Microplastic biofilm in fresh- and wastewater as a function of microparticle type and size class

Kathleen Parrish and N. L. Fahrenfeld *
Department of Civil and Environmental Engineering, Rutgers, The State University of New Jersey, 500 Bartholomew Rd., Piscataway, New Jersey 08854, USA. E-mail: nfahrenf@rutgers.edu; Tel: +848 445 8416

Received 12th October 2018 , Accepted 13th January 2019

First published on 14th January 2019


Microplastics are pollutants of concern in the freshwater and marine environments. These microparticles carry biofilm communities unique from the surrounding water. The objective of this study was to investigate the role of source water quality, microparticle type (i.e., different polymers and morphologies), and microparticle size on the resulting biofilm microbial communities. Of particular interest was determining whether microbial agents were concentrated in these biofilm communities. A series of batch reactors were prepared to investigate the microbial communities observed in wastewater or river water compared to those in biofilm on microplastic or glass microspheres via amplicon sequencing. Sampling was performed after 48 h. qPCR was also performed for a fecal indicator organism marker (BacHum) and sul1 antibiotic resistance gene. The biofilm community structures varied as a function of source water and as a function of microparticle type: polystyrene spheres had different microbial community structures from polyethylene microparticles. The differences observed between microparticle types may be due to the morphology/surface texture rather than polymer composition given that some glass microspheres had similar microbial communities to polystyrene microspheres. For a given microparticle type and source water, size (106–125 μm versus 355–425 μm for polystyrene microspheres, 125–250 μm versus 250–500 μm for polyethylene microparticles) did not significantly impact the microbial community structure. While the biofilm microbial communities differed from communities in the surrounding water, the biofilm did not have a higher relative abundance of the indicator organism marker or sul1 genes than the reactor filtrate.



Water impact

Microplastics are emerging contaminants and the potential hazard posed by their surface biofilm communities is poorly characterized, limiting our ability to develop mitigation strategies. We observed biofilm communities varied by source water (indicating potential utility for source tracking) and microparticle type, but not size. Select marker genes of public health interest were similar in the reactor filtrate and biofilm.

Introduction

Microplastics (MPs, defined as plastic particles <5 mm diameter)1 are emerging contaminants of concern in the freshwater environment2 and are better studied contaminants in the marine environment.3 MPs can each carry biofilm communities (referred to as the “plastisphere”) that are unique from the surrounding water.4,5 Given the detection of potentially pathogenic Vibrio spp. on polyethylene (PE), polypropylene (PP), and polystyrene (PS) particles from the North/Baltic Sea,6 there is a concern that plastics may offer unique transport mechanisms for harmful microbes. Of particular concern would be MPs incubated in wastewater that are released during sewer overflows7 that do not receive treatment including disinfection. For the small portion of MPs that escape wastewater treatment plants,8 it is worth noting that biofilms generally may be more resistant to disinfection than planktonic cells.9 Another potentially important feature of biofilms on MPs is that biofouling can increase the density and alter the fate of particles.10 Although these polymers are known to have long half-lives in the environment,10 MP biofilms would be of great interest if they harbor biodegrading species.11,12 Therefore, it is desirable to better understand microbial communities growing on MP particles and the factors that control biofilm growth.13

A challenge in investigating field samples of MP biofilm is that the methods to identify plastics may not be simultaneously compatible with methods used to study biofilms. Current methods to analyze MPs from water involve oxidation to remove other organics and surface contaminants followed by incubation in density separation solutions,14 which would be expected to remove or alter the biofilm communities. Confirmation of polymers by Fourier-transform infrared spectroscopy (FTIR) or micro-FTIR is generally considered accurate,15 but it is recommended that biofilms be removed prior to identification due to their potential to interfere with the surface scans.6 One could extract biofilms from microparticles before polymer confirmation, but given reports of particle misclassification by visual inspection,16,17 this approach would likely involve processing potential false positives and, thus, may not be practical. Despite these challenges, field studies have investigated microbial communities associated with MPs on particles >300 μm in size. Potentially pathogenic species have been detected in MP biofilms.5,18 Few field studies of microbial communities on MPs in freshwater environments have been undertaken.4,19 The source water quality is expected to impact both the seed community for the biofilm and the availability of nutrients and selecting factors to support and alter its development. Given the relative lack of data on the freshwater environment and the analytical challenges, there is motivation for a systematic laboratory investigation of biofilm formation on microparticle surfaces where the water source and the polymer type and size are controlled.

Studies of MPs in the laboratory should use particles representative of those observed in the environment. Several types of MPs have been identified in environmental samples with PE, PP, and PS often reported as the most abundant polymers among several others observed in marine20 and freshwater environments.21 Few studies have compared biofilm growth on different MP polymer types, and those that did focused on larger particles >300 μm.4,5,22 The potential for differences in biofilm observed on particles of different sizes is unknown but worth investigating given the large range of MP particle sizes observed in the environment.20 A range of morphologies and surface textures have also been observed on primary (i.e., those manufactured in small sizes) and secondary (i.e., particles derived from larger plastic waste) MPs in the environment.17 It is expected that biofilm growth may be affected by surface texture of the substrates because irregularly shaped surfaces may provide more attachment points for biofilms.23

The objective of this research was to improve understanding of the biofilm communities that grow on MPs in wastewater and freshwater environments as a function of microparticle type (i.e., different polymer and morphology) and particle size. Of particular interest was understanding the potential for biofilm communities on different microparticles to serve as reservoirs of microbial agents (antibiotic resistance genes, fecal indicators, pathogen marker genes) compared to the surrounding river or wastewater. To achieve this goal, PE MPs of irregular shape, PS microspheres, and glass microspheres were incubated in either river water or wastewater influent to allow for biofilm growth. The biofilm and water microbial communities were quantified using qPCR, targeting 16S rRNA gene copies and select genes encoding for microbial agents. Amplicon sequencing was performed to compare the microbial community of biofilms for each microparticle type to the surrounding river or wastewater. The results presented provide insight into the microbial ecology of biofilm communities on MP and potentially give guidance on microparticle type(s) and size class(es) to target during wastewater treatment processes.

Materials and methods

Batch reactors for biofilm growth

Continuously stirred 100 mL batch reactors were prepared in duplicate to understand the roles of water quality and microparticle type and size class on the biofilm microbial community (Fig. S1). Small “S” sized particles were <250 μm in diameter, and large “L” sized particles were >250 μm but <500 μm in diameter. These sizes were chosen because the majority of MPs found in personal care products (PCPs) and in the environment were <500 μm, particularly <250 μm, which is smaller than the minimum size (333 μm) captured in many MP studies.24,25 Polystyrene beads in two size classes (diameter of 106–125 μm “S” or 355–425 μm “L”) were purchased dry from a commercial vendor (PolySciences, Inc., Warrington, PA) and were directly added to the reactors. Polyethylene MPs (composition confirmed by attenuated total reflectance, ATR, Fourier-transform infrared spectroscopy, FTIR) were extracted from a facial scrub PCP by washing the product with tap water through a set of standard sieves to separate the particles into two size classes (125–250 μm “S” or 250–500 μm “L”) (a full ingredient list is included in the supplemental material and a spectrum for the particles as Fig. S2). Glass microspheres (diameter of 355–420 μm) were purchased from a commercial vendor (PolySciences, Inc., Warrington, PA) to serve as a material control. Aliquots of each microparticle type and size class were measured with a scoop (∼210 μL) and were incubated in 100 mL of pre-filtered (63 μm stainless steel wire mesh, TWP, Berkeley, CA) 24 h composited wastewater influent or freshwater grab samples (at the surface ∼0.5 m from the shore of the Raritan River) collected during baseflow conditions (Fig. S3). The number of particles per sample are included in Table S1. Wastewater influent reflects conditions for MPs released by systems without treatment during a sewer overflow. Duplicate reactors were incubated for two days at 25 °C with stir bars (∼90 rpm) to simulate the shear stress experienced in open channel flow conditions by creating turbulent flow. Sampling was performed after 48 h to reflect an incubation closer to the expected three to 12 h residence time of water in sewers.

MP and glass microsphere samples were recovered from the reactors on sterile stainless-steel wire mesh filters (63 μm) and rinsed gently with sterilized deionized (DI) water to remove planktonic cells without disrupting the biofilms.5 The microparticle samples were transferred directly to lysing tubes for biofilm DNA extraction using 200 μL of sterile DI water. The filtrate was reserved for filter concentration (0.22 μm nitrocellulose membrane). DNA was extracted from the microparticles and concentrated filtrate samples using a commercial kit (FastDNA Spin Kit for Soil, MP Biomedicals, Hercules, CA) following the manufacturer's directions. DNA extracts were preserved at −20 °C prior to biomolecular analyses.

Chemical analyses of water and morphological analyses of particles

Basic water quality data was collected for both the wastewater influent and river water. pH and conductivity were measured using a multimeter (Orion Star A329, Thermo Scientific). Chemical oxygen demand (COD) was measured according to Hach Method 8000 with Hach COD vials (20–1500 mg L−1 range) and a DR2700 spectrophotometer (Hach, Loveland, CO). Microscopic images of the particle surface prior to biofilm formation were obtained using a reflected light microscope (Stereo Zoom Microscope, Olympus, Japan) and images were captured using a 21 MP digital camera phone.

Biomolecular analyses

qPCR was performed for 16S rRNA gene copies26 as a surrogate for total bacterial population. These data were normalized to volume of particles or filtrate. qPCR was also performed for a sulfonamide antibiotic resistance gene (sul127) and fecal indicator maker gene BacHum.28 BacHum and sul1 gene copy numbers were normalized to volume and to 16S rRNA gene copies. Details about the primers and annealing temperatures are included in Table S2. Standard SybrGreen (5 μL SsoFast EvaGreen, BioRad, Hercules, CA) chemistry with 0.4 μM forward and reverse primers, and 1 μL diluted DNA extract (1[thin space (1/6-em)]:[thin space (1/6-em)]50 dilution to reduce inhibition) was used for 16S rRNA gene copies and sul1 in a 10 μL total reaction volume. For BacHum, 5 μL probes mix (SsoAdvanced Universal Probes Supermix, Bio-Rad, Hercules, CA, United States) with 0.22 μM of each primer, 0.07 μM of probe, and 1 μL diluted DNA extract was prepared in a 10 μL reaction. Thermocycler (BioRad CFX96 Touch, Hercules, CA) conditions are summarized in Table S2. qPCR samples and standards were analyzed in triplicate (technical replicates). A seven-point calibration curve and negative control were included in each run. A melt curve and/or gel electrophoresis were used to verify the specificity of qPCR products.

To evaluate the microbial community structure in the biofilms compared to filtrate, DNA extracts were sequenced at a commercial facility (Mr DNA, Shallowater, TX, USA). Amplicon sequencing was performed targeting the V3–V4 hypervariable region of the 16S rRNA gene using Illumina MiSeq (300 bp, paired end). Sequences were analyzed using Quantitative Insights in Microbial Ecology (QIIME) version 1.9 run through Oracle Virtual Box VM. Sequences were trimmed using Trimmoatic29 and stitched using PandaSeq.30 Sequences were otherwise analyzed following the tutorial for next generation sequencing.31 Briefly, after extracting barcodes (barcode_extract.py), samples were demultiplexed and quality filtered (split_libraries_fastq.py), followed by assigned operational taxonomic units (OTUs, pick_de_novo_otus.py). Samples were rarefied at 19[thin space (1/6-em)]117, the minimum number of sequences observed per sample (rarefaction curves are included as Fig. S4). Sequences are available in the NCBI database under Accession Number SRP150842.

Statistical analyses

OTUs determined using QIIME were analyzed using Primer 7 (Primer E, UK). Following log-transformation, a dissimilatory matrix was generated using Bray Curtis similarity. Hierarchical cluster analysis with a SIMPROF test (significance level 5%) was performed to provide a graphical representation of the similarity of the biofilm communities grown on different microparticle types and in different water sources. ANOSIM was performed to determine if different source waters, microparticle types, and/or sizes resulted in significantly different microbial communities. To compare qPCR data and diversity data among samples, non-normality of data was confirmed with a Shapiro test and then data were tested with a Kruskal–Wallis test followed by a post-hoc pairwise t-test with a Bonferroni correction for multiple comparisons. Significant differences were reported when p < α = 0.05. Chi squared, degrees of freedom and p-values are reported in Table S3. The Linear discriminant analysis effect size (LEfSe) tool32 was used to identify biomarkers using the relaxed settings (comparisons performed between subclasses, and one-against-all comparisons). Briefly, this tool applies an algorithm that determines significantly different taxa between conditions by a Kruskal–Wallis test, determines within condition differences via a Wilcoxon rank-sum test, then applies linear discriminant effect size analysis to reduce the dimensions.32

Results & discussion

Batch reactors were used to provide insight into the impact of water source and microparticle characteristics that drive microbial communities on MP pollution in freshwater systems. qPCR was used to detect genes of interest (16S rRNA, sul1, and BacHum gene copies) and amplicon sequencing was performed to provide insight into the microbial community structure of the samples. Both microparticle type and water quality impacted the growth of biofilm on microparticles, as discussed below.

Microparticle characteristics and water quality

Microscopic evaluation of the shape and surface texture of the microparticles used in this study was performed given that these characteristics can influence biofilm formation. The microparticles studied had different morphologies and surface textures. The glass particles had a smooth surface and regular spherical shape, similar in shape and surface texture to the polystyrene particles (Fig. 1b, c and e). The polyethylene particles extracted from PCP had microparticles with irregular shapes and rough textures (Fig. 1a and d). The three microparticle types investigated had different densities: all glass microparticles sank (ρ = 2.48 g cm−3), approximately 56% of the PS particles were suspended in the bottom half of the reactor or rested on the bottom, (density of PS can range from 0.96 to 1.05 g cm−3, manufacturer estimated ρ = 1.05 g cm−3), and all of the PE particles stayed in suspension in the reactors (density of PE can range from 0.857 to 0.975 g cm−3). Given that the reactors were stirred with a magnetic stir bar that rested on the reactor bottom, the differences in densities would be expected to impact the shear stress experienced by the microparticles.
image file: c8ew00712h-f1.tif
Fig. 1 Reflected light microscope images of (a) polyethylene microplastic extracted from personal care product with diameter 250–500 μm, (b) 355–425 μm diameter polystyrene microspheres, (c) 355–420 μm diameter glass microspheres, (d) 125–250 μm polyethylene microparticles, and (e) 106–125 μm diameter polystyrene spheres.

As expected, select chemical evaluations of the two water sources indicated different conditions for biofilm growth. Chemical oxygen demand (COD) was used to measure the amount of organic material in the water samples, a portion of which is biologically available organic carbon. The river water had a COD of 46 mg mL−1 and the wastewater influent had an average COD of 332 ± 24 mg mL−1. pH and conductivity gave insight into the ion composition of the solutions and oxidation reduction potential (ORP) gave insight into the redox conditions. The river water had a pH of 8.08 and a conductivity of 423.4 μS cm−1. The composite wastewater sample has a similar pH of 7.79 and over three times the conductivity at 1766 μS cm−1. The wastewater had an ORP of 210.9 mv at 21.7 °C.

qPCR for total microbial community

The bacterial density of the microbial communities was evaluated for the biofilm and reactor filtrate by qPCR of the 16S rRNA gene. Comparing biofilm grown in the two water sources, 16S rRNA gene copies on glass and large PS particles incubated in river water were lower than wastewater (based on gene copies/210 μL of particles, both p ≤ 0.002, Fig. 2a). As expected, the planktonic communities in wastewater also had elevated 16S rRNA gene copy numbers compared to river water. The differences in the gene copy numbers between particles incubated in wastewater versus river water is likely explained by the combination of greater planktonic community density and the higher COD.
image file: c8ew00712h-f2.tif
Fig. 2 Concentration of 16S rRNA gene copies observed on glass, polyethylene (PE), polystyrene (PS) microparticle biofilm or filtrate incubated in wastewater influent (WW) or raritan river water (RR). Results are normalized either to (a) volume of particles and filtrate or (b) number of particles. Two size classes of microplastics were analyzed: small and large. PE particles had irregular morphology, and PE and glass particles were smooth and spherical. Error bars represent high and low observations for samples from replicate reactors (n = 2).

Next, comparisons were made between particle type and size class. No differences in the 16S rRNA gene copies (per particle volume) were observed for biofilm extracted from different particle types and sizes classes incubated in wastewater (all p = 1.0). In contrast, for microparticles incubated in river water, the biofilm density grown on glass was significantly lower than small PE, large PE, and small PS MPs (all p < 0.03) but not large PS (p = 1.0). Comparing the biofilm growth on the two MP particle types, differences in 16S rRNA gene copies were not observed between the small particles. The large particles of PE had more 16S rRNA gene copies than the large PS particles (p = 0.003). Conclusions were similar when normalizing 16S rRNA gene copy numbers to the number of particles (Fig. 2b).

Differences in particle morphology, surface texture, and density may explain these observed differences in the 16S rRNA gene copy numbers. The glass microbeads most closely resembled the spherical shape and smooth texture of large PS microspheres under visual inspection, and in wastewater, glass samples were similar to large PS in 16S rRNA gene copy numbers (and community structure, as discussed below). These particles had smoother surfaces and would be expected to have less available surface area than textured particles of the same size that would have different availability and morphology of attachment sites. This interpretation is supported by findings from marine samples that attributed similarities between biofilm growth on macroplastic samples of poly(ethylene terephthalate) (abbreviated as PET or PETE) and glass to their hard smooth surface texture based on sequencing of V3 – V4 region of the 16S rRNA gene.33,34 Given that the majority of MPs observed in the environment are secondary, the impact of surface texture is notable. Furthermore, the density of the microparticles investigated varied and thus the biofilms may have experienced different shear forces or nutrient availability. [While it was not the focus of this study, MP buoyancy may also be impacted by biofilm accumulation.35 No sinking of particles was observed due to biofilm growth in this study (nor studies with 1 week incubations, data not shown)]. Finally, the microparticles studied were measured volumetrically and the packing impacted the number of particles of each type in a sample. Packing differences did not help explain differences in biofilm density, despite the fact that this would also affect the available surface area for biofilm growth.

qPCR for sulfonamide resistance and fecal markers

Of particular interest is the potential for the biofilms on MP surfaces to serve as reservoirs of fecal indicator organisms, pathogens, and antibiotic resistance genes, and thus, potentially pose a hazard. Quantification of sul1, an antibiotic resistance gene, and BacHum, a fecal indicator marker, via qPCR resulted in no differences between microparticle type, size classes, or water source (all p > 0.07, Fig. 3). BacHum gene copy numbers were near or below detection limits providing little evidence that these genes of concern are accumulating in MP biofilms. BacHum was above detection limits in the wastewater filtrate but only one replicate for the Raritan River filtrate. sul1 was above detection in filtrate from both water sources but below detection in some biofilm replicates for the river water. The 16S rRNA normalized concentrations for sul1 in the wastewater filtrate were within ranges previously reported for wastewater influent,36 indicating the source water contained sources for this gene. One other similar study incubating PS MPs in a chemostat demonstrated accumulation of the antibiotic resistance gene associated integrase 1 was only observed with increasing plastic concentration.37
image file: c8ew00712h-f3.tif
Fig. 3 (a) Sulfonamide resistance (sul1) and (b) human fecal indicator marker (BacHum) gene copies observed in microparticle biofilm (solid bars) and filtrate (textured bars) per unit volume of particles, wastewater (WW, gray bars) or raritan river (RR, white bars) on the primary y-axis. 16S rRNA gene copy normalized sul1 and BacHum gene copies are shown on the secondary y-axis and represented by circles. Error bars represent high and low values of samples from replicate reactors (n = 2).

Microbial community analysis by amplicon sequencing

Incubation of microparticles in wastewater or river water significantly impacted the biofilm microbial community structures observed on the surfaces with at most 64% similarity between the two water sources (Fig. 4). Analysis by ANOSIM indicated that source water (p = 0.001, R = 0.92) and microparticle type (p = 0.002, R = 0.86), but not size (both p = 0.07, R = 0.38), resulted in significantly different microbial communities. In wastewater, PS microsphere and PE microparticle biofilms shared 81% similarity, forming clusters by material without significant differences (>89% similarity). The microbial ecology of MP biofilms is of interest given that they may be conserved and travel long distances, potentially introducing new microbial community members into other ecosystems.38
image file: c8ew00712h-f4.tif
Fig. 4 (a) Cluster analysis and (b) relative abundance of highly abundant microbial classes based on 16S rRNA gene V3–V4 region. Dotted red lines indicate that there were no significant differences between the samples, while solid black lines indicate significant differences. Particle materials included polyethylene (PE), polystyrene (PS), or glass (G), unless the sample was from filtrate. Particle size was denoted by large (L) or small (S). Samples were incubated in wastewater influent (WW) or raritan river water (RR). Experimental replicates are indicated as “a” and “b.”

Microbial community in relation to source water and microparticle type

Analysis of class-level biomarkers with LEfSe was used to quantitatively investigate the community structures of biofilm and filtrate samples from the two waters. LEfSe was performed and identified 46 discriminating features among the samples (Fig. S5). Wastewater PS biofilms were marked by Deltaproteobacteria and Acidomicrobia (two classes common to wastewater39), as well as Saprospirae of Bacteroidetes. Gammaproteobacteria and Alphaproteobacteria have been described as early colonizers of marine biofilms, with Bacteroidetes as secondary colonizers.40 Observations of Bacteroidetes in this study, which are described as secondary colonizers, may not be surprising because biofilm growth can start developing within hours.18 Thus, the biofilms observed here may have been mature and developed following typical biofilm formation patterns observed on other substrates. Wastewater PS biofilms had 7.4 times more Gammaproteobacteria (a class containing many commensal organisms and some species identified as waterborne pathogens) and 33.7 times more Deltaproteobacteria than PS biofilms in river water. Polyethylene and PS MP biofilms grown in river water showed less similarity (64%, p = 0.001) between microparticle composition than wastewater biofilms for these materials (Fig. 4a). Other researchers reported 40% similarity in biofilm microbial community structure grown on larger (>333 μm) PE and PS MP samples from the marine environment analyzed using amplicon sequencing of the V3–V4 region.5 LEfSe analysis indicated that the PE microparticle biofilm in river water was marked by elevated Gammaproteobacteria, while the PS biofilm was marked by Betaproteobacteria. The differences in the biofilm community structure between particles incubated in wastewater versus river water was likely due to the attachment of the different planktonic bacteria in these water sources to surfaces. In another batch study of MP biofilm formation, Eckert et al.37 incubated MP squares for 15 days in a chemostat inoculated with a mix of wastewater effluent and lake water; the resulting MP biofilm community more closely resembled the wastewater source with increased plastic concentration.

Glass microspheres were used as a material control to understand whether plastic surfaces grew unique biofilm communities compared with other materials. Discriminating features were observed for glass microspheres between the two water sources. The wastewater glass biofilm microbial communities were marked by Firmicutes such as Clostridia and Bacilli, Bacteroidia, Coriobacteria, and Epsilonproteobacteria (a class previously identified as a biomarker of wastewater influent41). In wastewater, biofilm microbial community structures observed on glass microspheres formed unique clusters or clustered with those observed for PS microspheres. Alphaproteobacteria, Cyanobacteria, Verrucomicrobia, Spirochaetes were associated with river water glass biofilms. The glass biofilm communities clustered either with wastewater and river water filtrate or with the PS microspheres. ANOSIM indicated that glass biofilm was not significantly different from PE or PS biofilms across both water sources (both p > 0.11).

None of the microparticles investigated here were known to have surface coatings. It could be expected that surface coatings42 or sorbed compounds43 would also impact the biofilm communities and could be of interest for future study, particularly if they differ by microparticle type.

The effect of microparticle type on the formation of these unique biofilm microbial communities is of further interest given that microplastic biofilms may select for contaminant or plastic degrading species that could help mitigate the impacts of these pollutants.44 There have been reports of species that could degrade plastics, such as Pseudomonas spp.45 and Bacillus spp.12 in soil and Ideonella sakaiensis for poly(ethylene terephthalate).46 The action of plastic degrading microbial species is also of interest given that they could further impact physical properties of plastic substrates, such as buoyancy and the surface texture.47 In this study a significant amount of Pseudomonas sp. were observed, similar to McCormick et al.4 However, direct evidence would be needed (i.e., observations of pitting, release of carbon) to prove if degradation was occurring. Plastic polymers can take decades to centuries to degrade via oxidation by UV radiation in soil environments, and even longer in aquatic environments,10 therefore evidence of improved biodegradation would be of interest.

Particle size

MP biofilms in this study were not significantly impacted by the particle size for a given material type (>89% similarity for the two size classes tested, all p > 0.08). Particle size is an important consideration given that the majority of MP found in PCPs and released by wastewater treatment plants are smaller sized (<300 μm).48 Standard MP sampling techniques in surface water often apply nets with aperture size of 330 μm, thus not capturing MP biofilms on smaller size class particles.20,49 The fate of smaller sized microplastics after release to receiving waters may also be different.50 Small fish and invertebrates may consume smaller size classes of MPs (<500 μm, especially <100 μm),51 and MPs smaller than studied here (<10 μm) may be available for uptake by plankton.52 In this study, comparing 106–125 μm PS to 355–425 μm PS microspheres or 125–250 μm PE to 250–500 μm PE microparticles, no differences were observed in the microbial communities or the number of gene copies present between different size classes of the same material normalized to the volume of particles in the sample.

The results presented here indicate that if the two size classes in the range studied here of a given polymer and morphology are incubated in the same water source, then the resulting biofilm communities are similar. This could also imply that if different biofilm communities are observed on the same polymer in the environment, then the particles may be from different source waters or have undergone different weathering processes in the environment. Field observations from other studies of MPs measured in river and marine ecosystems had community structures in common with wastewater, notably containing Arcobacter spp., which were also observed in some of our PE and glass samples. Several Arcobacter spp. are enteropathogens, and are associated with fecal contamination in water, suggesting that wastewater was a potential source of these particles.4,38 This may indicate that analysis of biofilms may be useful for source tracking. The similarity between the two size classes for a given material is of interest because it is difficult to simultaneously identify the polymer type of small size class samples of MPs and study their biofilms. If different sized MP likely share a source, and if weathering of the biofilm is similar (which is not known), then one could choose to study the larger samples as an indication of the biofilm found on smaller size class particles in size ranges studied here. However, studies of plastic ingestion have used nanoplastics, plastics <100 μm, in sizes as small as 2–10 μm in diameter.51 While biofilms on particles this small have not been studied, it is suspected that surface area limitations at this size would significantly alter the microbes that could form the biofilm given that most bacteria are themselves ∼2 μm in size. Whether our observations hold for nanoplastics or a broader range of morphologies (i.e., fragments versus fibers) would require further study.

Comparison of biofilm and filtrate communities

The final filtrate microbial communities between the river and wastewater were distinct, as expected (Fig. 4). The diversity (Shannon diversity index) of wastewater filtrate was 1.3 times greater than the river water (p = 0.048, Table S4). Wastewater reactor filtrate was characterized by the class Sphingobacteriia as well as Chlamydiae. The river water reactor filtrate was characterized by Flavobacteria and Actinobacteria (Fig. S5). Proteobacteria and Bacteroidetes were common to all samples, particularly Betaproteobacteria and Gammaproteobacteria. For a given water type, biofilm community structures were distinct from the filtrate community structures (Fig. 4a), likely attributable to differences in the initial water quality. Despite differences in the microbial community structure, there were no differences in the diversity of the biofilm grown on PS, PE, and glass microparticles in wastewater compared to the wastewater filtrate microbial community (all p > 0.15). The river water filtrate microbial community was more diverse than the MP biofilm microbial communities grown in that water source (all p < 0.004) except the glass microbead biofilm (p = 1). The river water filtrate microbial communities clustered more closely with wastewater biofilms than river water biofilms. The relative abundance of different bacteria in the filtrate did not necessarily correspond to that observed in the biofilm. For example, biofilm grown in wastewater had 4.8 times greater relative abundance of Bacteroidetes than biofilm grown in river water (19.9 ± 4.9% versus 4.2 ± 2.6%) despite the fact that the wastewater filtrate had 0.7 times the relative abundance of Bacteroidetes observed in river water filtrate (23.5 ± 6.3% versus 34.0 ± 4.5%).

Evidence for indicator organisms, pathogens, and antibiotic resistance genes?

To further investigate the potential for MPs to serve as carriers of microbial agents, amplicon sequencing data was searched for OTUs containing waterborne pathogens. While it is important to consider that amplicon sequencing of the V3–V4 region is limited because it is based on short reads,53 this data can be used as an indicator of the species that could be screened for using targeted techniques (e.g., qPCR) but should not be used to make conclusions about pathogenicity.54 Sequencing results indicated at very low relative abundances (5.23 × 10−5–0.51) the presence of potentially pathogenic species of Klebsiella, Pseudomonas, and Legionella in our samples. Pseudomonas was detected in all samples, particularly wastewater biofilm. Klebsiella and Legionella were primarily found in wastewater samples, especially in the filtrate. Again, these findings must be interpreted with caution given the short reads and presence of commensal organisms in these groups. Nonetheless, it is worth noting that other studies have noted the detection of potentially pathogenic species of Vibrio,6Aeromonas,55Pseudomonas,4 and Campylobacteraceae.4 Zettler et al.5 first described potentially pathogenic Vibrio as part of “the plastisphere” of microbes colonizing MP samples of PE and PP (333 μm to 5 mm) in the marine environment using high-throughput sequencing of the V3–V4 region of the 16S rRNA gene. The presence of a pathogenic species of Vibrio, V. parahaemolyticus, was confirmed on several samples of marine MP (0.5–5 mm) using CHROMEAgar Vibrio and MALDI-TOF analysis.6 However, similar to our study (which had few annotations for the order Vibrionales in biofilm samples and none for filtrate), samples of PE and PS incubated in an estuary did not accumulate gene copies of V. aestuarianus compared to a wood pellet control.56 Another marine study used the Bac27F and Univ1492R primers and Sanger sequencing to analyze DNA from 30–50 mg MP samples captured with a 308 μm net, which likely contained the fish pathogen Aeromonas salmonicida.55 In fresh water, potentially pathogenic species of Pseudomonas and Campylobacteraceae were found colonizing MP caught using a 333 μm net in a Chicago river4 and rivers in Illinois.19 It is not surprising that these studies detected Pseudomonas spp., as did this study, since this is a common and diverse genus that includes pathogens and has been studied for biodegradation.57 Differences between our findings and similar studies could be attributed to differences in the water that the samples were incubated in and/or the length of time for biofilm formation prior to sampling.

Benefits and limitations of study design

The use of batch reactors spiked with microparticles allowed for a controlled environment to study the biofilms on MPs. The majority of MP biofilm studies have taken environmental samples of MPs to study biofilm,37 which makes it more difficult to control variables and to study smaller size classes. Spiking in MP extracted from PCPs and purchased from laboratory suppliers allowed for the investigation of smaller (<300 μm) MPs of known polymer type, which were abundant in PCPs prior to phaseout. Given the voluntary phaseout of primary MPs in PCPs followed by bans on their use in the USA, there would be an interest in the future to study biofilm growth on secondary MPs, which would be expected to have different morphology and surface texture from primary MPs (i.e., less likely to be smooth spheres like the PS MPs investigated here). The choice of source water and reactor conditions for the growth of biofilm is of interest. Here, wastewater influent was used as a source water which may reflect MPs released during sewer overflows (i.e., without treatment). However, during regular flow conditions most MPs would reach the wastewater treatment plant and it is not known if/how the MP biofilm changes during the wastewater treatment process, particularly disinfection. Biofilms are generally more difficult to disinfect than planktonic bacteria9 and preliminary disinfection studies with peracetic acid indicated that MP biofilm may be more resistant to disinfection than planktonic cells (results not shown), but this result requires confirmation. For this study and a previous bench scale investigation,37 MPs were incubated in wastewater for longer than the average hydraulic residence time expected in sewers, which is estimated to range from 30 min to 12 h. Given the benefits of studying MP biofilms in a controlled setting, further bench scale studies should be conducted, paying particular attention to the morphology of particles and testing a greater variety of different incubation times and water sources. In these studies, choices must be made as to how much MP to dose into reactors, we chose to use the same volume of MPs. This resulted in different concentrations of particles per volume and it is noteworthy that our conclusions were similar normalizing to the volume of particles and to number of particles. Here we chose MP concentrations to provide sufficient biofilm for biomolecular analysis which resulted in particle concentrations comparable to the higher reported concentrations in wastewater influent21 but significantly higher than reported in fresh surface water. Choosing surface water relevant concentrations is warranted in future studies.

Conclusions

This research indicates that microparticle type and source water were greater drivers of microbial community structures observed on microparticle surfaces than size class. Thus, one could focus on the larger and easier to study size ranges tested here if one could assume the microparticles were inoculated in similar source waters and underwent similar weathering. It is not known if the differences in the biofilm observed on different MP particle types studied here were due to differences in polymer composition or differences in morphology and particle density, although clustering of glass and PS microspheres suggests morphology and density may be driving factors. With respect to the potential for MP biofilm to carry microbial agents, the biofilms did contain sul1, BacHum, and OTUs containing potential pathogens, but results indicate that microparticle biofilm did not accumulate these maker genes relative to the filtrate for the incubation periods tested. With these results, one may consider targeting the removal of more abundant particle types and sizes during wastewater treatment. Given the differences in biofilm microbial community structure as a function of source water, the biofilm may be useful for source tracking MPs from different water sources. However, microparticle biofilm in the environment would experience more complex phenomena including UV radiation, exposure to new microbial communities, and further biofouling, which would impact particle density and can in turn impact the biofilm community.20 A challenge for future bench scale studies is finding suitable reference materials, particularly those representative of secondary MPs, which are expected to be more abundant in freshwater environments as bans and phaseouts of primary MPs proceed.

Conflicts of interest

There are no conflicts of interest to declare.

Acknowledgements

Funding for this research was provided in a Rutgers Aresty and Honors College research fellowship to KP and from a grant from the Rutgers Sustainable Raritan River Coalition. Laboratory assistance was provided by William R. Morales Medina and Kris Parker. Thanks to Sheri Elsaker for collecting the FTIR spectra with the help of Nick Stone-Weiss in Ashutosh Goel's laboratory.

References

  1. C. Arthur, J. Baker and H. E. Bamford, Proceedings of the International Research Workshop on the Occurrence, Effects, and Fate of Microplastic Marine Debris, 2009 Search PubMed.
  2. J. Li, H. Liu and J. Paul Chen, Microplastics in freshwater systems: A review on occurrence, environmental effects, and methods for microplastics detection, Water Res., 2018, 137, 362–374 CrossRef CAS PubMed.
  3. M. Cole, P. Lindeque, C. Halsband and T. S. Galloway, Microplastics as contaminants in the marine environment: A review, Mar. Pollut. Bull., 2011, 62, 2588–2597 CrossRef CAS PubMed.
  4. A. McCormick, T. J. Hoellein, S. A. Mason, J. Schluep and J. J. Kelly, Microplastic is an abundant and distinct microbial habitat in an urban river, Environ. Sci. Technol., 2014, 48, 11863–11871 CrossRef CAS PubMed.
  5. E. R. Zettler, T. J. Mincer and L. A. Amaral-Zettler, Life in the “plastisphere”: microbial communities on plastic marine debris, Environ. Sci. Technol., 2013, 47, 7137–7146 CrossRef CAS PubMed.
  6. I. V. Kirstein, S. Kirmizi, A. Wichels, A. Garin-Fernandez, R. Erler, M. Löder and G. Gerdts, Dangerous hitchhikers? Evidence for potentially pathogenic Vibrio spp. on microplastic particles, Mar. Environ. Res., 2016, 120, 1–8 CrossRef CAS PubMed.
  7. R. Dris, J. Gasperi and B. Tassin, in Freshwater Microplastics, Springer, 2018, pp. 69–83 Search PubMed.
  8. S. Ziajahromi, P. A. Neale, L. Rintoul and F. D. Leusch, Wastewater treatment plants as a pathway for microplastics: development of a new approach to sample wastewater-based microplastics, Water Res., 2017, 112, 93–99 CrossRef CAS PubMed.
  9. T.-F. C. Mah and G. A. O'toole, Mechanisms of biofilm resistance to antimicrobial agents, Trends Microbiol., 2001, 9, 34–39 CrossRef CAS.
  10. A. L. Andrady, in Marine Anthropogenic Litter, ed. M. Bergmann, L. Gutow and M. Klages, Springer International Publishing, Cham, 2015, pp. 57–72 Search PubMed.
  11. M. Shimao, Biodegradation of plastics, Curr. Opin. Biotechnol., 2001, 12, 242–247 CrossRef CAS PubMed.
  12. H. Auta, C. Emenike and S. Fauziah, Screening of Bacillus strains isolated from mangrove ecosystems in Peninsular Malaysia for microplastic degradation, Environ. Pollut., 2017, 231, 1552–1559 CrossRef CAS PubMed.
  13. A. Keswani, D. M. Oliver, T. Gutierrez and R. S. Quilliam, Microbial hitchhikers on marine plastic debris: human exposure risks at bathing waters and beach environments, Mar. Environ. Res., 2016, 118, 10–19 CrossRef CAS PubMed.
  14. J. Masura, J. E. Baker, G. D. Foster, C. Arthur and C. Herring, Laboratory methods for the analysis of microplastics in the marine environment: recommendations for quantifying synthetic particles in waters and sediments, 2015, NOAA Technical Memorandum NOS-OR&R-48 Search PubMed.
  15. L. Martin Günter Joachim, K. Mirco, M. Svenja, L. Claudia and G. Gunnar, Focal plane array detector-based micro-Fourier-transform infrared imaging for the analysis of microplastics in environmental samples, Environ. Chem., 2018, 12, 563–581 Search PubMed.
  16. M. G. Löder and G. Gerdts, in Marine anthropogenic litter, Springer, 2015, pp. 201–227 Search PubMed.
  17. V. Hidalgo-Ruz, L. Gutow, R. C. Thompson and M. Thiel, Microplastics in the marine environment: a review of the methods used for identification and quantification, Environ. Sci. Technol., 2012, 46, 3060–3075 CrossRef CAS PubMed.
  18. G. M. Quero and G. M. Luna, Surfing and dining on the “plastisphere”: Microbial life on plastic marine debris, Adv. Oceanogr. Limnol., 2017, 8, 199–207 Search PubMed.
  19. A. R. McCormick, T. J. Hoellein, M. G. London, J. Hittie, J. W. Scott and J. J. Kelly, Microplastic in surface waters of urban rivers: concentration, sources, and associated bacterial assemblages, Ecosphere, 2016, 7, e01556 CrossRef.
  20. A. L. Andrady, The plastic in microplastics: a review, Mar. Pollut. Bull., 2017, 119, 12–22 CrossRef CAS PubMed.
  21. N. L. Fahrenfeld, G. Arbuckle-Keil, N. Naderi and S. Bartelt-Hunt, Source tracking microplastics in the freshwater environment, TrAC, Trends Anal. Chem. DOI:10.1016/j.trac.2018.11.030 , (in press).
  22. M. Ogonowski, A. Motiei, K. Ininbergs, E. Hell, Z. Gerdes, K. I. Udekwu, Z. Bacsik and E. Gorokhova, Evidence for selective bacterial community structuring on microplastics, Environ. Microbiol., 2018, 20, 2796–2808 CrossRef CAS PubMed.
  23. Y. Ammar, D. Swailes, B. Bridgens and J. Chen, Influence of surface roughness on the initial formation of biofilm, Surf. Coat. Technol., 2015, 284, 410–416 CrossRef CAS.
  24. S. Estahbanati and N. Fahrenfeld, Influence of wastewater treatment plant discharges on microplastic concentrations in surface water, Chemosphere, 2016, 162, 277–284 CrossRef CAS PubMed.
  25. A. P. Barrows, C. A. Neumann, M. L. Berger and S. D. Shaw, Grab vs. neuston tow net: a microplastic sampling performance comparison and possible advances in the field, Anal. Methods, 2017, 9, 1446–1453 RSC.
  26. G. Muyzer, E. C. de Waal and A. G. Uitterlinden, Profiling of complex microbial populations by denaturing gradient gel electrophoresis analysis of polymerase chain reaction-amplified genes coding for 16S rRNA, Appl. Environ. Microbiol., 1993, 59, 695–700 CAS.
  27. R. Pei, S.-C. Kim, K. H. Carlson and A. Pruden, Effect of river landscape on the sediment concentrations of antibiotics and corresponding antibiotic resistance genes (ARG), Water Res., 2006, 40, 2427–2435 CrossRef CAS PubMed.
  28. B. J. Kildare, C. M. Leutenegger, B. S. McSwain, D. G. Bambic, V. B. Rajal and S. Wuertz, 16S rRNA-based assays for quantitative detection of universal, human-, cow-, and dog-specific fecal Bacteroidales: A Bayesian approach, Water Res., 2007, 41, 3701–3715 CrossRef CAS PubMed.
  29. A. M. Bolger, M. Lohse and B. Usadel, Trimmomatic: a flexible trimmer for Illumina sequence data, Bioinformatics, 2014, 30, 2114–2120 CrossRef CAS PubMed.
  30. A. P. Masella, A. K. Bartram, J. M. Truszkowski, D. G. Brown and J. D. Neufeld, PANDAseq: paired-end assembler for illumina sequences, BMC Bioinf., 2012, 13, 31 CrossRef CAS PubMed.
  31. 454 Overview Tutorial: de novo OTU picking and diversity analyses using 454 data, Online, http://qiime.org/tutorials/tutorial.html Search PubMed.
  32. N. Segata, J. Izard, L. Waldron, D. Gevers, L. Miropolsky, W. S. Garrett and C. Huttenhower, Metagenomic biomarker discovery and explanation, Genome Biol., 2011, 12, R60 CrossRef PubMed.
  33. S. Oberbeckmann, A. M. Osborn and M. B. Duhaime, Microbes on a bottle: substrate, season and geography influence community composition of microbes colonizing marine plastic debris, PLoS One, 2016, 11, e0159289 CrossRef PubMed.
  34. T. Hoellein, M. Rojas, A. Pink, J. Gasior and J. Kelly, Anthropogenic litter in urban freshwater ecosystems: distribution and microbial interactions, PLoS One, 2014, 9, e98485 CrossRef PubMed.
  35. D. Kaiser, N. Kowalski and J. J. Waniek, Effects of biofouling on the sinking behavior of microplastics, Environ. Res. Lett., 2017, 12, 124003 CrossRef.
  36. P.-Y. Hong, N. Al-Jassim, M. Ansari and R. Mackie, Environmental and public health implications of water reuse: antibiotics, antibiotic resistant bacteria, and antibiotic resistance genes, Antibiotics, 2013, 2, 367–399 CrossRef CAS PubMed.
  37. E. M. Eckert, A. Di Cesare, M. T. Kettner, M. Arias-Andres, D. Fontaneto, H.-P. Grossart and G. Corno, Microplastics increase impact of treated wastewater on freshwater microbial community, Environ. Pollut., 2018, 234, 495–502 CrossRef CAS PubMed.
  38. J. P. Harrison, T. J. Hoellein, M. Sapp, A. S. Tagg, Y. Ju-Nam and J. J. Ojeda, in Freshwater Microplastics, Springer, 2018, pp. 181–201 Search PubMed.
  39. A. Eramo, H. Delos Reyes and N. Fahrenfeld, Partitioning of antibiotic resistance genes and fecal indicators varies intra and inter-storm during combined sewer overflows, Front. Microbiol., 2017, 8, 2024 CrossRef PubMed.
  40. C. A. De Tender, L. I. Devriese, A. Haegeman, S. Maes, T. Ruttink and P. Dawyndt, Bacterial community profiling of plastic litter in the Belgian part of the North Sea, Environ. Sci. Technol., 2015, 49, 9629–9638 CrossRef CAS PubMed.
  41. S. L. McLellan, J. C. Fisher and R. J. Newton, The microbiome of urban waters, Int. Microbiol., 2015, 18, 141–149 Search PubMed.
  42. K. Bruellhoff, J. Fiedler, M. Möller, J. Groll and R. E. Brenner, Surface coating strategies to prevent biofilm formation on implant surfaces, Int. J. Artif. Organs, 2010, 33, 646–653 CrossRef CAS PubMed.
  43. J. Schmitt, D. Nivens, D. C. White and H.-C. Flemming, Changes of biofilm properties in response to sorbed substances-an FTIR-ATR study, Water Sci. Technol., 1995, 32, 149–155 CrossRef CAS.
  44. S. Klein, I. K. Dimzon, J. Eubeler and T. P. Knepper, in Freshwater Microplastics, Springer, 2018, pp. 51–67 Search PubMed.
  45. K. Kathiresan, Polythene and plastic-degrading microbes in an Indian mangrove soil, Rev. Biol. Trop., 2004, 51, 629–633 Search PubMed.
  46. S. Yoshida, K. Hiraga, T. Takehana, I. Taniguchi, H. Yamaji, Y. Maeda, K. Toyohara, K. Miyamoto, Y. Kimura and K. Oda, A bacterium that degrades and assimilates poly(ethylene terephthalate), Science, 2016, 351, 1196–1199 CrossRef CAS PubMed.
  47. C. D. Rummel, A. Jahnke, E. Gorokhova, D. Kühnel and M. Schmitt-Jansen, Impacts of Biofilm Formation on the Fate and Potential Effects of Microplastic in the Aquatic Environment, Environ. Sci. Technol. Lett., 2017, 4, 258–267 CrossRef CAS.
  48. K. Magnusson and F. Norén, Screening of microplastic particles in and down-stream a wastewater treatment plant, 2014, IVL Swedish Environmental Research Institute Report Search PubMed.
  49. Y. K. Song, S. H. Hong, M. Jang, J.-H. Kang, O. Y. Kwon, G. M. Han and W. J. Shim, Large accumulation of micro-sized synthetic polymer particles in the sea surface microlayer, Environ. Sci. Technol., 2014, 48, 9014–9021 CrossRef CAS PubMed.
  50. M. Kooi, E. Besseling, C. Kroeze, A. P. van Wenzel and A. A. Koelmans, in Freshwater Microplastics, Springer, 2018, pp. 125–152 Search PubMed.
  51. Y. Chae and Y.-J. An, Effects of micro-and nanoplastics on aquatic ecosystems: Current research trends and perspectives, Mar. Pollut. Bull., 2017, 124, 624–632 CrossRef CAS PubMed.
  52. O. Setälä, V. Fleming-Lehtinen and M. Lehtiniemi, Ingestion and transfer of microplastics in the planktonic food web, Environ. Pollut., 2014, 185, 77–83 CrossRef PubMed.
  53. P. Woo, S. Lau, J. Teng, H. Tse and K.-Y. Yuen, Then and now: use of 16S rDNA gene sequencing for bacterial identification and discovery of novel bacteria in clinical microbiology laboratories, Clin. Microbiol. Infect., 2008, 14, 908–934 CrossRef CAS PubMed.
  54. J. A. Bryant, T. M. Clemente, D. A. Viviani, A. A. Fong, K. A. Thomas, P. Kemp, D. M. Karl, A. E. White and E. F. DeLong, Diversity and activity of communities inhabiting plastic debris in the North Pacific Gyre, mSystems, 2016, 1, e00024 CrossRef PubMed.
  55. M. K. Viršek, M. N. Lovšin, Š. Koren, A. Kržan and M. Peterlin, Microplastics as a vector for the transport of the bacterial fish pathogen species Aeromonas salmonicida, Mar. Pollut. Bull., 2017, 125, 301–309 CrossRef PubMed.
  56. S. Oberbeckmann, B. Kreikemeyer and M. Labrenz, Environmental factors support the formation of specific bacterial assemblages on microplastics, Front. Microbiol., 2018, 8, 2709 CrossRef PubMed.
  57. M. Reddy, M. Deighton, R. K. Gupta, S. Bhattacharya and R. Parthasarathy, Biodegradation of oxo-biodegradable polyethylene, J. Appl. Polym. Sci., 2009, 111, 1426–1432 CrossRef CAS.

Footnote

Electronic supplementary information (ESI) available. See DOI: 10.1039/c8ew00712h

This journal is © The Royal Society of Chemistry 2019