Fast and Facile Biodegradation of Polystyrene by the Gut Microbial Flora of Plesiophthalmus davidis Larvae

PS is widely produced in the modern world, but it is robust against biodegradation. A few studies reported the biodegradation of PS, but most of them merely observed its weight loss; fewer were able to find its chemical modifications, which are rather direct evidence of biodegradation, by using limited organisms. Therefore, it is required to find an effective way to decompose PS using various kinds of organisms. Herein, we discovered a new PS-degrading insect species and bacterial strain, and we found that the genus that includes the PS-degrading bacterial strain occurs in significant amounts in the larval gut flora, and the proportion of this genus increased as the larvae were fed Styrofoam. Our research offers a wider selection of PS-degrading insects and the possibility of using a certain mixture of bacteria that resemble the gut flora of a PS-degrading insect to biodegrade PS, and thus could contribute to solving the global plastic crisis.

ABSTRACT

Polystyrene (PS), which accounts for a significant fraction of plastic wastes, is difficult to biodegrade due to its unique molecular structure. Therefore, biodegradation and chemical modification of PS are limited. In this study, we report PS biodegradation by the larvae of the darkling beetle Plesiophthalmus davidis (Coleoptera: Tenebrionidae). In 14 days, P. davidis ingested 34.27 ± 4.04 mg of Styrofoam (PS foam) per larva and survived by feeding only on Styrofoam. Fourier transform infrared spectroscopy confirmed that the ingested Styrofoam was oxidized. Gel permeation chromatography analysis indicated the decrease in average molecular weight of the residual PS in the frass compared with the feed Styrofoam. When the extracted gut flora was cultured for 20 days with PS films, biofilm and cavities were observed by scanning electron microscopy and atomic force microscopy. X-ray photoelectron spectroscopy (XPS) studies revealed that C-O bonding was introduced into the biodegraded PS film. Serratia sp. strain WSW (KCTC 82146), which was isolated from the gut flora, also formed a biofilm and cavities on the PS film in 20 days, but its degradation was less prominent than the gut flora. XPS confirmed that C-O and C=O bonds were introduced into the biodegraded PS film by Serratia sp. WSW. Microbial community analysis revealed that Serratia was in the gut flora in significant amounts and increased sixfold when the larvae were fed Styrofoam for 2 weeks. This suggests that P. davidis larvae and its gut bacteria could be used to chemically modify and rapidly degrade PS.

IMPORTANCE PS is widely produced in the modern world, but it is robust against biodegradation. A few studies reported the biodegradation of PS, but most of them merely observed its weight loss; fewer were able to find its chemical modifications, which are rather direct evidence of biodegradation, by using limited organisms. Therefore, it is required to find an effective way to decompose PS using various kinds of organisms. Herein, we discovered a new PS-degrading insect species and bacterial strain, and we found that the genus that includes the PS-degrading bacterial strain occurs in significant amounts in the larval gut flora, and the proportion of this genus increased as the larvae were fed Styrofoam. Our research offers a wider selection of PS-degrading insects and the possibility of using a certain mixture of bacteria that resemble the gut flora of a PS-degrading insect to biodegrade PS, and thus could contribute to solving the global plastic crisis.

INTRODUCTION

Plastic wastes have been a dangerous threat to human society and the environment since the invention of plastic materials. On a global scale, while approximately 8.3 billion metric tons of plastics were produced through 2017, less than 9% of the produced plastics had been recycled (1). In 2018 alone, 359 million metric tons of plastic was produced globally (2). A notable example of such threatening plastics is polystyrene (PS), which accounted for ∼6% of the world plastic production in 2017 (1). Currently, these plastic wastes are discarded by environmentally harmful methods, such as incineration, burial in landfills, and dumping into the ocean (3, 4). Once discarded, the plastic waste accumulates in the ecosystem due to its long decomposition time, and thus causes detrimental effect to both the environment and human society (5). Biodegradation, which is a sustainable method, is therefore considered an alternative for the disposal of plastic wastes.

PS has a linear carbon backbone, and alternating backbone atoms are attached to phenyl moieties. Because this unique structure makes the biodegradation of PS difficult, there have been limited reports on the biodegradation of PS (6–19). Certain microorganisms are capable of forming a biofilm on PS films, and a fraction of them could partially change the chemical properties (6–14). However, there were only a few reports, to the best of our knowledge, on the direct decomposition of PS. For example, a PS-ingesting yellow mealworm (larvae of Tenebrio molitor) mineralized PS as its energy source (15), and its gut bacterium Exiguobacterium sp. strain YT2 was able to degrade PS (16). Moreover, a PS-ingesting dark mealworm (larvae of Tenebrio obscurus) alongside with a PS-ingesting superworm (larvae of Zophobas auratus) and its gut bacterium Pseudomonas sp. strain DSM 50071 were also capable of degrading PS (17–19). During an incubation period of 28 days, approximately 5 × 10⁷ cells of Exiguobacterium sp. YT2 could change the surface properties of PS films by introducing C-O bonds, as confirmed by X-ray photoelectron spectroscopy (XPS) (16). However, the chemical conversion rate of PS was not prominent (∼9%). In addition, the PS-degrading insect was limited to the species that was previously known to ingest PS, even though the researchers imply that the PS-degrading ability is not limited to only mealworms (15, 17, 20, 21). Therefore, a more effective way to decompose PS by using more kinds of organisms is needed to achieve the practical biodegradation of PS.

Plesiophthalmus davidis is a species of darkling beetle from the family Tenebrionidae. They mainly inhabit the mixed forest of northeastern Asia, including the northern and middle regions of China and the Korean peninsula (see Fig. S1 in the supplemental material) (22). The larvae and adults of P. davidis are known to feed on rotten wood (22). PS-ingesting mealworms feed on various organic materials, including rotten wood in their natural habitat (23), and this is similar to the diet of P. davidis. Thus, the similarity of the diet might indicate the possibility of PS biodegradability in this P. davidis species.

In the present work, we report the biodegradation of PS by P. davidis larvae and their gut bacteria, which were not previously known to ingest PS. Such rapid degradation can be attributed to the interaction between bacteria that play a key role in plastic biodegradation and other gut microorganisms. This report features facile and prominent chemical conversion by introducing C-O and C=O bonds to PS.

RESULTS

When the larvae of P. davidis in their last stage or a molt before the last stage were kept with blocks of PS foam (Styrofoam), they ingested the Styrofoam and survived (Fig. 1A). To confirm the biodegradation of PS, larvae were fed Styrofoam as their sole diet for 14 days. As a result, 34.27 ± 4.04 mg of Styrofoam was degraded per larva (n = 7), and this was significantly greater than the control group (2.80 ± 2.15 mg, n = 4; P < 0.001 by Student’s t test). The mortality rate was zero within the inspection period. Furthermore, during a longer incubation period of 4 weeks, the mortality rate of the PS-fed larvae was zero, while the negative-control group had a higher mortality rate, 11.11% ± 19.25%.

FIG 1.

(A) Boring into commercial Styrofoam by P. davidis larvae. (B) FT-IR transmittance spectra of the control Styrofoam and the frass from 850 cm⁻¹ to 1,800 cm⁻¹. (C) FT-IR absorbance spectra of the control Styrofoam and the frass from 2,500 cm⁻¹ to 3,500 cm⁻¹. (D) MWD of the control Styrofoam and the extracted frass.

The frass of Styrofoam-fed larvae was analyzed with Fourier transform infrared spectroscopy (FT-IR) to evaluate the possible chemical modifications of PS. These results showed that new bands near 1,670 cm⁻¹ and 3,285 cm⁻¹ evolved in the frass compared with those seen in the control Styrofoam (Fig. 1B and C).

The residual PS from the frass was extracted with tetrahydrofuran (THF). The extractable percentage of the frass was 41.41% ± 0.13% (n = 3), while the control feed Styrofoam was 100%. Gel permeation chromatography (GPC) analysis revealed the decrease of weight-average molecular weight (M_w) and number-average molecular weight (M_n) (M_w = 147,254 ± 1,430, M_n = 77,161 ± 2,223; n = 3) and negative shift of molecular weight distribution (MWD) of the THF-extracted frass compared to the feed Styrofoam (M_w = 152,339 ± 1,430, M_n = 80,515 ± 2,586; n = 3) (Fig. 1D). Moreover, the GPC chromatogram revealed there was a lower molecular weight peak that has evolved from the extracted frass (see Fig. S2 in the supplemental material).

To confirm whether the gut flora was responsible for the chemical modifications and degradation of PS, scanning electron microscopy (SEM) was used to directly observe the surface morphology of the film. Pristine PS film was first incubated for 20 days in a liquid-phase carbon-free medium with the gut flora of P. davidis larvae. The SEM image of the treated film showed numerous cavities, and there were also biofilms formed by the bacteria in the cavities, while the control film did not have any of these structures (Fig. 2A and B; see also Fig. S3 and S4). Moreover, when the biofilm covering the biodegraded PS film was removed, the surface morphology was significantly different from that of the control film (Fig. 2C and D with atomic force microscopy [AFM] analyses and Fig. S5A and B with SEM analyses).

FIG 2.

(A and B) SEM images of the negative-control PS film (A) and the PS film biodegraded by the gut flora (B). (C and D) AFM images of the negative-control PS film (C) and the PS film biodegraded by the gut flora after removal of the biofilm (D).

Water contact angle (WCA) measurements showed that the WCA of PS film incubated with the gut flora significantly decreased from 94.5° ± 0.8° (n = 3, the control group) to 80.0° ± 2.5° (n = 3; P < 0.005 by Student’s t test) (Fig. 3A and B). Moreover, XPS confirmed the chemical modifications of the PS film by biodegradation. While the C-O peak was not found in the control group (Fig. 3C), the PS film incubated with the gut flora had a significant portion of C-O bonds with a relative peak area of 23.8% (Fig. 3D); this observation matches that of the FT-IR spectrum (Fig. 1B).

FIG 3.

(A and B) WCA values of the negative-control PS film (n = 3) (A) and the PS film biodegraded by the gut flora (n = 3) (B). (C and D) XPS C1s spectra of the negative-control PS film (C) and the PS film biodegraded by the gut flora (D).

The bacterium that could biodegrade PS was isolated within the gut flora. We identified this bacterium as a Serratia species and named it Serratia sp. strain WSW. The Serratia sp. WSW strain formed a biofilm on PS during a 20-day incubation period on a solid carbon-free medium compared to the control PS film as confirmed by SEM (Fig. 4A and B). The obtained partial 16S rRNA sequence from Serratia sp. WSW had 99.65% similarity to the bacterium Serratia plymuthica strain 31upmr (Fig. S6). The partial 16S rRNA sequence of Serratia sp. WSW was deposited in GenBank under accession number MN097873.1, and the type strain was deposited to Korean Collection for Type Cultures (KCTC) under the accession number KCTC 82146. The cells of Serratia sp. WSW were Gram negative and rod shaped with a width of 0.3 to 0.6 μm and length of 0.9 to 2.8 μm (Fig. S7). When Serratia sp. WSW was cultured in Luria-Bertani (LB) medium, it had a doubling time of 1.50 ± 0.32 h and grew optimally at both 24°C and 37°C. The Serratia sp. WSW had an extracellular polymeric substance (Fig. 4B and Fig. S8, yellow arrows) and cell appendages (Fig. 4B and Fig. S8, white arrows) attached to the PS film. When Serratia sp. WSW was removed from the PS film, topographic changes were observed by comparing it with the control PS film (Fig. 4C and D with SEM analyses), but the changes were not as drastic as for the gut flora (Fig. S5B). WCA measurement revealed that the control group had a WCA of 90.7° ± 0.4° (n = 3) (Fig. 5A) , which was significantly larger than that of the PS film incubated with Serratia sp. WSW, 76.0° ± 3.4° (n = 3; P < 0.001 by Student’s t test) (Fig. 5B). Moreover, XPS results showed that Serratia sp. WSW introduced both C-O and C=O bonds to PS (Fig. 5C and D).

FIG 4.

(A and B) SEM images of the negative-control PS film (A) and the PS film biodegraded by the isolated bacterium Serratia sp. WSW (B). The yellow arrows indicate the extracellular polymeric substance, and the white arrows indicate the cell appendages of Serratia sp. WSW attached to the PS film surface. (C and D) SEM images of the negative-control PS film (C) and the PS film biodegraded by the isolated bacterium Serratia sp. WSW after removal of the biofilm (D). To the left of the images, X indicates that the biofilm was not removed and O indicates that the biofilm was removed prior to the SEM imaging of the sample.

FIG 5.

(A and B) WCA values of the negative-control PS film (n = 3) (A) and the PS film biodegraded by the isolated bacterium Serratia sp. WSW (n = 3) (B). (C and D) XPS C1s spectra of the negative-control PS film (C) and the PS film biodegraded by the isolated bacterium, Serratia sp. WSW (D).

Microbial community analysis revealed that the gut floral microbiota community shifted as the wild-type larvae were fed PS for 2 weeks. The gut florae of the wild-type larvae and the PS-fed larvae had 108,963 and 106,726 reads, respectively. The percentile of Q30 (sequencing quality score of 30) was more than 95%, and the Good’s coverage was higher than 99% in both samples. The gut flora of the wild-type larvae had five operational taxonomic units (OTUs) of the genera Lactococcus, Aquabacterium, Buttiauxella, Raoultella, and Serratia, while the PS-fed larvae had one additional OTU, of the genus Enterococcus (Fig. 6). The genera Serratia and Lactococcus increased 6- and 10-fold, respectively, when the larvae were fed PS. In particular, the genus Serratia made up 33% of the gut flora of the PS-fed larvae (Fig. 6). Moreover, a representative sequence of Serratia sp. had 99.35% identity and 0 gaps compared with Serratia sp. WSW.

FIG 6.

Gut flora microbial community analysis of wild-type (WT) and Styrofoam-fed (PS-fed) P. davidis larvae. The species listed are the species that have high 16S rRNA sequence similarities with the representative sequence of each OTU.

DISCUSSION

P. davidis is a wood-boring beetle species, and so far, it had not been known to ingest PS to survive. In this study, the potential PS degradability of P. davidis larvae was evaluated. The results pointed out the obvious ability of P. davidis larvae and their gut flora to degrade PS; however, there are more points that could be deduced with the results.

PS degradation by P. davidis larvae was observed in the form of weight loss (Fig. 1B). The larvae were starved or prefed Styrofoam before the degradation experiment; therefore, the Styrofoam weight loss was probably not due to ingested and remaining Styrofoam in the gut of the larvae. The degraded Styrofoam is likely to have been converted to CO₂ or the biomass in the larvae, for the frass was also weighed with Styrofoam after ingestion. Moreover, when Styrofoam was given as the sole diet for the larvae for 4 weeks, all larvae survived, while the mortality rate of the control group (no diet) was high. Thus, it seems plausible that the larvae could use Styrofoam as their energy source.

The FT-IR spectra of the frass of Styrofoam-fed larvae had a relatively broad band at 1,670 cm⁻¹ that can be assigned to the carbonyl bond conjugated to the aromatic ring in PS (Fig. 1B) (24). The weak and broad band near 3,285 cm⁻¹ is attributed to a hydroxyl group (Fig. 1C). The presence of these moieties implies that the PS has been oxidized prior to or during biodegradation, possibly leading to the final conversion to CO₂ that can serve as an energy source for the larvae.

The THF-extractable percentage is one of the indirect indicators of PS degradability (25). The extractable percentage of the frass was significantly lower than for the control feed Styrofoam (Student’s t test, P < 0.001), indicating that the PS was degraded and converted into compounds that are not extractable by THF in the digestion process. Because there was no cannibalism in the experimental period, the digestion rate could be higher than 50%. Moreover, GPC results showed that the M_w and M_n decreased in the extracted frass compared to the feed Styrofoam, while the polydispersity index (PDI) was basically unchanged (1.89 for feed Styrofoam and 1.91 for frass). The PDI value and the negative shift of the MWD of frass compared to the feed Styrofoam suggest that the biodegradation process might occur evenly on different-sized chains. Compared to previous works using yellow and dark mealworms, the change in the PDI has a different tendency. While the PDI of the biodegraded PS from the yellow and dark mealworms drastically decreases (3.22 for feed Styrofoam,  2.56 for yellow mealworm, and 1.78 for dark mealworm) (17), the PDI of the biodegraded PS by P. davidis was almost the same as the PDI for the feed Styrofoam. This suggests that the degradation mechanism could be different according to PS-degrading insects. However, the P value by Student’s t test was lower than 0.05 only in the M_w results (the M_n difference had a P value of 0.08 by Student’s t test). This could be explained by observing the GPC chromatograms of the extracted frass and the feed Styrofoam. The GPC chromatogram of the extracted frass had a new fraction (see Fig. S2, fraction B, in the supplemental material) compared to the single fraction in the feed Styrofoam (Fig. S2, fraction A). Because the GPC was calibrated by using 800- to 400,000-Da-sized PS polymers, fraction B, which was smaller than 800 Da, could not be detected by the GPC software. By using the calibration plot, the molecular weight of fraction B was calculated to be around 500 to 600 Da, which was the weight of ∼5- to 6-mer PS. Because the smaller fraction was not added in the total molecular weight calculation, the decreases in M_w and M_n were not drastic. Moreover, fraction B indicates that the biodegradation of PS could occur in both terminal and nonterminal styrene.

This kind of molecular weight distribution of the frass could be explained if the degradation takes place only on the surface of the Styrofoam. When the bulk Styrofoam enters the digestive system of the larvae, it would first go through the masticatory process, becoming smaller pieces. Afterwards, the smaller Styrofoam pieces would go under biological and chemical degradation processes in the gut. If the smaller pieces do not degrade to significantly small or porous particles in the masticatory process so the degrading component could equally reach all of the polymers composing the particle, the degradation process would mostly take place on the surface of the piece. Therefore, if the frass is composed of the surface partially degraded Styrofoam, the GPC chromatogram would have nearly the same fraction as the feed Styrofoam and a smaller fraction of the degradation intermediate (oligomer).

The gut flora was able to form numerous cavities on the PS surface in 20 days (Fig. 2B and D and Fig. S3, S4, and S5B). Unknown PS oxidative enzymes (e.g., lignin peroxidase), which might be excreted by the bacteria, could create these cavities. The presence of the cavities on the surface as observed by SEM suggests that the excreted degradation enzyme(s) first oxidizes and leaves small holes by degradation on the PS film surface, making it more habitable for the bacteria (Fig. S4A). Then, the settled bacteria would accelerate the degradation process by forming a biofilm, resulting in a large cavity (Fig. S4B to D).

The XPS results show that C-O bonds were introduced to the PS film incubated with the gut flora (Fig. 3D). Because the penetration depth of the XPS is a few nanometers, these new C-O bonds cover the surfaces of the cavities. The cavities, where biodegradation is likely to take place, could have been oxidized by a degradation enzyme(s), leading to the depolymerization and utilization of PS as a carbon source. The increase in the hydrophilicity confirmed by WCA measurements (Fig. 3A and B) can be attributed to the increase in the roughness and formation of polar moieties such as C-O or O-H groups at the surface. The increased hydrophilicity can facilitate the biodegradation of the PS film (8, 16). Despite the small amount of bacteria, the chemical conversion of the PS film was 23.8%, based on the relative area of C-O bonds in C1s XPS (16).

Compared to the gut flora, Serratia sp. WSW made few topographical changes to the PS film (Fig. 4D and Fig. S5B). This can be attributed to the cooperative role of the bacteria in the gut flora. Certain microbial consortia are known to interact with other species to achieve such specific goals as survival or synthesis of novel molecules (26–29). The interaction could be in the form of sharing the metabolic pathway between the species and compartmentalizing the pathway (30). This is also reported to happen in biodegradation processes (31). While the oxidation and depolymerization of PS are attributed to Serratia sp. WSW, another microbial community(ies) in the gut flora could assist or accelerate the process by supplying factors required for the process or by utilizing the lower degradants more efficiently. Therefore, metabolic compartmentalization could occur and lead to an increased degradation efficiency in the gut flora compared to Serratia sp. WSW. In the same sense, ruling out any microbial community(ies) that participate in the processing of C=O bonding would facilitate the direct detection of C=O bonds in PS film incubated with Serratia sp. WSW (Fig. 5D).

Serratia sp. WSW had extracellular polymeric substances and cell appendages attached to the PS film (Fig. 4B and Fig. S8), which might help the cells attach to the surface and deliver degradation enzyme(s). Intriguingly, these structures were not found in the gut floral group (Fig. 2B and Fig. S3D and S4), which might be due to the difference in the degradation mechanisms between the two groups. If the gut flora could compartmentalize the degradation metabolism process, degradation can occur without having Serratia sp. WSW attached to the PS film surface. Therefore, the biofilm formed on the surface of the PS film in the gut floral group may be a different species from Serratia sp. WSW. The difference in the morphologies other than the extracellular polymeric substance and cell appendages can be explained by the same reason. These differences might also be derived from the different culture methods of the two groups.

Microbial community analysis revealed the clear shift in the gut flora microbial communities when the larvae were fed PS. The Q30 percentile and the Good’s coverage indicate that the extracted DNAs were high quality and that most of the reads were detected (32). Moreover, the arcing of the rarefaction curve near 3,000 sequences per sample indicates that there are enough amplified sequences for sequencing (Fig. S9) (33). Two of the genera found by the representative 16S rRNA sequence of each microbial community are known to be lignin biodegradable species (Raoultella [34, 35] and Serratia [36, 37]), and one genus includes a species that is suspected to be ligninolytic (Enterococcus [37]). The 16S rRNA sequence similarity between the representative sequence of Serratia sp. OTU and Serratia sp. WSW implies that Serratia sp. WSW could be in the Serratia sp. OTU. The relative abundances of Serratia sp. and Lactococcus sp. in the gut flora increased 6- and 11-fold, respectively, when the larvae were fed PS for 2 weeks. This indicates that Serratia sp. might play a central role in PS biodegradation in the gut flora and Lactococcus sp. could be the main assisting microbial community. To the best of our knowledge, there were no reports of detection of the isolated plastic-biodegrading bacterial species from the gut floral OTUs of the host organism. The simplicity of the gut flora and the presence of plastic-degrading strain in detectable amounts in the gut flora of P. davidis larvae could help the construction of an effective microbial consortium for PS biodegradation.

Such PS degradability of P. davidis can be explained by their diet. It is known that a similar group of wood-boring insects has the gut flora capable of utilizing lignin or cellulose as a carbon source (38). Because the structure of PS is similar to lignin or cellulose, it can be degraded and utilized as an energy source by the gut flora of wood-boring insects. Overall, these findings imply that the gut flora as a whole is more efficient at degrading PS than a single bacterium and that the ubiquity of PS degradability within wood-boring insects can lead to a new era in the plastic crisis around the world. Further studies are needed to define the cooperative role of multiple microorganisms in PS biodegradation and the full biochemical pathway of PS biodegradation.

MATERIALS AND METHODS

Collection of P. davidis larvae.

The hibernating larvae of P. davidis were collected from 35°59'36.8"N 129°18'22.8"E (Pohang, Republic of Korea), 36°02'45.9"N 129°21'42.4"E (Pohang), and 33°58'54.3"N 126°55′37.5"E (Yeoseo, Republic of Korea).

Preparation of Styrofoam material.

The Styrofoam material for the experiments was prepared by breaking down the Styrofoam box manufactured by Woolim (Gyeongi-do, Republic of Korea) to smaller pieces. Three of the randomly sampled pieces were determined for their density and molecular weight. The density was 16.676 ± 0.253 mg/ml, and GPC revealed M_n of 80,515 ± 2,586 and M_w of 152,339 ± 1,430. No additives were added to the Styrofoam according to the manufacturer.

Styrofoam degradation by P. davidis larvae.

Two experimental groups were prepared with the collected larvae. The first larval group was starved for 3 days after collection (n = 4), while the other group was fed Styrofoam as their sole diet for 14 days after starving for 3 days (n = 3). The treated larvae were placed in a weighed 50-ml tube with a Styrofoam piece (one larva per tube), which was rinsed with ethanol and then dried at 60°C overnight and weighed. In this setup, 0.5 ml of deionized water (DW) was added. Then, the tubes were half sealed and incubated in a dark environment at 24°C and 60% relative humidity for 14 days. After incubation, the larvae were removed, and the tubes were dried at 60°C overnight and processed for weighing. The negative-control group was also prepared with the same setup as the experimental groups but without the larvae. The removal process was replicated in the negative-control group (n = 4) to minimize the error caused by the loss of Styrofoam while removing the larvae from the tube. Statistical analysis was performed using Student’s t test.

Survival rate of Styrofoam-fed larvae.

Eighteen larvae were placed in two six-well plates (one larva per well, three larvae per each replication). Nine of the larvae were incubated with blocks of Styrofoam, while the others were given no food source, serving as a negative control. The larvae were incubated at 24°C with 60% relative humidity for 4 weeks. After the larvae were incubated, the numbers of living and dead larvae were counted to determine the survival rate over the incubation period.

FT-IR analysis of the frass from Styrofoam-fed larvae.

Fourteen larvae were placed in a petri dish (diameter = 20 cm) with a block of Styrofoam (n = 3). The frass was collected after incubating the petri dish with the larvae at 24°C with 60% relative humidity for 3 days. The collected frass was then washed with a 2% (wt/vol) sodium dodecyl sulfate (SDS) solution for 4 h to remove the residual biofilm or other miscellaneous debris from the gut. The SDS was removed by washing the frass with DW (repeated 4 times), and the frass was dried at 60°C overnight. The dried frass and the Styrofoam feed control, which went through the same process as the frass, were then analyzed using a Nicolet iS50 FT-IR device (Thermo Scientific, USA).

GPC analysis of frass from Styrofoam-fed larvae.

Five larvae were placed in a six-well plate (one larva per well) with blocks of Styrofoam. The larvae were incubated at 24°C with 60% relative humidity for 2 weeks. The Styrofoam debris was cleaned off the incubated larvae, and the larvae were moved to a new six-well plate (1 larva per well). The frass was collected in a preweighed glass vial after incubating the larvae at 24°C with 60% relative humidity for 3 days. The collected frass was weighed, and 5 ml THF was added. The added mixture was vigorously resuspended 20 times with a 10-ml syringe and was incubated at room temperature for 6 h to extract the residual PS in the frass. The extracted solution was filtered through a 0.2-μm polytetrafluoroethylene (PTFE) syringe filter and was placed in a preweighed vial, which was dried for 24 h in a fume hood. The dried weight of the filtered frass was measured, and the molecular weight was analyzed by GPC (e2695; Waters, USA) with a flow rate of 1 ml/min using THF as a solvent. PS with molecular weights of 800, 3,000, 10,000, 30,000, 50,000, 100,000, 216,000, and 400,000 Da were used as the standard polymers in the calibration process. The MWD was obtained from the raw data by first transforming the retention time into the logarithmic value of the molecular weight (Log M) using the standard curve. Then, the area under the curve between each consecutive Log M was calculated by using the trapezoidal method, and each value was divided by the total sum area to obtain the weight (wt) fraction [dwt/d(Log M)].

Preparation of PS film.

A total of 1.5 g of the same Styrofoam blocks that were used in larval degradation experiment was dissolved in 50 ml of xylene. The dissolved Styrofoam was poured into a glass dish and dried for 24 h in a fume hood. The dried film was then rinsed with methanol and DW. Afterwards, the PS film was dried in a fume hood for 3 days and kept in an aluminum foil-wrapped desiccator prior to usage.

Preparation of media.

Liquid carbon-free medium was prepared by dissolving 0.7 g of KH₂PO₄, 0.7 g of K₂HPO₄, 0.7 g of MgSO₄·7H₂O, 1.0 g of NH₄NO₃, 0.005 g of NaCl, 0.002 g of FeSO₄·7H₂O, 0.002 g of ZnSO₄·7H₂O, and 0.001 g of MnSO₄·H₂O in 1 liter of DW. LB medium was prepared by dissolving 25 g of LB Miller broth into 1 liter of DW. To make the solid version, 25 g of agarose was added to the medium. All of the media were autoclaved at 121°C for 30 min and cooled prior to usage, and the solid medium was solidified in a 90-mm-diameter petri dish.

Extraction of gut flora.

The larvae that were fed Styrofoam as their sole diet for 28 days were sterilized by submerging the individuals in 70% ethanol for 3 min. After sterilization, the midgut was extracted and placed in a 15-ml tube containing 3 ml of liquid carbon-free medium (three midguts in each tube). The tubes were then vortexed for 15 min, and the gut tissue was removed. The tubes were centrifuged at 8,000 rpm to collect the gut flora bacteria, and the supernatant was discarded and replaced with 5 ml of liquid carbon-free medium, which was vortexed to homogenize the solution. After repeating the centrifugation step 3 times, the gut flora mixture was used for further studies.

PS film biodegradation by gut flora.

Approximately 0.5 g of PS film, 100 ml of liquid carbon-free medium, and 1 ml of the gut flora mixture were placed in a sterilized 250-ml Erlenmeyer flask (n = 6). The negative-control group was prepared by excluding the gut flora mixture. The inoculum was incubated at 24°C with 60% relative humidity and 120 rpm for 20 days.

Isolation of Serratia sp. WSW.

After 20 days of incubation, 1 ml of the culture broth and the PS film were collected and placed in a 1.5-ml tube. The tube was vortexed for 5 min to detach the biofilm-forming bacteria, and the inoculum was serially diluted. Two hundred microliters of the diluted mixture was spread on solid LB medium. The mixture was incubated at 24°C with 60% relative humidity for 24 h, and the colonies obtained from the medium were classified by their morphology. Each type of colony was inoculated in 5 ml LB liquid medium and incubated at 24°C with 60% relative humidity for 24 h. The incubated mixture was centrifuged at 8,000 rpm to collect the bacteria, and the supernatant was discarded and replaced with 5 ml carbon-free liquid medium, which was repeated 3 times to ensure that no LB medium was left in the bacterial cells. Two hundred microliters of this bacterial mixture was spread on carbon-free solid medium. Two solid media were prepared for each type of bacterial single colony mixture. A sterilized PS film was placed on one of the two solid media, while the other one was incubated without a PS film to ensure that the bacteria cannot grow by using only agarose. Then, the bacterial colony that was able to form a biofilm on the PS surface was analyzed by colony PCR to amplify the 16S rRNA gene sequence. Four kinds of universal primers were used: 9F (5′-GAGTTTGATCCTGGCTCAG-3′), 1512R (5′-ACGGCTACCTTGTTACGACTT-3′), 518F (5′-CCAGCAGCCGCGGTAATACG-3′), and 800R (5′-TACCAGGGTATCTAATCC-3′). Three reactions were performed with the primer sets (primer sets 9F and 1512R, 9F and 800R, and 518F and1512R). The obtained sequence was analyzed with the Basic Local Alignment Search Tool (BLAST) created by the National Center for Biotechnology Information (Bethesda, MD, USA) and was deposited in GenBank.

Characterization of Serratia sp. WSW.

Two hundred microliters of Serratia sp. WSW liquid LB medium culture that reached the stationary phase was inoculated in 20 ml of liquid LB medium. The inoculum was incubated at 24°C (n = 3) and 37°C (n = 3) with shaking at 200 rpm. Starting from the inoculation point, 100 μl of the culture was sampled every 2 h and mixed with 900 μl liquid LB medium. The optical density (OD) of the mixture was measured at a 600-nm wavelength using a UV spectrophotometer (Shimadzu, Japan) until the culture reached the stationary phase. The exponential phase (2 to 4 h) was selected from the growth curve, and the exponential trend line was obtained. The natural logarithm of 2 was divided with the growth constant to obtain the doubling time. The mixture for the OD measurement at the incubation time of 12 h was sampled for Gram staining. The stained sample was imaged under ×1,000 magnification using an optical microscope (Olympus, Japan).

PS film biodegradation by Serratia sp. WSW.

The isolated single colony of Serratia sp. WSW from the solid LB medium was transferred to a 15-ml tube containing 5 ml of liquid LB medium. The 15-ml tube containing Serratia sp. WSW was incubated at 24°C with 60% relative humidity and shaking at 120 rpm for 12 h. Afterwards, the tube was centrifuged at 8,000 rpm for 10 min to collect the cells. The supernatant was removed and replaced with 5 ml of liquid carbon-free medium. Then, the tube was vortexed to homogenize the cells and centrifuged at 8,000 rpm for 3 min. This step was repeated 3 times to remove the residual LB medium. Then, approximately 5 × 10⁷ cells from the mixture were spread on a solid carbon medium. On top of the Serratia sp. WSW, five PS films of different sizes were placed. The negative-control group was prepared by excluding the Serratia sp. WSW from the setup. The experimental and control groups were then incubated at 24°C with 60% relative humidity for 20 days and used for analysis.

SEM sample preparation.

After incubation, the PS film was fixed in phosphate-buffered 2% glutaraldehyde solution for 1 h. Then, the film was dehydrated using 50% to 100% graded ethanol. The dehydrated film was then freeze-dried using a FD-1000 freeze dryer (Eyela, Japan). Afterwards, the film was coated with gold by using an ion coater PS-1200 (Paraone, Republic of Korea) and was then imaged by scanning electron microscopy (SEM) (XL30S FEG FE-SEM; Philips Electron Optics B.V., USA).

Physicochemical and topological characterizations of biodegraded PS film surface.

After incubation, the PS film was randomly selected and washed in 2% (wt/vol) SDS for 4 h. The washed films were then rinsed with DW several times and dried in a desiccator for 24 h. The dried films were analyzed by SEM to confirm the complete removal of the biofilm. Afterwards, the surface hydrophobicity, surface chemical composition, and surface topography of the films were analyzed. The surface hydrophobicity of the films was analyzed by dropping 1 μl of deionized water onto the surface of the PS film using a Smartdrop WCA measurement device (Smart Drop, Republic of Korea). The surface chemical composition of the films was analyzed using XPS with an AXIS ultradelay line detector (DLD) (Kratos, UK) equipped with monochromatic Al Kα (1,486.6 eV) as an X-ray source. The surface topography of the films was analyzed using atomic force microscopy (AFM) (MMAFM-2/393EX; Digital Instruments, USA).

Microbial community analysis of gut flora.

The gut flora of two larval groups were sampled for microbial analysis. The first group was fed only Styrofoam for 2 weeks, while the other group consisted of freshly caught specimens from the wild and was put into hibernation at 4°C until the extraction (<2 weeks). Each group had three larvae so that the individual variance of the gut flora could be pooled. The larvae were sterilized by submerging the individuals in 70% ethanol for 3 min. After sterilization, the midgut was extracted and placed into a 15 ml-tube containing 3 ml of carbon-free liquid medium. The tube was then vortexed for 15 min, and the gut tissue was removed. The samples were sent to Macrogen (Seoul, Republic of Korea) for the microbial community analysis service.

Data availability.

The partial 16S rRNA sequence of Serratia sp. WSW has been deposited in the GenBank database with accession number MN097873.1 .