Rapid Physicochemical Changes in Microplastic Induced by Biofilm Formation

Abstract

Risk assessment of microplastic (MP) pollution requires understanding biodegradation processes and related changes in polymer properties. In the environment, there are two-way interactions between the MP properties and biofilm communities: (i) microorganisms may prefer some surfaces, and (ii) MP surface properties change during the colonization and weathering. In a 2-week experiment, we studied these interactions using three model plastic beads (polyethylene [PE], polypropylene [PP], and polystyrene [PS]) exposed to ambient bacterioplankton assemblage from the Baltic Sea; the control beads were exposed to bacteria-free water. For each polymer, the physicochemical properties (compression, crystallinity, surface chemistry, hydrophobicity, and surface topography) were compared before and after exposure under controlled laboratory conditions. Furthermore, we characterized the bacterial communities on the MP surfaces using 16S rRNA gene sequencing and correlated community diversity to the physicochemical properties of the MP. Significant changes in PE crystallinity, PP stiffness, and PS maximum compression were observed as a result of exposure to bacteria. Moreover, there were significant correlations between bacterial diversity and some physicochemical characteristics (crystallinity, stiffness, and surface roughness). These changes coincided with variation in the relative abundance of unique OTUs, mostly related to the PE samples having significantly higher contribution of Sphingobium, Novosphingobium, and uncultured Planctomycetaceae compared to the other test materials, whereas PP and PS samples had significantly higher abundance of Sphingobacteriales and Alphaproteobacteria, indicating possible involvement of these taxa in the initial biodegradation steps. Our findings demonstrate measurable signs of MP weathering under short-term exposure to environmentally relevant microbial communities at conditions resembling those in the water column. A systematic approach for the characterization of the biodegrading capacity in different systems will improve the risk assessment of plastic litter in aquatic environments.

Keywords: biodegradation; biofilm; microbiome composition; microplastic; physicochemical characterization; polyethylene; polypropylene; polystyrene.

Figure 1

Changes in the physicochemical characteristics of the microplastic measured in this study: (A) Degree of crystallinity [X_c]; (B) Stiffness [k]; (C) Maximum compression [ε_max]; (D) Arithmetic roughness [R_a]; and (E) Diameter [d]. The data are presented as mean (horizontal notches) and 95% confidence interval (vertical bars) values of the bootstrapped differences between the treatment means (B for Biofilm and W for Water) and their respective controls. Asterisks (*) indicate significant difference between the treatment and the control indicated by the distributions with the confidence interval excluding zero. No statistical comparisons were possible for crystallinity in PS because this material is amorphous (X_c ≈ 0).

Figure 2

Degree of crystallinity, X_c, for Biofilm (solid squares) and Water (solid circles) treatments and their respective untreated controls (hollow symbols). The data points represent replicate samples (n = 3). The polymers tested were polyethylene (PE), polypropylene (PP), and polystyrene.

Figure 3

Tensile comparisons for PE, PP, and PS beads from the Biofilm and Water treatments in relation to their respective controls: (A) diameter (B) stiffness [k], and (C) maximum compression [ε_max]. The solid horizontal lines represent the means, error bars represent the 95% confidence intervals, and symbols represent individual observations. Each symbol represents an individual observation. Solid symbols represent experimental treatments (Biofilm or Water) and the hollow symbols of the same color represent their respective controls. The number of observations, n, ranges between 8–12, 3–10, and 3–17 for panels (A–C) respectively.

Figure 4

ATR-FTIR spectra of beads of polyethylene (PE), polypropylene (PP), and polystyrene (PP). For each polymer, the bottom three spectra show the spectra of the MP control (black), the MP Water (blue), and the MP Biofilm samples (red). These spectra were approximately normalized on the main polymer bands by dividing the original spectra by 5 for all PE spectra, by 1.8 for the PP Control spectrum, and by 2 for the PP Water and PS Control spectra. The top spectra for each polymer illustrate the spectral changes that are induced by incubation in seawater and sterilized water. They are subtractions of the respective control spectrum from the MP Water (blue) and the MP Biofilm (red) spectra. These spectra were multiplied by 2 for a clearer presentation. The dark red difference spectrum labeled PS Biofilm 5—PS Control was calculated from a PS Biofilm spectrum that significantly deviated from the other four PS Biofilm spectra and which therefore was not included in the averaged PS Biofilm spectrum and in the difference spectrum labeled PS Biofilm 1 to 4—PS Control.

Figure 5

The effects of incubation on the 1,800–1,500 cm⁻¹ spectral region of the infrared spectrum. The spectra are the difference spectra shown in Figure 4, but before multiplication by 2. See that figure for more information. The spectral position of the shoulder above 1,740 cm⁻¹ was determined from the second derivative spectrum.

Figure 6

Alpha diversity indices (mean ± SD, n = 3) for bacterial communities in the biofilm grown on the test polymers (PE, polyethylene; PP, polypropylene; and PS, polystyrene). Shannon-Wiener (Shannon H) and Fisher's alpha indices are shown on the left y-axis and Chao 1 and ACE estimators are shown on the right y-axis. Statistical comparisons are presented in Table 1; the differences between the polymers were significant (PE vs. PS and PP vs. PS) for Chao1 and ACE but not for the other two indices.