Investigating biodegradation of polyethylene and polypropylene microplastics in Tehran DWTPs

Plastics are synthetic polymers that usually contain chemicals to have better performance. They are produced from petrochemical refineries and vary in density and plasticity (Costa et al. 2016). However, in certain environmental conditions, they are susceptible to undergo weathering and lose some of their chemical bonds (Browne et al. 2007). The crushing and decomposition of these plastics turn them into tiny microscopic particles called microplastics (MPs; Lwanga et al. 2017), ranging from 1 to 5,000 μm in their longest end (Koelmans et al. 2019). Considering the increasing production of plastic products and the adverse effects of MPs on humans (gastrointestinal problems, endocrinal disruption, and respiratory problems) and other organisms (gut blockage, oxidative stress, reduced fecundity, and behavioral alteration), addressing such pollutants has become more crucial than ever (Nanthini devi et al. 2022; Emenike et al. 2023). These particles can also reduce the nitrogen cycle in soil, increase soil pH, and enter the food chain (Li et al. 2023a). Generally, MPs have been investigated in multiple freshwater and saline water resources including rivers (Cook et al. 2021; Stride et al. 2023a), wetlands (Mahdian et al. 2023), urban water infrastructures (Stride et al. 2023b), or nearshore regions (Abolfathi et al. 2020).

As drinking water treatment plants (DWTPs) are considered a reliable source of water supply for people, it is imperative to examine whether treated water is contaminated with MPs. Recent studies have detected MPs in DWTPs and tap water, raising concerns about the effectiveness of DWTPs in removing MPs at a desirable degree (Pivokonsky et al. 2018; Novotna et al. 2019; Jinkai et al. 2021). Therefore, it is necessary to investigate MPs reduction in DWTPs. Growing evidence indicates that microorganisms play a crucial role in decomposing plastics across various environments, including soil, sediments, seawater, and compost (Chen et al. 2023; Li et al. 2023b; Peng et al. 2023). In fact, through biodegradation, microorganisms are used to decompose plastic. Currently, different microorganisms are being explored in order to enhance the rate of plastic degradation (Li et al. 2023c).

Microorganisms can grow in diverse conditions and use carbon as their energy source. It has been observed that some strains can produce enzymes to degrade synthetic polymers, and the nature and catalytic activity of these enzymes vary depending on the microbial species and strains (Auta et al. 2018a, 2018b). For instance, Auta et al. (2018a, 2018b) observed that Bacillus sp. and Rhodococcus sp., isolated from sediments near mangrove plant roots, could grow in the presence of MPs and reduce their mass. After 40 days of contact, Bacillus sp. and Rhodococcus sp. achieved a 6.4 and 4% reduction in MPs mass, respectively. This analysis was further confirmed using a Fourier transform infrared (FTIR) spectroscopy and scanning electron microscope (SEM), which revealed morphological and structural changes on the surface of MPs. Finally, the researchers concluded that these types of bacteria can utilize polypropylene (PP) as a carbon source.

In another study, Habib et al. (2022) investigated the growth potential and degradation of PP MPs by a strain of bacteria found in the Antarctic, specifically Pseudomonas sp. ADL 15 and Rhodoccocus sp. ADL 36. Over a 40-day period in the Bushnell Haas medium, the weight loss and degradation rate of MPs were analyzed, and the biodegradation of MPs was assessed by analyzing structural changes via FTIR spectroscopy. After 40 days, the weight loss rate of MPs was reported 17.3% for ADL 15 and 7.3% for ADL 36. FTIR analysis revealed significant changes in the functional groups of PPs.

This study aimed to identify microbial strains in the raw water environment from Tehran DWTPs and investigate the interaction between microorganisms and MPs. This paper also aims to investigate the ability of microbial communities to biodegrade MPs. For the first time, native microorganisms present in raw water were used in this study to assess the potential of DWTPs to remove MPs.

In this study, PP and PE (0.92 g cm⁻³) granules were purchased from Marun Petrochemical Company (Tehran, Iran). Then, the granules were immersed in 0.1 mol HCl for 12 h to decompose the potential residues. After drying in an oven at 60°C for 2 h, granules were immersed in liquid nitrogen until they reached −196°C. Afterward, they were rapidly milled with an ultra-centrifugal mill (ZM 200, Retsch^®, Germany) to achieve micron-sized MPs (Munno et al. 2018; Pivokonsky et al. 2018). Finally, MPs were sieved into three different size categories including d > 75 μm, 75 > d > 150 μm and d < 150 μm. After sieving, prepared MPs were sent to be exposed to selected microbial colonies. The culture media used for microbial growth were made with 1 g L⁻¹ ammonium chloride (NH₄Cl), 3 g L⁻¹ monopotassium phosphate (KH₂PO₄), 7 g L⁻¹ disodium phosphate (Na₂HPO₄), 0.5 g L⁻¹ sodium chloride (NaCl), 0.25 g L⁻¹ magnesium sulfate (MgSO_4·7H₂O), 1 g L⁻¹ PP or PE, 40 μg L⁻¹ copper sulfate (CuSO_4·5H₂O), 0.2 μg L⁻¹ ferric chloride (FeCl₃·6H₂O), 0.4 μg L⁻¹ magnesium sulfate (MnSO₄·5H₂O), 0.4 μg L⁻¹ zinc chloride (ZnCl₂), 0.2 μg L⁻¹ ammonium molybdate ((NH₄)₆Mo₇O₂₄·7H₂O), and 0.5 μg L⁻¹ boric acid (H₃Bo₃) (Shirazi et al. 2023).

The sampling method in this study is based on Bonetta et al. (2022) with minor changes according to the purpose of this study. From June to July 2022, three 100 mL samples were taken from the raw water of three different DWTPs in Tehran and fed from different water resources. Sampling bottles were made of glass in a dark color. In total, 9 samples comprising 2.7 L of samples were collected for bacterial identification. Samples were kept refrigerated within 24 h before being sent to microbial-related analysis.

Regarding the biodegradation of the identified polymers, the biodegradation process was carried out in a synthetic environment. Initially, after preparing the culture medium and sterilizing MPs using an autoclave, they were inoculated in a 500 mL Erlenmeyer flask along with 1 mL of water sample from the raw water of DWTPs and 100 mL of prepared culture medium. The flasks were placed on a shaker at a speed of 140 rpm and incubated for 1 month at 37°C in pH 7 ± 0.5 (Gong et al. 2018). After this period, colonies consisting of bacteria and fungi were observed in the culture medium. These colonies were purified and isolated, and the process of MPs inoculation and bacterial colony growth without the water sample was repeated for the second and third times. The samples were placed on the shaker for 9–12 weeks. The consortium of bacteria exhibiting the highest consumption of MPs was selected through primary screening based on the greatest weight loss of MPs. Among 20 bacteria tested, the three samples showing the highest efficiency were chosen for deoxyribonucleic acid (DNA) extraction and polymerase chain reaction (PCR) testing. Next, three samples containing the selected pure bacteria and MPs were inoculated into the culture medium and placed on a shaker for 3 months. In addition, two control samples were prepared alongside the main samples, with all conditions being similar except for the presence of bacteria. After the specified time, the weight of MPs was compared to that of before the exposure to determine the mass reduction and calculate the degradation rate. To assess the molecular bonds and analyze the structure of MPs, samples were filtered through a 0.45 μm membrane filter. In this regard, to eliminate possible bacteria on the filter samples, 96% ethanol diluted with 20 mL of distilled water was utilized. Next, the filters were placed in a glass petri dish for 30 min in an oven at 60°C. For further analysis, samples containing MPs were exposed to bacteria, as well as two control samples of PP and PE MPs without bacteria were prepared. Raman spectroscopy, FTIR analysis, and thermal gravimetric analysis (TGA) were performed to identify changes in molecular bonds. Additionally, a SEM was utilized to identify the morphology of the decomposed MPs (Rizzarelli et al. 2016; Mintenig et al. 2019).

For molecular identification and genomic DNA extraction, the desired isolates were first inoculated into a nutrient broth liquid culture medium and heated until reaching a suitable biomass. Subsequently, the cells were physically disrupted by beating the frozen biomass with liquid nitrogen, and the DNA was isolated through phenol–chloroform extraction (Oßmann et al. 2018). To achieve molecular identification of the isolates, the extracted DNA was amplified using PCR with the general primers 27F and 1492R. Horizontal electrophoresis was performed using a 1% agarose gel to assess sample purity. Finally, the target gene and its sequence were compared to sequences in the National Center for Biotechnology Information (NCBI) databases for identification (Oßmann et al. 2018).

The growth kinetic of selected strains in dried weight in the culture medium comprising 1,000 mg of MPs signifies the consumption of MPs by the inoculated strains. Since the only carbon source in the medium is MPs, the amount of created biomass can indirectly mean the biodegradation of these particles. Hence, selected strains were inoculated in the culture medium containing 1,000 mg of MPs and incubated at 28°C for 30 days on a shaker at the speed of 150 rpm. After this step, the flask contents were centrifuged at 8,000 rpm and supernatants were disposed of. To measure the weight loss, the flask contents were poured on pre-weighed glass plates. Next, the plates were kept in 65°C for 48 h and weighed again. The weight loss was calculated using Equation (1), where W₀ represents the initial weight in grams and W represents the weight of the remaining MPs (Shirazi et al. 2023).

Measuring produced protein from the MPs biodegradation can indirectly represent the ability of selected bacteria in plastic degradation. For this process, the method developed by Lowry et al. (1951) was employed. In brief, four reagents were prepared including reagent A (2% Na₂CO₃ in 0.1 mol NaOH), reagent B (0.5% CuSO₄·5H₂O in 1% sodium tartrate), reagent C (mixture of 50 mL of reagent A with 1 mL of reagent B), and reagent D (same as reagent C, except for the omission of NaOH). A 100 μL of the prepared culture medium was mixed with 1 mL of reagent C in a 1.5 mL vial and kept at room temperature for 15 min. Then, 0.1 mL of reagent D was added to the mixture and mixed for 45 s with a vortex. To create a protein assay standard curve, different concentrations of bovine serum albumin from 0.1 to 1 mg mL⁻¹ in deionized water were prepared, and the optical absorption of the medium was measured at a wavelength of 650 nm. Mentioned experiments were conducted in triplicate.

After evaluating the effectiveness of 20 different strains of bacteria, 3 samples with the highest efficiency were selected for DNA extraction and PCR testing. The microbial community analysis, focusing on the biodegradation of PP and PE MPs, was carried out using the NCBI database. The analysis revealed the presence of the strains Pseudomonas protegens (A), Bacillus amyloliquefaciens (B) and Bacillus cereus (C) with a 99, 99 and 100% similarity, respectively. Strain A has been observed as a biofilm on MPs surface found in Ganjiang River, China (Hu et al. 2021). In another study, bacterial strain B was observed to be effective on polystyrene (PS) MPs biodegradation (Kang et al. 2023). Moreover, strain C has been effective in the biodegradation of low-density polyethylene in other studies. However, they have found the bacteria in the soil environment (Jayan et al. 2023).

In Figure 1, the weight loss observed corresponds to the thermal decomposition of the PP chemical chain. In the control sample, according to the derivative thermogravimetric (DTG) curve, this decomposition occurred at approximately 457°C. This weight loss at this temperature aligns with findings reported in similar articles (Jung et al. 2010; Żwierełło et al. 2020). It is evident from the figure that the thermogram of sample C closely resembles that of the control sample. This indicates that the bacteria in this sample probably has a minimal impact on the structure of the polymer. The microbial strain caused the breakdown of some long chains, resulting in a slight reduction in thermal stability. The weight loss of this polymer at 800°C increased slightly from 7.97 to 8.13 mg. In the TGA test results of sample A, it is clear that the bacterial strain in this sample attacked the PP polymer structure, leading to significant changes. One notable change observed in the thermograms is an additional weight loss in the TGA curve, accompanied by an endothermic peak in the DTG curve between 50 and 150°C (DTG peak at 97°C). This weight loss and endothermic peak can be attributed to eliminating water absorbed on the surface. PP is inherently hydrophobic due to its long hydrocarbon chains. Therefore, no water absorption effect was observed in the TGA curve of the control sample at temperatures lower than 200°C. However, when the PP MPs were exposed to the bacteria in sample A, structural changes occurred, resulting in the creation of polar groups on the polymer's surface. As a result, the polymer became hydrophilic and exhibited water absorption. As depicted in the thermograms, sample A experienced weight loss in the temperature 440–520°C (DTG peak at 460°C), corresponding to the thermal decomposition of intact PP structure. Another weight loss stage was observed in the temperature of 220–375°C (DTG peak at 281°C), indicating the thermal decomposition of PP chemical chains broken by bacteria in this sample.

Moving on to the TGA and DTG curves for the PE shown in Figure 2, it is evident that the thermal decomposition of this polymer resulted in a single stage of weight loss in the temperature of 400–530°C (DTG peak at 476°C). In the TGA curve of strain B, although the weight loss occurred at almost the same temperature as the control sample, the weight loss in strain B was greater than that of the control sample. Specifically, the weight loss increased from approximately 3.9 mg in the control sample to about 5.9 mg in sample B. These results confirm that the selected strains used in sample B did not destroy the main structure of the PE but rather produced products with lower thermal stability than the control sample.

In the obtained spectra, the stretching vibrations of C–C bonds are at 1,059, 1,079 and 1,126 cm⁻¹. Additionally, the rocking vibration of CH₂ groups and the twisting vibration of C–H bonds have made peaks at 1,166 and 1,293 cm⁻¹. The peak at 1,367 cm⁻¹ corresponds to the wagging vibration of CH₃ bonds (Visentin et al. 2006). In addition, the bending vibrations of CH₂ bonds have created peaks at 1,415, 1,438 and 1,459 cm⁻¹. These spectra are consistent with the previous study on PE (Furukawa et al. 2006), confirming the existence of the PE structure.

Figure 3 illustrates that in sample B, exposed to bacteria, there is a reduction in peak intensity without any shift in peak positions, suggesting the potential breakage of PE chains. However, the bacteria have not been able to change the structure of PE and form new structures. Figure 4 presents the Raman spectra of PE samples.

In the Raman spectrum of the PP control sample, the stretching vibrations of C–C bonds appeared at 805 and 1,034 cm⁻¹. Also, the rocking vibration of CH₃ groups resulted in peaks at 837 and 969 cm⁻¹ (Jung et al. 2018). In addition, the bending vibrations of CH₂ bonds were evident in peaks at 994, 1,149, and 1,164 cm⁻¹ (Cerqueira et al. 2013). The peak at 1,216 cm⁻¹ can also be attributed to the combined effects of the wagging vibration of C–H bonds, the twisting vibration of CH₂ groups, and the stretching vibration of C–C bonds. The twisting vibration of CH₂ groups and the wagging vibration of CH₃ groups created peaks at 1,327 and 1,358 cm⁻¹, respectively. The bending vibration of CH₂ groups intensified the peaks at 1,433 and 1,457 cm⁻¹ (Jung et al. 2018). Figure 4 indicates that the same peaks were also present in sample C, suggesting that the overall polymer structure remained unchanged in the presence of bacteria. However, in sample A, the spectrum was completely changed, and a new chemical structure was obtained due to the combination of polymer and bacteria. In the spectrum of sample A, the bending vibration of C–H bonds in Raman showed shifts at 631 and 691 cm⁻¹. Furthermore, due to the structural changes induced by the bacteria, original peaks at 1,149 and 1,433 cm⁻¹ in PP were shifted to 1,162 and 1,436 cm⁻¹, respectively (Jung et al. 2018).

In Figure 5, the spectrum of the obtained samples implies a broad peak in the range of 3,000–3,700 cm⁻¹ wave number, which can be attributed to the stretching vibration of hydroxyl bonds (O–H). This peak exists due to the surface absorption of water on these polymers (Muthuselvi et al. 2018). Also, multiple absorption peaks observed in wave numbers ranging from 2,700 to 3,000 cm⁻¹ correspond to the stretching vibration of C–H bonds in CH, CH₂, and CH₃ groups, as reported in similar studies (Shahmoradi et al. 2020). Furthermore, according to the findings of a similar study (Jung et al. 2018), the peak at approximately 1,600 cm⁻¹ is related to the bending vibration of O–H bonds (surface adsorbed water), while the peak at around 1,380 cm⁻¹ is attributed to the bending vibration of C–H bonds. Also, the wagging vibration of C–H bonds is indicated by the presence of an absorption peak at wave number 1,120 cm⁻¹, while the stretching vibration of C–C bonds illustrates an absorption peak at wave number 1,000 cm⁻¹ (Shahmoradi et al. 2020). The broad peak at a wave number below 800 cm⁻¹ may be attributed to the stretching vibration of C–CH bonds. By comparing these spectra, it is evident that the FTIR spectrum of sample C closely resembles that of the reference sample, albeit with reduced peak intensities for certain vibrations such as C–H and C–C. This suggests that in the presence of bacteria in sample C, changes in PP chemical chains have been created, resulting in a decrease in peak intensity. The absence of new peaks confirms the absence of new bonds in the structure of PP. However, in sample A, notable changes in peak intensities can be observed, along with the creation of a peak at approximately 1,050 cm⁻¹. This peak can be attributed to the stretching vibration of C–O bonds in this sample. This peak can confirm the results of Raman analysis and TGA, which indicated a complete change in PP structure and the formation of new functional groups in the presence of bacteria. Figure 6 shows the FTIR spectrum of PE. According to Figure 6, the spectrum of the studied samples reveals several characteristics: (1) a broad peak in the wave number of 3,000–3,700 cm⁻¹ is observed, which corresponds to the stretching vibration of the hydroxyl bonds (O–H). This peak is attributed to the surface absorption of water on the polymers. Additionally, there are absorption peaks between wavenumber 2,800 and 3,000 cm⁻¹, which are related to the symmetric and asymmetric stretching vibration of C–H bonds in CH groups.

The peak at approximately 1,600 cm⁻¹ is related to the bending vibration of O–H bonds (surface absorbed water), while the peak at 1,350 cm⁻¹ corresponds to the bending vibration of C–H bonds. In this sample, the wagging vibration of C–H bonds is represented by a peak at around 1,115 cm⁻¹, and the stretching vibration of C–C bonds appears as an absorption peak at approximately 1,010 cm⁻¹. Furthermore, the broad peaks observed at wave number below 800 cm⁻¹ can be attributed to the stretching vibration of C–CH bonds. By comparing these spectra, it is evident that the FTIR spectrum of sample B closely resembles that of the control sample, with a reduction in intensity observed for certain peaks, such as those associated with C–H and C–C vibrations. Therefore, it can be concluded that due to the breaking of PE chemical chains in the presence of bacteria in sample B, only the intensity of the peaks has been affected, and the absence of a new peak confirms the absence of new bonds in the structure of this material. This result is in line with the results of Raman analysis and TGA.

The results of the cell weight measurement test showed that the average weight loss of the PP sample exposed to bacterial strains A and C were 38.6 and 25.6%, respectively. The greater weight loss observed in the presence of microbial strain A compared to strain C indicates that strain A has a more significant effect on the degradation of the chemical structure of PP MPs. Furthermore, the average cell weight loss for PE MPs exposed to strain B was 13.5%, which was the lowest decrease among the investigated strains. In a recent study, Tiwari et al. (2024) observed that Achromobacter xylosoxidans reduced PE MPs dry weight by 26.7%. This indicates that strain C is not effective enough to biodegrade PE MPs and other strains from other locations may be more effective. In addition, the test results for the amount of CO₂ produced for the samples exposed to strains A, C, and B were 3.03, 2.36, and 0.83 g L⁻¹, respectively. Also, the results of the test measuring the growth of the strains based on protein production indicated average protein yields of 4.1, 3.01, and 1.48 g L⁻¹ for the samples exposed to strains A, C, and B, respectively. Table 1 shows MPs weight loss exposed to strains A, B, and C.

Weight loss of MPs was investigated based on 1,000 mg of the initial addition of particles to the samples. Based on Table 1, strain A had the most effect on biodegradation among the other two strains with a 32.6% decrease in the weight of PP MPs, while strain B had a lower impact on PE MPs weight loss with a 13.3% decrease. These findings indicate that strain A has a greater effect on the destruction of MPs chemical structures than the other two strains. In a similar study, Auta et al. (2018a, 2018b) exposed PP nanoplastics to Bacillus sp. and Rhodococcus sp. after 40 days of incubation, only 6%. Likewise, in another study by Tiwari et al. (2023), PE MPs were incubated with BreviBbacillus brevis for 35 days, which resulted in a 19.8% reduction in MPs weight. So, the obtained result in this study is comparable to that of mentioned similar studies.

This study investigated the most abundant polymer types in MPs; however, other polymers that are predominantly found in DWTPs need to be investigated in terms of biodegradation capacity including PS, polyethylene terephthalate, and polyvinyl chloride. Moreover, bacterial strain detection should be investigated in DWTPs of other geographical locations to find out whether it exists another bacterial strain that has a more efficient effect on MPs biodegradation. In addition, this study collected raw water samples in a single period of time within a year, while sampling in multiple times of a year may alter the results in bacterial selection for biodegradation.

It can be concluded from the overall test results that exposure of MPs to strains A, B, and C for 3 months demonstrates that these strains adhere to the surface of both PE and PP MPs, leading to chemical damage and erosion. Various irregularities and unevenness of pits were observed in the samples containing PP MPs treated with strain A, and the chemical structure of PP MPs was changed according to FTIR and Raman analysis and underwent a complete degradation with the creation of new functional groups. On the other hand, in the samples containing PE MPs treated with strain B, noticeable changes were observed to some extent and a lower ability of this strain was observed to break down the PE structure compared to the other strains used in this study. In samples with strains B and C, the general structure of the polymer was not changed by the bacteria present. The results of cell dry weight loss also indicated that the weight loss of MPs in the presence of microbial strain A was more significant compared to microbial strain C. The average weight loss of cells observed for PE MPs exposed to strain B was 13.5%, the lowest among the strains examined in this research. In addition, after 3 months of incubation, MPs exposed to bacteria experienced weight loss. PP MPs weight decreased by 32.6% when exposed to strain A, while the decrease rate was lower when exposed to strain B (25.6%). On the other hand, strain B could not have a significant impact on PE MPs biodegradation and caused a 13.3% decrease in their weight.

We want to thank all the professors and friends who, with complete honesty and sincerity, supported us scientifically and technically during the implementation of this research. All the expenses of this study were supported by Fatemeh Tabatabaei. We appreciate Islamic Azad University, West Tehran Branch for providing equipment to conduct this study.

All authors equally contributed to preparing this article.

All relevant data are included in the paper or its Supplementary Information.

The authors declare there is no conflict.