Effects of Polycyclic Aromatic Hydrocarbons on Soil Bacterial and Fungal Communities in Soils

Abstract

Soil organic pollution (such as heavy metals, PAHs, etc.) has caused serious environmental problems, which have resulted in unexpected effects on contaminated soil ecosystems. However, knowledge of the interactions between environmental PAHs and bacterial and fungal communities is still limited. In this study, soil samples from different PAH-contaminated areas including non-contaminated areas (NC), low-contaminated areas (LC), and high-contaminated areas (HC) were selected. Results of toxic equivalent quantity (TEQ) indicated that Benzo[a]pyrene (BaP) and Dibenzo[a,h]anthracene (DBahA) constituted the main TEQs of ∑16PAHs. Incremental lifetime cancer risk (ILCR) assessment revealed that the main pathway of exposure to soil PAHs was dermal contact in adults and children. Furthermore, adults faced a higher total cancer risk (including dermal contact, ingestion, and inhalation) from soil PAHs than children. The microbial community composition analysis demonstrated that soil PAHs could decrease the diversity of bacterial and fungal communities. The relative abundance of Acidobacteriota, Gemmatimonadota, Fimicutes, Bacteroidota, Ascomycota, and Basidiomycota exhibited varying degrees of changes under different concentrations of PAHs. Benzo[a]anthracene (BaA) and Chrysene (Chr) drove the bacterial community composition, while BaP and DBahA drove the fungal community compositions. Co-occurrence network analysis revealed the high contamination levels of PAHs that could change the relationships among different microorganisms and reduce the complexity and stability of fungal and bacterial networks. Overall, these findings provide comprehensive insight into the responses of bacterial and fungal communities to PAHs.

1. Introduction

Polycyclic aromatic hydrocarbons (PAHs) belonging to persistent organic pollutants (POPs) are semi-volatile, chemically stable, and hydrophobic organic compounds [1,2]. Through hundreds of PAHs exist in the environment, 16 PAHs are priority pollutants according to the US Environmental Protection Agency (US EPA) [3]. Seven PAHs (seven carcinogenic PAHs) including BaA, Chr, BbF, BkF, BaP, IDP, and DBahA are probable human carcinogens among the sixteen PAHs [4]. The main sources of PAHs are human activities, especially industrial production activities, such as the incomplete combustion of fossil fuels, etc. [5]. The emissions of PAHs reached approximately 32,720 tons in China because of the incomplete combustion of fuel based on a survey in 2016 [6,7]. China is considered to have the greatest emissions of PAHs, which account for 21% of the total emissions in the world [8]. The emissions of PAHs can reach 120,000 tons per year [9]. The generated PAHs are released into the atmosphere, then transported and deposited on soil, vegetation, and surface water [10]. In particular, soil is considered one of the most important reservoirs of PAHs [11]. According to the data from the National Soil Contamination Survey Bulletin, sampling sites with PAHs exceeding the standard reach 1.4% of the total surveyed sites [12]. Based on differences in geographic distribution, Northeast China has the highest levels of PAHs in surface soils (1467 ng·g⁻¹), followed by North China (911 ng·g⁻¹), East China (737 ng·g⁻¹), South China (349 ng·g⁻¹), and West China (209 ng·g⁻¹) [13]. PAHs can eventually become bound to soil particles. Furthermore, PAHs in soils can accumulate in vegetables and enter the human body through the food chain [14,15]. PAHs can induce DNA damage and disturbances in adults and children [16]. In particular, chronic diseases and cancer can be triggered by PAHs in children [17]. In view of the hazards of PAHs, the concentrations of PAHs in contaminated soils and their health risks to humans need to be well studied.

Soil microorganisms (bacteria, fungi, viruses, and protozoa), an important part of soil ecosystems, are sensitive to changes and disturbances in the soil environment [13,16]. Meanwhile, it is reported that they also can drive most of the transformations of macro-elements and micro-elements [18]. Hence, soil microorganisms are considered crucial indicators for the quality of soil environments [19,20]. Recent studies have pointed out that some soil microorganisms play an active part in the biological balance (transformation, degradation, etc.) of soil pollutants [21,22,23]. In addition, those effects may also have an impact on the circulation of other substances (carbon, nitrogen, phosphorus, sulfur, etc.) in soils [24]. The individual soil contaminant pyrene can decrease soil microbial diversity, abundance, and metabolic functions [25]. However, there is still a lack of studies on the effects of multifarious PAH pollution on microorganisms. It is necessary to elucidate the response of soil microorganisms to PAH stress at the community level in PAH-contaminated soils.

The overall objectives of this study were (1) to perform a health risk assessment of adults and children exposed to different PAH-contaminated soils; (2) to characterize the diversity and community composition of bacterial and fungal communities under the stress of PAH pollution; (3) to clarify the key drivers shaping the bacterial and fungal community structure; (4) to identify keystone species based on a co-occurrence network analysis.

2. Materials and Methods

2.1. Study Area and Collection of Soil Samples

The study area is around a coking plant situated in the suburbs of an industrial city in Liaoning Province, Northeast China (120°21′12″–120°23′45″ E, 41°31′10″–41°32′52″ N). The coking plant has almost 15 years of production activities. Its main products include coke, coarse benzene, coal tar, etc. A total of 27 surface soil samples, about 500 g (0–20 cm) each, were taken from three different PAH-contaminated areas, including non-contaminated areas (NC), low-contaminated areas (LC), and high-contaminated areas (HC). The determination of contamination levels was based on the pre-test of measuring PAHs in soils. The division of PAH contamination levels was on the basis of the study conducted by Maliszewska-Kordybach B [26]. After collection, three replicates of soil subsamples were mixed to form 9 composite samples of NC, LC, and HC, respectively. Rocks, pebbles, and plant litter were removed from the composite soil samples. Then, soil samples were sieved through 2 mm sieves. Each composite soil sample was separated into two portions. One portion was stored at −80 °C for further DNA extraction, while the other one was used for the analysis of PAHs.

2.2. Determinations of PAHs in Soils

A total of 16 PAHs (Nap, Ace, Acy, Flu, Phe, Ant, Fla, Pyr, BaA, Chr, BbF, BkF, BaP, IDP, DBahA, and BghiP) specified as U.S. EPA priority pollutants were determined in this study. Each soil sample (5 g) was Soxhlet-extracted for 24 h with 120 mL of an acetone and dichloromethane mixture (1:1, v/v) in NC, LC, and HC areas. The extracts were concentrated using rotary evaporation and solvent-exchanged with n-hexane, then purified with silica gel column chromatography [27]. The purified extracts were solvent-exchanged with n-hexane and concentrated under a stream of nitrogen, then analyzed by gas chromatography–mass spectrometry (GC7890/5975, Agilent, Santa Clara, CA, USA) equipped with an HP5-MS column [27,28]. Duplicates and method blanks were used for quality control. The tagged PAH recoveries for the spiked blanks and samples were in the range of 65–123%.

2.3. Health Risk Assessment

2.3.1. Toxic Equivalent Quantity (TEQ)

In this study, TEQ was selected to evaluate human exposure to PAHs [3]. The mixture of PAHs was considered to be carcinogenic and could be expressed as TEQ, which was calculated as follows:

$$\mathrm{T}\mathrm{E}\mathrm{Q}={\sum }_{i=1}^n(C_i\times {TEF}_i)$$

(1)

C_i is the concentration of individual PAHs. TEF_i is the toxic equivalent factor of individual PAHs. The TEF values were 0.001 for Nap, Ace, Acy, Flu, Phe, Fla, and Pyr; 0.01 for Ant, Chr, and BghiP; 0.1 for BaA, BbF, BkF, and IDP; and 1 for BaP and DBahA [29].

2.3.2. Incremental Lifetime Cancer Risk (ILCR)

The USEPA standard exposure models including children (childhood) and adults (adulthood) exposed to PAHs via the pathways of oral ingestion (ILCR_Ingestion), dermal contact (ILCR_Dermal), and inhalation (ILCR_Inhalation) were selected to assess the ILCR [30,31]. The calculation of ILCR equations was as follows:

$${\mathrm{I}\mathrm{L}\mathrm{C}\mathrm{R}}_{\mathrm{I}\mathrm{n}\mathrm{g}\mathrm{e}\mathrm{s}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}}={\displaystyle \frac{{\mathrm{C}}_{\mathrm{S}}\times ({\mathrm{C}\mathrm{S}\mathrm{F}}_{\mathrm{I}\mathrm{n}\mathrm{g}\mathrm{e}\mathrm{s}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}}\times \sqrt[3]{\frac{\mathrm{B}\mathrm{W}}{70}})\times {\mathrm{I}\mathrm{R}}_{\mathrm{I}\mathrm{n}\mathrm{g}\mathrm{e}\mathrm{s}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}}}{\mathrm{B}\mathrm{W}\times \mathrm{A}\mathrm{T}\times {10}^6}}\times \mathrm{E}\mathrm{F}\times \mathrm{E}\mathrm{D}$$

(2)

$${\mathrm{I}\mathrm{L}\mathrm{C}\mathrm{R}}_{\mathrm{D}\mathrm{e}\mathrm{r}\mathrm{m}\mathrm{a}\mathrm{l}}={\displaystyle \frac{{\mathrm{C}}_{\mathrm{S}}\times ({\mathrm{C}\mathrm{S}\mathrm{F}}_{\mathrm{D}\mathrm{e}\mathrm{r}\mathrm{m}\mathrm{a}\mathrm{l}}\times \sqrt[3]{\frac{\mathrm{B}\mathrm{W}}{70}})\times \mathrm{S}\mathrm{A}\times \mathrm{A}\mathrm{F}\times \mathrm{A}\mathrm{B}\mathrm{S}}{\mathrm{B}\mathrm{W}\times \mathrm{A}\mathrm{T}\times {10}^6}}\times \mathrm{E}\mathrm{F}\times \mathrm{E}\mathrm{D}$$

(3)

$${\mathrm{I}\mathrm{L}\mathrm{C}\mathrm{R}}_{\mathrm{I}\mathrm{n}\mathrm{h}\mathrm{a}\mathrm{l}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}}={\displaystyle \frac{{\mathrm{C}}_{\mathrm{S}}\times {\mathrm{C}\mathrm{S}\mathrm{F}}_{\mathrm{I}\mathrm{n}\mathrm{h}\mathrm{a}\mathrm{l}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}}\times \sqrt[3]{\frac{\mathrm{B}\mathrm{W}}{70}}\times {\mathrm{I}\mathrm{R}}_{\mathrm{I}\mathrm{n}\mathrm{h}\mathrm{a}\mathrm{l}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}}}{\mathrm{P}\mathrm{E}\mathrm{F}\times \mathrm{B}\mathrm{W}\times \mathrm{A}\mathrm{T}}}\times \mathrm{E}\mathrm{F}\times \mathrm{E}\mathrm{D}$$

(4)

$$\mathrm{TCR}={\mathrm{I}\mathrm{L}\mathrm{C}\mathrm{R}}_{\mathrm{D}\mathrm{e}\mathrm{r}\mathrm{m}\mathrm{a}\mathrm{l}}+{\mathrm{I}\mathrm{L}\mathrm{C}\mathrm{R}}_{\mathrm{I}\mathrm{n}\mathrm{g}\mathrm{e}\mathrm{s}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}}+{\mathrm{I}\mathrm{L}\mathrm{C}\mathrm{R}}_{\mathrm{I}\mathrm{n}\mathrm{h}\mathrm{a}\mathrm{l}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}}$$

(5)

The parameters and descriptions involved in the above equations are displayed in Supplementary Table S1. The selection of those parameters was based on the studies conducted by Chen et al. [31] and Qi et al. [32]. The different risk levels of ILCR are as follows: very low cancer risk (below 10⁻⁶), low cancer risk (between 10⁻⁶ and 10⁻⁴), moderate cancer risk (between 10⁻⁴ and 10⁻³), high cancer risk (between 10⁻³ and 10⁻¹), and very high cancer risk (above 10⁻¹) [33].

2.4. Amplification and Sequencing

DNA was extracted from 0.5 g soil samples using the DNeasy^® PowerSoil^® Pro Kit (QIAGEN, Hilden, Germany) in NC, LC, and HC areas. The extracted DNA concentrations (between 18.74 ng·μL⁻¹ and 93.57 ng·μL⁻¹) were determined by a spectrophotometer (NanoDrop2000, Thermo Fisher Scientific, Waltham, MA, USA). DNA was purified, then checked by 1% agarose gel electrophoresis. The primer pairs of 338F/806R for the bacterial PCR amplifications and ITS1F/ITS2R for the fungal PCR amplifications were selected [33,34]. The bacterial reaction system had a final volume of 20 μL, including 4 μL 5× FastPfu Buffer, dNTPs (2 μL 2.5 mM each of dCTP, dTTP, dATP, and dGTP), 0.8 μL forward primer (5 μM), 0.8 μL reverse primer (5 μM), 0.4 μL FastPfu Polymerase, 0.2 μL BSA, 10 ng template DNA, and ddH₂O. The fungal reaction mixtures consisted of 2 μL 10× buffer, dNTPs (2 μL 2.5 mM each of dCTP, dTTP, dATP, and dGTP), 0.8 μL forward primer (5 μM), 0.8 μL reverse primer (5 μM), 0.2 μL rTaq Polymerase, 0.2 μL BSA, 10 ng template DNA, and ddH₂O (supplemented to 20 μL). The amplifications were conducted as follows: an initial denaturation step of 95 °C for 3 min, 27 cycles for bacterial PCR conditions (27 cycles for fungal PCR conditions) at 95 °C for 30 s, annealing at 55 °C for 30 s, and elongation at 72 °C for 45 s, followed by a final extension at 72 °C for 10 min. After successful amplifications, the PCR products were extracted from a 2% agarose gel and purified with the AxyPrep DNA Gel Extraction Kit (Axygen Biosciences, Tewksbury, MA, USA), then sent to Majorbio Co., Ltd. (Shanghai, China) for Illumina MiSeq sequencing [35,36].

2.5. Data Analysis and Co-Occurrence Network Analysis

The raw reads of 16S and ITS were processed by DADA2 and incorporated into QIIME2 [37,38,39]. Those processes included quality filtering, denoising, merging, and dereplication [37,40]. The DADA2-denoised sequences were considered the amplicon sequence variants (ASVs) [39]. The analysis of data was conducted by the diversity cloud analysis platform (QIIME2 process) [39]. Statistical analysis, such as one-way analysis of variance (ANOVA), was performed using SPSS v.26.0 software. The raw sequencing data have been deposited in the NCBI Sequence Read Archive (SRA) under accession no. PRJNA882593. Redundancy analysis (RDA) was conducted by Canoco 5.0 [41]. The co-occurrence network analysis of bacterial and fungal communities was conducted based on the abundance of data at the genus level [42,43]. Network attributions, such as the number of edges and average degree, were calculated using the Psych package in the R program (V.3.6.1) [44,45]. The visual network graphs were created using Gephi (version 0.9.2).

3. Results

3.1. PAH Contamination Levels in Soils and Health Risk Assessment

The contamination levels of ∑7PAHs and ∑16PAHs in the soil samples from NC, LC, and HC areas are presented in

Table 1. Concentrations of soil PAHs in the soil samples from NC, LC, and HC areas.

Aromatic Ring	16 PAHs	Abbreviations	NC (μg/kg)	LC (μg/kg)	HC (μg/kg)
Two rings	Naphthalene	Nap	7.3 a ± 1.0	44.2 b ± 7.1	86.9 c ± 15.0
Three rings	Acenaphthene	Ace	1.4 a ± 0.2	21.2 b ± 5.2	31.9 b ± 9.4
	Acenaphthylene	Acy	3.0 a ± 0.5	11.6 b ± 2.2	28.1 c ± 5.7
	Fluorene	Flu	8.2 a ± 1.2	52.9 b ± 8.7	189.3 c ± 29.7
	Phenanthrene	Phe	10.5 a ± 2.0	36.9 a ± 6.5	134.3 b ± 23.4
	Anthracene	Ant	3.8 a ± 0.7	13.0 b ± 2.5	30.1 c ± 5.5
Four rings	Fluoranthene	Fla	3.9 a ± 0.4	20.7 b ± 5.3	34.0 c ± 6.4
	Pyrene	Pyr	13.6 a ± 2.4	54.6 a ± 10.4	175.0 b ± 34.1
	Benzo[a]anthracene	BaA	11.4 a ± 2.0	38.5 b ± 7.5	133.6 c ± 10.6
	Chrysene	Chr	12.4 a ± 2.2	49.8 b ±9.3	148.2 c ± 11.7
Five rings	Benzo[b]fluoranthene	BbF	13.9 a ± 2.8	47.1 b ± 5.7	156.1 c ± 17.3
	Benzo[k]fluoranthene	BkF	10.1 a ± 1.9	40.2 b ± 4.9	130.9 c ± 15.6
	Benzo[a]pyrene	BaP	15.0 a ±2.9	44.4 b ± 6.4	131.6 c ± 19.1
	Indeno[1,2,3-cd]pyrene	IDP	11.8 a ±1.9	34.1 b ± 4.6	73.9 c ± 9.4
Six rings	Dibenzo[a,h]anthracene	DBahA	4.2 a ± 0.9	14.2 b ± 2.4	30.5 c ± 5.2
	Benzo[g,h,i]perylene	BghiP	10.9 a ± 1.2	20.7 b ± 2.8	42.8 c ± 7.7
Total 7 PAHs d			78.8	268.3	804.8
Total 16 PAHs			141.4	544.0	1557.0

The superscripts denote (a, b, and c) differences based on one-way ANOVA (p < 0.05). ^d Concentrations of 7 carcinogenic PAHs (BaA, Chr, BbF, BkF, BaP, IDP, and DBahA).

. The total ∑16PAH concentrations ranged from 141.4 μg/kg to 1557 μg/kg (dry weight, dw), with an average value of 747.5 μg/kg dw in all soil samples. The ∑7PAH concentrations were between 78.8 μg/kg and 804.8 μg/kg in all soil samples. The seven carcinogenic PAH concentrations could account for 49.3–55.7% of the total ∑16PAH concentrations. In addition, BaA, Chr, BaP, and DBahA were the most abundant PAHs in ∑7PAHs and ∑16PAHs. For example, BaP was one of the most abundant PAHs, and it contributed 26.9–30.0% of the ∑7PAH concentration and 8.6–10.9% of the ∑16PAH concentration.

As shown in

Table 2. TEQs (μg/kg) of PAH contamination in all soil samples from NC, LC, and HC areas.

PAHs	TEF	NC-1	NC-2	NC-3	LC-1	LC-2	LC-3	HC-1	HC-2	HC-3
Nap	0.001	0.0085	0.0068	0.0066	0.0468	0.0361	0.0496	0.0749	0.0821	0.1038
Ace	0.001	0.0015	0.0012	0.0016	0.0158	0.0217	0.0262	0.0409	0.0326	0.0221
Acy	0.001	0.0026	0.0029	0.0035	0.0105	0.0101	0.0141	0.0323	0.0216	0.0305
Flu	0.001	0.0095	0.0073	0.0077	0.0587	0.0571	0.0429	0.1583	0.2174	0.1922
Phe	0.001	0.0120	0.0112	0.0082	0.0442	0.0318	0.0346	0.1318	0.1123	0.1589
Ant	0.01	0.039	0.031	0.045	0.143	0.101	0.145	0.289	0.253	0.361
Fla	0.001	0.0035	0.0039	0.0043	0.0192	0.0266	0.0163	0.0282	0.0328	0.0409
Pyr	0.001	0.0108	0.0147	0.0152	0.0446	0.0653	0.0539	0.1897	0.1993	0.1361
BaA	0.1	0.93	1.32	1.18	3.04	3.99	4.51	12.42	14.51	13.15
Chr	0.01	0.102	0.145	0.125	0.401	0.507	0.586	1.368	1.602	1.476
BbF	0.1	1.21	1.71	1.26	4.21	4.59	5.33	14.11	17.51	15.22
BkF	0.1	0.83	1.21	0.98	3.65	3.84	4.58	11.61	14.71	12.94
BaP	1	12.3	18.1	14.5	38.8	43.1	51.3	118.2	153.5	123.1
IDP	0.1	0.98	1.36	1.19	3.02	3.29	3.91	6.39	8.26	7.51
DBahA	1	3.4	5.1	4.1	12.1	13.7	16.9	25.3	35.7	30.6
BghiP	0.01	0.098	0.122	0.108	0.179	0.206	0.235	0.351	0.505	0.427
∑7PAHs a		19.752	28.945	23.335	65.221	73.017	87.116	189.398	245.792	203.996
∑16PAHs		19.9374	29.1460	23.5351	65.7829	73.5727	87.7336	190.6941	247.2481	205.4685

^a 7 carcinogenic PAHs (BaA, Chr, BbF, BkF, BaP, IDP, and DBahA).

, the TEQs of ∑16PAHs varied from 19.9 μg/kg to 247.2 μg/kg, with an average concentration of 104.8 μg/kg. In addition, both the TEQs of carcinogenic ∑7PAHs and the TEQs of ∑16PAHs were below the safe value of 600 μg/kg. The TEQs of carcinogenic ∑7PAHs could account for 99.1% to 99.4% of the TEQs of ∑16PAHs. Those results indicated that the seven carcinogenic PAHs were the most important contributors to the total carcinogenic potency of PAHs. The different contributions of the seven carcinogenic PAHs to the TEQs of ∑16PAHs were as follows: BaP (between 58.58% and 62.1% with an average of 60.6%) > DBahA (between 13.27% and 19.26% with an average of 16.76%) > BbF (between 5.35% and 7.41% with an average of 6.43%) > BaA (between 4.53% and 6.51% with an average of 5.35%) > BkF (between 4.15% and 6.3% with an average of 5.2%) > IDP (between 3.34% and 5.06% with an average of 4.28%) > Chr (between 0.5% and 0.72% with an average of 0.62%). In consequence, BaP and DBahA were the major TEQ contributors to the TEQs of ∑16PAHs in all soil samples.

The ILCR values for different exposure pathways (ILCR_Ingestion, ILCR_Dermal, and ILCR_Inhalation) and TCR values are shown in

Table 3. ILCR values for different pathways of exposure to PAHs in all soil samples from NC, LC, and HC areas.

Sampling Sites	PAHs	Cs (TEQs)	Adulthood				Childhood
			${\mathbf{I}\mathbf{L}\mathbf{C}\mathbf{R}}_{\mathbf{I}\mathbf{n}\mathbf{g}\mathbf{e}\mathbf{s}\mathbf{t}\mathbf{i}\mathbf{o}\mathbf{n}}$	${\mathbf{I}\mathbf{L}\mathbf{C}\mathbf{R}}_{\mathbf{D}\mathbf{e}\mathbf{r}\mathbf{m}\mathbf{a}\mathbf{l}}$	${\mathbf{I}\mathbf{L}\mathbf{C}\mathbf{R}}_{\mathbf{I}\mathbf{n}\mathbf{h}\mathbf{a}\mathbf{l}\mathbf{a}\mathbf{t}\mathbf{i}\mathbf{o}\mathbf{n}}$	TCR	${\mathbf{I}\mathbf{L}\mathbf{C}\mathbf{R}}_{\mathbf{I}\mathbf{n}\mathbf{g}\mathbf{e}\mathbf{s}\mathbf{t}\mathbf{i}\mathbf{o}\mathbf{n}}$	${\mathbf{I}\mathbf{L}\mathbf{C}\mathbf{R}}_{\mathbf{D}\mathbf{e}\mathbf{r}\mathbf{m}\mathbf{a}\mathbf{l}}$	${\mathbf{I}\mathbf{L}\mathbf{C}\mathbf{R}}_{\mathbf{I}\mathbf{n}\mathbf{h}\mathbf{a}\mathbf{l}\mathbf{a}\mathbf{t}\mathbf{i}\mathbf{o}\mathbf{n}}$	TCR
NC1	∑7PAHs	19.752	9.51 × 10−8	1.69 × 10−7	6.45 × 10−12	2.64 × 10−7	9.92 × 10−8	1.24 × 10−7	2.10 × 10−12	2.23 × 10−7
	∑16PAHs	19.9374	9.60 × 10−8	1.71 × 10−7	6.51 × 10−12	2.67 × 10−7	1.00 × 10−7	1.25 × 10−7	2.12 × 10−12	2.25 × 10−7
NC2	∑7PAHs	28.945	1.39 × 10−7	2.48 × 10−7	9.46 × 10−12	3.87 × 10−7	1.45 × 10−7	1.81 × 10−7	3.07 × 10−12	3.27 × 10−7
	∑16PAHs	29.1460	1.40 × 10−7	2.49 × 10−7	9.52 × 10−12	3.90 × 10−7	1.46 × 10−7	1.83 × 10−7	3.10 × 10−12	3.29 × 10−7
NC3	∑7PAHs	23.335	1.12 × 10−7	2.00 × 10−7	7.63 × 10−12	3.12 × 10−7	1.17 × 10−7	1.46 × 10−7	2.48 × 10−12	2.63 × 10−7
	∑16PAHs	23.5351	1.13 × 10−7	2.01 × 10−7	7.69 × 10−12	3.15 × 10−7	1.18 × 10−7	1.47 × 10−7	2.50 × 10−12	2.66 × 10−7
LC1	∑7PAHs	65.221	3.14 × 10−7	5.58 × 10−7	2.13 × 10−11	8.72 × 10−7	3.28 × 10−7	4.09 × 10−7	6.93 × 10−12	7.36 × 10−7
	∑16PAHs	65.7829	3.17 × 10−7	5.63 × 10−7	2.15 × 10−11	8.79 × 10−7	3.31 × 10−7	4.12 × 10−7	6.99 × 10−12	7.43 × 10−7
LC2	∑7PAHs	73.017	3.52 × 10−7	6.25 × 10−7	2.39 × 10−11	9.76 × 10−7	3.67 × 10−7	4.57 × 10−7	7.75 × 10−12	8.24 × 10−7
	∑16PAHs	73.5727	3.54 × 10−7	6.29 × 10−7	2.40 × 10−11	9.84 × 10−7	3.70 × 10−7	4.61 × 10−7	7.81 × 10−12	8.31 × 10−7
LC3	∑7PAHs	87.116	4.19 × 10−7	7.45 × 10−7	2.85 × 10−11	1.16 × 10−6	4.38 × 10−7	5.46 × 10−7	9.25 × 10−12	9.83 × 10−7
	∑16PAHs	87.7336	4.22 × 10−7	7.50 × 10−7	2.87 × 10−11	1.17 × 10−6	4.41 × 10−7	5.50 × 10−7	9.32 × 10−12	9.90 × 10−7
HC1	∑7PAHs	189.398	9.12 × 10−7	1.62 × 10−6	6.19 × 10−11	2.53 × 10−6	9.52 × 10−7	1.19 × 10−6	2.01 × 10−11	2.14 × 10−6
	∑16PAHs	190.6941	9.18 × 10−7	1.63 × 10−6	6.23 × 10−11	2.55 × 10−6	9.58 × 10−7	1.19 × 10−6	2.03 × 10−11	2.15 × 10−6
HC2	∑7PAHs	245.792	1.18 × 10−7	2.10 × 10−6	8.03 × 10−11	3.29 × 10−6	1.23 × 10−6	1.54 × 10−6	2.61 × 10−11	2.77 × 10−6
	∑16PAHs	247.2481	1.19 × 10−7	2.11 × 10−6	8.08 × 10−11	3.31 × 10−6	1.24 × 10−6	1.55 × 10−6	2.63 × 10−11	2.79 × 10−6
HC3	∑7PAHs	203.996	9.82 × 10−7	1.74 × 10−6	6.67 × 10−11	2.73 × 10−6	1.02 × 10−6	1.28 × 10−6	2.17 × 10−11	2.30 × 10−6
	∑16PAHs	205.4685	9.89 × 10−7	1.76 × 10−6	6.71 × 10−11	2.75 × 10−6	1.03 × 10−6	1.29 × 10−6	2.18 × 10−11	2.32 × 10−6

. All the values of ILCR and TCR of childhood and adulthood were below 10⁻¹. The ILCR_Inhalation values were below 10⁻⁶ and much lower than the ILCR_Ingestion and ILCR_Dermal values of childhood and adulthood, indicating that the cancer risk through inhalation was almost negligible. The values of ILCR_Dermal were higher than those of ILCR_Ingestion and ILCR_Inhalation in childhood and adulthood. These results indicated that dermal contact was the main exposure pathway to soil PAHs in adults and children. Adults had higher ILCR_Dermal values than children, but lower ILCR_Ingestion values. The collected data suggested that the cancer risk of exposure to soil PAHs by ingestion in adults was lower than in children, whereas the risk of dermal contact in adults was higher than in children. For all soil samples from NC and LC areas, the average ILCR_Ingestion and ILCR_Dermal values were below 10⁻⁶. These results indicate that children and adults faced a very low cancer risk from the above soil samples through ingestion and dermal contact. Based on the analysis of the average values of ILCR_Ingestion and ILCR_Dermal in HC areas (between 10⁻⁶ and 10⁻⁴), it could be concluded that the human cancer risk caused by PAHs via ingestion and dermal contact in the soil samples from HC areas was low. Through the comparison of TCR values between childhood and adulthood, it was shown that the total cancer risk (including dermal contact, ingestion, and inhalation) suffered by adults was higher than that suffered by children. The average TCR values in all soil samples from NC, LC, and HC areas (adulthood, 1.40 × 10⁻⁶; childhood, 1.18 × 10⁻⁶) ranged from 10⁻⁶ to 10⁻⁴, indicating that a low cancer risk was shown in childhood and adulthood.

3.2. Soil Bacterial and Fungal Diversity and Community Composition Analysis

A sum of 525,221 bacterial raw reads (from 622 to 1120 ASVs) and 608,780 fungal raw reads (from 196 to 583 ASVs) were identified in this study. The dominant composition of bacterial and fungal communities is shown in

Table 4. The dominant composition of bacterial and fungal communities in all soil samples.

Soil Microbial Communities	Phylum	Class	Order	Family	Genus	Species	ASVs
Bacterial communities	35	100	232	391	792	1404	4919
Fungal communities	15	37	88	188	405	620	1784

.

The alpha diversity indices of soil bacterial and fungal communities were characterized by the Chao1 index and Shannon index in this study (Figure 1). There were no significant differences in the bacterial Chao1 index and Shannon index between the soil samples of NC and LC areas (Figure 1a,b). The bacterial Chao1 index and Shannon index in NC and LC samples were higher than those in HC samples. As illustrated in Figure 1c, the fungal Chao1 index showed significant differences under different contamination levels of PAHs (NC, LC, and HC). The fungal Shannon index in the soil samples of NC areas was considerably higher than that in the soil samples of LC and HC areas (Figure 1d). These results indicated that the existence of PAHs in soils could decrease the diversity of bacterial and fungal communities.

In this study, PCoA was conducted to compare the bacterial and fungal community structure under different levels of PAHs. As shown in Supplementary Figure S1, the cumulative loads of the first two axes of bacterial and fungal communities were 79.65% and 83.88%, which could distinctly separate the bacterial and fungal communities in the soil samples of NC, LC, and HC areas, thereby indicating distinct variations in bacterial and fungal community composition along with the contamination levels of PAHs.

The Circos analysis of bacterial and fungal communities is shown in Figure 2, while the soil bacterial and fungal community composition analysis is presented in Figure 3. The six most predominant phyla of bacterial communities were Actinobacteriota (33.39–34.18% in the soil samples of NC areas; 20.36–25.27% in the soil samples of LC areas; 29.4–36.54% in the soil samples of HC areas), Proteobacteria (26.05–29.83% in the soil samples of NC areas; 22.97–31.49% in the soil samples of LC areas; 27.13–29.82% in the soil samples of HC areas), Chloroflexi (7.97–9.1% in the soil samples of NC areas; 8.29–13.55% in the soil samples of LC areas; 10.29–11.85% in the soil samples of HC areas), Firmicutes (2.48–4.85% in the soil samples of NC areas; 13.26–16.05% in the soil samples of LC areas; 7.5–9.2% in the soil samples of HC areas), Acidobacteriota (13.54–16.01% in the soil samples of NC areas; 3.9–10.73% in the soil samples of LC areas; 0.94–2.03% in the soil samples of HC areas), and Bacteroidota (2.71–3.65% in the soil samples of NC areas; 3.02–3.85% in the soil samples of LC areas; 7.27–8.94% in the soil samples of HC areas) (Figure 3a). Those predominant phyla prevailed in all soil samples and could account for 83.9% to 92.9% of the total bacterial abundance. Significant differences existed in the relative abundance of Acidobacteriota and Gemmatimonadota in NC, LC, and HC areas. On the contrary, the changes in the relative abundance of Proteobacteria, Chloroflexi, and Myxococcota were not obvious under different contamination levels of PAHs. In addition, the highest relative abundance of Fimicutes and Gemmatimonadota was shown in the soil samples of LC areas, compared with those in the soil samples of NC and HC areas. In particular, the high contamination levels of PAHs in soils could reduce the relative abundance of Acidobacteriota and increase the relative abundance of Bacteroidota.

As shown in Figure 3b, the dominant fungal groups were Ascomycota and Basidiomycota at the phylum level. The relative abundance of Ascomycota and Basidiomycota in NC and HC areas showed significant differences with that in LC areas. Compared with the soil samples of NC and LC areas, the relative abundance of Mortierellomycota decreased sharply under the condition of high contamination levels of PAHs in the soil samples of HC areas. The relative abundance of Chytridiomycota considerably enhanced with the increase in the concentration of PAHs. Overall, the above results indicate that the changes in common and unique bacterial and fungal phyla within soil microbial communities hinted at the influence of stress from PAHs. The existence of PAHs in soils could affect the bacterial and fungal community composition.

3.3. Relationships between Bacterial and Fungal Community Composition and Soil PAHs

RDA was performed to identify the relative contributions of PAHs to the variation in bacterial and fungal community composition. Comprehensive consideration of the concentrations and TEF of 16 PAHs, BaA, Chr, BbF, BkF, BaP, IDP, DBahA, and BghiP were selected for RDA. RDA 1 and 2 could explain 53.8% and 36.2% of the total variations in Figure 4a, and 89.7% and 0.07% of the total variations in Figure 4b, respectively. In addition, RDA results of the interactive forward selection revealed that BaA (Pseudo-F = 3.3, p < 0.05) and Chr (Pseudo-F = 12.8, p < 0.05) could explain 32.2% and 46.2% of the overall variation in soil bacterial community compositions. These results indicated that BaA and Chr were the key factors affecting the soil bacterial community compositions. For the soil fungal communities, BaP and DBahA played a vital role in driving community structure, which could explain 65.7% (Pseudo-F = 13.3, p < 0.05) and 12% (Pseudo-F = 21.9, p < 0.05) of the total variation based on the interactive forward selection of RDA.

3.4. Network Analysis

In this study, bacterial and fungal networks were constructed to explore the co-occurrence patterns of bacterial and fungal communities (Figure 5). The greatest proportion of nodes was contributed by Actinobacteriota (30.0%) and Proteobacteria (31.0%) in the bacterial networks, and Ascomycota (70.0%) and Basidiomycota (17.9%) in the fungal networks. Therefore, the bacterial phyla Actinobacteriota and Proteobacteria were considered keystones, whereas the fungal phyla Ascomycota and Basidiomycota also acted as keystone taxa in the bacterial and fungal networks under the different contamination levels of PAHs. The results indicated that the above microorganisms played critical roles in the bacterial and fungal co-occurrence networks of the PAH-polluted soils. The modularity indexes of all co-occurrence networks were >0.4, implying a modular structure in the bacterial and fungal networks. Interestingly, the positive links of bacterial networks in the soil samples of NC and LC areas were more abundant than those in the soil samples of HC areas. Meanwhile, the positive and negative links of fungal networks in the soil samples of NC, LC, and HC areas showed the same trends. These results suggested that the high contamination levels of PAHs could change the relationships between different microorganisms in soils. Compared with the connections (network edges) in the soil samples of LC and HC areas, more connections (network edges) among different nodes existed in the soil samples of NC areas. The above results indicated that PAHs in soils could influence the co-occurrence networks and degree of connectivity, while the high contamination levels of PAHs could reduce the complexity and stability of bacterial and fungal networks.

4. Discussion

4.1. Soil Bacterial and Fungal Network Analysis

In this study, the soils with different contamination levels around the coking plant were selected to explore the response of bacterial and fungal communities to PAHs. Through the analysis of ∑₁₆PAHs in all soil samples, it could be found that the total ∑₁₆PAH concentrations in this study (average 747.7 μg/kg) were slightly higher than those in the abandoned coking plant soils from the Fatou area of Chaoyang District in Beijing (average 735.3 μg/kg) [46], but much lower than those in the coking plant soils from Xiangyuan in Changzhi City (average 857,000 μg/kg) [47], those in the area with the largest abundance of coking plants from Bohai Bay (average 8160 μg/kg) [3], and those in a coking plant field in Shijiazhuang City (average 385,200 μg/kg) [48]. In this study, the average TEQ of ∑₁₆PAH concentrations was 104.8 μg/kg, which was lower than that in soil samples from the steel industry in East China (637 μg/kg) [49] and coal mining in the same area (769 μg/kg) [50]. The TEQs of carcinogenic ∑₇PAHs could account for more than 99% of the TEQs of ∑₁₆PAHs, which were considered the main carcinogenic contributor of ∑₁₆PAHs. Those results agreed well with data from previous studies [51,52]. Through the comparison of TEQ values, BaP and DBahA were the main TEQ contributors to the TEQs of ∑₁₆PAHs. This phenomenon could be explained by the high TEF of BaP (TEF = 1) and DBahA (TEF = 1).

Dermal contact, ingestion, and inhalation were considered the main exposure pathways to PAHs in soils [3,53]. Based on the calculation and comparison of ILCR and TCR, the cancer risk via soil inhalation was almost negligible in this study. Meanwhile, both adults and children withstood the highest cancer risk via soil dermal contact. These results were in line with previous studies [32,54]. In this study, adults had a lower cancer risk than children when exposed to soil PAHs by ingestion, but the risk from dermal contact in adults was higher than that in children. This could be explained by the larger dermal surface exposure for adults than for children [54]. Furthermore, the hand-to-mouth activity of children, by which soil PAHs could be readily ingested, resulted in a higher cancer risk by ingestion in children than in adults [32,55]. Through the comparison of TCR values between childhood and adulthood, it was shown that the total cancer risk arising from soil PAH exposure suffered by adults is always higher than that suffered by children. The same results were also reported by Dreij et al. [56] and Xu et al. [57]. The soil PAHs in all samples from NC, LC, and HC areas showed a low cancer risk in children and adults based on the average TCR values (adulthood, 1.40 × 10⁻⁶; childhood, 1.18 × 10⁻⁶). These TCR values were lower than those obtained by Xie et al. [49] (adulthood, 7.58 × 10⁻⁶; childhood, 6.60 × 10⁻⁶) and Bigović et al. [58] (adulthood, 1.59 × 10⁻⁵; childhood, 1.16 × 10⁻⁵).

4.2. Diversity and Community Composition of Bacterial and Fungal Communities in PAH-Contaminated Soils

The alpha diversity indices (Chao1 index and Shannon index) of the soil bacterial and fungal communities decreased when the PAH concentrations increased, which indicated that the richness of soil bacterial and fungal communities decreased gradually with the existence of PAHs. It was reported that PAHs in soils could negatively affect microbial diversity [59,60]. Furthermore, some microbial species might be lost due to the presence of PAHs [60,61]. In sum, the toxic effect of PAHs on soil microorganisms could be the main reason for the decrease in microbial diversity [62].

Through the results of PCoA, it could be concluded that there were distinct variations in bacterial and fungal community compositions among different levels of PAHs in this study. Previous studies showed that the presence of PAHs in soils played a key role in the changes in bacterial and fungal communities [28,63,64]. In this study, the bacterial communities Actinobacteriota, Proteobacteria, Chloroflexi, Firmicutes, Acidobacteriota, and Bacteroidota and the fungal communities Ascomycota and Basidiomycota were the dominant microbial communities under PAHs in soils. They were commonly found in soils contaminated by PAHs [28,60,61,63,65,66]. The relative abundance of the phyla Acidobacteriota and Bacteroidota exhibited differences under different PAH concentrations. Those findings were consistent with previous observations [67,68]. The relative abundance of Proteobacteria in soils with different contamination levels of PAHs revealed no obvious differences. This was because of the resistance of Proteobacteria to PAHs [65]. Meanwhile, Chloroflexi showed the same trend with PAHs. This finding is consistent with the report by Sazykina et al. [69]. It has been reported that Chloroflexi was usually related to the degradation of PAHs in soils [28].

The relative abundance of Ascomycota and Basidiomycota showed differences with changes in PAH concentrations. These differences might be related to the link of Ascomycota and Basidiomycota with PAHs’ degradation [70,71,72,73,74]. In this study, Mortierellomycota decreased sharply under high contamination levels of PAHs. Similarly, Li et al. [44] noted that the growth of Mortierellomycota was inhibited in petroleum-polluted soils compared with control soils (low concentrations of PAHs). Therefore, it can be included that PAHs in soils could affect the bacterial and fungal community composition due to the changes in common and unique bacterial and fungal groups under the stress of PAHs.

4.3. The Key Factors of PAHs Driving Microbial Community Structure

By analyzing the results of the RDA, this study illustrated that BaA and Chr were considered the key factors driving the soil bacterial community compositions, while DBahA played a vital role in driving fungal community structure. It was difficult for most of the microbes to degrade BaA and Chr; only several specific bacterial species could live with and utilize BaA and Chr, such as Sphingobium strain [75]. Thus, BaA and Chr drive the compositions of soil bacterial communities. Similarly, the fungal community structure being driven by BaP and DBahA could be due to the dominant BaP-degradable and DBahA-degradable fungi in the endogenous microbial community. Further study is needed to clarify the degradation mechanism of BaP and DBahA by the newly formed fungal community, which will help find more microbial resources for the degradation of PAHs.

4.4. Co-Occurrence Patterns of Microbial Communities in PAH-Contaminated Soils

Co-occurrence networks can represent the response of microbial communities to environmental contamination [76,77]. Their nodes, edges, and links can also reflect the relationships in microbial communities. Previous studies have demonstrated that higher positive interactions in microbial networks indicated cooperative or reciprocal symbiosis relationships among different microorganisms, whereas lower positive interactions indicated antagonism relationships [78]. In this study, the relationships in positive and negative links of bacterial and fungal networks changed with an increase in PAHs. These results indicated that there were cooperative or reciprocal symbiotic relationships between bacterial and fungal communities in NC and LC areas, while in HC areas, there was an antagonistic relationship. Our study also revealed that the high contamination levels of PAHs reduced the connections among different nodes, which could lead to a reduction in the complexity and stability of bacterial and fungal networks and a reduction in resistance to disturbances [79].

In fact, there are close connections between PAHs and microbial communities in the soil environment. On one hand, soil microbial communities can affect the distribution pattern of PAHs in soils [66]. Meanwhile, PAHs can be degraded by specific microbial communities in soils [63]. On the other hand, soil PAHs can affect the microbial community composition [80,81]. Soil PAHs are also considered the carbon and energy sources for the growth of microbes [82]. The results of this study confirmed that PAHs had an impact on soil bacteria and fungi.

5. Conclusions

BaA, Chr, BaP, and DBahA were the most abundant PAHs in this study. BaP and DBahA contributed to the TEQs of ∑₁₆PAHs. The cancer risk through inhalation was almost negligible, while dermal contact was the main exposure pathway for adults and children. Based on the community composition analysis, the diversity of bacterial and fungal communities decreased due to the existence of PAHs. In addition, the differences in microbial community composition existed under different contamination levels of PAHs, of which BaA and Chr could drive the bacterial community composition and BaP and DBahA drove the fungal community composition. Actinobacteriota, Proteobacteria, Ascomycota, and Basidiomycota were the keystones in microbial network analysis. Furthermore, high contamination levels of PAHs reduced the complexity and stability of microbial networks. In conclusion, PAHs could affect the soil bacterial and fungal communities in soils. In the future, exploring the mechanism of these soil PAHs’ influence on microbes, especially in the mechanism of action of single PAHs, would be beneficial to conduct the in situ remediation of PAH-contaminated soil by indigenous microorganisms.