Since industrial production began in the 1950s, plastic debris has become ubiquitously distributed in our environment; it is even found in remote locations, from isolated mountain catchments in the French Alps (Allen et al., 2019) to the depths of the Mariana Trench in the western Pacific Ocean (Jamieson et al., 2019). An estimated 19 to 23 million metric tons of plastic waste amassed in our oceans in 2016 alone (Borrelle et al., 2020), increasing societal concerns regarding the accumulation and impact of plastics on the marine environment (Schnurr et al., 2018). As a result, the harm inflicted on macrobiota such as birds, fish and mammals, by ingestion or entanglement has been well documented (Laist, 1997; Wilcox et al., 2013; Besseling et al., 2015). However, the ecological impacts of plastic polymers and their additives on the ecology of marine microbiota are less explored and therefore not well understood (Lear et al., 2021).

Following preliminary observations of bacterial and diatom colonisation on plastic debris from the Sargasso Sea in the 1970s (Carpenter and Smith, 1972), it has become clear that complex, diverse microbial and metazoan communities rapidly colonise marine plastic (Zettler et al., 2013). Many studies have since characterised the bacterial and eukaryotic communities formed on environmental plastics, often referred to as the ‘plastisphere’, providing evidence that taxonomically distinct plastisphere communities develop as compared to communities in the surrounding environment (Lear et al., 2021). Whilst it has been shown that microbes such as the fungal phylum Chytridiomycota (Kettner et al., 2017) and the bacterial phyla Bacteriodetes, Proteobacteria (Oberbeckmann et al., 2014) and Cyanobacteria (Bryant et al., 2016) are highly abundant in the plastisphere, the specificity of plastisphere taxa remains enigmatic as none are known to colonise specific plastics alone. It has yet to be concluded whether different communities associate with, and develop on different types of plastic, which could impact the broader ecosystem’s functionality, with unknown and unforeseen consequences (Oberbeckmann et al., 2017; Pinto et al., 2019).

When studying the plastisphere microbiome, the focus has predominantly been to assess mature biofilm, or biofilm of undeterminable age. Whilst there have been limited investigations into the pioneer species during the early formation of plastisphere biofilms, a recent study investigated the microbiomes of plastic pre-production pellets in flowing seawater, revealing compositionally distinct communities forming on plastic between 8 and 70 days (Ward et al., 2022). Erni-Cassola et al. (2020) suggested that obligate hydrocarbon-degrading bacteria may dominate the preliminary stages of colonisation if they possess the ability to utilise hydrocarbons which can leach from weathered plastics, as a carbon source. As surface hydrophobicity influences microbial biofilm formation (Teughels et al., 2006; Rummel et al., 2017), early colonisers may be better adapted to colonising plastic surfaces, before succession results in domination by more ‘generic’ biofilm community members (Wright et al., 2020b). Due to the extensive variability in the physicochemical dynamics of plastic surfaces and particularly the presence of different additives and contaminants, microbial communities are likely to be impacted both positively and negatively by plastics, depending on their combined responses to these substrates as either sources or carbon and nutrients, or as sources of toxicants. The paucity of studies investigating microbial colonisation of marine plastics in situ leaves a significant knowledge gap in understanding which taxa are involved in the early stages of plastic colonisation and at different stages of biofilm succession. This is of particular concern since the composition of even the most commonly used plastics is highly variable (Roosen et al., 2020; Wiesinger et al., 2021), meaning microbial responses to marine plastics are likely to be very specific to the particular substrates and environments under study. Currently, we understand relatively little with regard to whether generalist biofilm communities form on plastics, or if different taxa colonise different plastics, as well as how these communities vary over time and the implications this may have on ecosystem health and function.

Previous studies have shown that plastisphere communities are significantly influenced by their geographic location (Hoellein et al., 2014; Oberbeckmann et al., 2014), highlighting the need for investigations to take place over a more extensive range of spatial scales to determine any geographic factors shaping plastisphere assembly and to identify a core microbiome within plastisphere communities, should one exist. Our study is the first to characterise marine plastisphere communities in Aotearoa-New Zealand (A-NZ), to determine the presence of taxa previously identified as putative plastic biodegraders in global studies and conversely, to determine whether there are taxa in our ocean that might be sensitive to plastics, with potentially unintended consequences. To assess the identities and abundance of plastic-associated bacteria and fungi, we deployed a structure containing two different plastic types along with an inert glass control into the marine environment at Te Whakaraupō-Lyttelton Harbour, A-NZ, for a short period of time (sampling from 2 – 12 weeks), allowing assessment of the early stages of microbial plastic colonisation. The polymers in our study – nylon-6 (also called polyamide-6, abbreviated in our study as PA) and linear low-density polyethylene (LLDPE), were selected based on their high abundance in marine plastic debris (Guzzetti et al., 2018; Erni-Cassola et al., 2019) and different chemical compositions as PA contains heteroatoms in the polymer backbone, LLDPE does not. They also contained additives typically used in commercial plastics (e.g., tris(2,4-di-tert-butylphenyl)phosphite, or Irgafos 168^®) which have been previously reported as hazardous or toxic (Hammond et al., 2013; Li et al., 2016; Groh et al., 2019). First, we hypothesised that microbial communities associating with these substrates differ taxonomically from microbes present in ambient seawater. Second, since the diversity of microbial communities is thought to be influenced by the degree of hydrophobicity of the substrate (PA is hydrophilic (Huang and McCutcheon, 2014) whereas LLDPE is hydrophobic (Lu et al., 2004)), we predicted to observe significant differences in the composition, evenness and diversity of bacterial and fungal communities between the biofilms that formed on these plastics and also on glass. Further, we predicted to see greater differences between the communities of younger plastic-associating biofilms compared to communities growing on the glass, indicating the presence of pioneer substrate-specific taxa capable of biodegrading plastics. We hypothesised that the community differences among substrates would reduce as the biofilm aged since we expect more ‘generalist’ marine taxa would dominate the biofilm over time.

A total of 36 marine samples were collected, comprising three replicates of each substrate, sampled 2, 6 and 12 weeks after deployment, and triplicate water samples taken at each timepoint. After quality filtering, a total of 5142 and 1773 unique taxa were obtained for bacterial and fungal comparative analyses, respectively. Rarefaction curves based on the ASVs of the bacterial communities ( Figure S3A ) indicated adequate sequencing depth was achieved to ensure the majority of species were represented. However, many fungal rarefaction curves did not reach saturation ( Figure S3C ); further sequencing would therefore give a better indication of diversity. Following rarefaction, 4431 bacterial and 688 fungal unique taxa were retained for further analysis.

The microbial colonisation of plastics established in the marine environment has been well-documented (Carpenter and Smith, 1972) but there is a lack of study identifying the community establishment over time. To address this knowledge gap, we explored temporal changes in bacterial and fungal communities following the deployment of two plastic polymer substrates (as well as glass slides) into the marine environment of Te Whakaraupo-Lyttelton Harbour, Aotearoa-New Zealand. We found distinct differences between communities in the water column and surface colonisers forming a biofilm at all timepoints, supporting the findings of Zettler et al. (2013) and Dussud et al. (2018). No significant difference in diversity was observed between communities on glass and the two plastic types tested at any timepoint. We did however, establish that surface associated microbial communities changed over time across all surfaces, as previously found by Oberbeckmann et al. (2016) and Xu et al. (2019). More taxa related to putative-plastic biodegraders were found on glass, LLDPE and PA at weeks two and six, for bacteria and fungi, respectively. However, we found no evidence that putative plastic degrading taxa were more abundant on any one substrate.

Biodegradation of polymers by microbial communities can include biodeterioration, bio-fragmentation, mineralisation and assimilation, as defined by Zeenat et al. (2021). We identified the presence of certain species and genera within our marine samples that might be capable of biodegrading these plastics via, noting that mineralisation and assimilation have yet to be directly determined in the marine environment. These ASVs were shared amongst glass, LLDPE and PA-associated communities, with the most abundant putative biodegraders being identified on multiple substrates (e.g., Vibrio spp.). Biofilm age appeared to have a significant impact on the diversity of proposed plastic-degraders present, rather than substrate type. Once immersed in the marine environment, organic matter builds up rapidly forming a conditioning film on any solid substrate, promoting microbial colonisation (Marshall, 1985; Galloway et al., 2017). This has been shown to occur within a matter of minutes following immersion (Harrison et al., 2014, Wright et al., 2020b) and provides a nutrient-enriched environment for biofilm-forming taxa. Factors including surface roughness and hydrophilicity may impact rates of surface conditioning and also subsequent rates of colonisation on different substrates (Bhagwat et al., 2021). However, the results of our study suggest immersion time, rather than substrate composition, was the more important determinant of biofilm community composition.

Although the plastics used in our study were not artificially aged by exposure to elevated UV light or temperature, it is likely that some carbon present on, and within, the virgin plastic surface is biologically accessible to colonising microbes in the form of short chain oligomers and monomers (Gewert et al., 2018; Romera-Castillo et al., 2018; Erni-Cassola et al., 2020). Monomers and short-chain polymers may leach out of plastic quickly (Kwon et al., 2015). Further, organic additives to plastic substrates added during manufacturing are not bound to the polymer and may be leached into the environment and used as a carbon or nutrient source by the plastisphere community (Wright et al., 2020a). The additives present in the PA and LLDPE used in this study (Nylostab S-EED and Irganox B215 (Irganox 1010 and Irgafos 168)), which are typically included to promote UV light stability and prevent physicochemical degradation, are not readily biodegradable according to their MSDS data. However, such additives may impact microbes trying to establish on the plastics, influencing the communities that are able to survive. An enormous diversity of chemical additives are currently incorporated into plastics such as LDPE (Thomas et al., 2012) including, in the case of ‘oxy-degradable’ plastics, metal salts of iron or manganese (Thomas et al., 2012) intended to catalyse abiotic plastic degradation. The sheer diversity of plastic additives used means that very different microbial community responses may be witnessed, even in reference to the same base polymer in the same environment. In addition, diverse contaminants within the ambient seawater may sorb to plastics, impacting community composition and functioning variably depending on their concentrations in the environment under investigation. Unfortunately, the diversity of environmental contaminants present was not assessed in the present study and so any related impacts caused by variation across substrate type and sampling time could not be quantified.

Research has previously indicated that plastic degradation, where observed, is likely to slow after the early stages of colonisation or immersion (Datta et al., 2016; Enke et al., 2018; Wright et al., 2019, Wright et al., 2020b), which could explain the greater relative abundance of reported bacterial plastic-biodegraders in the first weeks after immersion of plastics into seawater. Some studies demonstrate biofilm maturation after days to over a week (Erni-Cassola et al., 2020, Wright et al., 2020b), with the fast rates of colonisation and succession perhaps explaining the low relative abundances of putative plastic-biodegrading taxa in our samples as generalist taxa out-compete pioneer species for space and resources. Further, as succession occurred, the biofilms increased in thickness due to an increased abundance of microalgae (represented in Figure S9 , not all data shown) as is commonly observed (Eich et al., 2015). The increased biomass of taxa such as algae, that are not known to mineralise and assimilate plastics, would provide an alternative attachment substrate for the bacteria. This means a decreased proportion of the bacterial community would be in direct contact with the plastic surface over time, and they would also be less exposed to any toxic substances associated with the plastic polymer, as proposed by Ward et al. (2022). It is likely that a combination of these explanations is responsible for shaping the communities of putative plastic-degrading taxa in the marine environment. Temporal abundance patterns of putative plastic-degrading microbes were similar on plastic substrates as to those observed on inert glass slides. Thus, it remains unlikely that observed increases of presumed plastic-degrading taxa on the submerged plastic substrates were driven by community changes in response to the metabolism of the plastic polymers or their additives.

While providing a useful avenue for further exploration, we did not seek to confirm evidence of plastic biodegradation in the present study. At least 286 bacteria and 150 fungi have been reported as plastic biodegraders (Gambarini et al., 2021) yet it is often unclear whether or not these microbes are responsible for the biodegradation of the plastic polymer itself, monomers and short-chain polymers that are quickly leached from the plastic, or any additives present. To assess the putative plastic biodegradation potential of these communities would require additional physicochemical assessment of the plastics, as outlined in Lear et al. (2021). However, the identification of putative plastic biodegrading taxa, or closely related species and strains by taxonomic analysis, as used in this study, nevertheless provides a useful mechanism to identify environments, and timescales, for more targeted analyses of plastic biodegrading organisms and enzymes. PCR biases prevent the amplification of DNA from certain taxa, including putative plastic-biodegrading strains but the inclusion of mock-community extraction and PCR controls as recommended by Lear et al. (2018) might help to quantify the extent of such biases. Omitting the need for PCR and its associated biases, future investigations into the metagenomes or metatranscriptomes of these microbes and communities may reveal additional taxonomic details. Using genomics and proteomic profiling, Oberbeckmann et al. (2021) reported that the properties of polyethylene and polystyrene pellets had little influence on their associated microbial communities, they were similar on both substrate types. Future community-based assessments, including from metagenomics and proteomics data, as used by Oberbeckmann et al. (2021), may further help to identify appropriate sites for the later enrichment and isolation of potential plastic biodegrading organisms. In particular, confirmation of the presence of complete plastic biodegradation pathways, along with their corresponding regulatory mechanisms could help to identify environments and sites for the mining of novel plastic degrading genes and taxa.

We were not able to identify any fungal ASVs indicative of plastic-specific biofilms (i.e. none were exclusively or reliably found on any one plastic substrate). However, the presence of the previously reported plastic biodegrading bacterial genera Alcanivorax and Shewanella suggests that they attach more readily to LLDPE or grow at a faster rate on this substrate. Whilst Delacuvellerie et al. (2019) propose Alcanivorax borkumensis as a biodegrader of low-density polyethylene (LDPE), many other studies were unable to determine the identity of putative bacterial or fungal biodegraders of plastic past the genus level (Sekiguchi et al., 2011). Alcanivorax, unidentified at the species level, have been reported to biodegrade polycaprolactone (PCL) (Sekiguchi et al., 2011) and poly-β-hydroxybutyric acid (PHB) (Suzuki et al., 2017). Sekiguchi et al. (2011) also reported an unidentified Shewanella with the ability to biodegrade PCL in the marine environment of Japan. Shewanella isolated from soil has also been reported to biodegrade PHB by producing clear zones in agar infused with a PHB emulsion (Sung et al., 2016; Suzuki et al., 2017). Without clear taxonomic classification to the species level in both the samples from this study and reported putative biodegraders, it remains challenging to verify that the species we detected are the same as those reported in other studies. This limits our ability to elucidate that species present in our samples have the potential to be plastic biodegraders. Further, since the abundance of putative plastic-degraders was equally high in glass-associated biofilms it remains possible that none of the taxa identified have the capability to metabolise PA or LLDPE. Hundreds of publications identify putative PA and LDPE-degrading taxa, yet Lear et al. (2022) recently postulated that, in fact, there remains no strong evidence for their microbial degradation.

Gangoiti et al. (2012); Jeon and Kim (2013) and Montazer et al. (2018) reported the mesophilic bacterium Stenotrophomonas maltophilia to be a potential biodegrader of poly(3-hydroxydecanoate) (P3HO), PLA and LDPE, respectively, after isolating this bacterium from soil or landfill. S. maltophilia was identified in PA biofilm samples from this study, suggesting that this bacterium may form specific associations with PA in the marine environment. The fungus Aureobasidium pullulans was also present in seawater, glass and LLDPE biofilm samples from this study, and has previously been reported to biodegrade five different plastics: polyurethane (PU) (Crabbe et al., 1994), PCL (Fields et al., 1974), PHB, PLA (Jeszeová et al., 2018) and a polyvinyl chloride (PVC) blend (Webb et al., 2000). The pathogenic fungus Cladosporium cladosporioides, present in our samples across all substrates at 6 and 12 weeks, is reported to biodegrade polyethylene (PE) (Bonhomme et al., 2003), polyethylene terephthalate (PET) (Breuker et al., 2003) and PU (Brunner et al., 2018). Additionally, Gonda et al. (2000) reported that the fungus Debaryomyces hansenii, found in the North Sea, biodegraded PHB, as did Matavulj and Molitoris (2009), who reported the biodegradation of poly-3-hydroxyalkanoate (PHA) by D. hansenii. Sarocladium kiliense is another fungus identified within our samples, which has been reported to biodegrade PE (Karlsson et al., 1988), whilst Trichoderma viride, also found in soil and landfill material, has been reported to biodegrade PLA (Lipsa et al., 2016) and LDPE (Munir et al., 2018). In general, we found organisms previously reported to degrade plastics to only be present in extremely low relative abundances and they were shared throughout the glass, LLDPE and PA communities at each timepoint, indicating their presence and abundance were not substrate specific. Despite identifying numerous taxa related to proposed plastic degrading bacteria and fungi among our samples, it is important to note that many presumed plastic degraders were previously exposed to plastics subjected to quite specific laboratory conditions, including exposure to elevated temperature (e.g., Koutny et al., 2006) and UV light (e.g., Karlsson et al., 1988). These conditions are chosen to purposefully alter the physicochemical structure of the polymer. Significant polymer degradation is rarely or never demonstrated under ambient environmental conditions (Lear et al., 2022) and might not be possible under the natural conditions to which our plastics were exposed.

Studies with a much shorter range of incubation lengths [i.e. sampling from days to weeks, rather than over seasons or months as conducted by Xu et al. (2019) and De Tender et al. (2017)] are desirable to determine if microbial substrate specificity for different plastics could occur during the very early stages of colonisation and succession. Indeed, Ward et al. (2022) suggest that some early- to mid-stage colonizers associated with hydrocarbon degradation may contribute to the degradation of plastic leachate before early colonisers are replaced by ‘cheaters’ (Datta et al., 2016) which may not produce plastic- (or additive) degrading enzymes, but scavenge the products of their degradation. In addition to conducting studies over a broader range of incubation times, future investigations may benefit from including a wider range of plastic substrates, with additive combinations to control for their effects. This would help to determine whether plastisphere communities are unique to each plastic or are largely consistent across all substrates, as indicated here. Including different plastic conditions such as ‘aged’ plastics may also be useful since plastic weathering causes carbon availability to increase, as aged plastics contain a greater abundance of short chain polymers, compared to that of non-aged plastics (Gewert et al., 2018; Romera-Castillo et al., 2018; Erni-Cassola et al., 2020).

There remains a limited number of studies regarding the functional potential of the plastisphere (Lear et al., 2021). Oberbeckmann et al. (2021) determined proteomic and genomic profiles from microbial communities were not specific to PE and PS polymers, suggesting plastic type has a limited influence on plastisphere functional potentials. However, variability in the chemical composition of plastics, including additives and contaminants, may be expected to elicit different functional profiles in the communities colonising plastic surfaces. As we propose that immature plastispheres are more likely to contain plastic-biodegrading microbes, it is important to investigate the expression of known genes conferring plastic-biodegradation potential within these earlier communities. via Metatranscriptomic studies focussing on earlier timepoints would allow for the investigation of gene expression during colonisation, such as those relating to biofilm formation, adherence/surface attachment, type IV pilus expression, quorum sensing, flagella production and motility (Vu et al., 2009; Belas, 2014; Rampadarath et al., 2017). Analysis of differential expression to first determine whether the expression of certain genes is being up/down-regulated in the presence of plastics, as well as between plastic types, followed by establishing at which biofilm age(s) are these patterns being observed, would provide further evidence as to if taxa previously identified as putative plastic biodegraders do have the potential to biodegrade plastic. The isolation and testing of individual taxa identified for their plastic-biodegrading attributes would then be essential for characterisation of novel plastic-biodegrading organisms. Even if plastic biodegraders with limited degradation capacity are identified, prior studies indicate that enzymatic degradative capabilities might be enhanced following genetic modification (Yoshida et al., 2016; Austin et al., 2018). Such investigations would widen our understanding of the plastisphere and bring us closer to identifying the impact of these communities on the microbial ecology and functioning of our oceans, as well as identifying and characterising further microbes with plastic-biodegradation potentials.

We investigated the influence of two plastic polymers on the diversity and composition of surface-associated microbial communities in the marine environment using 16S rRNA gene and ITS2 region amplicon sequencing. Biofilm-associated communities (glass and plastic) significantly differed from communities found in ambient seawater, and age of the biofilm appeared to be a significant factor in determining community composition, as also recently reported by Ward et al. (2022). In particular, within our plastisphere communities, bacteria and fungi appeared to be more unique at two and six weeks of immersion, respectively, with these communities containing microbes previously reported as having plastic biodegradation potential. To our knowledge, this is the first study to investigate the plastisphere in the marine environment of Aotearoa-New Zealand. Our results suggest that communities attaching to plastic in the marine environment include species reported to have the potential for plastic biodegradation, yet these were no more abundant than in glass-associated biofilm communities. Rather, our research indicates that microbial degradation is not likely to be a dominant process in marine environments. To gain a more comprehensive understanding of the potential of the marine plastisphere, analysis of communities from shorter immersion durations and incorporating more plastic substrates and conditions is necessary, alongside analysis of metagenomic data, to confirm the presence of genes known or suspected to confer plastic biodegradation traits.

The datasets presented in this study can be found in the Sequence Read Archive (SRA) of National Centre for Biotechnology Information (NCBI) under the accession number PRJNA764817 and ID 764817.

All authors contributed to the research, writing and editing of this manuscript and provided their full consent for publication.

This work was conducted as part of the Aotearoa Impacts and Mitigation of Microplastics (AIM²) project, in receipt of funds from a New Zealand Ministry of Business, Innovation and Employment (MBIE) Endeavour Research Programme grant (C03X1802).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.