Dissolved organic carbon leaching from plastics stimulates microbial activity in the ocean. — TEI-tagged view

Cristina Romera-CastillopersName, Maria PintopersName, Teresa M. LangerpersName, Xosé Antón Álvarez-SalgadopersName, Gerhard J. HerndlpersName (2018date). “Dissolved organic carbon leaching from plastics stimulates microbial activity in the ocean..” Nature Communicationstitle·j.

Object of study — dissolved organic carbonterm·dom leachingterm·key from marine plasticsterm·key and its use by heterotrophicterm·key microbes

Research question — what impact plasticterm·key pollution has on the lowest trophic levels of the marine food web

Contribution — measurements and a global estimate of plastic-derived DOCterm·dom and its bioavailability to marine bacteria

Full text — tagged

The reference list is not part of the text we hold for this paper. Our copy was extracted from the publisher's page and the bibliography did not survive that extraction, so the article below ends before its references. Nothing has been removed by us.

Dissolved organic carbonterm·dom leachingterm·key from plasticsterm·key stimulates microbial activityterm·dom in the ocean.

Abstract Approximately 5.25 trillion plasticterm·key pieces are floating at the sea surface. The impact of plasticterm·key pollution on the lowest trophic levels of the food web, however, remains unknown. Here we show that plasticsterm·key release dissolved organic carbonterm·dom (DOCterm·dom) into the ambient seawater stimulating the activity of heterotrophicterm·key microbes. Our estimates indicate that globally up to 23,600 metric tons of DOCterm·dom are leachingterm·key from marine plasticsterm·key annually. About 60% of it is available to microbial utilization in less than 5 days. If exposed to solar radiationterm·key, however, this DOCterm·dom becomes less labile. Thus, plasticterm·key pollution of marine surface waters likely alters the composition and activity of the base of the marine food webs. It is predicted that plasticterm·key waste entering the ocean will increase by a factor of ten within the next decade, resulting in an increase in plastic-derived DOCterm·dom that might have unaccounted consequences for marine microbes and for the ocean system.

Introduction Plasticterm·key debris represents a contemporary environmental problem affecting marine fauna from small copepods to large mammals 1 . Some animals mistake plasticterm·key for food 2 and, among other impacts, the ingestion of plasticterm·key might cause stress, false satiation, reduced growth rates and can affect reproduction 3 . However, the impact of plasticsterm·key on the lower levels of the food web, dominated by marine microbes, remains largely unknown. Plasticterm·key can undergo natural abiotic degradation processes of chemical and mechanical nature, collectively called weatheringterm·dom, leading to its fragmentation. Moreover, biodegradation processes by microbes might also contribute to the degradation of plasticsterm·key 1 . Plasticterm·key litter on beaches and floating in seawater is exposed to solar radiationterm·key resulting in the formation of surface cracks and fragmentation into progressively smaller particles, ultimately reaching microscopic sizes 1 , 4 . Around 35,000 metric tons (MT) of all the plasticsterm·key floating in the ocean are smaller than 5 mm (microplasticsterm) 5 . Microplasticsterm can also be directly derived from personal care and cosmetic products or textile fibers entering the marine environment via wastewater discharge 6 . To increase their performance and durability, commercial plasticterm·key frequently contains additives. In contact with water, these additives can leach from the plasticterm·key into the surrounding water 7 . In this paper, the term leachateterm·dom is used for any compound (additive or organic substance derived from polymer degradation including any nano-particle that might break off from the plasticterm·key surfaces) released from the plasticterm·key into the seawater. It has been shown that photo-degradation of plasticterm·key can release submicron particles into the aquatic media 8 . With decreasing particle size, the surface to volume ratio of the particle increases, potentially increasing also the concentration of leachatesterm·dom in the surrounding water.

The oceanic DOCterm·dom pool is one of the largest reduced carbon pools on Earth (662 Pg C) 9 and it is of similar size as the atmospheric CO 2 (828 Pg C) 10 . Oceanic DOCterm·dom is mainly derived from phytoplankton and forms the basic substrate for the microbial food web fueling micro-heterotrophic growth 9 , 11 . In this study, we hypothesize that the estimated 250,000 MT of plasticsterm·key currently floating in the ocean 12 , 13 also contribute to the oceanic DOCterm·dom pool via leachingterm·key. If bioavailable these plastic-leachates might stimulate microbial carbon flux in the ocean, especially in the surface layer. Here we use the term bioavailable DOCterm·dom for the DOCterm·dom which is used by heterotrophicterm·key bacteria within days. In this regard, it is expected that particles of synthetic polymers are not bioavailable in the timescale of the experimentterm·mth but only the truly dissolved leachateterm·dom. Most of the marine studies on plasticsterm·key focus on the distribution of plasticterm·key size classes in the surface ocean 1 , 3 . The role of plasticsterm·key releasing DOCterm·dom into the ambient seawater, their biodegradation and effects on the marine biota, however, remain largely unknown.

In the present study, the potential contribution of DOCterm·dom leachingterm·key from microplasticsterm to the surface oceanic DOCterm·dom pool is estimated and the bioavailability of the plastic-derived DOCterm·dom determined. We calculate that up to 23,600 MT of DOCterm·dom is annually released by marine plasticterm·key litter. In highly contaminated areas, where plasticsterm·key can be found in concentrations up to 2500 g km −2 , the leached DOCterm·dom could make up to 10% of the DOCterm·dom in the surface microlayer (top 40 µm of the water column). About 60% of the DOCterm·dom leached from plasticsterm·key is bioavailable in less than 5 days. However, if exposed to solar radiationterm·key, this plastic-derived DOCterm·dom becomes less labile. Plasticterm·key leachingterm·key likely gives rise to local hot spots of DOCterm·dom that can be rapidly remineralized by marine microbes. However, we do not know the fate of the 40% that is not taken up rapidly by bacteria. Taken together, we conclude that increasing plasticterm·key waste entering the oceans will have consequences for the marine microbial food web and for the carbon cyclingterm·key.

Results DOCterm·dom leachingterm·key from plasticsterm·key Experimentsterm·mth were performed using commercially available low- and high-density polyethyleneterm·key (LDPEterm and HDPEterm) as well as polypropyleneterm·key (PP) and polyethyleneterm·key (PE) from supermarket packaging (see Methods section). PP and PE were used because they represent the most abundant plasticsterm·key found in the ocean 5 , 14 . The plasticsterm·key were added to autoclaved artificial seawater (ASW) and exposed to artificial solar radiationterm·key over a period of 6 days and/or 30 days. Dark treatments and controls without plasticsterm·key were also performed at the same time in all the experimentsterm·mth. DOCterm·dom was measured before and after the irradiation period from unfiltered samples. Therefore, in this study DOCterm·dom includes all the organic carbon compounds released by plasticterm·key, from truly dissolved substances to any polymeric nano-plastic that might have been also fragmented from the plasticterm·key surface. All plasticterm·key types used here leached DOCterm·dom into the ambient water whether they were irradiated or held in the dark. To calculate the total DOCterm·dom leachingterm·key from the plasticsterm·key, the DOCterm·dom concentration of the ASW prior to adding the plasticsterm·key was used as initial concentration (Table 1 ). Table 1 DOCterm·dom concentration leached from plasticsterm·key during the light and dark incubations Plasticterm·key type Incubation time (d) Treatment Total DOCterm·dom leached (µg cm −2 ) Instantaneous DOCterm·dom leached (µg cm −2 ) Incubation DOCterm·dom leached (µg cm −2 ) DOCterm·dom consumption during microbial incubation (µmol L −1 ) LDPEterm 6 Light 5.8 ± 0.51 3.50 ± 0.51 2.3 ± 0.07 74.0 ± 4.39 Dark 6.02 ± 1.75 2.81 ± 1.80 3.21 ± 0.41 94.6 ± 16.74 HDPEterm 6 Light 2.41 ± 0.94 2.03 ± 0.94 0.39 ± 0.05 53.7 ± 5.62 Dark 2.74* 1.56* 1.18 ± 0.5 104.1 ± 1.50 LDPEterm 30 Light 6.67 ± 0.90 0.67 ± 1.39 6.00 ± 1.06 92.1 ± 3.74 Dark 8.92 ± 0.29 8.13 ± 0.59 0.79 ± 0.52 150.4 ± 17.62 HDPEterm 30 Light 6.28 ± 0.96 5.16 ± 0.96 1.12 ± 0.03 136.3 ± 0.54 Dark 2.79 ± 0.34 2.42 ± 0.45 0.37 ± 0.29 69.6 ± 17.34 PE packaging 30 Light 0.31 ± 0.16 0.09 ± 0.17 0.22 ± 0.02 27.2 ± 1.16 Dark 0.26 ± 0.20 0.52 ± 0.39 −0.26 ± 0.33 30.9 ± 3.49 PP packaging 30 Light 2.17 ± 0.13 1.10 ± 0.42 1.07 ± 0.40 55.0 ± 1.77 Dark 1.61 ± 0.15 1.10 ± 0.20 0.51 ± 0.14 52.5 ± 2.25 Instantaneous DOCterm·dom leached is the DOCterm·dom released right after the addition of the plasticterm·key to the artificial seawater and prior to the exposure to the artificial solar radiationterm·key. Total DOCterm·dom leached: DOCterm·dom at the end of the irradiation period minus DOCterm·dom in artificial seawater before the plasticterm·key addition; Incubation DOCterm·dom leached: DOCterm·dom at the end of the irradiation period minus DOCterm·dom after the plasticterm·key addition and prior incubation. DOCterm·dom consumption during microbial incubation: DOCterm·dom at the end of the microbial experimentterm·mth minus DOCterm·dom at the initial time of the microbial experimentterm·mth. * only one sample of the three replicates. HDPEterm high-density polyethyleneterm·key, LDPEterm low-density polyethyleneterm·key, PE polyethyleneterm·key, PP polypropyleneterm·key LDPEterm and HDPEterm, either after 6 days or after 1 month of sterile incubation, leached between 2.4 and 8.9 μg C cm −2 of plasticterm·key surface (total DOCterm·dom leached; Table 1 ). Plasticsterm·key used as packing material released less DOCterm·dom amounting to 0.26–0.31 μg C cm −2 and 1.6–2.2 μg C cm −2 plasticterm·key surface for PE and PP, respectively. A time course experimentterm·mth conducted in the dark using LDPEterm added to artificial seawater confirmed that the highest leaching of DOCterm·dom from plasticsterm·key occurs initially, when plasticsterm·key get in contact with seawater (Fig. 1 ). Thereafter, DOCterm·dom concentration exponentially decreased (about 45%) until reaching a constant leachingterm·key of about 5.5 μg C cm −2 , roughly similar to the total DOCterm·dom leached in our irradiation experimentsterm·mth (Table 1 ). The first-order constant of DOCterm·dom lost was −0.0135 h −1 , i.e., half of the missing leached DOCterm·dom was lost in about 50 h (Fig. 1 ). Fig. 1 Loss of DOCterm·dom over time using LDPEterm. The original DOCterm·dom concentration of the artificial seawater before plasticterm·key addition was subtracted from all the points. Thus, the first point represents the maximum leachingterm·key after the plasticterm·key addition. Error bars show the standard deviation of the mean of three replicates. First order rate constant = −0.013 h −1 ± 0.003; R 2 = 0.97; p < 0.001. LDPEterm low-density polyethyleneterm·key, DOCterm·dom dissolved organic carbonterm·dom Accordingly, the effect of artificial solar radiationterm·key on the leaching of DOCterm·dom from plasticsterm·key floating in sterile seawater was also determined (Incubation DOCterm·dom leached; Table 1 ). The DOCterm·dom concentration immediately after adding the plasticsterm·key was used as initial DOCterm·dom concentration. The effect of artificial solar radiationterm·key varied depending on the plasticterm·key material and the duration of exposure. Over a 6-day incubation period under artificial solar radiationterm·key, we measured less DOCterm·dom leached from LDPEterm and HDPEterm (2.30 and 0.39 μg C cm −2 of plasticterm·key surface, respectively) than from their corresponding dark treatments (3.21 and 1.18 μg C cm −2 , respectively). For HDPEterm, however, the differences were not significant ( t test, p > 0.05). In contrast, over 1-month incubation period, irradiated LDPEterm and HDPEterm released significantly more DOCterm·dom ( t test, p < 0.05) (6 and 1.12 μg C cm −2 , respectively) than their corresponding dark treatments (0.79 and 0.37 μg C cm −2 , respectively). The packaging plasticterm·key (PE and PP) did not exhibit significant differences in the DOCterm·dom leachingterm·key between artificial solar radiationterm·key and the corresponding dark controls ( t test, p > 0.05).

Microbial utilization of DOCterm·dom leached from plasticsterm·key As plasticsterm·key floating at the sea surface release DOCterm·dom, we aimed at determining the bioavailability of plastic-derived DOCterm·dom and its influence on the microbial community. Therefore, after removing the plasticterm·key pieces from the seawater incubated in the previous experimentsterm·mth, we inoculated this water with a natural bacterial community from surface waters of the Adriatic Sea (nine parts of water with plastic-leachate: one part of inoculum). The samples were incubated in the dark until microbes reached stationary phase.

In the treatments previously containing plasticsterm·key (thereafter plasticterm·key treatments) and held in the dark, bacteria grew faster than in the previously irradiated plasticterm·key treatments or in the controls without plasticsterm·key (Fig. 2 ). In general, in the irradiated plasticterm·key samples, bacterial abundance reached higher values (Fig. 2 c, d) than in the irradiated control without plasticterm·key but in some cases, the differences were low or not significant ( t test, p < 0.05; Fig. 2a, b, e, f ). The higher bacterial abundance in the plasticterm·key treatments held in the dark regarding the irradiated and no-plastic treatments was also consistent with higher leucine incorporation rates (a proxy for biomass production) (Fig. 3 ). After 24 h of incubation, leucine incorporation in LDPEterm and HDPEterm held in the dark (0.26 nmol Leu L −1 h −1 and 0.34 nmol Leu L −1 h −1 , respectively) was at least one order of magnitude higher ( t test, p < 0.05) than in the irradiated treatments (0.75 × 10 −2 and 1.67 × 10 −2 nmol Leu L −1 h −1 , Fig. 3a ). This trend was also observed after 48 h of incubation (Fig. 3b ). Fig. 2 Bacterial abundance during incubation with plasticterm·key leachatesterm·dom. Bacterial abundance (cell mL − 1 ) during the incubation with DOCterm·dom leached from plasticsterm·key previously irradiated for 6 days using a LDPEterm and b HDPEterm; for 30 days using c LDPEterm and d HDPEterm; for 30 days using e PE and f PP. Error bars represent the standard deviation of the mean of the triplicate cultures. HDPEterm high-density polyethyleneterm·key, LDPEterm low-density polyethyleneterm·key, PE polyethyleneterm·key, PP polypropyleneterm·key Fig. 3 Leucine incorporation in the 30-days plasticterm·key leachingterm·key experimentsterm·mth. Leucine incorporation (nmol Leu L −1 h −1 ) for: LDPEterm and HDPEterm treatments after a 24 h and b 48 h of incubation; PE and PP packaging plasticterm·key after c 24 h and d 48 h of incubation. Error bars represent the standard deviation of the mean of the triplicate cultures. HDPEterm high-density polyethyleneterm·key LDPEterm low-density polyethyleneterm·key, PE polyethyleneterm·key, PP polypropyleneterm·key At the end of the bacterial incubation experimentsterm·mth, there were no significant differences ( p < 0.05) in bacterial abundance between plasticterm·key treatments and the controls without plasticsterm·key (Fig. 2 ). Also, there was no clear pattern in the maximum bacterial abundance reached among the different types of plasticsterm·key. All the treatments reached a maximum bacterial abundance of about 6×10 6 cells mL −1 . Bacteria started to grow slightly earlier in the LDPEterm and packaging PE and PP than in the HDPEterm treatments (Fig. 2 ). After 24 h of incubation, bacterial leucine incorporation was significantly higher ( p < 0.05) in the plasticterm·key treatments than in the non-plastic treatments, especially under dark conditions. When packaging plasticterm·key was used, leucine incorporation rates measured after 24 h of incubation were lower than in the other plasticterm·key types. Nevertheless, leucine incorporation in the dark plasticterm·key treatments (7.41 × 10 −2 and 0.12 × 10 −2 nmol Leu L −1 h −1 for PE and PP, respectively) was higher than in the corresponding irradiated (4.25 × 10 −3 and 4.97 × 10 −3 nmol Leu L −1 h −1 for PE and PP, respectively) treatments (Fig. 3c ). After 48 h of incubation, bacterial leucine incorporation in the dark treatments was significantly higher ( p < 0.05) than in the irradiated treatments. However, the dark control was not significantly different from the dark treatments containing plasticsterm·key (Fig. 3d ).

The bacterial growth efficiencyterm·mth (BGEterm·mth, as %) was calculated from the increase in bacterial carbon biomass and the concomitant decrease in DOCterm·dom in the incubations over time. An average BGEterm·mth of 12% ± 10% was obtained for plasticsterm·key (irradiated and dark treatments) and 19% ± 8% for the control treatments (irradiated and dark) without plasticsterm·key. No significant differences were found in the BGEterm·mth between treatment with and without plasticterm·key neither for irradiated and dark treatments. These BGEterm·mth values are comparable with those of coastal waters 15 .

A strong positive linear relationship between DOCterm·dom leachingterm·key and bacterial DOCterm·dom consumption ( R 2 = 0.87; p < 0.001; N = 46, all plasticterm·key experimentsterm·mth included, Model II regression 16 , Fig. 4 ) indicated that DOCterm·dom leachingterm·key from plasticsterm·key, either irradiated or held in the dark, is utilized by marine microbes. The regression slope indicates that 58% ± 3% of the leached DOCterm·dom is taken up by microbes at the time scale of the incubations. Fig. 4 Relation between DOCterm·dom consumption by microbes and DOCterm·dom leachingterm·key from plasticsterm·key. DOCterm·dom in µmol C L − 1 . Linear regression equation (Model II) using all data: Y = 0.58 (±0.03)⋅ X + 45.43 (±3.03) ( R 2 = 0.87; p < 0.001; N = 46). Open circles: irradiated samples; Gray circles: non-irradiated samples. DOCterm·dom dissolved organic carbonterm·dom Discussion All plasticterm·key types used here leached DOCterm·dom into seawater with a major fraction being released in the initial phase when plasticsterm·key first have contact with seawater (60% ± 26%; see Instantaneous DOCterm·dom leached in Table 1 ). About 45% of the leached DOCterm·dom is progressively lost during the first 200 h as shown in the kinetic experimentterm·mth conducted in the dark (Fig. 1 ). Since in this experimentterm·mth photo-oxidation cannot be the cause of the DOCterm·dom decline, the loss of DOCterm·dom could be due to sorption onto the plasticterm·key, a loss of volatile leachatesterm·dom or a combination of both. The sorption–desorption capacity of plasticsterm·key is well documented 17 ; thus a part of the DOCterm·dom might re-adsorb onto plasticsterm·key until an equilibrium is reached. The sorption of organic compounds onto plasticterm·key can be a fast process. Bakir et al. 18 found that some organic pollutants absorbed onto polyethyleneterm·key floating in seawater reaching an equilibrium within 24–48 h. However, other studies determined a longer period (>20 days) to reach equilibration for organic pollutants absorbing onto polyethyleneterm·key and polypropyleneterm·key 17 . In our study, the equilibrium was reached after about 200 h. That could include the sorption of the DOCterm·dom that was already present in seawater. However, in our study, this amounted to only 5% of the total DOCterm·dom before it started to decrease. So, if some sorption of the DOCterm·dom from seawater happened in our experimentsterm·mth, this was minimal compared to the DOCterm·dom leachateterm·dom. In nature, the sorption of natural DOCterm·dom onto plasticsterm·key could compete with microbial DOCterm·dom uptake disrupting the natural lower trophic level processes. Another explanation for the observed loss of DOCterm·dom might be the production of volatile organic compounds resulting from degradation of plasticsterm·key 19 . The strong plasticterm·key specific smell at the end of our incubation experimentsterm·mth (both in the irradiated and non-irradiated) suggests that some of the leached DOCterm·dom is volatile. In irradiated samples, in addition to the possible direct release of organic volatiles from the plasticsterm·key, solar radiationterm·key might have caused photo-degradation of the released non-volatile compounds resulting in CO, CO 2 , or organic volatiles not measurable with our method to determine DOCterm·dom. Possible photo-degradation products from plasticterm·key are propane, propene, ethane, ethylene, butane, and hexene 20 among others. The method for measuring DOCterm·dom applied in this study only accounts for the non-volatile fraction because the seawater samples are sparged with N 2 gas prior to the analysis to remove all the inorganic carbon which, however, also results in the loss of the volatile DOCterm·dom fraction (see Methods section). Thus, it is likely that plasticsterm·key exposed to solar radiationterm·key leach more DOCterm·dom than is measured with the approach used here.

It has been estimated that between 4.8 × 10 12 and 12.7 × 10 12 MT of plasticsterm·key entered the ocean in the year of 2010 21 . Assuming that floating plasticsterm·key have an average density of 0.96 g cm −3 (see Methods section) and given our range of DOCterm·dom leachingterm·key (0.26–8.92 µg C cm −2 ; Table 1 ), we estimate that between 260 and 23,600 MT of DOCterm·dom year −1 leach from those plasticsterm·key. Previous estimates indicate that 5.25 × 10 12 pieces of plasticterm·key particles weighing 268,940 MT are currently floating at the sea surface 5 , 12 , i.e., much less than the plasticterm·key entering the ocean annually 21 . This indicates that the majority of the plasticsterm·key entering the ocean is not floating at the surface. Missing plasticterm·key likely sunk in the water column and is buried in marine sediments or it is too small to be caught by the commonly used nets to collect marine plasticsterm·key (about 200 µm) 5 . In the subtropical gyres, plasticterm·key concentrations can be as high as 2500 g km −2 5 , 12 . By extrapolating our results to the concentration of plasticsterm·key in the ocean, we estimate that 2500 g km −2 of plasticsterm·key would leach between 0.14 and 4.65 g DOCterm·dom km −2 . Considering a 40 µm thick surface microlayer in which these plasticsterm·key are floating and leachingterm·key, 2500 g km −2 of plasticsterm·key leach between 0.28 and 9.68 µmol C L −1 within 5 days. DOCterm·dom concentrations of about 110 µmol L −1 have been measured in the surface microlayer of the oligotrophic gyre of the North Atlantic 22 . Thus, up to 10 ± 0.3% of the DOCterm·dom in the surface microlayer might originate from plasticsterm·key. This could create local hot spots of high DOCterm·dom concentrations stimulating microbial activityterm·dom.

Even if the plasticterm·key types used in our experimentsterm·mth are the most abundant plasticsterm·key found in the ocean (polyethyleneterm·key and polypropyleneterm·key) 14 , there is a high diversity of plasticterm·key polymers in the ocean. Other plasticterm·key types might release different amounts of DOCterm·dom than those examined here. Therefore, the values presented here are only tentative estimates. Aging and degradation of plasticterm·key causes its breakdown into small pieces 23 , a mechanism, which could potentially increase DOCterm·dom leachingterm·key due to the increase in surface area relative to its volume. The plasticsterm·key used in our experimentsterm·mth did not show any signs of aging to the naked eye after the exposure to artificial solar radiationterm·key. Moreover, since the missing plasticterm·key mentioned above is not taken into account here, our estimates of DOCterm·dom leachingterm·key from plasticsterm·key are likely very conservative.

Plasticsterm·key could also contribute to the DOCterm·dom measurements through their direct presence in the samples. Nanoplastics (<100 nm) are either derived from fragmentation of larger particles or directly from products for personal care, cosmetics, or textile fibers 6 , 24 , 25 . A recent study showed that plasticterm·key particles smaller than 2 µm are generated upon exposure of polystyrene to solar radiationterm·key 8 . A high number of these polymer particles were found to be smaller than 800 nm 8 . In our experimentsterm·mth, any nanoparticle formed would be included in the DOCterm·dom pool since we did not filter the sample before and after the irradiation experimentterm·mth. However, we did not find significant differences in DOCterm·dom leachingterm·key between irradiated and dark samples in most of the experimentterm·mth, indicating that sub-micrometer particles from photo-degradation were not generated to a significant extent in our experimentsterm·mth. DOCterm·dom is operationally defined as the fraction of organic matterterm·dom passing through a 0.2–0.7 µm filter 26 . Therefore, any particle smaller than 700 nm would pass the most commonly used filters for DOCterm·dom analysis (Whatman GF/F, 0.7 µm pore size) and consequently, will contribute to the measured DOCterm·dom. Thus, special care should be taken when collecting DOCterm·dom samples in plasticterm·key contaminated areas.

Leaching of DOCterm·dom from plasticsterm·key influences the microbial activityterm·dom and carbon cyclingterm·key in the ocean. The bioavailable fraction of the leached DOCterm·dom was (insignificantly) higher when the plasticterm·key was kept in the dark than under artificial solar radiationterm·key (61% ± 3% in the dark vs. 55% ± 5% in the light treatments). Even in the treatments with only low levels of DOCterm·dom leachingterm·key (e.g., PE packaging plasticterm·key), dark conditions stimulated the response of the bacterial community, as indicated by the DOCterm·dom utilization and by an earlier increase in microbial abundance in the plasticterm·key incubations than in the controls without plasticterm·key. Even where no differences in the DOCterm·dom leachingterm·key between the light and dark treatments were observed, microbial growth started earlier in the dark treatments containing plasticsterm·key indicating that irradiated DOCterm·dom derived from plasticsterm·key might undergo photochemical transformation affecting microbial growth. Photodegradation of plasticsterm·dom is known to produce free radicalsterm·dom 20 potentially inhibiting bacterial growth 27 . Therefore, the lower bacterial abundance and leucine incorporation in the treatments exposed to artificial solar radiationterm·key might be due to the formation of microbial inhibitors. The large amount of plasticsterm·key potentially present in the deeper water layers or in sediments 5 is not exposed to solar radiationterm·key. Thus, plastic-derived DOCterm·dom could also potentially enhance microbial growth in layers well below the oceanic surface layer. Our leachingterm·key experimentsterm·mth were performed under abiotic conditions. In the ocean, however, microbes will have access to the plastic-derived DOCterm·dom immediately upon its release, resulting in ~45% more DOCterm·dom, potentially available, than the estimates given in Table 1 (Total leached DOCterm·dom). Thus, re-sorption of leached DOCterm·dom onto the plasticsterm·key would be largely prevented if microbes would efficiently take up the DOCterm·dom as soon as it is released from the plasticsterm·key.

DOCterm·dom leachingterm·key from plasticsterm·key and its effect on microbial activityterm·dom might be important in areas with high plasticterm·key concentrations (e.g., subtropical gyres or near shore waters), especially in the surface microlayer. The quantity of plasticterm·key waste entering the oceans is predicted to increase by up to one order of magnitude by the year 2025 (20) resulting in plastic-derived DOCterm·dom of up to 236,000 MT per year in the global ocean, with potential major consequences for marine microbes and the carbon cyclingterm·key in the oceanic system.

Methods Plasticterm·key leachingterm·key experimentsterm·mth A total of six experimentsterm·mth were performed using four different types of plasticsterm·key (low density polyethyleneterm (LDPEterm), high density polyethyleneterm (HDPEterm), polyethyleneterm·key (PE), and polypropyleneterm·key (PP)) and two exposure times with artificial solar radiationterm·key, 6 days and 30 days. An additional experimentterm·mth to study the kinetics of DOCterm·dom leachingterm·key from plasticsterm·key was performed in the dark using LDPEterm.

Artificial photosynthetic active radiationterm·dom (PARterm·dom) was provided by a HQI-T Powerstar lamp (250 W, Osram), UV-A radiation by 2 Philips TL100W/10 R fluorescent tubes, and UV-B radiation by 2 UVA-340 fluorescent lamps (Q-Panel Company, UK). The radiation intensity for each wavelength or wavelength range was as follows: PARterm·dom (400–700 nm), 700 µmol m −2 s −1 ; 380 nm, 28.47 µW cm −2 nm −1 ; 340 nm, 16.31 µW cm −2 nm −1 ; 320 nm, 7.95 µW cm −2 nm −1 ; 305 nm, 1.09 µW cm −2 nm −1 . The radiation dose rate represents the solar radiationterm·key in the subtropical North Atlantic Gyre measured at noon at 15 m depth 28 . Artificial solar radiationterm·key was measured at 305, 320, 340, 380 nm, and PARterm·dom with a Biospherical PUV-510 radiometer using a correction factor for the 305 nm channel as suggested by Kirk 29 . The light treatments received continuous artificial solar radiationterm·key.

Polyethyleneterm·key plasticterm·key (LDPEterm and HDPEterm) was obtained from GoodFellow as plasticterm·key film (LDPEterm, 0.5 mm thickness) and pellets (HDPEterm, ~4 mm diameter). PE from fruit bags and PP from fruit packaging, both translucent, were obtained from a local supermarket in Vienna (Austria). LDPEterm, PE, and PP were cut into ~7 mm squares. Plasticterm·key pieces (63 squares for LDPEterm, PE and PP, and 123 pellets for HDPEterm experimentsterm·mth) were added to 250 mL of sterilized (autoclaved at 121 °C for 20 min) artificial seawater in quartz tubes for the treatments receiving artificial solar radiationterm·key. Dark controls were established in the same way but using borosilicate bottles wrapped in aluminum foil. Light and dark controls without plasticsterm·key were also performed in the same way as the samples with plasticsterm·key. All the treatments and controls were performed in triplicate. Light and dark samples were placed in the solar simulator under the radiation conditions described above during 6 days (LDPEterm and HDPEterm) and 30 days (LDPEterm, HDPEterm, PE and PP). A flow-through water bath maintained the temperature of the incubation flasks at 23 °C.

DOCterm·dom was measured from the unfiltered artificial seawater before and after the incubation period to estimate the DOCterm·dom leachingterm·key from the plasticterm·key material and the effect of the radiation on it. DOCterm·dom released was normalized to the surface area of the whole plasticterm·key used in every flask. Bacterial abundance was measured at the end of the experimentterm·mth to check whether bacterial growth took place in the plasticterm·key leachingterm·key experimentsterm·mth. However, bacterial abundance in these experimentsterm·mth was always below 10 4 mL −1 .

Bioavailability of DOCterm·dom leachingterm·key from plasticsterm·key Following the exposure to artificial solar radiationterm·key, plasticterm·key pieces were removed from the incubation flasks and 0.8 µm filtered surface seawater from the North Adriatic Sea was added to the incubation water at a ratio of 9:1 (incubation water: 0.8 µm filtered). The treatments were amended with NH 4 Cl and NaH 2 PO 4 to a final concentration of 10 and 2 µm L −1 , respectively, to avoid growth limitation by either nitrogen or phosphorus availability. The flasks were incubated in the dark at 23 °C until the mixed microbial community reached stationary phase. Samples for microbial abundance were collected every 6–12 h for the LDPEterm and HDPEterm experimentsterm·mth and daily for the PE and PP experimentsterm·mth. DOCterm·dom samples were collected at the beginning of the incubation and at the end of the exponential growth phase of the microbial community. At the end of this incubation experimentterm·mth, DOCterm·dom samples were filtered through combusted Whatman GF/F filters to remove bacteria. In the 30 days experimentsterm·mth with different types of plasticsterm·key (LDPEterm, HDPEterm, PE, and PP), heterotrophicterm·key microbial production was measured 24 and 48 h after inoculating the microbial community as described below.

An additional experimentterm·mth was performed to study the kinetics of the DOCterm·dom leachingterm·key from plasticsterm·key in which 63 LDPEterm squares were added to 300 mL of artificial seawater. The samples were kept in the dark at 23 °C and DOCterm·dom samples were collected at different times over a 2-weeks period, with a higher frequency in the initial phase of the kinetic study.

DOCterm·dom and microbial analysis DOCterm·dom samples were collected unfiltered for the initial and final time of the irradiation experimentterm·mth as well as the kinetic experimentterm·mth. However, they were filtered at the end of the microbial incubation experimentterm·mth (Whatman GF/F filters pre-combusted at 450 °C for 4 h) to remove bacteria. Water for DOCterm·dom analyses was collected in pre-combusted 20 mL glass vials, acidified with concentrated HCl to pH < 2 and stored at 4 °C until analysis. DOCterm·dom was measured with a Shimadzu TOC-V organic carbon analyzer 30 after removal of CO 2 by vigorous sparging with high purity N 2 . Therefore, only the non-volatile fraction of DOCterm·dom is measured. The accuracy was tested daily with the DOCterm·dom reference materials provided by D.A. Hansell (University of Miami). We obtained average concentrations of 43.3 ± 0.3 μmol L −1 for the deep ocean reference (Batch 16—2016) minus blank reference materials. The nominal DOCterm·dom value provided by the reference laboratory was 43–45 μmol L −1 .

The global budget of DOCterm·dom leachingterm·key from plasticterm·key in the ocean was calculated assuming a plasticterm·key density of 0.96 g cm −3 , a plasticterm·key thickness of 0.1 mm and the two sides of the plasticterm·key film.

Bacterial abundance was measured collecting aliquots of 1.5–2 mL from each bottle, fixed with 0.5% glutaraldehyde (final concentration, Sigma-Aldrich), stored in the dark for 10 min and subsequently frozen at −80 °C until further processing. Bacterial abundance was measured by flow cytometry (BD FACSAria IIu) following the method described elsewhere 31 .

Leucine incorporation rates of the microbial communities were measured by adding 20 nmol L −1 [ 3 H]-leucine (final concentration, specific activity 120 Ci mmol L −1 ) to triplicate 1.5 mL samples. Duplicate TCA (trichloroacetic acid)-killed blanks (5% final concentration) were treated in the same way as the samples 32 . Samples and blanks were incubated in the dark at 20 °C for 2 h. Incubations were terminated by adding TCA (5% final concentration) to the samples. Bacterial proteins were precipitated by two successive centrifugation steps (12,000 × g for 10 min), including a washing step with 1 mL of 5% TCA. The samples were air-dried before adding 1 mL of liquid scintillation cocktail. After 24 h, the radioactivity was determined in a scintillation counter (Packard Tri-Carb 1600TR). The disintegrations per minute (DPMs) of the blanks were subtracted from the mean DPMs of the respective samples and the resulting DPMs converted into leucine incorporation rates.

Data availability The authors declare that the data supporting the findings of this study are available within the article.