and extracellular enzyme assays support a plausible LDPE catabolic pathway, these do not demonstrate in vivo expression or catalytic activity for individual enzymes. Future studies should therefore validate key enzyme functions by transcriptomic analysis, proteomic profiling, and targeted biochemical assays. Also, soil microcosm studies and field trials are required to validate the degradation performance of these strains under realistic environmental conditions.