The diversity and function of the in-situ fungal communities in response to polycyclic aromatic hydrocarbons in the urban wetland

PAHs (polycyclic aromatic hydrocarbons) increases the potential harm to ecosystem and human health. The fungi is considered as a powerful choice for degradation of PAHs. The researches on the effect of PAHs on fungal population in sediment/soil mostly stayed in the laboratory simulation that is based on extreme pollution. This study investigated the fungal population of the urban wetland by high-throughput sequencing in-situ micro-pollution state. Our statistical analysis revealed significant difference in the whole fungal population at the phylum among three land use types in typical urban wetland. Among them, Ascomycota was the dominant fungi at the phyla in three land use types. Fungal genus of degrading PAHs were significantly correlated with Dibenz[a, h]anthracene (P = 0.018) in ditch wetland, Total Organic Carbon (P = 0.02) and Fluoranthene (P = 0.04) in riverine wetland, and Electrical Conductivity (P = 0.018) in agricultural land. PICRUSt (Phylogenetic Investigation of Communities by Reconstruction of Unobserved States) suggested that 20 enzymes were present related to PAHs metabolism in three land use types. Specifically, monoxygenase, dehydrogenase, and laccase were most abundant among inferred enzymes, indicating that the urban wetland had potential for the degradation of PAHs. This study contributed to in-depth understanding of the structure and function of fungal population and provided a theoretical basis for PAHs microbial remediation in the in-situ environment.


Introduction
Polycyclic aromatic hydrocarbons (PAHs) are a group of organic pollutants that are composed of two or more fused aromatic rings (Page et al. 1999). PAHs cause serious global environmental concerns for both ecosystems and human health because of their potential toxicity and carcinogenicity (Liang et al. 2017;Nasrin et al. 2018). Worryingly, PAHs are closely related to human activities such as settlement, transport, and industrial development (Yang et al. 2014;Banan et al. 2018) and prevalent in urbanized regions (Ren et al. 2015). About 43,000 t/a of PAHs are discharged into the environment by human in the world, and the emission of China is more than 25,000 t/a (Zhang et al. 2007), and the proportion of samples over national standard of PAHs concentration in domestic soil reached 1.4% (Ministry of Ecology and Environment 2014).
PAHs are difficult to centralized control, and microorganisms are placed great expectations as the main executors of purification (Samanta et al. 2002). Among them, fungi have become the main force of PAHs degradation. They have advantages over bacteria because of the capabilities, which are to grow on a variety of substrates, to produce extracellular hydrolytic enzymes and to hydroxylate PAHs. In addition, these polar intermediates produced after hydroxylation can be mineralized by soil bacteria and/or detoxified Communicated by Erko Stackebrandt. to simpler non hazardous compounds (Wang et al. 2008). The in-situ degradation of PAHs may involve the synergy of a variety of microorganisms (De Lorenzo. 2008), which is the long-term interaction between microorganisms and the natural environment. Therefore, exploring the relationship between environmental factors and fungal population is conducive to further exploring the mechanism of degradation of PAHs by fungi.
Next generation sequencing, taxonomic reference data bases and bioinformatics tools provide a great opportunity to characterize the complex relationship between microbial population and PAHs in detail. Compared with metagenomic research, PICRUSt (Phylogenetic Investigation of Communities by Reconstruction of Unobserved States) based on highthroughput sequencing is more convenient, faster, cheaper (Liangille et al. 2013). At present, this method has been applied in the research of lake and marine microorganism (Ward et al. 2017).
According to the classification established by Maliszewska-Kordybach (1996) in soil and Baumard (1998) in sediment, the total concentration of PAHs exceeding 5000 ng/ g −1 belongs to severe pollution. At present, the research on the effect of PAHs on fungal population in sediment/ soil mostly in the laboratory, based on extreme pollution (Mukherjee et al. 2014). 800 mg/kg −1 PAHs in soil was researched to determine the degradation rate of PAHs by fungi (Potin et al. 2004). The maximum concentration of 16 PAHs was 9180.05 ng/L −1 in the water bodies from the seven basins in China (Fan et al. 2019). The biodegradation of 15 mg/L Benzo [a] pyrene was studied in water (Yao et al. 2015).The degradation of 50 and 100 mg/L Naphthalene has been reported in water (Jiang et al. 2018). Compared with the pollutant concentration in the laboratory, PAHs in the actual environments keep in a "micro-pollution state". At this stage, the research results of indoor extreme pollution or high concentration pollution are difficult to explain the degradation and transformation of PAHs in the in-situ environment. Soil and sediment were the main carriers of PAHs in the environment. The content could reach the order of mg/kg −1 and more than 100 times higher than that in water when the concentration of PAHs in water was converted from mg/L to mg/kg (Li et al. 2015;Ma et al. 2015;Fan et al. 2019). Moreover, PAHs in soil and sediment have larger distribution coefficient and smaller fugacity . Chaohu Lake is the fifth largest freshwater lake in China. The Shiwuli River is the principal source of pollutant flowing into Chaohu Lake. Half of the water source of the Shiwuli River currently comes from the tail water of the sewage treatment plant . The total design scale of the treatment plant is 100,000 tons of sewage per day, and the planned total service area is 131.4 square kilometers. The Shiwuli River is a typical urban river. The wetland in the lower reaches of Shiwuli River is an important local water landscape and urban wetland. Urban wetland is an important part of urban ecological environment and a typical area of the interaction between water and land (Li et al. 2020). It has been disturbed by complex hydrodynamic and human activities for a long time, resulting in increased wetland pollution, which seriously affects the ecological function of urban wetland (Li et al. 2009). To explore fungi of degrading PAHs in the in-situ environment and the potential of degrading PAHs by fungi, soil and sediment in three land use types of the lower reaches of Shiwuli River were studied.
Baohe is the junction of the built-up area in the upper reaches and the agricultural area in the lower reaches of Shiwuli River. There is a sewage treatment plant nearby. The sampling point here can detect the pollution situation from the upper reaches. The estuary is directly connected with Chaohu Lake, which can directly detect the pollutants entering the lake. Baohe and estuary are selected as sampling points. The aims of the present study were (1) to investigate fungal population among three land use types; (2) to determine the diverse response patterns of PAH-degrading fungi to environmental factors; (3) to predict the inherent potential of fungi to degrade PAHs.

Sample collection
We extracted 6 columnar samples from sites representing three land use types (ditch wetland, agricultural land, and riverine wetland) ( Fig. 1; Table 1) with a TC-600H piston columnar mud collector (Qingdao Daneng Environmental Protection Equipment Co., Ltd., China). Cores were collected from the lower reaches of the Shiwuli River in Chaohu Lake, China, from July 20-27, 2016. Core sampling was conducted in triplicate from each sampling site. Sampling sites were selected and referenced relative to the location of intersection of Shiwuli River and Baohe Avenue, which is the border between the urban and rural areas of City (Fig. 1). Sites 1, 2, and 3 were 2 km downstream of the intersection (31° 45′ N and 117° 19′ E). Sites 4, 5, and 6 were 8 km downstream of the intersection (31° 44 N′ and 117° 23′ E). Every sediment/soil core from each site was stratified into six layers (0-5 cm, 5-10 cm, 10-15 cm, 15-20 cm, 20-25 cm, 25-30 cm). We collected a total of 18 cores (6 sites × 3 cores), and the same layer of the triplicate cores were mixed together, creating a composite core sample for each depth of each sampling site on the basis of the sampling methods (Li et al. 2017;GB17378.5-2007 andHJ/T 91-2002). Every sample was then separated into two sub-samples. All samples were placed in polyethylene bags, transported on ice, delivered to the laboratory within 24 h, and stored at − 80 °C until analysis.

Analysis of sediment/soil physiochemical parameters and PAHs
pH was measured after shaking a sediment: water (1:5 wt/vol) suspension for 30 min (Shen et al. 2013). Total organic carbon (TOC) was determined by the potassium dichromate oxidation-external heating method according to the national standard (GB7857-1987 and LY/T1237-1999). An electrical conductivity meter determined the electrical conductivity (EC), which was determined using a 1:2.5 wt/ vol sediment-to-distilled-water mix . Total nitrogen (TN) was measured by continuous flow analyzer (BRAN + LUBBEIII, Germany). PAHs were extracted from soil and sediment samples by accelerated solvent extractor after freeze-drying and sieving. The extraction was purified, concentrated, and analyzed by gas chromatography mass spectrometry (GC-MS, Agilent7890A/5975C, Agilent Technologies Inc., CA, USA). Details of PAHs analyses were based on previous studies (Wu et al. 2019). Equipment details, calibration, quantification, and quality control procedures were represented in Supplementary Text.

DNA extraction and PCR amplification
Microbial community genomic DNA was extracted from soil/sediment samples using the EZNA® soil DNA Kit (Omega Bio-tek, Norcross, GA, US) according to manufacturer's instructions. The DNA extract was checked on 1% agarose gel, and DNA concentration and purity were

Statistical analysis
Variance inflation factor (VIF) was measured to judge the colinearity among different factors, and environmental factors with VIF > 10 were removed from the following analysis. PLS-DA (Partial Least Squares Discriminant Analysis) was applied to group the samples. The differences of the abundance of fungal phylum and predicted enzymes among the three land use types were tested by One-way ANOVA. Redundancy analysis (RDA) was used to assess the relationship between fungal population, individual PAHs degrading fungus and the selected environmental factors. The software CANOCO (version 4.5) was used to perform RDA. As the abundance of predicted functional genes changed greatly, the abundance was transformed (ln). Heatmap illustrator (heml 1.0) displayed the abundance of predicted function genes.

PAHs concentrations and physicochemical properties
The total concentrations of PAHs, which referred to the group average, ranged from 36.5-1031.8 ng/g, with the highest value in the Baohe ditch wetland. Overall, the total PAHs in the soil/sediment followed this order: ditch wetland (765.97 ng/g group average) > riverine wetland (181.73 ng/g group average) > agricultural lands (59.19 ng/g group average). According to the classification established by Maliszewska-Kordybach (1996), most of sediment/soil in the urban wetland in this study was moderately contaminated.
In particular, some sampling points in the ditch wetland were heavily contaminated. 16 priority PAHs were detected in the soil/sediment, indicating that PAHs were distributed widely throughout the lower reaches of Shiwuli River (Table S1). pH varied from 6.40 to 7.24 at the study area. The mean value of TOC and Nitrate nitrogen in the riverine wetland (10.88, 1.67 mg/kg) were higher than that in ditch wetland (8.87, 1.54 mg/kg) and agricultural land (5.33, 1.54 mg/kg). Among the three land use types, EC changed greatly. The electrical conductivity of ditch wetland was higher than that of agricultural land. PAHs and physicochemical properties were derived from the previous research (Wu et al. 2021).

Diversity and composition of the fungal population
To investigate the diversity and structure of the fungal population, we used Illumina high-throughput sequencing technology to obtain the sequence of the ITS. After filtering the low-quality reads and trimming the adapters and barcodes, there were 2,165,949 effective sequences with an average length of 242 bp of microbiota generated from the soil/ sediment. Based on a 97% sequence similarity threshold, 5193 OTUs were identified, of which 8.4% was found in only a single sample. The shape of the rarefaction tended to approach the saturation plateau, indicating near complete sampling of endemic fungal richness (Figs. S1and S2). Based on the OTUs number, the sediment in the ditch wetland was bound to have the richest diversity, with OTUs number of 1082 ± 481, followed by the samples in the Riverine wetland and in the agricultural land with OTUs numbers of 458 ± 223 and 352 ± 134, respectively. In detailed, fungal population in the ditch wetland had the maximum diversity, with Ace index of 1440 ± 671, Chao1 index of 1392 ± 649, Shannon index of 4.07 ± 0.93 and Simpson index 0.11 ± 0.1. In this study, Alpha-diversity of fungal population based on OTUs in the riverine wetland and the agricultural land was close (Table 2).
The results of PLS-DA show that the samples of three land use types could be distinguished and clustered into three groups. Nevertheless, our statistics analyses revealed significant difference of the main dominant phylum among the three land use types (Fig. 2). In this study, we found that Ascomycota was the first dominant phylum in the Riverine wetland and the Agricultural land while Rozellomycota was the first dominant phylum in the ditch wetland. Ascomycota was the main dominant phylum in three land use types.

Response of fungal population to environmental factors
Environmental factors were selected by function of vif. cca, and environmental factors with vif > 10 were removed from the following analysis. In this study, six PAHs and four physicochemical parameters, including pH, EC, TN, and TOC, were taken into consideration to evaluate the relative contributions to fungal population (Fig. 3). The CCA of the fungal population showed that average 18.88%  of the variability in the data was explained by the first two principle components (CCA1 and CCA2). The first axis and the second axis could be well explained by the environmental variables. The first axis mainly reflected the environmental information other than TOC, and the second axis reflects TOC information. PAHs associated with anthropogenic activities can be transported and mixed through river runoff during the interaction of rivers and lands (Wu et al. 2021). Owing to rapid urbanization, the PAHs contamination to the environment is accelerating. Microorganisms are the main executors of degradation. PAHs can be used as carbon sources by microorganisms, that can result in the increase of abundance and species richness. Some studies have shown this effect, especially on fungi (Jones 2000;Cordova-Kreylos et al. 2006). In the study, the concentrations of PAHs in ditch wetland were significantly higher than that in the other two land use types. This might explain the significant effect of PAHs on the fungal population of ditch wetland sediment. There was a strong correlation between TOC and fungal population in riverine wetland. Some sediment samples of the riverine wetland showed high TOC with average concentrations > 20 mg/g. TOC was significantly higher than that in other two land use types. Fungi may also use different C sources more efficiently than bacteria (Ellen et al. 2019). It was reasonable to suspect that these dynamics might lead to a strong correlation between fungal population and TOC in riverine wetland.
In addition, pH significantly affected the fungal population in agricultural land consistent with previous studies. It suggested that soil pH was an important environmental factor affecting fungal population, which had significant effects on the growth and reproduction of soil fungi and the secretion of extracellular enzymes (Sheng et al. 2018).

Correlation and response of individual PAH-degrading fungi to environmental factors
According to the previous research results of PAHs degrading fungi, 10 PAH-degrading fungal genera were identified in this study (Table 4). We specifically focused on these genera in the following discussion. The correlation between fungus which could degrade PAHs and environmental factors among different sampling points was illustrated (Fig. 4). Partial RDA based on a Monte Carlo permutation (n = 499) kept only the significant parameters in the models, indicating that these environmental factors might be important for explaining the fungi. Among PAHs degrading fungal genera reported in literatures, Talaromyces and Aspergillus were abundant PAHs degrading fungal genera in ditch wetland. Talaromyces and Fusarium were abundant PAHs degrading fungal genera in riverine wetland. Talaromyces and westerdykella were abundant PAHs degrading fungal genera in agricultural land. Fungal genus, such as Fusarium, Talaromyces, Trichoderma and Westerdykella, were the dominant PAH-degrading fungi in this study. This was consistent with a previous study showing that Trichoderma and Fusarium were representative PAH-degrading fungal genus (Cordova-Kreylos et al. 2006). DhA (P = 0.018) was more important (Fig. 4a; Table 5). In contrast, other variables (P > 0.05) had no significant correlations with fungi which could degrade PAHs. The first and second axes explained 72.3% and 16.9% of the total variance, respectively, in the ditch wetland. DhA was the most important environmental factor related to the distribution of PAH-degrading fungi in the ditch wetland. Previous study indicated PAHs in contaminated soil solution was used as the unique carbon source (Sheng et al. 2018). DhA was positively correlated with Talaromyces, Trichoderma, and Westerdykella, suggesting that these taxon could degrade DhA or the increasing input of DhA linked with urbanization. Most fungi may co-metabolize PAHs to a wide variety of oxidized products and in some cases to CO 2 (Ghizelini et al. 2019). However, DhA was negatively correlated with Fusarium, and Fusarium had been identified as important  HMW (high-molecular weight) PAH degraders (Cordova-Kreylos et al 2006). In riverine wetland, there were 2 variables TOC (P = 0.02), FLT (P = 0.04), that explained 61% of the total variation of fungi which could degrade PAHs ( Fig. 4b; Table 5). In contrast, other variables (P > 0.05) had no significant correlations. The first axis explained 72.3% of the total variance. The second axis explained 16.9% of the total variance for the riverine wetland. TOC and FLT were also important environmental factors related to the distribution of PAH-degrading fungi in the riverine wetland. A study showed that the quality and quantity of organic substrates could affect fungal diversity and composition (Waldrop et al. 2006). The Baohe samples exhibited high TOC values, with average concentrations > 20 mg/g. Consequently, these high values might underlie the strong correlation between PAHdegrading fungi and TOC in riverine wetland. In addition, the concentration of FLT was the highest among 16 PAHs in the riverine wetland (10% average). This might be the reason why FLT was positively correlated with the main degradation PAH-degrading fungi.
EC (P = 0.018) was more influential for fungi compared to other environmental factors ( Fig. 4c; Table 5). The others did not have significant correlations. In the agricultural land, the first axis explained 41.1% of the total variance and the second axis explained 18.9%. In contrast to the other sediment, EC was the most important environmental factor related to PAH-degrading fungi in the agricultural land. Soil electrical conductivity is closely related to soil properties. Soil electrical conductivity can reflect soil salinity, moisture, porosity and other parameters (Smiarowski et al. 2011).
PAH-degrading fungi were nevertheless primarily associated with organic compounds or physicochemical variables in three land use types. FLT was the most studied PAHs. Compared with FLT, DhA was less studied. They were listed as priority pollutants due to known or suspected carcinogenicity, teratogenicity and acute toxicity by the US Environmental Protection Agency (Juhasz et al. 2000). 16 PAHs were divided into low-cyclic PAHs (LPAHs, 2-and 3-ring PAHs) and high-cyclic PAHs (HPAHs, 4-, 5-, and 6-ring PAHs) (Yunker et al. 2002). FLT and DhA belong to HPAHs. HPAHs are more stable in the natural environment ).
Degradation of indigenous microorganisms has been recognized as a cost-effective alternative to remove HPAHs pollution in the environment . As a result of such a large experience, the fungi emerge as a powerful choice for degradation of PAHs. They have advantages over bacteria due to their capability to grow on a large spectrum of substrates at the same time (Sheng et al. 2018). Fungi have more obvious advantages in the degradation of HPAHs. These might the reasons.
PAH-degrading fungi responded to the presence of FLT of the ditch and DhA of the riverine wetland in this study.

Prediction of degradation of PAHs by fungi
To understand fungal degradation of PAHs in the lower reaches of urban rivers, the metabolic pathway of PAHs in the Kyoto Encyclopedia of Genes and Genomes (KEGG) database was predicted by PICRUSt (Phylogenetic Investigation of Communities by Reconstruction of Unobserved States). PICRUSt was applied to predict functional metagenomes of three land use types based on the retrieved ITS information.
It was speculated that there were 20 enzymes related to PAHs metabolism in three land use types, indicating that there might be abundant genes related to PAH metabolism in the samples (Fig. 5). The relative abundance of inferred enzymes varied greatly (Fig. 6), and 15 of the 20 enzymes were significantly different among the three land use types (P < 0.01). Enzymes inferred from the three In general, we observed that under the same disturbance conditions the enzyme related to PAH degradation was different with the whole fungal population and the main PAH-degrading fungi. Although the PAH-degrading fungi changed under different conditions, functional redundancy might mitigate any turnover in functional capacity of fungi. In other words, the response of PAH-degrading fungi to external disturbances might be apparently more sensitive than the enzyme. This was similar to PAHs degrading bacteria and consistent with previous studies (Wu et al. 2021;Berga et al. 2012;Wang et al. 2017).
In different land use types, monooxygenase was the most abundant PAH-degrading related enzyme, which was more abundant than dehydrogenase and laccase. Monooxygenase, dehydrogenase and laccase played an important role in the degradation of organic matter (Hadibarata et al. 2012;Haritash et al. 2009) indicating that the urban wetland had a certain degradation potential of organic pollutant.

Conclusions
In this study, statistical analyses suggested that PAHdegrading fungi responded to PAHs and other physicochemical variably among the three land use types. Fungal population functional inference by PICRUSt suggested that diverse enzymes related to PAHs metabolism were present among the three land use types, indicating that the urban wetland might have degradation potential for organic pollutant.
Future work will focus on the functional groups of PAHs degrading fungi. The correlation between these dominant fungal genus of degrading PAHs and environmental factors will be quantitatively studied in the laboratory.
This experiment was only for in-situ characterization of fungi that might be involved in PAHs degradation. The generalizability/transferability of the findings reported in this manuscript to broader researches is unclear and needs a lot of experimental exploration in future.

Fig. 6
Heatmap of functional profiles inferred by PICRUSt in three land use types. Red colors represent higher abundance and blue colors lower abundance. Because of the large difference of the value, the abundance value are logarithmic. The left side color bars and the right side, respectively, represents the type and the name of the enzyme of the KEGG pathway of PAH degradation Author contributions HW, SP and BS conceived and designed the experiments. HW performed the experiments. HW, BS and JL analyzed the data and wrote the original draft preparation. JL revised the manuscript.
Funding Research in this study was funded by the National Key Scientific Instrument and Equipment Development Grant of China (2012YQ2011308) and Natural Science Foundation of Anhui Provincial Education Department (KJ2019A0553).
Data availability Raw amplicon sequence data related to this study were deposited in the NCBI Sequence Read Archive (NCBI SRA) under Bioprojects PRJNA817351.