Abstract
Microplastics (MPs) represent persistent pollutants in marine environments and may act as substrates for complex microbial biofilms, collectively referred to as the plastisphere. However, the extent to which microplastic-associated communities differ from surrounding water microbiota and contribute to the environmental dissemination of potentially pathogenic taxa and antibiotic resistance genes (ARGs) remains incompletely understood. In this study, we investigated environmental MPs collected in the Northern Adriatic Sea together with polyethylene MPs experimentally aged for eight months under mesocosm conditions. Using 16S rRNA gene amplicon sequencing, shotgun metagenomics and droplet digital PCR, we compared MP-associated biofilms with the corresponding marine and lagoon water microbiota and assessed selected ARG profiles. Microplastic-associated biofilms displayed significantly lower alpha-diversity and distinct beta-diversity patterns compared with free-living communities, indicating that MPs harboured microbial communities differing from the surrounding water microbiota. Both environmental and experimentally aged MPs showed a higher relative presence of potentially pathogenic genera, with Vibrio representing the dominant genus within the potentially pathogenic component of the plastic microbiota. Shotgun metagenomic analysis revealed a richer and more diverse resistome on environmental MPs compared to seawater, including resistance classes exclusively detected on MPs. Digital droplet PCR confirmed higher normalized abundances of key ARGs (tetA, tetB, sul2) on MPs, together with the preferential detection of clinically relevant genes such as mcr-1 and qnr variants. By integrating environmental MPs, experimentally aged polyethylene MPs, 16S rRNA gene amplicon sequencing, shotgun metagenomics and ddPCR validation, this study provides a combined environmental and experimental assessment of MP-associated microbial communities and selected ARGs in marine and lagoon environments. Together, these findings demonstrate that marine MPs may act as persistent biologically active substrates for biofilm-forming microbial communities that differ from surrounding free-living communities and may contribute to the spread of ARGs, highlighting the need for integrated monitoring of MPs pollution and antimicrobial resistance.
Supplementary Information
The online version contains supplementary material available at https://doi.org/10.1186/s40793-026-00940-5.
Keywords: Microplastics, Marine environment, Microbial community, Pathobiome, Food safety, Public health
Introduction
The increase in worldwide plastic production, mainly driven by the extensive use of single-use plastics, inadequate waste management and low recycling rates, has led to the accumulation of plastic debris in aquatic and marine environments [24, 27, 37]. Microplastics (MPs), commonly defined as plastic particles between 1 μm and 5 mm, may be directly manufactured as primary MPs or originate from the degradation of larger plastic debris as secondary MPs [27, 37].
A variety of MP polymers can be found in the environment, even though polyethylene (PE) and polypropylene (PP) are the most commonly reported polymers in coastal ecosystems, thus reflecting the global plastic production of the market and everyday use [23]. Moreover, polyvinyl chloride (PVC), polystyrene (PS) and polyethylene terephthalate (PET) can frequently be found [1].
Different polymers provide different surface chemistries and physical properties which can influence the initial adhesion of microorganisms and the subsequent establishment of biofilms [7]. In addition to the polymeric composition, MP aging introduces another layer of complexity to the development of the fouling communities. Indeed, these characteristics raise several questions about the potential impacts of MPs on both environmental and human health. Beyond MPs intrinsic potential toxicity, their effects may be amplified through trophic transfer in the marine food webs and, potentially from seafood to humans [30]. Due to their buoyant nature, MPs can be transported over long distances by ocean and river currents. This dispersal capacity may allow MPs to act as rafts for microorganisms and small aquatic organisms, potentially contributing to their transport across aquatic environments. MPs offer hydrophobic surfaces where nutrients and organic matter from the surrounding environment adsorb, creating suitable substrates for microbial colonization and biofilm formation [32], including bacteria, algae and small invertebrates. The microbial colonization of MPs can lead to the establishment of surface-associated biofilm communities and may influence the spatial distribution of microbial taxa and ARGs [48]. The adhesion of microorganisms to MPs surfaces facilitates close cell-cell interactions and horizontal gene transfer, further supporting the persistence and exchange of ARGs. These MP-associated microbial communities are commonly referred to as the “plastisphere”, a term first introduced by Zettler and colleagues in 2013 [46]. Previous studies have shown that MP-associated biofilms can harbor ARGs and bacterial genera of potential sanitary relevance [48], [14, 33, 42]). Several studies have shown that polymer type and environmental exposure can influence the composition of MP-associated microbial communities, including experimental comparisons of plastics with different chemical compositions and field observations from the Tara-Mediterranean expedition [15], [19], [26]. However, the actual ecological and public-health relevance of these findings remains uncertain, because genus-level detection does not demonstrate pathogenicity, viability or virulence, and ARGs detection does not necessarily imply gene expression. Notable differences were detected between microbial communities on MPs and those on particulate organic carbon (phycospheres and detritospheres) [20]. More broadly, particle-attached and biofilm-associated communities are known to differ from surrounding free-living communities in aquatic environments, and environmental biofilms may act as reservoirs of antibiotic-resistant bacteria and ARGs independently of the specific substrate involved [4, 11]. Conversely, other research indicates high similarity between MPs microbiomes and those of surrounding natural particles, challenging the notion of MPs as an additional ecological niche for bacteria [14]. Therefore, MP-associated biofilms should be investigated within the broader framework of particle-associated microbial ecology, particularly in coastal environments where persistent and mobile substrates may contribute to the spatial distribution of selected microbial taxa and ARGs.
Although previous studies have reported the occurrence of potentially pathogenic taxa and ARGs in MP-associated biofilms, integrated studies combining environmental MPs, experimentally aged MPs, shotgun metagenomics and ddPCR-based validation remain limited, particularly in coastal marine and lagoon environments. Therefore, the novelty of the present study lies in the combined environmental and experimental design and in the use of complementary molecular approaches to compare MP-associated biofilms with the corresponding water microbiota. In this context, the present study aimed to investigate whether MP-associated biofilms host microbial communities that differ in diversity, composition and functional potential from surrounding free-living microbial communities in marine and lagoon environments. Specifically, we investigated microbial composition, potentially pathogenic genera and ARGs to assess whether MPs may act as distinct microbial habitats with potential implications for food safety and public health. We hypothesized that: (i) microbial communities colonizing MPs differ taxonomically from free-living communities in the surrounding water; (ii) MPs-associated biofilms are selectively enriched in potentially pathogenic bacterial taxa compared with water samples; and (iii) MPs harbor a broader ARG profile compared to the surrounding water. To address the role of environmental exposure and aging processes, we further explored whether laboratory-made PE MPs subjected to long-term environmental aging develop microbial communities comparable to those colonizing environmental MPs. To test these hypotheses, we combined 16S rRNA gene amplicon sequencing, shotgun metagenomics, and droplet digital PCR (ddPCR) to characterize the plastisphere of environmental and aged MPs and compare it with the surrounding marine and lagoon water microbiota.
Materials and methods
Sampling and analyses of environmental MPs (E-MPs)
In order to characterise the microbial community colonizing environmental MPs (E-MPs), MPs samples were collected during two surveys in April 2021 and April 2022. Sampling was performed in spring, a period generally characterized by increased microbial growth in the marine environment [17]. Floating MPs were collected in the coastal area in front of Pellestrina Island (Venice), North-Western Adriatic Sea (Italy) using a Manta trawl with a 330-µm mesh net. Three replicates were performed during each sampling campaign. Each replicate was carried out for 20 min at 1–3 knots, in calm conditions to allow reliable sample collection on the sea surface. The Manta trawl was provided with a flowmeter to calculate the volume of water passing through the net. After sampling, the Manta net was thoroughly washed to collect all the debris particles stuck to the mesh and samples were transferred into 1-L glass jars using filtered seawater (31 μm). In addition to MPs collection, 1 L of seawater samples (n = 6) were also collected to characterize the planktonic microbial community present in the surrounding marine environment. During sample transport and processing, precautions were taken to avoid contamination: only glass and metal (preferably stainless steel) tools, cotton coats, and nitrile gloves were used. Once in the lab, samples were immediately analysed to preserve the integrity of the microbial community adhering to the MPs surface. Briefly, samples were shaken to facilitate detachment of the MPS from the walls of the 1-L glass jar and then sieved through two metal sieves, 5 mm and 300 μm mesh-size, rinsing the glass container several times with filtered sea water (1 μm). The plastic fragments were then sorted by microscopic inspection in white room [39] and counted to obtain the environmental concentration. Values were expressed as mean values ± s.d. of the number of particles / m3. Blank filters were used during the procedures to check possible contaminations. After microscopic sorting, MP sub-samples were divided into two independent subsets: one subset was used for DNA extraction, excluding fibers, and a separate subset was used forµ-FT-IR polymer characterization. Since DNA extraction targeted the microbial biofilm associated with each particle and did not allow reliable recovery of the same particle for subsequent µ-FT-IR analyses, polymer composition was assessed on the representative µ-FT-IR subset rather than on each individual particle used for molecular analyses. After checking for the normal distribution and homogeneity of variance, evaluated using Shapiro–Wilk’s test and Bartlett test, respectively, the t-test was applied to highlight the possible statistical difference between the first and second sampling campaign. The statistical analysis was performed using the program R (R Core Team 2024, Austria).
Seawater samples for the characterization of the planktonic microbial community were filtered onto 0.45 μm pore-size metallic filters to retain the bacterial fraction. This pore size was selected as a practical compromise between bacterial recovery and the need to limit filter clogging and excessive retention of fine particulate or colloidal material during the filtration, and has been shown to effectively capture microbial diversity in coastal and lagoon environments [22]. Filtered samples were rinsed with sterile seawater and stored in glass tubes at 4 °C until DNA extraction, which was performed within 24 h of collection.
Manufacturing and aging of laboratory-made MPs
Polyethylene aged MPs (A-MPs) were obtained by cryogenic milling according to Lievens et al., [28]. PE was chosen as previous studies performed in the Northern Adriatic coastal area have shown that this is the most common polymer found in sea water, beaches and sediments [10, 39]. Particles were sieved through stainless-steel sieves and those of dimension between 45 and 250 μm were selected and aged for 8 months (from September 2021 to May 2022) using a flow-through system set up at the Hydrobiological Station “Umberto D’Ancona” (University of Padova) in Chioggia (Italy), that was supplied by waters from the Venice Lagoon.
In order to mimic a natural water-column environment, seawater pumped continuously from the lagoon fed three truncated tanks (70 L capacity) at a constant flow of 100 mL/min. In two tanks (i.e. experimental tanks), 12 g of homemade polyethylene MPs were added, while the third tank acted as a control, to understand the possible contamination from the tank itself and from the environment. Before the experiment, the tanks were thoroughly washed with fresh water and hypochlorite to remove any residual of organic matter.
An inlet seawater cascade filtration system provided the tanks with 5 μm-filtered seawater, thus eliminating particulate matter but allowing microbial passage. The system consisted of (i) absorbent cotton (Riwalon®, Dari GmbH, Germany), (ii) one 125-µm filter, (iii) two 25-µm filters, (iv) two 5-µm filters (Atlas Filtri Italia Srl, Italy). Filters were cleaned or replaced as needed. At the outlet of the truncated tanks, 21 μm mesh drum filters were placed to block MPs from exiting the experimental tanks. To ensure oxygenation and movement of the water, each tank was equipped with an aerator. In addition, to prevent aerial contamination, each tank was closed with transparent plastic sheets (Fig. 1). Throughout the experiment, the seawater physico-chemical parameters in the experimental setup were checked daily (for the data see: https://chioggia.biologia.unipd.it/banche-dati/parametri-laguna/).
Fig. 1.

Schematic representation of the inlet seawater cascade filtration system used in the MP aging experiment. Lagoon seawater was continuously pumped at a flow rate of 100 mL/min through a cascade filtration system consisting of absorbent cotton, one 125-µm filter, two 25-µm filters and two 5-µm filters. At the tank outlet, a 21-µm mesh drum filter was placed to prevent MPs from leaving the system
During the aging period, five samples were collected approximately every two months (from September 2021 to May 2022 to monitor the temporal dynamics of the microbial MPs colonization and to identify late-stage biofilm development.
After an 8-month aging period, corresponding to the final sampling point, six 2-L aliquots of water containing the aged MPs were collected from the experimental tanks and were filtered using 45 μm pore-size metallic filters. A sub-sample was analysed using µFT-IR to verify the reliability of the experimental set up and possible contaminations. Filtered MPs were washed with sterile seawater and stored in glass tubes at 4 °C until DNA extraction, which was performed within 24 h of collection. A similar procedure was followed to collect samples (n = 4) of lagoon water from the control tank to characterize the microbial community of the seawater.
MP polymer characterization
The identification of environmental and aged MPs was performed via Micro-FTIR (Nicolet iN10 Thermofisher, Thermo Fisher Scientific, Madison, WI, USA), a Fourier Transformed Infrared Spectrometer coupled with a microscope (LOD = 5 μm; RSD < 5%), equipped with MCT detector (Mercury Cadmium Telluride; cooled with liquid nitrogen). The following parameters were used: transmittance mode, spectral range 4000–650 cm− 1, 100-mm step size scanning (spatial resolution) at 100–100 mm aperture, and 128 co-added scans at the spectral resolution of 4 cm− 1. Identification is considered optimal from a match percentage of 70% (similarity algorithm) [9, 35].
DNA extraction
Nine individual E-MPs were selected for DNA extraction, including five particles from the first sampling campaign and four particles from the second sampling campaign. The three Manta trawl replicates performed during each campaign were used to estimate environmental MP concentrations, whereas DNA extraction was performed on non-fibrous MP particles. In addition, five individual A-MPs collected during the aging process and six individual A-MPs collected at the end of the aging period were processed for DNA extraction. MPs were recovered by centrifugation at 12,000 g for 5 min. Total DNA for the analyses of the plastisphere was extracted from the pellet using the commercial column-based kit QIAmp DNA Mini Kit (QIAGEN, Hilden, Germany), following the manufacturer’s instructions. In particular, the thermal lysis of the bacterial component was carried out at 56 °C for 2 h and RNaseA (100 mg/mL) was added to each sample to guarantee a preparation free from RNA contamination. Total DNA was resuspended in 200µL of nuclease-free water and stored at − 20 °C until library preparation.
Water samples were centrifuged at 12,000 g for 5 min to collect the microbial component. Total DNA for microbiota analyses was extracted from pellet samples after centrifugation, as described for the plastisphere [46].
16 S rRNA gene amplicon sequencing
DNA extracted from E-MPs, A-MPs, marine water and lagoon water was used as a template for amplification of the V3-V4 hypervariable regions of the 16S rRNA gene. The 16S library was prepared according to the Illumina 16S Metagenomic sequencing Library Preparation protocol, using the primers Bact341F and Bact785R (Fwd: CCTACGGGNGGCWGCAG and Rev: GACTACHVGGGTATCTAATCC) and following the instructions of the Nextera XT DNA Library Prep kit (Illumina). The following conditions were applied for the PCR: 95 °C for 3 min, followed by 30 cycles at 95 °C for 30 s, 55 °C for 30 s and 72 °C for 30 s. A final elongation step was performed for 5 min at 72 °C. The amplification was checked by 2% TAE agarose gel electrophoresis to identify a DNA fragment accounting for 550 bp length. Libraries were checked for both concentration and quality using Qubit and 2200 TapeStation (Agilent), respectively. Samples were pooled at equimolar concentration, and sequencing was performed with an Illumina MiSeq platform using a MiSeq 600V3 cartridge (600 cycles, 2 × 300 bp, paired-end reads).
16S rRNA sequence analysis
After sequencing, the raw data generated were further processed in RStudio using the DADA2 package [6]. The sequences were first filtered and trimmed utilizing a truncation length of 240 bp and 180 bp for the forward and reverse reads, respectively, based on the quality profiles of the reads. The filtered sequences were then dereplicated to combine all identical sequencing reads into unique sequences. Then, the forward and the reverse reads were merged to obtain the full denoised sequences. In order to increase the accuracy of the results, chimeras were removed. SILVA ribosomal database (version 138.1) [34] was used to assign taxonomy to the identified Amplicon Sequence Variant (ASV). The ASV table provides a more accurate representation of the microbial community as compared to the Operational Taxonomic Unit (OTU) tables because it can resolve sequence differences as short as a single nucleotide. Statistical analysis was conducted by phyloseq package [31]. To calculate alpha-diversity (observed ASVs, Shannon and Pielou’s index), the count-table was normalized using Geometric Mean of Pairwise Ratio p-values method (GMPR package) [8]. These metrics were used to describe microbial richness, diversity and evenness, rather than absolute microbial abundance or biomass. The non-parametric Kruskal-Wallis test was used to compare alpha diversity among samples types. When significant differences were detected,, the pairwise comparison Wilcoxon test was performed adjusting the p-value (padj-value) for multiple comparisons using the false discovery rate (FDR) method. Beta-diversity was evaluated with Bray-Curtis distance and visualized by Non-metric Multi-dimensional Scaling (NMDS) plot. Permutational Multivariate Analysis of Variance (PERMANOVA) test was used to compare beta-diversity parameters between groups. Tests were considered significant with p-value < 0.05.
The pathobiome concept refers to potentially disease-associated microorganisms considered within the broader context of microbial communities rather than as isolated pathogens [38]. An operational pathobiome screening was performed at genus level by filtering the ASV table according to the comprehensive list of bacterial genera including species associated with human infections reported by Bartlett et al., [5]. Specifically, ASVs assigned to genera that included at least one species listed by Bartlett et al., as associated with human infection were retained. In this study, the term “pathobiome” is therefore used operationally to indicate genera of potential sanitary relevance. Because this classification was performed at genus level, it does not imply that the detected ASVs correspond to pathogenic, viable or virulent strains, nor does it indicate a direct health risk. Results were visualized using barplots and Venn diagrams.
Shotgun metagenomics sequencing
Shotgun metagenomic sequencing was performed only on environmental MPs and the corresponding marine water samples, in order to characterize the resistome associated with MPs collected under real environmental exposure conditions and to compare it with the surrounding planktonic water community. Aged MPs were not subjected to shotgun metagenomic sequencing; they were included in the 16S rRNA gene amplicon sequencing analyses to investigate microbial colonization during experimental aging and in the ddPCR assay to quantify selected ARGs across sample types. The protocol followed was performed using the Nextera XT DNA Library Prep Kit (Illumina), following the manufacturer’s instructions. The quality and quantity of the resulting libraries were assessed using Qubit and the Agilent 2200 TapeStation, respectively. Once normalized to the same molarity, the samples were pooled and diluted to the appropriate loading concentration according to the instructions in the NextSeq 550 Sequencing System Guide. Sequencing was performed using a shotgun metagenomic approach on the Illumina NextSeq 550 platform with a NextSeq 550 High Output Kit v2.5 cartridge (300 cycles, 2 × 150 bp, paired-end).
Resistome characterization
Metagenomic characterization of the resistome was conducted on both E-MPs and the corresponding seawater samples to assess the distribution and diversity of ARGs within the associated microbial communities. Raw reads were processed using TrimGalore (version 0.6.10), applying a Phred quality score cutoff of 20 and retaining only sequences longer than 20 bp. On average, 14 billion base pairs and 50 million reads were processed per sample, of which 99.5% passed trimming and quality-control filtering. Quality data for each sample is reported in Supplementary Data file. The resistome was extracted from the filtered sequences using RGI (Resistance Gene Identifier, version 6.0.3) with the RGIbwt module, employing the CARD (Comprehensive Antibiotic Resistance Database) as reference and default parameters. The comparison was based on protein homology models. In addition, because the metagenomic analysis was designed as a read-based ARG profiling approach, the genomic context of ARGs, including their possible physical association with mobile genetic elements or metal resistance genes, was not assessed. For downstream analyses, ARG richness was defined as the number of distinct ARGs detected in each sample after CARD/RGI annotation. ARG count tables were analysed by phyloseq package [31]. For alpha-diversity measures (richness, Shannon and Pielou’s index), the count-table was normalized using Geometric Mean of Pairwise Rations method (GMPR package) [8]. The pairwise comparison Wilcoxon test was used to compare richness and diversity. Tests were considered significant with p-value < 0.05.
ddPCR quantification
Following the in-silico identification of the resistome in the E-MPs and seawater, quantitative analysis of a selected panel of ARGs was performed by ddPCR on DNA extracted from E-MPs, A-MPs, marine water and lagoon water, to determine their prevalence across all samples and results were normalized using 16S rRNA gene abundance to account for differences in microbial biomass among sample types.
The following resistance genes were tested: blaCTXM, blaSHV (representative β-lactam resistance genes); blaOXA48 (representative carbapenemase gene, class D β-lactamase); qnrS, qnrA (quinolone resistance genes); sul2, sul3 (resistance genes against sulphonamides); tetA, tetB, tetM (conferring resistance to tetracyclines, respectively); ermB (representative of genes conferring resistance to macrolides, lincosamides, and streptogramins); strB (representative resistance gene against aminoglycosides) and vanA (representative glycopeptide resistance gene). In addition, mcr-1 was included as a clinically relevant antibiotic resistance gene. The primers used are listed in Table S1. ddPCR reactions were prepared in 22 µl total volume using the QX200 ddPCR EvaGreen Supermix, primers at 3 µM, 2 µl of DNA, and nuclease-free water. 20 µl of each mixture was loaded into a DG8 cartridge along with 70 µl of QX200 droplet generation oil (Bio-Rad). The DG8 cartridge was placed in the QX200 Droplet Generator (Bio-Rad). Generated droplets were carefully transferred to a 96-well PCR plate for amplification on a C1000 Touch thermal cycler (Bio-Rad). Each run included positive controls (target gene amplified) and no-template controls (NTC). The PCR program recommended by Bio-Rad was: 95 °C for 5 min, 40 cycles of 95 °C for 30 s and annealing/extension (optimized for each target gene) for 1 min, with a 2 °C/s ramp rate, followed by two final steps at 4 °C for 5 min and 90 °C for 5 min. Plates were then read using the QX200 Droplet Reader (Bio-Rad), and only reactions generating more than 10,000 droplets were analyzed. Thresholds distinguishing positive from negative droplets were set manually, and only samples with at least three positive droplets were considered positive. To account for differences in microbial biomass, the 16S rRNA gene was used as a proxy for total microbial community abundance, and counts of each resistance gene were normalized to the number of 16S rRNA gene copies, resulting in values expressed as gene copies per 16S rRNA gene copy. Normalized abundances were compared among samples types using non-parametric tests. For each target gene, pairwise Wilcoxon rank-sum tests were performed when sufficient positive detections were available, and p-values were adjusted for multiple comparisons using the FDR method. Genes with sparse detections were interpreted descriptively. Tests were considered significant with p-value < 0.05.
Results
Quantification and characterization of environmental MPs (E-MPs)
Sampling locations, environmental conditions and the concentrations of E-MPs collected during the surveys performed in 2021 and 2022 are reported in Supplementary Table S1.
The mean MPs concentrations were 0.51 ± 0.29 particles / m3 and 2.97 ± 0.87 particles / m3 for the first and second survey, respectively. The results from the two surveys were significantly different (p < 0.01). Fibers were the prevailing type of MPs, followed by flat and irregular fragments (Supplementary Fig. S1A). The µFT-IR characterization of E-MPs showed that PE is the most common polymer, followed by polyoleofin and PP. The “composite material” was made of wood and Formica® (a laminated plastic) (Supplementary Fig. S1B).
Characterization of aged MPs (A-MPs)
The presence of only PE A-MPs was confirmed in the experimental tanks through the µFT-IR analysis. In the first tank, the MPs length ranged from 550 μm to 180 μm, and the width from 178 μm to 89 μm. In the second tank, the optimally identified particles were also mainly polyethylene, with the length ranging from 330 μm to 120 μm, and the width ranging from 80 μm to 40 μm.
Plastic particles were also found in the tank used as control. The only identified polymers were rayon, polyester, and sea silk (likely originating from the filtration system and / or from the environment), with the length ranging from 510 μm to 65 μm and the width ranging from 115 μm to 29.2 μm.
Microbial colonization dynamics during MPs aging
The analysis of the microbial community colonizing MPs over the 8-month aging period showed that alpha-diversity indices varied over time, with no further marked increase at the final sampling point. (Fig. 2A–B).
Fig. 2.

Alpha diversity during microplastics aging A. B. Boxplots of the number of identified ASVs and Shannon diveristy in MP-A surface during the aging period. The significance of differences between the samples is indicated by asterisks: * p < 0.05; ** p < 0.01; *** p < 0.001; **** p < 0.0001
In particular, the ASVs increased over time from T1 to T4, then slightly decreased, whereas the Shannon index reached the highest value at T3. The final community exhibited a consistent taxonomic structure at the class level, dominated mainly by Gammaproteobacteria, Cyanobacteria, Bacteroidia, Alphaproteobacteria, and Planctomycetes (Fig. 3). These results suggested that, after 8 months of aging, the MP-associated microbial community had developed a late-stage biofilm structure characterized by a consistent taxonomic profile at the class level. Moreover, since T5 corresponded to May 2022, close to the environmental sampling, A-MPs and E-MPs were exposed to partially overlapping seasonal conditions. However, comparisons between A-MPs and E-MPs should be interpreted cautiously because the particles differed in origin, size, polymeric composition and exposure history.
Fig. 3.

Relative abundance during microplastics aging. Stacked bar charts showing relative abundance (%) of microbial communities, annotated to the taxonomic level of class derived from experimental microplastics during aging period
Identification of microbial communities on MPs and surrounding waters
To test whether microplastic-associated biofilms host microbial communities distinct from those inhabiting the surrounding aquatic environment, alpha- and beta-diversity analyses were performed comparing E-MPs, A-MPs, marine water and lagoon water. As no significant differences were observed in the microbial community colonizing E-MPs collected in 2021 and 2022, data from the two sampling campaigns were merged to increase statistical robustness. Sequencing of MPs samples collected after the aging period yielded a total of 3,707,514 raw reads (mean per sample = 137,315 reads). After quality filtering, 3,663,558 reads (mean per sample = 135,687 reads) were retained, resulting in 1198 unique Amplicon Sequence Variants (ASVs suitable for comparative analyses). Alpha-diversity analyses revealed significant differences between microplastic-associated and free-living microbial communities (Fig. 4A–C). For observed ASVs, the Kruskal-Wallis test showed significant differences among samples types (χ² = 17.18, df = 3, p = 0.0006). Pairwise Wilcoxon tests with FDR correction indicated significant differences between lagoon water and A-MPs (p-adj = 0.0143), marine water and E-MPs (p-adj = 0.0051), marine water and A-MPs (p-adj = 0.0100), and A-MPs and E-MPs (p-adj = 0.0051), whereas no significant differences was observed between marine and lagoon waters (Fig. 4A). Similar patterns were observed for Shannon diversity, which also differed significantly among samples types (Kruskal–Wallis test, χ² = 15.8821, df = 3, p = 0.0012). Significant pairwise differences were detected between lagoon water and A-MPs (p-adj = 0.0143), marine water and E-MPs (p-adj = 0.0024), marine water and A-MPs (p-adj = 0.0065), and A-MPs and E-MPs (p-adj = 0.0143) (Fig. 4B). Pielou’s evenness index also differed significantly among samples types (Kruskal–Wallis test, χ² = 13.5293, df = 3, p = 0.0036), with significant pairwise differences between marine water and lagoon water (p.adj = 0.0190), marine water and E-MPs (p.adj = 0.0065), and marine water and A-MPs (p.adj = 0.0065). (Fig. 4C).
Fig. 4.

Diversity’s analysis of microbial community A. B. C. Boxplots of the number of identified ASVs and alpha-diversity (Shannon and Pielou indices) in the four kinds of samples: lagoon water, cryo-milled aged microplastics (A-MP), environmental microplastics (E-MP) and marine water. Extreme values are indicated by dots. D. non-metric multidimensional scaling (NMDS) plot of beta-diversity. Each point represents the microbial community of a sample. Samples with similar microbiota composition tend to be in the same area of the plot, while points far apart from each other represent samples with dissimilar microbiota. Kinds of samples are identified by the same colour used in the boxplots. The significance of differences between the samples is indicated by asterisks: * p < 0.05; ** p < 0.01; *** p < 0.001; **** p < 0.0001
Beta-diversity analysis based on Bray–Curtis dissimilarity showed a clear separation between microplastic-associated and free-living microbial communities, as visualized by non-metric multidimensional scaling (NMDS) (Fig. 4D). Samples clustered according to matrix type, with E-MPs, A-MPs, marine water and lagoon water forming four distinct community groups. PERMANOVA confirmed that sample type significantly explained variation in microbial community composition (999 permutations, df = 3, R² = 0.2799, F = 2.7211, p = 0.0010).
Taxonomic profiling at the family level further supported these differences in community composition (Fig. 5). Flavobacteriaceae and Pseudoalteromonadaceae were consistently abundant across all samples types. However, pairwise comparisons revealed a higher relative abundance of Vibrionaceae in A-MPs compared to lagoon water (Fig. 5A), and an enrichment of Alteromonadaceae in E-MPs compared to marine water (Fig. 5B), indicating matrix-specific taxonomic signatures associated with microplastic surfaces.
Fig. 5.

Relative abundance of microbial community. Stacked bar charts showing relative abundance (%) of microbial communities, annotated to the taxonomic level of family, derived from experimental (A) and from environmental conditions (B)
Analysis of shared and unique taxa at the family level (Fig. 6) highlighted both overlap and matrix-specific microbial assemblages among the different samples types. Five bacterial families: Pseudoalteromonadaceae, Vibrionaceae, Alteromonadaceae, Alcanivoracaceae, Flavobacteriaceae, were shared across all matrices, collectively accounting for approximately 10–20% of the total microbial community. In contrast, several families were uniquely associated with specific environments. Desulfonatronaceae, Xanthomonadaceae, Sandaracinaceae, Nitrosomonadaceae, Balneolaceae, Ectothiorhodospiraceae, Verrucomicrobia subdivision 3, Saccharospirillaceae, Coxiellaceae, Spirosomaceae, Reichenbachiellaceae, Puniceicoccaceae, Marinilabiliaceae, Rikenellaceae, Thiotrichaceae and Burkholderiaceae were exclusively detected in lagoon water. Pirellulaceae and Parachlamydiaceae were uniquely associated with A-MPs, whereas Methylococcaceae, Piscirickettsiaceae and Cryomorphaceae were detected only in marine water. Listeriaceae, Staphylococcaceae, Enterococcaceae and Desulfobacteraceae were exclusively identified in E-MPs.
Fig. 6.

Unique and shared bacterial taxa. Venn diagram showing the number of unique and shared bacterial taxa in the four sample groups at the family taxonomic level. Values within the intersections represent the shared families, while values outside the intersections are families unique to each group
Distribution of potentially pathogenic genera in MP-associated biofilms and surrounding waters
To assess the distribution of genera of potential sanitary relevance, an operational pathobiome screening was performed at genus level by filtering the ASV table according to Bartlett et al., [5]. This analysis retained ASVs assigned to genera including at least one species previously associated with human infection. The screening revealed both shared and matrix-specific patterns within this operationally defined subset (Table 1; Fig. 6). Only one genus, Vibrio, was detected across all sample types and represented the dominant genus within the MP-associated pathobiome subset. In both aged and environmental MPs, the higher relative contribution of this subset was mainly driven by Vibrio. In contrast, other genera included in the pathobiome, such as Salmonella, Pseudomonas and Flavobacterium, were more frequently associated with water samples than with MP-associated communities.
Table 1.
Pathobiome composition. For each kind of samples, the percentages of pathogens, at the genus level, are reported
| Bacterial genus | Lagoon water | A-MP | E-MP | Marine water |
|---|---|---|---|---|
| Vibrio | 32% | 97% | 91% | 40% |
| Legionella | 4% | 2% | 0% | 0% |
| Flavobacterium | 0% | 1% | 0% | 19% |
| Bacillus | 0% | 0% | 0% | 0% |
| Psychrobacter | 0% | 0% | 4% | 2% |
| Salmonella | 36% | 0% | 0% | 19% |
| Pseudomonas | 16% | 0% | 4% | 9% |
| Alistipes | 5% | 0% | 0% | 0% |
| Lactobacillus | 3% | 0% | 0% | 0% |
| Coxiella | 3% | 0% | 0% | 0% |
| Photobacterium | 1% | 0% | 0% | 2% |
| Escherichia | 0% | 0% | 0% | 0% |
| Enterococcus | 0% | 0% | 0% | 0% |
| Listeria | 0% | 0% | 0% | 0% |
| Moraxella | 0% | 0% | 0% | 9% |
| Staphylococcus | 0% | 0% | 0% | 0% |
Some potentially pathogenic taxa displayed matrix-specific occurrences. Members of the family Moraxellaceae were detected only in marine water samples, whereas Bacillaceae were identified exclusively in A-MPs. No potentially pathogenic genera were uniquely associated with E-MPs (Fig. 7A). The relative contribution of potentially pathogenic taxa was higher in microplastic-associated communities than in water samples (Fig. 7B). In particular, potentially pathogenic genera accounted for approximately 15% of the total microbial community in A-MPs and 5% in E-MPs, whereas lower proportions were observed in lagoon water (4%) and marine water (1%).
Fig. 7.

Pathobiome distribution. (A) Venn diagram showing the number of unique and shared bacterial taxa in the four sample groups at the genus taxonomic level. Values within the intersections represent the shared genera, while values outside the intersections are genera unique to each group. (B) Stacked bar chart depicting % of taxa belonging to pathobiome and non-pathobiome. Red bars show the % of pathobiome, and blue bar represents other taxa
Resistome profiles in environmental MPs and surrounding seawater
To investigate whether microplastic-associated microbial communities harbor an enriched resistome compared to free-living communities in the surrounding water column, shotgun metagenomic analyses were performed on environmental MPs (E-MPs) and corresponding marine water samples.
A total of 605 ARGs were identified across all samples, indicating the presence of a diverse and complex resistome in the studied system. Overall, E-MPs exhibited higher ARG richness compared to marine water (Fig. 8), revealing clear and statistically significant differences between the two matrices. This pattern indicates a broader resistome profile associated with E-MPs biofilms.
Fig. 8.

Antibiotic resistance gene richness in environmental microplastics and marine water samples. Barplot showing the number of distinct ARGs detected in the two different type of samples and grouped by resistance class
The resistome composition was dominated by genes conferring resistance to macrolides, fluoroquinolones, carbapenems, aminoglycosides, penams and tetracyclines. Despite this shared core composition, differences emerged in the distribution of specific resistance classes between MPs and seawater. In addition to hosting a richer resistome, E-MPs harbored several ARG classes that were not detected in the surrounding marine water. These included genes associated with resistance to aminocoumarins, bicyclomycin-like antibiotics, antibacterial free fatty acids and pleuromutilin antibiotics, highlighting a selective enrichment of specific antibiotic resistance determinants on microplastic surfaces.
Quantification of ARGs by ddPCR
To validate and quantify the enrichment of ARGs detected by shotgun metagenomics, (ddPCR was used to quantify a selected panel of clinically and environmentally relevant ARGs across MPs and water samples. ddPCR analysis revealed marked differences in ARG abundance between MPs and the surrounding water matrices (Fig. 9). Lagoon water and marine water samples generally showed low and sporadic detections of ARGs, mostly limited to tetracycline- and sulfonamide-resistance genes, with normalized values ranging between 10−5 and 10−3 copies per 16S rRNA gene copy.
Fig. 9.

Quantification of ARGs. Dot-heatmap showing the normalized abundance of selected ARGs in water and microplastic samples. Dot size represents ARG abundance expressed as gene copies normalized to 16S rRNA gene copies, while dot colour indicates the sample category
In contrast, microplastic-associated samples displayed a broader ARG profile and higher normalized abundances. Aged MPs (A-MPs) frequently harbored quantifiable levels of tetA, tetB, sul2, and strB, with several samples showing abundances up to one order of magnitude higher than those observed in lagoon water. Environmental MPs (E-MPs) showed the highest ARG number of detected ARGs within the ddPCR panael and higher normalized abundances of selected ARGs compared with aged MPs and water samples.
E-MP samples showed quantifiable levels of tetA, tetB and sul2, as well as low but detectable levels of qnrA, despite its absence or near absence in marine water samples. Notably, clinically relevant resistance determinants, including mcr-1 and the genes sul3, qnrA and qnrS, which were only sporadically detected in lagoon or marine water, were more frequently recovered from both A-MPs and E-MPs. Therefore these specifically patterns are reported descriptively and should be interpreted cautiously.
Discussion
Microplastic particles were detected in all the samples collected during the two sampling campaigns with a marked interannual variability between 2021 and 2022 surveys. Concentrations observed in 2021 were significantly lower than those measured in 2022 and slightly lower than values previously reported for landward stations in the same coastal area [39], while remaining higher than those reported for the coasts of Italian minor islands [12]. Such variability is consistent with the well-documented multiple factors influencing MPs distribution and abundance, including source pressure, hydrodynamic conditions and meteorological events [40]. The lower concentrations observed in 2021 might be partly explained by reduced freshwater inputs preceding the sampling period, which was characterized by exceptionally low rainfall, whereas the higher concentrations detected in 2022 coincided with increased precipitation, consistently with the role of rainfall-driven runoff in transporting MPs to aquatic environments [2]. Rainfall and runoff may therefore contribute to temporal variability in MP inputs to coastal waters [2]. Additionally, the reduced urban and touristic activities associated with COVID-19 restrictions in 2021 might have contributed to lower plastic littering. In this context, a study focused on the analysis of beach litter datasets, highlighted that public litter, i.e. domestic and touristic waste deriving from public activities, is the most abundant litter source in the coastal area facing the Lagoon of Venice [18].
Polyethylene was the dominant polymer among E-MPs, consistently with previous observations in the Northern Adriatic Sea, and was therefore selected for the laboratory aging experiment [10, 39]. However, environmental MPs showed a more heterogeneous polymeric composition than A-MPs. Polymer chemistry, surface hydrophobicity, roughness and porosity can influence the initial adhesion of microorganisms and subsequent biofilm development. Therefore, part of the observed differences between E-MPs and A-MPs may reflect not only differences in environmental exposure and aging history, but also differences in polymeric composition and surface properties.
The present study shows that MP-associated biofilms differed in diversity and taxonomic composition from the surrounding free-living water microbiota. Both environmental and laboratory-aged MPs displayed lower alpha-diversity and distinct beta-diversity patterns compared with marine and lagoon waters, indicating that MP surfaces supported surface-associated microbial communities that were compositionally different from the surrounding planktonic communities. However, because natural particle or non-plastic biofilm controls were not included, these differences cannot be attributed exclusively to plastic-specific properties. These findings are consistent with previous studies describing the plastisphere as a unique ecological niche shaped by the physicochemical properties of plastics, including surface hydrophobicity, roughness and adsorption of organic matter [26, 46, 47]. The observed differences at the family level, together with the presence of both shared and matrix-specific taxa, suggest that microplastic-associated communities arise from a combination of stochastic colonization and selective processes driven by surface characteristics and environmental conditions.
The stabilization of microbial communities observed after eight months of experimental aging further indicates that plastisphere assemblages may undergo successional dynamics leading to a late-stage biofilm community with a relatively consistent taxonomic structure, in agreement with previous experimental and field-based observations [29]. Although environmental MPs differ from laboratory-aged particles in size, shape and exposure history, the consistent separation between microplastic-associated and water-associated microbiota indicate that MPs supported biofilm communities that differed from the surrounding free-living microbial communities. Altered metabolic capacities associated with microplastic biofilms may influence nutrient availability, microbial competition and energy flow within marine food webs, ultimately affecting ecosystem structure and function [25]. In this context, the establishment of distinct plastisphere communities observed in the present study may have broader ecological implications beyond microbial community composition. The uneven distribution of Gammaproteobacteria between MP-associated biofilms and surrounding waters is noteworthy from an ecological perspective, as several members of this class are commonly associated with marine particles, surface colonization and biofilm lifestyles. This class also includes genera of potential sanitary relevance, such as Vibrio, which was further detected in the genus-level operational pathobiome screening. However, taxonomic resolution at the class level does not allow inference of pathogenicity, virulence or health risk. Therefore, the higher relative contribution of Gammaproteobacteria should be interpreted as a broad taxonomic signal potentially linked to surface-associated lifestyles, while genus-, species- or strain-level analyses are required to assess its sanitary relevance.
The operational pathobiome screening showed a higher relative contribution of genera of potential sanitary relevance in MP-associated biofilms compared with surrounding waters. However, because this analysis was performed at the genus level, these results should not be interpreted as evidence of pathogenic strains, virulence, or direct health risk. In particular, both environmental and aged MPs exhibited a higher relative proportion of potentially pathogenic taxa, with Vibrio representing the dominant genus within this defined subset. The predominance of Vibrio on MPs has been widely reported in marine environments and is often attributed to the affinity of this genus for particulate surfaces and biofilm-associated lifestyles ([26],[48]). However, it is important to emphasize that genus-level detection does not allow inference of pathogenicity, virulence or health risk, as many Vibrio species are non-pathogenic or opportunistic. The presence of other potentially pathogenic genera, such as Salmonella and Pseudomonas, predominantly in water samples rather than on MPs further highlights the complexity of interpreting pathobiome data in environmental studies. Nevertheless, the enrichment of specific potentially pathogenic genera on microplastic surfaces supports the hypothesis that MPs may act as selective substrates favouring taxa with biofilm-forming capabilities and surface-associated lifestyles. These findings align with recent studies suggesting that MPs can influence the spatial distribution of microbial taxa of potential relevance for food safety, particularly in coastal environments [5, 14].
A major strength of this study is the combined use of shotgun metagenomics and ddPCR to characterize ARG profiles in MP-associated biofilms and surrounding waters. Shotgun metagenomic analysis revealed a broader ARG profile on environmental MPs compared to surrounding seawater, with dominance of genes conferring resistance to macrolides, fluoroquinolones, carbapenems, aminoglycosides, penams and tetracyclines. Several resistance classes were detected exclusively on MPs, suggesting that these surfaces may act as reservoirs for rare or environmentally relevant ARGs. Quantitative ddPCR analysis confirmed these patterns, demonstrating higher normalized abundances of selected ARGs on both environmental and aged MPs relative to water samples. The detection of clinically relevant resistance determinants, including mcr-1, qnrA and qnrS, predominantly on microplastic-associated biofilms is consistent with growing evidence that MPs can act as reservoirs or accumulation sites for antimicrobial resistance determinants in aquatic systems [3, 14, 41]. These findings are also in line with recent studies showing that plastisphere-associated communities can harbor ARGs, mobile genetic elements and taxa of potential sanitary relevance in aquatic environments [36, 42, 43, 45]. However, these studies also highlight that the ecological and public-health relevance of MPs depends on the environmental context, the surrounding water microbiota, the presence of mobile genetic elements and the comparison with other particle-associated biofilms. In this framework, the added value of the present study lies in the integration of environmental and experimentally aged MPs with complementary molecular approaches, including shotgun metagenomics and ddPCR validation of selected ARGs. Although the present read-based approach revealed differences in ARG profiles between MP-associated biofilms and surrounding waters, the mechanisms underlying these patterns cannot be resolved from these data alone. Biofilm-associated features, such as increased cell density, close cell–cell contact and organic matter accumulation, may create conditions compatible with ARG persistence or exchange; however, assembly-based analyses integrating mobile genetic elements, metal-resistance genes and biocide-resistance genes would be required to investigate potential ARG mobilization or co-selection mechanisms [44].
Together, these factors support the hypothesis that MPs may function as persistent surface-associated microhabitats where antibiotic resistenze determinants can be retained or detected more frequently than in the surrounding water column. Previous studies have shown that plastisphere-associated biofilms are enriched in genes related to plastic degradation, resistance to toxic substances and other stress-response mechanisms [13, 21]. These functional traits suggest that microplastic-associated microbial communities are well adapted to chemically complex substrates and may contribute to the persistence of ARGs observed in the present study.
Despite the clear patterns observed, this study has some limitations that should be considered when interpreting the results. The relatively low number of individual MP particles processed for molecular analyses should also be considered when interpreting the results, particularly for comparisons among MP types and sampling conditions. Environmental and laboratory-aged MPs differed in size, shape, polymeric composition and exposure history, and absolute microbial biomass was not quantified for 16S rRNA sequencing data, limiting direct comparisons of microbial load between matrices. In addition, the absence of natural particle or non-plastic biofilm controls, such as organic aggregates, wood fragments, shell fragments or inert glass particles, prevents us from determining whether the observed differences are specifically attributable to plastic surfaces or instead reflect broader ecological differences between free-living and surface-associated microbial communities. This interpretation is supported by previous studies showing that particle-attached microbial communities can differ substantially from surrounding free-living communities in marine environments [11, 20], and that bacterial communities and resistomes associated with MPs can resemble those found on natural organic particles [14]. Therefore, the higher relative contribution of potentially pathogenic taxa and ARGs should not be interpreted as uniquely plastic-specific. These patterns may also reflect ecological features common to particle-associated and biofilm dynamic, including increased cell density, organic matter accumulation, close cell–cell contact and enhanced opportunities for gene exchange. Environmental biofilms have also been recognized as potential reservoirs of antibiotic-resistant bacteria and ARGs, independently of the specific substrate involved [4, 16]. In addition, because the metagenomic analysis was reads-based ARG detection, the genomic context of ARGs was not assessed. Therefore, we could not evaluate their physical association with mobile genetic elements or metal- and biocide-resistance genes, which would be necessary to investigate potential ARG mobilization or co-selection mechanisms.
Nonetheless, the combined evidence presented here suggests that MPs may act as mobile, persistent microbial platforms capable of hosting distinct microbial communities rich in potentially pathogenic taxa and ARGs. These findings have important implications for marine ecosystem functioning, food web dynamics and environmental surveillance, particularly in coastal areas subjected to multiple anthropogenic pressures. Overall, the combined use of 16S rRNA gene amplicon sequencing, shotgun metagenomics and ddPCR showed that environmental and laboratory-aged MPs supported microbial biofilms that differed in diversity, composition and resistome profile from the surrounding water microbiota. These findings suggest that MPs can act as persistent microbial substrates in coastal environments and may contribute to the spatial distribution of selected microbial taxa and ARGs. However, in the absence of natural particle controls, MPs should be interpreted within the broader framework of particle-associated microbial ecology rather than as definitively plastic-specific hotspots. Nevertheless, the integration of 16S rRNA gene amplicon sequencing, shotgun metagenomics and ddPCR provides a robust description of microbial community structure and selected ARG profiles associated with MPs in coastal environments. The main take-home message is that MPs are not inert pollutants, but biologically active surfaces that can host distinct microbial and resistome profiles, supporting their inclusion in environmental surveillance of coastal ecosystems.
Supplementary Information
Below is the link to the electronic supplementary material.
Supplementary Material 1. Characterization of environmental microplastics (E-MPs) Percentage of the various MPs according to their shapes (A) and polymer types (B).
Acknowledgements
The authors acknowledge Luigi Stocco (IZSVenezie) for his valuable help in plastic cryomilling and Fabiana Corami (CNR-ISP) for MPs analyses service.
Author contributions
AP: Data curation, Formal analysis, Investigation, Visualization, Writing – original draft; DA: Investigation, Writing – original draft; VM: Sampling, Writing – original draft; GB: Formal analysis; ETM: Formal analysis; GF: Formal analysis; TM: Sampling; MGM: Methodology, Writing – original draft; CL: Conceptualization, Funding acquisition, Methodology, Project administration, Supervision, Writing – original draft.
Funding
This work was funded by Italian Ministry of Health, Grant N. RC_IZSVE_02/20 and Grant N. RF-2021-12372336.
Data availability
The datasets for this study can be found in the Sequence Read Archive (SRA) database with the BioProject: : PRJNA1084257.
Declarations
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Ahmed MB, Rahman MS, Alom J, Hasan MS, Johir MA, Mondal MJ, Yoon MH. Microplastic particles in the aquatic environment: a systematic review. Sci Total Environ. 2021;775:145793. [DOI] [PubMed] [Google Scholar]
- 2.An Z, Chen F, Hou L, Chen Q, Liu M, Zheng Y. Microplastics promote methane emission in estuarine and coastal wetlands. Water Res. 2024;259:121853. [DOI] [PubMed] [Google Scholar]
- 3.Arias-Andres M, Kethtner MT, Miki T, Grossart HP. Microplastics: new substrates for heterotrophic activity contribute to altering organic matter cycles in aquatic ecosystems. Sci Total Environ. 2018;635:1152–9. [DOI] [PubMed] [Google Scholar]
- 4.Balcázar JL. The role of biofilms as environmental reservoirs of antibiotic resistance. Front Microbiol. 2015;6:1216. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Bartlett A, Padfield D, Lear L, Bendall R, Vos M. A comprehensive list of bacterial pathogens infecting humans. Microbiology. 2022;168:001269. [DOI] [PubMed] [Google Scholar]
- 6.Callahan BJ, McMurdie PJ, Rosen MJ, Han AW, Johnson AA, Holmes SP. DADA2: High-resolution sample inference from Illumina amplicon data. Nat Methods. 2016;13:581–3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Caruso G. Microbial colonization in marine environments: overview of current knowledge and emerging research topics. J Mar Sci Eng. 2020;8(2):78. [Google Scholar]
- 8.Chen L, Reeve J, Zhang L, Huang S, Wang X, Chen J. GMPR: a robust normalization method for zero-inflated count data with application to microbiome sequencing data. PeerJ. 2018;6:e4600. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Corami F, Rosso B, Morabito E, Rensi V, Gambaro A, Barbante C. Small microplastics (< 100 µm), plasticizers and additives in seawater and sediments: oleo-extraction, purification, quantification, and polymer characterization using Micro-FTIR. Sci Total Environ. 2021;797:148937. [DOI] [PubMed] [Google Scholar]
- 10.Cozzarini L, Buoninsegni J, Corbau C, Lughi V. Characterization of large microplastic debris in beach sediments in the Po Delta area. Microplastics. 2023;2(1):147–57. [Google Scholar]
- 11.Dang H, Lovell CR. Microbial surface colonization and biofilm development in marine environments. Microbiol Mol Biol Rev. 2016;80(1):91–138. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.De Lucia GA, Vianello A, Camedda A, Vani D, Tomassetti P, Coppa S, Matiddi M. Sea water contamination in the vicinity of the Italian minor islands caused by microplastic pollution. Water. 2018;10(8):1108. [Google Scholar]
- 13.Debroy A, George N, Mukherjee G. Role of biofilms in the degradation of microplastics in aquatic environments. ournal Chem Technol Biotechnol. 2022;97(12):3271–82. [Google Scholar]
- 14.Di Cesare A, Sathicq M, Sbaffi T, Sabatino R, Manca D, Breider F, Corno G. Parity in bacterial communities and resistomes: Microplastic and natural organic particles in the Tyrrhenian Sea. Mar Pollut Bull. 2024;203:116495. [DOI] [PubMed] [Google Scholar]
- 15.Dussud C, Heder C, George M, Fabre P, Higgs P, Bruzaud S, Ghiglione JF. Colonization of Non-biodegradable and Biodegradable Plastics by Marine Microorganisms. Front Microbiol. 2018;9. 10.3389/fmicb.2018.01571. [DOI] [PMC free article] [PubMed]
- 16.Flores-Vargas G, Bergsveinson J, Lawrence JR, Korber DR. Environmental biofilms as reservoirs for antimicrobial resistance. Front Microbiol. 2021;12:766242. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Fuhrman JA, Cram JA, Needham DM. Marine microbial community dynamics and their ecological interpretation. Nat Rev Microbiol. 2015;13:133–46. [DOI] [PubMed] [Google Scholar]
- 18.Ghezzo M, Petrizzo A, Madricardo F, Folegot T, Gallou R, Clorennec D, Picciulin M. Natural and shipping underwater sound distribution in the Northern Adriatic Sea basin and possible application on target areas. Mar Pollut Bull. 2024;207:116852. [DOI] [PubMed] [Google Scholar]
- 19.Ghiglione JF, Barbe V, Bruzaud S, Burgaud G, Cachot J, Eyheraguibel B. Mission Tara Microplastics: a holistic set of protocols and data resources for the field investigation of plastic pollution along the land-sea continuum in Europe. Environ Sci Pollut Res. 2023;32:10032–49. the Mission Tara Microplastics consortium. [DOI] [PMC free article] [PubMed]
- 20.Ghiglione JF, Conan P, Pujo-Pay M. Diversity of total and active free-living vs. particle-attached bacteria in the euphotic zone of the NW Mediterranean Sea. FEMS Microbiol Lett. 2009;299(1):9–21. [DOI] [PubMed] [Google Scholar]
- 21.Gouin T. Toward an improved understanding of the ingestion and trophic transfer of microplastic particles: critical review and implications for future research. Environ Toxicol Chem. 2020;39(6):1119–37. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Gregori I, Martino F, Zane L, Pallavicini A, Vezzi A. (2025). Filtering microbial diversity: evaluation of different pore sizes for eDNA community profiling. Environ DNA, 7(6), e70221.
- 23.Gündoğdu S. Polymer types of microplastic in coastal areas. Microplastic pollution: environmental occurrence and treatment technologies. Springer International Publishing; 2022. pp. 77–88.
- 24.Joyce H, Frias J, Kavanagh F, Lynch R, Pagter E, White J, Nash R. Plastics, prawns, and patterns: microplastic loadings in Nephrops norvegicus and surrounding habitat in the North East Atlantic. Sci Total Environ. 2022;826:154036. [DOI] [PubMed] [Google Scholar]
- 25.Khalid N, Aqeel M, Noman A, Hashem M, Mostafa YS, Alhaithloul HS, Alghanem SM. Linking effects of microplastics to ecological impacts in marine environments. Chemosphere. 2021;264:128541. [DOI] [PubMed] [Google Scholar]
- 26.Kirstein IV, Wichels A, Gullans E, Krohne G, Gerdts G. (2019). The plastisphere–uncovering tightly attached plastic specific microorganisms. PLoS ONE, 14(4), e0215859. [DOI] [PMC free article] [PubMed]
- 27.Li WC, Tse HF, Fok L. (2016). Plastic waste in the marine environment: a review of sources, occurrence and effects. Sci Total Environ, 566–7, 333–349. [DOI] [PubMed]
- 28.Lievens S, Vervoort E, Covaci P, A., Van Der Borght M. (2023). A production and fractionation protocol for polyvinyl chloride microplastics. Methods Protocols, 15. [DOI] [PMC free article] [PubMed]
- 29.Luo H, Liu C, He D, Xu J, Sun J, Li J, Pan X. Environmental behaviors of microplastics in aquatic systems: a systematic review on degradation, adsorption, toxicity and biofilm under aging conditions. J Hazard Mater. 2022;423:126915. [DOI] [PubMed] [Google Scholar]
- 30.Marfella R, Prattichizzo F, Sardu C, Fulgenzi G, Graciotti L, Spadoni T, Paolisso G. Microplastics and nanoplastics in atheromas and cardiovascular events. N Engl J Med. 2024;390(10):900–10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.McMurdie PJ, Holmes S. (2013). phyloseq: An R package for reproducible interactive analysis and graphics of microbiome census data. PLoS ONE, e61217. [DOI] [PMC free article] [PubMed]
- 32.Miao L, Wang P, Hou J, Yao Y, Liu Z, Liu S, Li T. Distinct community structure and microbial functions of biofilms colonizing microplastics. Sci Total Environ. 2019;650(2):2395–402. [DOI] [PubMed] [Google Scholar]
- 33.Pedròs-Aliò. The rare bacterial biosphere. Annual Rev Mar Sci. 2012;4:449–66. [DOI] [PubMed] [Google Scholar]
- 34.Quast C, Pruesse E, Yilmaz P, Gerken J, Schweer T, Yarza P, Glöckner FO. (2013). The SILVA ribosomal RNA gene database project: improved data processing and web-based tools. Nucleic Acids Res, D590–6. [DOI] [PMC free article] [PubMed]
- 35.Rosso B, Corami F, Barbante C, Gambaro A. Quantification and identification of airborne small microplastics (< 100 µm) and other microlitter components in atmospheric aerosol via a novel elutriation and oleo-extraction method. Environ Pollut. 2023;318:120889. [DOI] [PubMed] [Google Scholar]
- 36.Stevenson EM, Buckling A, Cole M, Hayes A, Lindeque PK, Murray AK. Sewers to Seas: exploring pathogens and antimicrobial resistance on microplastics from hospital wastewater to marine environments. Environment International; 2025. p. 109944. [DOI] [PubMed]
- 37.Thompson R, Courtene-Jones W, Boucher J, Pahl S, Raubenheimer K, Koelmans A. Twenty years of microplastic pollution research—what have we learned? Science. 2024;386:6720. [DOI] [PubMed] [Google Scholar]
- 38.Vayssier-Taussat M, Albina E, Citti C, Cosson JF, Jacques MA, Lebrun MH, Loir L, Ogliastro Y, Petit M, Roumagnac MA, P., Candresse T. Shifting the paradigm from pathogens to pathobiome: new concepts in the light of meta-omics. Front Cell Infect Microbiol. 2014;4:29. 10.3389/fcimb.2014.00029. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Vianello A, Da Ros L, Boldrin A, Marceta T, Moschino V. First evaluation of floating microplastics in the Northwestern Adriatic Sea. Environ Sci Pollut Res. 2018;25(28):28546–61. [DOI] [PubMed] [Google Scholar]
- 40.Welden N, Lusher A. Welden, N. A., & LuImpacts of changing ocean circulation on the distribution of marine microplastic litter. Integr Environ Assess Manag. 2017;13(3):483–7. [DOI] [PubMed] [Google Scholar]
- 41.Witsø IL, Baral AL, Aspholm K, M., Myrmel MW. Plastispheres as reservoirs of antimicrobial resistance: Insights from metagenomic analyses across aquatic environments. PLoS ONE. 2025;20(9):e0330754. [DOI] [PMC free article] [PubMed]
- 42.Xu C, Lu J, Shen C, Wang J, Li F. Deciphering the mechanisms shaping the plastisphere antibiotic resistome on riverine microplastics. Water Res. 2022;225:p119192. [DOI] [PubMed] [Google Scholar]
- 43.Yan W, Bai R, Zhang Q, Jiang Y, Chen G, Zhang Y, Wu Y, Guo X, Xiao Y, Zhao F. Metagenomic insights into ecological risk of antibiotic resistome and mobilome in riverine plastisphere under impact of urbanization. Volume 190. Environment International; 2024. p. 108946. [DOI] [PubMed]
- 44.Yu S, Lu H. Marine microbial biofilms on diverse abiotic surfaces. Front Mar Sci. 2025;12:1482946. [Google Scholar]
- 45.Zadjelovic V, Wright RJ, Borsetto C, Quartey J, Cairns TN, Langille MG, Wellington EM, Christie-Oleza JA. Microbial hitchhikers harbouring antimicrobial-resistance genes in the riverine plastisphere. Microbiome. 2023;11(1):225. [DOI] [PMC free article] [PubMed]
- 46.Zettler ER, Mincer TJ, Amaral-Zettler LA. (2013). Life in the plastisphere: microbial communities on plastic marine debris. Environ Sci Technol, 7137–46. [DOI] [PubMed]
- 47.Zhurina MV, Bogdanov KI, Gannesen AV, Mart’yanov SV, Plakunov VK. (2022). Microplastics as a new ecological niche for multispecies microbial biofilms within the plastisphere. Microbiology, 107–23.
- 48. Wu P, Huang J, Zheng Y, Yang Y, Zhang Y, He F, Chen H, Quan G, Yan J, Li T, Gao B. Environmental occurrences, fate, and impacts of microplastics. Ecotoxicol Environ Saf. 2019;184:109612. 10.1016/j.ecoenv.2019.109612 [DOI] [PubMed]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplementary Material 1. Characterization of environmental microplastics (E-MPs) Percentage of the various MPs according to their shapes (A) and polymer types (B).
Data Availability Statement
The datasets for this study can be found in the Sequence Read Archive (SRA) database with the BioProject: : PRJNA1084257.
