Purification capacity of natural biofilms and their physiochemical and biological properties: a case study in the Jishan River, a heavily polluted river

Although the river is the sink for the discharge of terrestrial pollutants, it has certain self-purification which can be used to maintain the ecological health and balance of the river. Self-purification of water refers to the reduction of pollutant concentration and partial or complete restoration to original river levels through natural processes, including physical, chemical, and biological purification and their interactions (McColl 1974; Vagnetti et al. 2003; Semenov et al. 2019; Cai et al. 2021). Among them, the physical self-purification mainly includes dilution, adsorption, sedimentation, volatilization, etc. Chemical self-purification contains neutralization reactions, REDOX reactions, precipitation reactions, adsorption, and condensation reactions. Notably, biological self-purification includes plant absorption, enrichment, and transformation as well as microbial assimilation and dissimilation, microorganism secretion of extracellular polymer flocculation, sedimentation, polymerization, etc. (Wu et al. 2012; Han et al. 2015; Li et al. 2021b). Since physical and chemical processes are affected by biological factors, biological self-purification is the core of the overall self-purification process (Ostroumov 2017). The participation of microorganisms in the process of biological self-purification is the main reason for the removal of pollutants (Battin et al. 2016).

Approximately 95% of all aquatic microorganisms can adhere to the inner surface of pipe materials, forming a biofilm, and only 5% are floating in the water column (Hemdan et al. 2019). Natural biofilms, which are widely distributed in various aquatic environments, are microbial assemblages of algae, fungi, bacteria, protozoa, metazoans, and other micro-biotic and abiotic matters in which cells are frequently embedded in their self-produced matrix of extracellular polymeric substances (EPSs) (Costello et al. 2016; Zhang et al. 2022). As a vital component of metabolic products, EPSs have been considered to be representative of a biofilm structure and function (Felz et al. 2019). Organic matter biodegradation, primary production, and cycling of nitrogen, phosphorus, and sulfur are directly regulated by processes driven by biofilms (Costello et al. 2016; Zhang et al. 2022). A previous study has demonstrated that contaminants could be adsorbed and degraded by biofilms through different mechanisms, including electrostatic, cation exchange, complexation, hydrophobic, and micropore filling interactions (Huang et al. 2018). Hence, biofilms have been proposed as natural bioindicators to reflect the spatial distribution of anthropogenic pesticides (Tien et al. 2013; Zhang et al. 2022). Furthermore, biofilms provide major food sources that fuel the secondary production of invertebrates and fishes, thus making the related pollutants trophically transferable to the food chain (Costello et al. 2016; Zhang et al. 2022). More importantly, biological contact oxidation technology based on natural biofilm technology has been used for heavily polluted water treatment with excellent results (Wu et al. 2012). These suggest that biofilms play important roles in the transport and transformation of pollutants in aquatic environments.

Previous studies have proved that river biofilms contribute significantly to the self-purification of the river (Tien et al. 2013; Hou et al. 2022). Moreover, several studies have paid attention to the variation in the microbial community structure of river biofilms during the biodegradation processes and the effects of hydrodynamics on biofilm purification abilities (Tien et al. 2013; Hou et al. 2022). However, little was known about the inherent purification capacity of natural biofilms for different pollution indicators of heavily polluted water bodies and their physiochemical and biological properties. We hypothesized that natural biofilms had strong wastewater purification capacity. Herein, we explored the purification effects of natural biofilms by outdoor sampling and indoor experiments and further investigated the physiochemical and biological properties of natural biofilms to reveal their purification mechanism (Supplementary material, Figure S2). The main goals of this study were (1) the inherent purification capacity of natural biofilms; (2) the microbial community structure and function of natural biofilms; and (3) the physiochemical properties of natural biofilms. These results provide a new perspective on the in situ remediation function of natural biofilms and reveal the potential biological mechanisms of natural biofilms for the remediation of polluted water bodies.

The experiment described in this paper was conducted in the Jishan River (Supplementary material, Figure S1) of Jinmen, Hubei province, China (Zhu et al. 2022b). The length of the river is 4,150 m (latitude 30°29′N, longitude 112°10′E). With several chemical factories and soybean processing factories, residential areas, and rapeseed fields along its banks, this river is an important source of drinking and irrigation for local residents, as well as a major discharge area for industrial effluent, which plays an important role in the normal life and economic development of local residents. There are a lot of aquatic plants downstream of the river. A large number of clearly visible loose porous, mature white flocculent biofilms were attached to the plant's surface and between the plants and the substrate above the water surface. Five representative surface water samples were collected along the river in November 2020 (Supplementary material, Figure S1). The collection, preservation, and transportation of water samples were carried out in accordance with the Standard Method for Water and Wastewater Monitoring (Ministry of environmental protection of China 2002). Duplicate natural biofilm samples were scraped from aquatic plants and collected in sterile centrifuge tubes. All samples were refrigerated and brought back to the laboratory. The water samples were divided into two parts: one was stored in a refrigerator at 4 °C for the determination of physiochemical parameters, and the other was filtered through a 0.22-μm membrane and then stored at −20 °C for molecular analysis.

The laboratory experiments on the self-purification ability of natural biofilms were constructed in duplicate in 500-mL conical flasks containing 250 mL of synthetic black-odor water and incubated at 30 °C with moderate shaking. Four treatments were conducted: sterilized black-odor water (SW), black-odor water (W), black-odor water + 0.1 g biofilms (BW), black-odor water + 0.1 g sterilized biofilms (SBW). Approximately 4.0 mL of mixtures was taken out at a flexible interval of 1.0–5.0 days for detecting physiochemical parameters. To ensure the reproducibility of the experimental results, the experiments were conducted with synthetic black-odor water. Working guidelines for the treatment of urban black-odor water were issued by the Chinese Ministry of Housing and Urban-Rural Development in 2015 (Chinese Ministry of Housing and Urban-rural Development 2015), defining black-odor waterbodies as those that present with unpleasant colors and/or emit stinky odor. Synthetic black-odor water formulation was modified based on He et al. (2021), and its specific protocol was described in the Appendix. The physiochemical characteristics of this black-odor water are shown in Supplementary material, Table S1. The sterilized biofilms were prepared by autoclaving at 121 °C for 25 min.

Physiochemical parameters were analyzed after filtering the water samples through 0.45-μm membranes. The concentrations of ammonia nitrogen -N), nitrite nitrogen (⁠-N), and nitrate nitrogen (⁠-N) were determined using Nadler's reagent colorimetry (GB/T 7479-87), sulfanilamide coupled with N-(1-naphthyl)-ethylenediamine (GB/T 7493-87), phenoldisulfonic acid spectrophotometry (GB/T 7480-87), respectively. Total phosphorus (TP) and Chemical oxygen demand (COD) were measured using the ammonium molybdate spectrophotometry method (GB 11893-89) and the potassium dichromate method (GB/T 34500.2-2017).

To explore the community structures and diversities of natural biofilms, high-throughput sequencing of the 16S rRNA and internal transcribed spacer-1 (ITS1) gene of biofilms were conducted (Zhu et al. 2022a). The total DNA was extracted from the biofilms with the DNeasy^® PowerSoil^® Kit (100 T, Qiagen, Germany) following the manufacturer's instructions. Electrophoresis and NanoDrop 2000 UV–vis spectrophotometer (Thermo Scientific, USA) were used to quantify the DNA quality and concentration. The V3-V4 hypervariable regions of the bacteria 16S rRNA gene were amplified to profile bacterial community structures of natural biofilms with primers (338F: 5′-ACTCC-TACGGGAGGCAGCAG-3′, 806R: 5′-GGACTACHVGGGTWTCTAAT- 3′). The fungal ITS1 was amplified with primers (ITS1F: 5′-CTTGGTCATTTAGAGGAAGTAA-3′, ITS2R: 5′-GCTGCGTTCTTCATCGATGC-3′). After qualified library preparation, DNA sequencing was performed on the Illumina HiSeq PE300 platform at Shanghai Meiji Biomedical Technology Co., Ltd. Raw reads of gene amplicons were quality-controlled and analyzed using QIIME2 0.

The total structural EPSs was extracted from the biofilms with a mild temperature Na₂CO₃ extraction method as described previously (Felz et al. 2016). In brief: a biofilm sample of 6.0 g (wet weight) was added into a 0.5% (w/v) Na₂CO₃ solution up to 100 mL and subsequently stirred for 35 min at 80 °C in a water bath., followed by centrifuge at 4,000 × g and 4 °C for 20 min. The organics in the supernatant comprised the total extractable EPSs. The protein contents (PN) were measured by a modified Lowry method with bovine serum albumin (Sigma, USA) as standard (Griebe & Nielsen 1995; Frølund et al. 1996). The polysaccharide concentration (PS) was determined by phenolsulfuric acid assay using glucose (Sigma, USA) as standard. The DNA content was calculated by a diphenylamine colorimetric method (Zhao et al. 2013).

The microstructure of biofilm was observed and analyzed by scanning electron microscope (SEM; TESCAN MIRA). 5.0 g biofilm samples were taken out and washed with phosphate buffer solution (pH 8.0) for three times. The supernatant was removed to retain the precipitation. The precipitated samples were immersed in 2.5% glutaraldehyde solution at 4 °C overnight. The samples were then washed three times with phosphate buffer solution (pH 8.0) to remove glutaraldehyde for 10 min each time. The biofilm samples were dehydrated in a gradient of 30, 50, 70, 90, and 100% ethanol solutions (v/v) with a concentration gradient of 15–20 min each, adding a 1:1 solution of anhydrous ethanol: isoamyl acetate, then removing the supernatant and adding isoamyl acetate for 15 min to remove the precipitates from the supernatant. The samples were dried in a vacuum freeze dryer, followed by gold spraying in an ion sputtering apparatus (SC 7620), and finally, the samples were observed and photographed using SEM.

The main functional groups of natural biofilms were conducted by using Fourier transform infrared spectrometer (FTIR, Cary 630, Agilent, USA). Lyophilization of natural biofilms was compressed with potassium bromide (KBr) in the mass ratio of 1:200 into tablets. FTIR measurements consisted of 16 scans with a resolution of 4 cm⁻¹ from 4,000 to 400 cm⁻¹ (Li et al. 2020; Ugya et al. 2021).

Experiment data were presented statistically as mean and standard deviation. Excel 2016 (Microsoft, USA), OriginPro 2022 (OriginLab, USA) were used for data processing and mapping. The sequences obtained by Miseq sequencing were first splintered according to FLASH (1.2.11), and the quality of the sequences was controlled and filtered. UPARSE (7.0.109) was used for OUT clustering and annotation. MOTHUR software was used for bacterial Alpha diversity analysis, Uparse is used to process OUT data, and the community structure of the sample was analyzed by R software. FAPROTAX and was used to predict bacterial metabolic functions. FUNGuild was used to distinguish communities and predict fungi functions.

Table 1 shows the detailed water quality physiochemical characteristics, such as COD, -N, -N, -N, and TP at 6 different sampling sites along the Jishan River. Results revealed that the water quality of the Jishan River is worse than Grade V based on the Environmental Quality Standards for China (GB3838-2002). The concentrations of COD, -N, -N, -N, and TP ranged from 52.4 to 1,020 mg/L, 2.10 to 13.60 mg/L, 0 to 0.38 mg/L, 0.47 to 2.50 mg/L, and 0.34 to 0.84 mg/L, respectively. Site A, as the source of this river, showed the highest concentrations of COD, -N, -N, and -N. Site B, as the second sampling site next to site A had the highest concentration of TP. It was worth mentioning that the content of these physiochemical characteristics generally showed a decreasing trend with the course of this river. Therefore, it could be presumed that this river had natural self-purifying capacities. Moreover, site F had the lowest levels of COD, -N, -N, and TP. Observation in the field also revealed a large amount of white flocculent biofilms attached to aquatic plants at site F (Supplementary material, Figure S2). We initially hypothesize that these biofilms have a strong connection with the natural self-purification ability of the river. Biofilm samples were taken for indoor test verification.

Table 1

Water quality physiochemical characteristics of six different sampling sites along the Jishan River

.	A .	B .	C .	D .	E .	F .
COD (mg/L)	1,020 ± 16	222 ± 24	187.60 ± 8	127.79 ± 16	108 ± 0	52.4 ± 0
-N (mg/L)	13.6 ± 0.12	7.08 ± 0.34	5.24 ± 0.20	6.63 ± 0.37	4.94 ± 0.14	2.10 ± 0.76
-N (mg/L)	0.38 ± 0	0.01 ± 0	0 ± 0	0 ± 0	0 ± 0	0 ± 0
-N (mg/L)	2.50 ± 0.07	1.21 ± 0.04	1.27 ± 0.02	1.30 ± 0.01	0.54 ± 0.03	0.47 ± 0.02
TP (mg/L)	0.44 ± 0.01	0.84 ± 0.24	0.78 ± 0.01	0.80 ± 0	0.47 ± 0.01	0.34 ± 0.03

.	A .	B .	C .	D .	E .	F .
COD (mg/L)	1,020 ± 16	222 ± 24	187.60 ± 8	127.79 ± 16	108 ± 0	52.4 ± 0
-N (mg/L)	13.6 ± 0.12	7.08 ± 0.34	5.24 ± 0.20	6.63 ± 0.37	4.94 ± 0.14	2.10 ± 0.76
-N (mg/L)	0.38 ± 0	0.01 ± 0	0 ± 0	0 ± 0	0 ± 0	0 ± 0
-N (mg/L)	2.50 ± 0.07	1.21 ± 0.04	1.27 ± 0.02	1.30 ± 0.01	0.54 ± 0.03	0.47 ± 0.02
TP (mg/L)	0.44 ± 0.01	0.84 ± 0.24	0.78 ± 0.01	0.80 ± 0	0.47 ± 0.01	0.34 ± 0.03

View Large

The observation of SEM (Figure 4(b)–4(d)) showed the microscopic morphology and the formation of natural biofilms. The lyophilized biofilm samples were white sponge-like with uniform and elastic textures. The morphology of the biofilms at 100 μm was shown in Figure 4(b), which can be seen in the form of bands and blocks, with an overall loose and porous appearance. When this biofilm sample was magnified to 10 μm, the presence of a large number of rod-shaped microorganisms, more discontinuous pores, and complex embedding of extracellular polymers with bacterial colonies to form a dense layer of colonies in various forms such as blocks or long strips could be observed. Continuing to zoom in to 2 μm, the different sizes of bacilli can be seen more clearly attached to the extracellular polymer in a lumpy form, and it can also be seen that the extracellular polymer is the backbone of the entire biofilm, providing a habitat for microorganisms.

Most studies at present have only focused on the structural properties of EPSs extracted from activated sludge in different wastewater treatment processes (Felz et al. 2019; Lotti et al. 2019; Guo et al. 2020). However, few studies have addressed the physiochemical properties of nature biofilms in heavily polluted rivers (Cheng et al. 2018; Ramazanpour Esfahani et al. 2021). Compared with previous studies, our study comprehensively reveals the unique physiochemical and biological properties of natural biofilms for the first time. As found in previous studies, the main components of natural biofilms are polysaccharides and proteins (Guo et al. 2020). Moreover, the FTIR spectra depicted the functional groups of natural biofilms also contained carbohydrates, lipids, amines, methyl, and methylene (Figure 5). The interaction and combination of these specific functional groups were beneficial to the flocculation, aggregation, and adhesion of microorganisms (Jiang et al. 2021). As parts of the active composition of biofilms, these microorganisms play an important role in the biogeochemical cycling and biodegradation of pollutants (Manirakiza et al. 2022). Therefore, the biological properties of natural biofilms require further analysis. Previous studies have reported the microbial community structures of naturally grown biofilms and biofilms formed in wastewater treatment processes, respectively (Supplementary material, Table S2). Compared to these biofilms dominated by bacterial phylum Proteobacteria, Firmicutes, Actinobacteria, and Bacteroidetes, natural biofilms have unique bacterial communities, the most abundant populations belonged to the phyla Campilobacterota, Proteobacteria, Bacteroidota, Firmicutes, and Desulfobacterota (Figure 2(a)). In addition, the fungal community of natural biofilms appeared to be particularly distinctive compared to previously reported studies (Supplementary material, Table S2). The unique physiochemical and biological properties of natural biofilms might lead to its special functions.

We further analyzed the potential mechanisms of natural biofilms for the remediation of polluted waterbodies based on previously reported studies. The primary microbial communities of natural biofilms at the phylum level were Campilobacterota, Proteobacteria, Bacteroidota, Firmicutes, and Desulfobacterota (Figure 2). Campilobacterota were rarely found in either naturally grown biofilms or biofilms formed during wastewater treatment processes but were well known for degrading organic matter (Zhu et al. 2022b). Proteobacteria typically occupy dominant proportions in the aquatic ecosystem and various municipal wastewater treatment reactors (Yuan et al. 2020; Zhang et al. 2020). Moreover, Proteobacteria were directly associated with ammonia oxidization and denitrification processes (Cai et al. 2016; Song et al. 2021), which might be the reason why indoor experiments showed efficient ammonium and nitrate removal performance. Bacteroidota is always involved in the degradation of carbohydrates (Li et al. 2021a). Moreover, a previous study has found that Proteobacteria and Bacteroidota play important roles in denitrifying phosphorus removal (Qu et al. 2021). Firmicutes, which participated in refractory organic compounds decomposition and microbial nitrogen fixation, could be promoted to grow by a high concentration of organic matter (Li et al. 2021a). At the genus level, Pseudarcobacter, unclassified_f_Arcobacteraceae, Sulfuricurvum, unclassified_f_Comamonadaceae, Sulfurimonas, Acinetobacter, Maikia, Paludibacter, and Flacobacterium were the most abundant genera in natural biofilms. Pseudarcobacter, which are responsible for the transmitted electrons, are significantly related to nutrient elements metabolisms (Luo et al. 2022). Previous studies demonstrated that many members of the Arcobacteraceae family are nitrate-reducing bacteria (Gulliver et al. 2020). Sulfuricurvum and Sulfurimonas play important roles in the removal of toxic organic matter and nitrate (Li et al. 2019; Fida et al. 2021). Acinetobacter and Comamonadaceae are reported as typical phosphorus-accumulating organisms (PAOs) that are widely applied in wastewater treatment because of their abilities to remove phosphorus (Zhao et al. 2022). Paludibacter were proposed as common sulfate and refractory organics degraders (Liang et al. 2013; Wang et al. 2019a). The presence of Flacobacterium can promote the denitrification ability of natural biofilms at low temperatures (Qu et al. 2021).

At the fungal phylum level, Chytridiomycota have been used in the production of antibiotics, organic acids, hormones, vitamins, and in the brewing industry (Money 2016). Most species in the Ascomycota phylum can decompose cellulose and chitin (Czaplicki et al. 2018; Challacombe et al. 2019). At the fungal genus level, the beneficial contributions of Kazachstania are well-established in various food processes (Spanoghe et al. 2017; Urubschurov et al. 2018). Plectosphaerella, which have previously been shown to be potential biocontrol agents, are effective against potato cyst nematodes (Kusstatscher et al. 2019). These results confirmed that natural biofilms contain a large number of functional microorganisms that can purify polluted water bodies, and are worthy of our further exploration of patterns and strategies for their extensive application in polluted water.

In this study, the inherent purification capacity, physiochemical, and biological properties of natural biofilms collected from a heavily polluted river were synchronously investigated. Outdoor sampling and indoor experiments demonstrated the superior purification effect of natural biofilms on heavily polluted water. Indoor experiments showed that the purification capacity of natural biofilms was dominated by the microbial biodegradation rather than physical biosorption of EPSs, and after 14.0 days of incubation, the removal rates of COD, TP, -N, and -N could reach 93.6, 80.83, 85.93, and 81.03%, respectively. The SEM, FTIR spectra, and components analysis revealed that natural biofilms were mainly composed of polysaccharides and proteins, and also contained small amounts of functional groups such as carbohydrates, lipids, amines, methyl, and methylene. High-throughput Illumina MiSeq sequencing analysis illustrated that the dominant phyla in the bacterial community structure were Campilobacterota (52.65%), Proteobacteria (28.45%), Bacteroidota (9.51%), Firmicutes (4.56%), and Desulfobacterota (1.99%), and the major phyla in fungal community structure were Chytridiomycota (49.42%) and Ascomycota (37.69%). These microorganisms might be the main degraders of riverine pollutants. However, this study only focused on the removal of limited contaminants by natural biofilms and did not identify which microorganisms were involved in the purification processes. Further investigations are recommended to pay attention to the purification effects of natural biofilms on novel pollutants, such as heavy metals and refractory organic pollutants, and to determine the changes in microbial community structures during the purification processes.

W.D. studied methodology, did data analysis and data curation, wrote, reviewed and edited the article. X.Z. conceptualized the study, performed methodology, investigated the study, wrote the original draft, wrote, reviewed and edited the article. W.X. studied methodology. H.P. did project administration, acquired funds, and supervised the study. H.Y. acquired funds and supervised the study. Y.D. studied methodology.

This work is supported by the Educational Commission of Hubei Province of China (No. Q20211310), the Hubei Key Laboratory of Intelligent Yangtze and Hydroelectric Science (No. ZH2002000113, ZH2102000113), the Department of Ecology and Environmental of Hubei Province of China (No. 2015HB17), and the Natural Science Foundation of Hubei Province of China (No. 2020CFB750).

All authors have reviewed, approved, and consented to the publish, and they are accountable for all aspects of its accuracy and integrity.

All relevant data are included in the paper or its supplementary information.

The authors declare there is no conflict.