The global diversity of the plastisphere

An estimated seven billion tons of plastic waste has now been produced globally, 79% of which has been discarded in landfills or the environment. There are a range of routes for plastic to end up in the environment, such as mismanagement of waste, lost or discarded fishing gear, fibres released during the washing of clothes or microplastics in cosmetic products. Currently between 19 and 23 million tons of plastics are predicted to enter the oceans each year. Once in the oceans, these plastics experience a range of biotic and abiotic processes. They can be transported long distances by wind and currents, they may be fragmented by wave action, or undergo weathering through processes such as UV photodegradation. As a result, they can cause a range of negative environmental impacts, both directly, through entanglement or ingestion, or indirectly, through the transfer of toxic chemicals. Although the fate and durability of plastics is unknown, some suggest a persistence of hundreds of years or fragmentation rates as low as 1-5% per year. This has led us to look towards microbes for a solution to this problem.

As with any surface that enters aquatic environments, plastics are rapidly colonised by living organisms. These are typically microbial communities that are composed of diverse bacteria, single-celled algae and fungi, but can also include macro-organisms such as barnacles, bryozoans, hydroids or multicellular algae. Collectively, these biofilms have been termed the ‘plastisphere’. There are now over 100 marine plastisphere studies, which have used a range of experimental and methodological conditions to characterise the plastisphere, and microbial members of the plastisphere have been suggested to be:(i) different from communities that colonise other surfaces (or from surrounding free-living communities);(ii)not different from communities that colonise other surfaces; (iii) only different from communities colonising other surfaces under specific environmental conditions or at specific time points; (iv) more diverse than other microbial communities; (v) less diverse than other microbial communities; (vi) potentially degrading the plastics; (vii) potentially degrading the additives of plastics; or (viii) pathogenic and/or carrying antimicrobial resistance genes. So how do we know which of these are true, or under which circumstances they occur?

All of the studies that use Next Generation Sequencing to characterise the plastisphere are theoretically comparable, however, the different methods that they use for data analysis mean that they are not directly comparable. With the aims of investigating the environmental and methodological factors that shape the plastisphere, as well as whether potential plastic degraders or pathogens are present within the plastisphere, we collected all of the marine plastisphere studies that had available sequencing and metadata (at the beginning of 2020). This meant that we were able to perform a meta-analysis that included 21 marine studies and almost 1200 samples that characterised the 16S rRNA gene using amplicon sequencing (as well as a further 14 studies conducted in freshwater, other aquatic or terrestrial environments). Most of these studies collect plastics after unknown environmental residence times and are conducted around Europe or in other temperate environments. There were no studies conducted in the Southern Hemisphere and many of the studies don’t determine plastic type or include biofilm controls (e.g. rocks, glass or shells). There were also very few studies that characterised 18S or ITS2 rRNA genes or used shotgun metagenomics.

We found that samples from the same study or environment were the most similar to one another, with the Proteobacteria almost always dominating in terms of relative abundance. We used random forest models (machine learning) on 20 metadata categories to show that overall environmental variables have the largest impact on shaping the plastisphere. The top five variables were: (1) light availability (ambient or modified); (2) whether experiments were carried out in the laboratory or in the field; (3) whether plastics were incubated in the water column or the sediment; (4) the environment that the study was performed in; and (5) the primer pair that was used for sequencing.

While plastic type (to varying levels of specificity) did not produce accurate random forest models when samples from all environments were considered together, when we split samples to the environment that they were from we found that the community composition could be used to accurately predict the general plastic type up to 83% of the time in the marine environment. We found that many of the taxa that were most informative for building these models, and that were also more abundant on plastics than control biofilms across the included studies, were from hydrocarbonoclastic, or hydrocarbon-degrading, groups. In particular, the Oceanospirillales were more abundant on aliphatic plastics (polyethylene or polypropylene) and the Alteromonadales were more abundant on other plastics (e.g. polyethylene terephthalate or polystyrene). These are taxa that have previously been suggested to be degrading plastics or the additives of plastics but are also known to be capable of degrading substrates that share structural similarity with plastics (as well as the additives of plastics) like alkanes or components of crude oil.

There were also several taxa that were more abundant on plastics than other substrates that are potentially pathogenic. There are limits, however, to what we can establish solely based on the taxonomic information gained from sequencing of the 16S rRNA gene. The ability of a bacterium to be pathogenic is likely dependent on the presence of specific virulence factors which are often in mobile genetic elements. Furthermore, the amplicon sequencing used by most plastisphere studies to date does not give the resolution to differentiate between closely related strains of the same species, such as between pathogenic and non-pathogenic Vibrio species or hydrocarbon degrading and non-degrading Pseudomonas putida strains. These results confirmed that general principles in marine microbial ecology govern the colonisation dynamics of plastics but highlighted the need for further work that characterises the functional capacity of the plastisphere.

It is clear that plastics are not going anywhere anytime soon, and the potential for them to be degraded by environmental microbes has captured the imaginations of scientists and members of the public alike. Whilst some studies have shown promising results on the biodegradation of plastics by isolated microorganisms, there is still much work to be done before we can definitively determine what happens in the environment.