G. mellonella larvae were purchased from Huiyude Co. (Tianjin, China). PS foam board was obtained from Nannan Building Materials Co. (Zhejiang, China) and contained polystyrene purity over 98%. The number-average molecular weight (Mn) and weight-average molecular weight (Mw) of the PS were 64400 and 144,400 Da, respectively, as measured by gel permeation chromatography. The PS film for microbial degradation was prepared using the previous method (Yang et al., 2015) and fine-tuned for this study: the foam PS was dissolved in dichloromethane (0.03 g/ml) and then the solution was spread on a glass plate, after 5 h, the resulting film was removed from the glass plate and immobilized in a fume hood for 3 days at room temperature. The film was then rinsed with de-ionized water and dried before use. The thickness of the prepared film was approximately 0.02 mm.

The composition of the mineral salts medium (MSM, pH = 7.01) was as follows: 4.54 g/L KH₂ PO₄, 11.94 g/L Na₂HPO₄·12H₂O, 1.0 g/L NH₄Cl, 0.5 g/L MgSO₄, 0.005 g/L CaCl₂, 0.002 g/L FeSO₄, 0.001 g/L MnSO₄ and 0.002 g/L ZnSO₄. The Luria-Bertaini (LB) medium was prepared by dissolving 10 g NaCl, 10 g tryptone, and 5 g yeast extract in 1 L deionized water. 15 g agar was added to prepare the LB agar medium. 0.9 g NaCl was dissolved in 100 ml deionized water to prepare saline. All media buffers and solutions were subjected to high-pressure steam sterilization (121°C, 103.4 kPa, 20 min). All other chemical reagents used in this study were analytical reagent grade and obtained from commercial sources.

Larvae (n = 200) were fed with PS foam for 21 days, and then 10 larvae were collected. After sterilization, the larvae were dissected and then the intestinal tissue was placed into a 1.5 ml centrifuge tube containing 1 ml of normal saline, and then shaken on a vortex mixer for 5 min. A pure intestinal cell suspension was obtained and used as a bacterial inoculum to enrich PS-degrading bacteria. A 1 ml aliquot of the suspension was transferred into a 250-ml Erlenmeyer flask with 80 ml MSM and 0.15 g PS film (1 × 3 mm), which was shaken on a rotary shaker (120 rpm) at room temperature. After 60 days, the remaining PS film was removed, and the enrichment culture was dispersed on LB agar plates. After culturing for 24 h at room temperature, colonies were picked and then spread on fresh LB agar plates until a pure colony was finally obtained according to the standard methodology of bacterial isolation (Yang et al., 2015). After the pure bacterial isolates were grown in liquid LB medium for 12 h, the cells were collected by centrifugation (10000 rpm) and then washed with sterile water to remove residual medium. This step was repeated until there were no remaining nutrients. Next, the collected cells were resuspended in sterile water and then diluted 100 times to obtain a cell suspension (Yang et al., 2015). The cell suspension (0.4 ml) was distributed evenly on the surface of a MSM plate, which was then covered with PS film. Two control groups were set up. One was only inoculated with bacterial culture, while the other was only covered with PS film. All plates were cultured in triplicate at 30°C for 30 days. Place an open Petri dish with distilled water in the incubator to prevent the plates from drying. Changes in the film and bacterial growth were regularly observed. The remaining larvae were refrigerated at 4°C after pupation for subsequent use.

Genomic DNA used for 16S rDNA amplification was extracted from the cells during the logarithmic growth stage using the TIANamp Bacteria DNA Kit (Tiangen Biotech Co., Ltd., Beijing, China). Next, PCR amplification and agarose gel electrophoresis were performed to verify successful extraction. The gene was amplified using the universal primers 27-F (5′-AGA​GTT​TGA​TCC​TGG​CTC​AG-3′) and 1492-R (5′-GGT​TAC​CTT​GTT​ACG​ACT​T-3′). Amplicons in the gel were recovered using TIANquick Midi Purification Kit (Tiangen Biotech Co., Ltd., Beijing, China), and then sequenced using a kit from Sangon Biotech Co., Ltd. (Shanghai, China). The obtained sequences were aligned with known organisms in the GenBank database using the Basic Local Alignment Search Tool (BLAST) created by the National Center for Biotechnology Information (NCBI, Bethesda, MD, United States). A phylogenetic tree was constructed with MEGA5.0 software using the neighbor-joining method. Bootstraps of the supporting tree branches were constructed with 1,000 replications and the default settings. A total of 17 16S rRNA sequences were used for the phylogenetic analyses, and the accession numbers of these sequences are shown in Figure 1.

The determination of microbial physiological and biochemical indicators was carried out using the method of bacterial identification in the “Berger’s Bacterial Identification Manual” (eighth edition) and Identification Manual of Common Bacteria System (Buchanna & Gibbons 1984; Dong & Cai 2001).

An SEM (SU8010, Hitachi, Tokyo, Japan) was used to inspect the biofilm that formed on the PS film and the holes in the PS film after degradation. To observe the presence and growth of microorganisms, the PS films that covered the agar plates were removed after 30 days, fixed with 2.5% glutaraldehyde for 3 h and dehydrated with an ethanol gradient (30, 50, 70, 90 and 100%) for 15 min (Wang et al., 2020). To detect changes in the physical and chemical properties of the PS film, the films were immersed in a 2% (w/v) SDS aqueous solution for 4 h, and then washed with de-ionized water to completely remove the biofilm from the surface. The PS film from the uninoculated control was also treated in the same way (Sivan et al., 2006). To verify changes in the elemental composition of the PS film surface during the degradation process, a module connected to the SEM (EDS (SU8010, Hitachi, Tokyo, Japan) was used to evaluate the differences in the composition of the carbon and oxygen elements before and after degradation.

The WCA (DSA100, KRUSS, Hamburg, Germany) was used to analyze changes in the hydrophilicity of the PS film surface. The contact angle was measured under static conditions at room temperature. The PS film was washed three times with SDS solution and then de-ionized water, and the contact angle measurement was estimated five times.

To investigate degradation and determine the variations of functional groups on the surface of the PS film, XPS (ESCALAB Xi+, Thermo Fisher, Massachusetts, United States) was used to measure the binding energy. The PS film (1 cm × 1 cm) was fixed on a carbon ribbon and measured within the energy range of 2 + P−300eV, C1s.

After overnight incubation in 4 ml liquid LB medium, bacteria were collected by centrifugation for 10 min at 10,000 rpm, and then washed three times with 4 ml of saline. PS film (1 cm × 3 cm, 0.15 g) was added into a 250-ml Erlenmeyer flask containing 80 ml MSM. A 1 ml suspension of the bacterial culture was added into the flask, which was then incubated on an orbital shaker (150 rpm) at 30°C. This method was a slight modification of the following reported scheme (Wang et al., 2020) because the PS powder was replaced with a film; the reaction system volume was doubled; the temperature and speed were fine-tuned higher; and the reaction time was 30 days. The PS film incubated in MSM without bacteria served as a control. The weight loss assay was replicated three times. After a 30-days incubation, the PS films were harvested (Mor and Sivan, 2008).

The weight loss of the PS film was calculated using the following formula: Percentage of weight loss  =  100% ×  ( Initial PS weight - Final PS weight ) /Initial PS weight

Statistical ANOVAs were performed using SPSS 20.0 (SPSS Inc., Chicago, United States) to evaluate the differences in contact angle changes and weight loss produced by bacteria. Pairwise comparisons were analyzed with the student’s t-test, as all date were normally distributed. All error values are reported as the mean value ±standard deviation.

G. mellonella larvae were fed with PS for 21 days. By the third day of feeding, the PS foam block was already full of holes and a small amount of block was gnawed and attached to the culture container. By the fifth day, the whole culture container was covered with PS debris (Figure 2). After consuming the PS foam diet for 21 days, the larva appeared to be normal and they produced a normal amount of silk. These results suggested that the PS diet did not harm the larvae over a short-term feeding period.

After 2 months enrichment culture, when compared with the control group, a turbid liquid was observed in the experimental group, indicating that some intestinal bacteria with the ability to degrade and use PS as a nutrition source may have been enriched. After cultivation and purification, a pure colony named as FS1903was finally obtained. When cultivated on solid LB medium, FS1903 were round with clear edges and a slightly yellowish color (Figure 3A). The shape of the bacterial cells examined with an SEM. The results showed that FS1903 cells are an elliptical rod shape approximately 1.5 μm long and 0.5 μm in diameter (Figure 3B). To verify the PS degrading ability of FS 1903, a PS film (50 mm × 50 mm) was supplied as the sole carbon source on top of an MSM plate that was spread with the cell suspension. After 30 days cultivation, when compared with control group (Figure 3C), some small colonies were observed under the PS film and a linear bacterial lawn appeared at the edge of the film in the experimental group (Figure 3D). This provides evidence that FS1903 isolated from the gut of G. mellonella larvae can degrade PS.

The result of strain culturing showed that the tested strain is short rod-shaped bacterium, is motile by a terminal flagellum; forms smooth, round yellow-white colonies on LB medium; and optimum growth at 30°C and pH 7.0. Physiological and biochemical characteristics of the strain were list at Table 1. According to the physiological and biochemical characteristics results, the strain is similar to the microorganisms of the genus Massilia (Shen et al., 2013; Orthova et al., 2015). In addition, bacteria of this family are often isolated from the rumen or the large intestine of humans and animals (Kämpfer et al., 2012), this is also in line with our research.

A phylogenetic tree of FS1903 and other Massilia bacteria was constructed based on the 16S rRNA gene (Figure 1). According to the phylogenetic tree, FS1903 and Massilia suwonensis 5414s-25 were on the same branch and closely related with high sequence similarity of 79%. Combined with the physiological and biochemical test, tested strain can be identified as the genus Massilia, and then the strain was named Massilia sp. FS 1903. The 16S rRNA gene sequence of this bacterium was deposited in the GenBank database under accession number MW138062.

The biofilm was scanned with an SEM to more closely examine the effects of the colonization of the PS film surface by FS 1903. After 30 days of cultivation, the PS film surface was examined before and after the microorganisms were completely washed off, and the results showed that the PS film was damaged by the bacteria (Figure 4A,B). In contrast, the surface of the uncultured control was smooth (Figure 4C) without any defects. These observations revealed that the bacteria caused pits and cavities on the film surface. The maximum width of a typical cavity was approximately 4 μm (Figure 4D).

The carbon and oxygen composition on the PS surface where the bacteria grew was analyzed by EDS to determine the effects of the bacteria on the composition of the film. There were no obvious differences in the number of carbon atoms between the PS in the control group and the experimental group (Figure 5). However, more oxygen atoms were detected in the experimental group, which shows that oxidization occurred.

Oxidation was confirmed by measuring changes in the contact angle of water droplets on the PS surface, which is indicative of the surface hydrophobicity of the PS film. The results show that the contact angle of the experimental group was 66.1 ± 5.1^o, while that of the control group was 96.0 ± 3.8^o (Figure 6). Compare to the control, the contact angle of the experimental group was significantly decreased (p < 0.01). This decrease in the contact angle shows that the surface tension of water decreased, because of the insertion of oxygen on the surface of PS during the oxidation process. Oxidation during PS degradation converts hydrophobic regions to hydrophilic ones, thereby changing the chemical properties of the PS surface (Gu, 2003). FS1903 colonization reduced the hydrophobicity of the PS film and at the same time the film was damaged, the decreased hydrophobicity reduced the resistance to bacterial cells to subsequent PS degradation.

XPS was used to analyze changes in the surface chemical composition and functional groups. Figure 7 shows the comparison (0–900 eV) of the XPS scanning spectra of the PS film for the inoculated experimental and non-inoculated control groups. The control group had only surface carbon at peak of 284.8eV, while the PS inoculated with strain FS 1903 except 284.8eV had another obvious peak at 532.3 eV that represented the amount of surface oxygen (Figure 7A). The XPS spectra of C1s on the PS film surface inoculated with strain FS1903 versus the control was compared (Figure 7B), and it showed that culturing with the FS1903 caused a notable decrease of the C-C group (284.8eV). Meanwhile, compare to the control group with one peak, the PS inoculated with strain FS 1903 had another peak at 286.5 eV, which was assigned to the C−O group. This implies that part of the C−C groups in the PS was oxidized to alcohol and carboxylic-acid-like compounds (Shang et al., 2003). In addition, on the PS film surface inoculated with strain FS 1903, another new peak appeared at 288 eV, which was assigned to the −C=O group. These observations all demonstrate a transition from a C−C bond to −C=O and C−O bonds during the degradation process. Furthermore, these results indicate that strain FS1903 is capable of attacking or oxidizing the PS structure to produce more polar derivatives. This change in binding energy confirms that oxidation occurs during FS1903-mediated degradation. The results are in accordance with previous work (Yang et al., 2015; Kim et al., 2020).

To confirm the degradation ability of strain FS 1903, we calculated the weight loss after infestation with the bacteria. After cultivation with the bacteria for 30 days, the weight loss of PS treated with strain FS1903 was 12.97 ± 1.05%, which was significantly decreased compared with the control group (p < 0.01). Furthermore, FS1903 degraded the PS without being supplied with nutritional supplements, such as yeast extract or gelatin.