Effects of Polyethylene Microplastics and Phenanthrene on Soil Properties, Enzyme Activities and Bacterial Communities

: Microplastics (MPs) or polycyclic aromatic hydrocarbons (PAHs) pollution has received increasing concern due to their ubiquitous distribution and potential risks in soils. However, nothing is known about the inﬂuences of PAHs-MPs combined pollution on soil ecosystems. To address the knowledge gap, a 1-year soil microcosm experiment was conducted to systematically investigate the single and combined effect of polyethylene (PE) /phenanthrene (PHE) on soil chemical properties, enzymatic activities and bacterial communities (i.e., diversity, composition and function). Results showed that PE and PHE-PE signiﬁcantly decreased soil pH. The available phosphorus (AP) and neutral phosphatase activity were not considerably changed by PHE, PE and PHE-PE. Signiﬁcant enhancement of dehydrogenase activity in a PHE-PE amended system might be due to the degradation of PHE by indigenous bacteria (i.e., Sphingomonas , Sphingobium ), and PE could enhance this stimu-lative effect. PHE and PHE-PE led to a slight increase in soil organic matter (SOM) and ﬂuorescein diacetate hydrolase (FDAse) activity but a decrease in available nitrogen (AN) and urease activity. PE signiﬁcantly enhanced the functions of nitrogen cycle and metabolism, reducing SOM/AN contents but increasing urease/FDAse activities. There were insigniﬁcant impacts on overall community diversity and composition in treated samples, although some bacterial genera were signiﬁcantly stimulated or attenuated with treatments. In conclusion, the addition of PHE and PE inﬂuenced the soil chemical properties, enzymatic activities and bacterial community diversity/composition to some extent. The signiﬁcantly positive effect of PE on the nitrogen cycle and on metabolic function might lead to the conspicuous alterations in SOM/AN contents and urease/FDAse activities. This study may provide new basic information for understanding the ecological risk of PAHs-MPs combined pollution in soils.


Introduction
Plastics and their products are widely applied in production and peoples' lives because of their durability and low cost. Improper disposal of plastic waste causes severe environmental contamination. The fragmentation of plastics might occur in the environment by a physical, chemical and biological process to smaller sizes (<5 mm), defined as microplastics (MPs) [1]. Moreover, microparticles and microbeads used in industrial products, such as raw materials, drug delivery particles in medicines and personal care products, are the primary sources of MPs [2,3]. The terrestrial environment is an important reservoir of MPs [4]. Lv et al. found that the MPs abundance was 16.1 ± 3.5 items kg −1 in rice soils in Shanghai [5]. In the farmed soil of southwestern China, 95% of the sampled plastic particles were in the microplastic size (0.05~1 mm), with an average abundance of 18,760 items kg −1 [6]. In addition, the contents of polycyclic aromatic hydrocarbons (PAHs) increase dramatically in the environment as the demand and consumption of fossil fuels increase with the development of industry [7]. According to the survey, agricultural soils have

Soil Incubation Experiment
Tested soil was collected from the surface (0-20 cm) of agricultural land in Zhaoqing (112 • 30 23 E, 23 • 9 33.40 N). The soil had been used for vegetable cultivation. Soil samples were randomly selected by "S" pattern form, and they were thoroughly mixed to form one composite sample. These soils were air-dried and sieved through a mesh (2 mm) to remove stones and gravel and then evenly mixed and stored in the dark at 4 • C before incubation. The microcosm experiments were performed in a climate-controlled chamber at Zhaoqing University, China. Soil, PHE-soil, PE-soil or PHE-PE-soil were mixed and then homogenized thoroughly to incubate in a glass container. The specific treatments are Processes 2022, 10, 2128 3 of 18 as follows: (1) control soil (CK), no PE and PHE were added to the soils; (2) PHE, PHE was added into the soil with a content of 20 mg kg −1 ; (3) PE, polyethylene was added into the soil at 5% dose (w/w); (4) PHE-PE, PHE (20 mg kg −1 ) + PE (5%). Each treatment was performed in triplicate. The incubations were maintained at a consistent humidity (70%) and temperature (30 • C). After exposure for 1 year, soils were sampled for chemical properties, enzymatic activities and DNA extraction.

Soil Chemical Properties and Enzymatic Activities Analysis
Soil chemical properties were analyzed based on the method as described by Lu [26]. Briefly, soil pH was measured in water-soil (2.5:1) based on the potentiometric method. Available nitrogen (AN) was determined by the alkaline hydrolysis diffusion method. The available phosphorus (AP) was analyzed by sodium bicarbonate extraction molybdenum antimony resistance colorimetry. The content of soil organic matter (SOM) was determined by the hydrothermal potassium dichromate oxidation colorimetric method.
Soil enzymatic activities included urease, neutral phosphatase, fluorescein diacetate hydrolase (FDAse) and dehydrogenase, which were determined with the method described by Yi et al. [27] and Wallenstein et al. [28]. Urease activity was determined by the sodium phenol-sodium hypochlorite colorimetric method and expressed as the mass (mg) of NH 3 -N in 1 g soil after 24 h. Benzene disodium phosphate colorimetric method was used for studying the neutral phosphatase activity and expressed as the mass (mg) of phenol released in 1 g soil after 24 h. The absorbance of the released fluorescein was used to determine FDAse activity and expressed as fluorescein (ug) measured in 1 g soil after 20 min. Dehydrogenase activity was determined by 2,3,5-triphenyltetrazolium chloride (TTC) reduction colorimetry and expressed as the production of 1,3,5-triphenylformazan (TPF) in 1 g of soil.

Soil DNA Extraction and 16S Amplicon Sequencing
DNA was extracted from 0.5 g of soil using the E.Z.N.A™ Mag-Bind Soil DNA Kit (Omega, Norcross, GA, USA) according to the manufacturer's instructions. The quality and concentration of the isolated DNA were evaluated. A Qubit3.0 DNA detection kit was used to accurately quantify genomic DNA, determining the amount of DNA to be added in the polymerase chain reaction (PCR) reaction. The primer pair 341F (5 -CCTACGGGNGGCWGCAG-3 ) and 805R(5 -GACTACHVGGGTATCTAATCC-3 ) were designed to amplify the V3-V4 hypervariable regions of bacterial 16S rDNA gene by polymerase chain reaction (PCR). The first round PCR reaction conditions were as follow: 94 • C for 3 min, 5 cycles at 94 • C for 30 s, 45 • C for 20 s and 65 • C for 30 s, followed by 20 cycles at 94 • C for 20 s, 55 • C for 20 s, 72 • C for 30 s and a final extension at 72 • C for 5 min. A second round of PCR amplification was performed following 3 min at 95 • C, 5 cycles of 20 s at 94 • C, 20 s at 55 • C, 30 s at 72 • C and a final extension at 72 • C for 5 min.
The PCR products were purified and quantified and then sequenced using the Illumina Miseq platform by Sangong Bioengineering Co., LTD (Shanghai, China), and the raw reads were generated. After removing the primer connector sequence, the pairs of reads were merged into a sequence according to the overlap relationship between reads. Sequence quality control and filtering were carried out. The operational taxonomic units (OTUs) were selected at similar levels of 97% to perform statistical analysis by biological information.

Analysis of Microbial Community Diversity, Composition and Function
The diversity of the microbial communities in all treatments was presented as Shannon, Chao, Ace and Simpson. Chao and Ace are the indexes commonly used to estimate the total number of species. Shannon and Simpson are the indexes for estimating microbial diversity in samples. Statistical analysis was used to observe the community structure of the samples at the genus level. According to the full sequence of the 16S rDNA gene of the tested microbial genome, the gene function spectrum was deduced. The sequence data of 16S rDNA obtained by sequencing were compared with the Greengenes database to search for Processes 2022, 10, 2128 4 of 18 reference sequences and classify them as reference OTU. The OTUs abundance matrix was normalized in accordance with the rDNA gene copy numbers. Function predictions were categorized into "Kyoto Encyclopedia of Genes and Genomes (KEGG)" and "Functional Annotation of Prokaryotic Taxa (FAPROTAX)" to determine the metabolic pathways and functions of soil microorganism, respectively, using Phylogenetic Investigation of Communities by Reconstruction of Unobserved States (PICRUSt). The phenotypic traits of bacterial communities were predicted with BugBase.

Soil Chemical Properties
Compared to the CK, soil pH was significantly decreased with the addition of PE and PHE-PE (p < 0.05). Moreover, PE addition significantly decreased the pH in the soil with PHE supplementation. The addition of PE significantly lowered the SOM content (p < 0.05), while PHE and PHE-PE exerted a slightly positive effect on SOM. Both PHE and PE decreased AN content singly or jointly, with a more pronounced effect by PE (p < 0.05), compared to CK. No significant changes in soil AP content were observed in all treatments (PE, PHE, PE-PHE) ( Table 1).

Soil Enzymatic Activities
Soil enzymes are mainly secreted by microorganisms and play crucial roles in catalyzing various biochemical reactions in the soil. Four soil enzymes, including urease, FDAse, dehydrogenase and neutral phosphatase, were measured in this study, as shown in Table 2. Compared to CK, a slight reduction or increase of the urease or FDAse activity was observed for PHE and PHE-PE treatments, PE amendment significantly increased the urease and FDAse activities in the soil. The presence of PHE/PE and PHE-PE slightly and considerably increased dehydrogenase activity, respectively. PHE, PE and PHE-PE had insignificant effect on the neutral phosphatase activity ( Table 2).

Microbial Community Diversity and Composition
The effects of PHE, PE and PHE on bacteria diversity were assessed, as shown in Table 3. Compared with the CK, no significant difference in Shannon, Chao, Ace and Simpson indexes was observed in all treatments (p > 0.05). At the genus level, taxonomic profiling showed that the bacterial communities were dominated by Sphingomonas, Gemmatimonas, Acidobacteria (Gp4, Gp7 and Gp6), Ohtaekwangia, Betaproteobacteria and Burkholderiales in all treatments, as shown in Figure 1. The Venn diagram generated from genus is shown in Figure S1. CK shared a majority of the genus with PHE, PE and PHE-PE-treated soils. In terms of the bacterial community composition, no specific clustering pattern was observed between all samples based on non-metric multidimensional scaling (NMDS) (Figure 2). ANOSIM analysis confirmed that addition of PHE and PE had no significant effect on the overall bacterial community composition (CK vs. PHE, p = 0.3; CK vs. PE, p = 0.2; CK vs. PHE-PE, p = 0.5) (Table S3). Only minor changes of the Sphingomonas, Gemmatimonas and Acidobacteria (Gp3, Gp4, Gp6, Gp7 and Gp16) were observed ( Figure 1). Statistical analysis of metagenomic profiles (STAMP) method was performed to determine whether the bacterial taxa significantly changed in specific treatment ( Figure 3). In this study, we found some bacterial genera (i.e., Azohydromonas, Sorangiineae, Sphingobium, Cupriavidus, Rhodospirillales, Devosia, Nannocystineae and Comamonadaceae) were significantly influenced with the addition of PHE and PE solely or jointly. However, the changes in these genera were not sufficient to alter the bacterial community composition, as described below (Section 4.3.).
The effects of PHE, PE and PHE on bacteria diversity were assessed, as shown in Table 3. Compared with the CK, no significant difference in Shannon, Chao, Ace and Simpson indexes was observed in all treatments (p > 0.05). At the genus level, taxonomic profiling showed that the bacterial communities were dominated by Sphingomonas, Gemmatimonas, Acidobacteria (Gp4, Gp7 and Gp6), Ohtaekwangia, Betaproteobacteria and Burkholderiales in all treatments, as shown in Figure 1. The Venn diagram generated from genus is shown in Figure S1. CK shared a majority of the genus with PHE, PE and PHE-PE-treated soils. In terms of the bacterial community composition, no specific clustering pattern was observed between all samples based on non-metric multidimensional scaling (NMDS) (Figure 2). ANOSIM analysis confirmed that addition of PHE and PE had no significant effect on the overall bacterial community composition (CK vs. PHE, p = 0.3; CK vs. PE, p = 0.2; CK vs. PHE-PE, p = 0.5) (Table S3). Only minor changes of the Sphingomonas, Gemmatimonas and Acidobacteria (Gp3, Gp4, Gp6, Gp7 and Gp16) were observed ( Figure 1). Statistical analysis of metagenomic profiles (STAMP) method was performed to determine whether the bacterial taxa significantly changed in specific treatment ( Figure 3). In this study, we found some bacterial genera (i.e., Azohydromonas, Sorangiineae, Sphingobium, Cupriavidus, Rhodospirillales, Devosia, Nannocystineae and Comamonadaceae) were significantly influenced with the addition of PHE and PE solely or jointly. However, the changes in these genera were not sufficient to alter the bacterial community composition, as described below (Section 4.3.).

BugBase
Based on 16S rDNA gene sequences, the phenotypic change of bacterial communities was investigated using BugBase ( Figure 4). The bacterial phenotypes of all samples were mainly oxygen utilizing (i.e., aerobic, anaerobic) and Gram-negative. BugBase predictions indicated that PHE and PE exposure hardly altered the aforementioned phenotypes.

BugBase
Based on 16S rDNA gene sequences, the phenotypic change of bacterial communities was investigated using BugBase ( Figure 4). The bacterial phenotypes of all samples were mainly oxygen utilizing (i.e., aerobic, anaerobic) and Gram-negative. BugBase predictions indicated that PHE and PE exposure hardly altered the aforementioned phenotypes.

FAPROTAX
FAPROTAX can annotate the functions of prokaryotes in elemental cycles (i.e., carbon, nitrogen, sulfur), nutritional types and degradation ( Figure 5). Modifications in the relative abundance of the bacterial community associated with elemental cycles were observed for the treatment groups (PHE, PE, PHE-PE). Chemoheterotrophya and aerobic chemoheterotrophy were enhanced with the addition of PHE, PE and PHE-PE. The nitrate reduction, nitrification, nitrogen/nitrate/nitrite respiration, nitrite oxidation, denitrification and nitrogen fixation functions were mostly connected to the nitrogen cycle. PE showed a significantly influence on nitrogen cycle. Bacterial functions for sulfur cycles were dominantly composed of dark oxidation of sulfur compounds and sulfate/sulfur compounds respiration. Although the microbial functions according to FAPROTAX were stimulated by PHE-PE to some degree, the CK-, PHE-and PHE-PE-treated samples have a close relationship, as indicated by principal component analysis (PCA) ( Figure S2). Whereas the treatment with PE clustered together at the right side of the PC1 axis and separated from the CK, PHE and PHE-PE.

FAPROTAX
FAPROTAX can annotate the functions of prokaryotes in elemental cycles (i.e., carbon, nitrogen, sulfur), nutritional types and degradation ( Figure 5). Modifications in the relative abundance of the bacterial community associated with elemental cycles were observed for the treatment groups (PHE, PE, PHE-PE). Chemoheterotrophya and aerobic chemoheterotrophy were enhanced with the addition of PHE, PE and PHE-PE. The nitrate reduction, nitrification, nitrogen/nitrate/nitrite respiration, nitrite oxidation, denitrification and nitrogen fixation functions were mostly connected to the nitrogen cycle. PE showed a significantly influence on nitrogen cycle. Bacterial functions for sulfur cycles were dominantly composed of dark oxidation of sulfur compounds and sulfate/sulfur compounds respiration. Although the microbial functions according to FAPROTAX were stimulated by PHE-PE to some degree, the CK-, PHE-and PHE-PE-treated samples have a close relationship, as indicated by principal component analysis (PCA) ( Figure S2). Whereas the treatment with PE clustered together at the right side of the PC1 axis and separated from the CK, PHE and PHE-PE.

KEGG
The distribution of the microbial functions assessed by KEGG pathway composition is shown in Table 4. The KEGG pathways at Level 2 annotations were mainly related to metabolism (e.g., amino acid, carbohydrate, lipid, nucleotide, xenobiotics), biosynthesis of secondary metabolites, membrane transport, replication and repair, translation, cell motility, nucleotide metabolism, genetic information processing, folding / sorting / degradation, transcription, signal transduction, enzyme families, cell growth and death, transport and catabolism, environmental adaptation and excretory system. The abundance of the aforementioned pathways in all treated soils (especially PE, p < 0.05) were higher than those in CK.

KEGG
The distribution of the microbial functions assessed by KEGG pathway composition is shown in Table 4. The KEGG pathways at Level 2 annotations were mainly related to metabolism (e.g., amino acid, carbohydrate, lipid, nucleotide, xenobiotics), biosynthesis of secondary metabolites, membrane transport, replication and repair, translation, cell motility, nucleotide metabolism, genetic information processing, folding / sorting / degradation, transcription, signal transduction, enzyme families, cell growth and death, transport and catabolism, environmental adaptation and excretory system. The abundance of the aforementioned pathways in all treated soils (especially PE, p < 0.05) were higher than those in CK.   Significant differences (p < 0.05) between the treatments and CK are indicated with asterisk (*).

Effects of PHE and PE on Soil Chemical Properties
Soil pH is an important index of soil chemical variables, which plays a crucial role in the availability of nutrients [29]. We found that PE and PHE-PE significantly decreased soil pH, which is in accordance with previous findings that the agricultural soil exposed to HDPE is 0.62 units lower than the controls [30]. However, a different result has also been reported in which the MPs (i.e., HDPE, PLA) alone and jointly with multiwall carbon nanotubes (MWCNTs) could increase soil pH [31]. This may be attributed to the differences in MPs (type, dosage and size) and soil properties among these studies. Due to the strong adsorption capacity, MPs could change soil adsorption to regulate the acid-base balance. Various additives released from MPs during soil incubation might modify soil pH. Additionally, previous studies found that soil pH correlated negatively with the relative abundance of Acidobacteria [32]. Although PE and PHE-PE hardly affected the relative abundance of Acidobacteria ( Figure S3, p > 0.05), the soil pH was markedly reduced. This suggested that soil microbes are not responsible for the changes in soil pH in our research. In sum, PE may singly or jointly with PHE decrease soil pH, while the underlying mechanisms need further investigation.
Due to its high organic carbon content, PHE can contribute to organic carbon accumulation in soil. Moreover, PHE can serve as carbon source for the growth of bacteria, such as Sphingomonas and Sphingobium, which suggested that PHE might be preferentially metabolized, thus reducing the soil original SOM decomposition in PHE and PHE-PE amended soil. PHE may consume soil oxygen (O 2 ) during its degradation process and further inhibit the mineralization of SOM. All of these probably resulted in a slight increase of SOM with the addition of PHE and PHE-PE. Previous studies have demonstrated that soil without additional carbon sources could decompose SOM for energy and carbon substrates [33]. Here, the energy required for microbial metabolism may be a key driver of native SOM decomposition in the PE-treated system, which showed significantly decreased SOM content compared with the PHE-PE (p < 0.05) ( Table 1). SOM decomposition can be regulated by competition among bacterial populations [34]. The effects of MPs on SOM decomposition may be controlled by the balance between microbial anabolism and catabolism [32]. The "plastisphere" formed in the MPs surface could be a microbial activity hotspot that accelerated the decomposition of SOM [35]. These suggested that MPs induced the higher microbial activity and metabolic capability of probably availed SOM. As shown in Sections 3.4.3. and 4.4.3., the addition of PE significantly enhanced the metabolic processes of the bacterial community, which can explain the significant decrease in SOM content when exposed to PE. Consistent results were reported by Xiao et al., who found that low doses of PE (0.01%, w/w) strongly promote SOM decomposition in paddy soil [35]. The SOM decomposition was positively consistent with the carbon dioxide (CO 2 ) efflux rate in soils [33]. This indicated that the addition of PE would significantly induce the CO 2 efflux, while PHE and PHE-PE showed less influence.
Zhou et al. concluded that carbon source supply from biodegradable material could stimulate microbial biomass and intensify nitrogen immobilization [36]. This suggested that PHE as the carbon source of some bacteria in the soil could lead to the increase of biological nitrogen-fixation, resulting in the reduction of AN content in PHE and PHE-PE groups. Previous researcher pointed out that the decomposition of SOM could facilitate microbial growth [35]. We can speculate that increased nitrogen is required to support bacterial growth and thus accelerate AN demand from the soil, which can partly contribute to the significantly lowered AN with the addition of PE (p < 0.05). In another aspect, the accumulation of AN (including N-NH 4 + and N-NO 3 -) are dependent on their balance between production and depletion, such as ammoniation, nitrification, denitrification and immobilization. The Acidobacteria and Pseudomonas have been reported to perform nitrate reduction, nitrification and denitrification [37,38]. Here, soil with PHE, PE and PHE-PE changed the abundance of this bacterial taxon, which might also lead to a decrease in AN. As shown in Sections 3.4.2. and 4.4.2., PE significantly influences the functions related to nitrification, aerobic nitrite oxidation, denitrification, nitrite respiration and nitrite denitrification of the microbial communities, and this may be another reason why AN markedly decreased with the presence of PE.
A previous study demonstrated that the change of soil AP induced by MPs (i.e., PS, polytetrafluorethylene) was consistent with the change of phosphatase activity [39]. Similarly, we observed consistent (negligible) changes in AP content and neutral phosphatase activity in the present study ( Table 2). The research of Satyaprakash et al. [40] and Qu et al. [41] found that the change of AP was related to the dissolution of inorganic phosphorus and organic phosphate mineralization mediated by microorganisms. Bergkemper et al. pointed out that Acidobacteria contributed significantly to phosphorus turnover and phosphorus availability in soils [42]. Here, PHE, PE and PHE-PE did not significantly change the relative abundance of Acidobacteria ( Figure S3), which may partly explain the insignificant variation in AP content. However, Wang et al. discovered that 10% polylactic acid (PLA) significantly decreased the soil AP content, which might arise from its enhancement in soil pH [31]. Findings from Yan et al. confirmed that polyvinyl chloride (PVC) (0.1% and 1%) had no distinct effect on overall bacterial community diversity and composition while causing a significant change in soil AP content [43]. These controversial results may be attributed to the differences in MPs types and soil properties. The mechanisms of AP transformation in PE-and PHE-amended soils should be explored in the future.

Effects of PHE and PE on Enzymatic Activities
As the enzymes that catalyze the hydrolysis of nitrogen-containing organic matter (e.g., urea), urease plays an important role in regulating the nitrogen cycle. Previous studies confirmed that the urease activities were increased with the addition of PE [12,44], which was also observed in our current study. Compared to CK, a slight reduction of the urease activity was observed for PHE and PHE-PE treatments, whereas PE amendment significantly increased the urease activity in the soil, which was consistent with the results of FAPROTAX analysis (Section 3.4.2.) that the functions of ureolysis were stimulated or inhibited by PE or PHE/PHE-PE. Additionally, as a hydrolase, urease is important for transforming organic matter. Negative correlation between the urease activity and SOM content was found by Kompała-Bąb et al. [45], which was consistent with our findings in the present study. Wang et al. indicated that urease activity was correlated significantly with some bacterial genera (e.g., Aeodermatophilus, Blastococcus, Bacillus, Marmoricola, Nitrospira) [31], while these bacteria have not been found in our present study. Therefore, it is speculated that the microbial communities did not play an important role in regulating urease activity in this research.
FDAse activity can represent the overall metabolic activity of bacteria, and it is a good indicator of microbial activity [46]. FDAse activity was significantly promoted with the addition of HDPE and PP [21,31]. Here, we further observed higher FDAse activity in the presence of PHE, PE and PHE-PE (Table 2). PE addition dramatically increased FDAse activity relative to CK in the present study, indicating that PE could significantly enhance microbial metabolic activity, which was further verified in KEGG metabolic pathways determined by PICRUSt analysis (as shown in Sections 3.4.3. and 4.4.3.). This probably resulted in increased mineralization of SOM and assimilation of AN by bacteria with the PE treatment. The positive effect of PHE and PHE-PE on FDAse activity was weaker than that of PE addition. Previous studies showed that MPs might reduce soil bulk density and increase air circulation, resulting in the accumulation of aerobic microorganisms and then increased microbial activity [32,47]. Inversely, the decomposition of PHE by the soil bacteria (e.g., Sphingomonas, Sphingobium) would consume O 2 , leading to the enrichment of anaerobic microorganisms with lower activity in PHE and PHE-PE systems than PE.
Dehydrogenase is directly involved in the degradation of PAHs (PHE, pyrene, naphthalene). Previous studies found that the removal percentage of PAHs is positively correlated with dehydrogenase activity in the soil [48,49]. The presence of PHE-PE considerably boosted dehydrogenase activity compared to the CK or PHE (Table 2), indicating that increased natural biodegradation of PHE by indigenous bacteria in soil (i.e., Sphingomonas, Sphingobium) and PE could facilitate the degradation of PHE. A similar result was reported by Yi et al. [27] in which membranous PE improved the dehydrogenase activity in soil to the extent of 21%. On the other hand, PHE and PE had no discernible effect on the activity of neutral phosphatase singly or jointly (Table 2). Wang et al. pointed out that phosphatase activity was significantly associated with some bacterial genera, including Phenylobacterium, Pseudonocardia, Ramlibacter, Marmoricola and Saccharimonadales [31]. However, these aforementioned bacteria have not been found in our present study, leading to no discernible alterations in phosphatase activity. Previous research revealed that the inorganic phosphorus availability for microorganisms was related to the activity of phosphatase [50,51]. This indicated that the addition of PHE, PE and PHE-PE did not influence the demand for phosphorus by bacteria in the soil.

Effects of PHE and PE on Microbial Diversity and Community Composition
The high Chao and Ace values denote superior richness of the soil bacterial community. The greater Shannon or smaller Simpson values indicate a higher community diversity. The results of Section 3.3. indicated that PE, PHE and PHE-PE showed negligible effect on microbial community richness and diversity under our experimental conditions. Similarly, Sun et al. revealed that 1%PE (w/w) did not show a significant influence on soil bacterial diversity [52]. They conducted a thorough literature search and found that 11 of the 14 previous studies indicated that MPs amendment hardly influences the bacterial diversity in soil. The indices obtained for bacteria showed no significant differences between the CK and PHE (50 mg kg −1 ) groups [53]. However, Fei et al. reported that the addition of PE (1% and 5%) declined the richness and diversity of the bacterial communities in an acid farmed soil [12]. A significant increase of microbial richness and diversity was observed with the addition of PHE [54]. Judy et al. reported that MPs-mixed waste organic output (MWOO) showed minimal effect on bacterial diversity [55], while polystyrene (PS) markedly alleviated sulfamethazine's adverse effects on bacterial diversity [56].
Sphingomonas was characterized by its ability to biodegrade organic compounds, such as PHE, petroleum, allethrin, phenols and dioxins [57,58]. Acidobacteria have appeared to tolerate or even degrade various pollutants (e.g., heavy metals, polychlorinated biphenyls, petroleum compounds, PAHs) [37]. Compared with PHE, PHE-PE increased the abundance of Sphingomonas and Acidobacteria, suggesting that the biodegradation of PHE might be enhanced with the addition of PE in our study. This deserves to be assayed in the future. In contrast, a significant reduction of Sphingomonas and Acidobacteria abundance with the addition of MPs (PE, PVC) was found by Fei et al. [12], which may be attributed to diverse soil properties and MPs types. Moreover, Acidobacteria plays significant ecological roles in soil, as evidenced by their active participation in carbon, nitrogen, phosphorus and sulfur cycling [59]. The abundance of Gemmatimonas was related to the organic carbon, nitrogen and phosphorus levels in soil [60]. Owing to insignificant changes in Acidobacteria and Gemmatimonas, they probably did not contribute to the variation in soil nutrients.
Compared to the CK, Azohydromonas and Sorangiineae showed a significantly positive response to PHE. The negative response group in PE comprised members affiliated to Sphingobium, Cupriavidus and Comamonadaceae, the positive response group was mostly from Rhodospirillales, Devosia and Sorangiineae. The relative abundances of Nannocystineae and Opitutus were significantly lower in the PHE-PE-treated sample than those in CK. Azohydromonas, Rhodospirillales and Devosia have the function of nitrogen fixation, which can transform atmospheric nitrogen into ammonium [61][62][63]. Sorangiineae are known to produce a number of protein clusters, including amino acid/ lipid/carbohydrate transport [64]. Nannocystineae is recognized for the production of long-chained polyunsaturated fatty acids and unusual steroids [65]. Opitutus might play a role in fermenting complex organics to simple ones [66]. The results potentially suggested that the addition of single or conjoint PE and PHE probably influenced elemental cycles and the metabolic process by affecting the abundance of some bacteria. However, these effects may not be pivotal in regulating soil chemical properties and microbial community function, which is confirmed in Sections 3.1. and 3.4.
In summary, the PHE and PE addition influenced the microbial community to some extent. However, no crucial effects of single or co-contaminated PHE and PE were observed in altering bacterial community diversity and composition. This is not supported by the variational soil pH, AN, AP, SOM and enzymatic activities. Presumably, the variations of soil chemical properties incurred by PHE and PE were not enough to change soil bacterial diversity and composition. In addition, soils generally harbor a vast diversity of bacteria with an intrinsic capacity to cope with varied disturbances [20]. In addition to that, soil texture/structure/permeability are the important physical properties [67,68], and MPs could induce alterations of soil structure or permeability [69,70]. Such impacts may be associated with changes in bacterial communities [69]. Thus, the effects of PHE/PE on soil physical properties need to be determined to test these hypotheses in the future.

Microbial Phenotype Prediction Based on BugBase
The bacterial cell membrane damage could induce the efflux of cytoplasmic substances and then facilitate biofilm formation, which was tolerant to the adverse environmental stresses. However, the relative abundance of biofilm-forming and stress tolerance were slightly changed in the presence of PHE, PE and PHE-PE, compared to the CK (p > 0.05) ( Figure 4). In this regard, the relative abundance of Gram-negative phenotype (i.e., Acidobacteria, Bacteroidetes, Chloroflexi, Gemmatimonadete, Planctomycetes, Proteobacteria) was significantly higher than Gram-positive (i.e., Actinobacteria, Armatimonadetes, Chloroflexi). Previous research pointed out that Gram-negative bacteria with an outer membrane were more resistant to membrane damage than the Gram-positive strains [71]. This demonstrated that PHE, PE and PHE-PE probably did not cause damage to the bacterial communities in our study.

Functional Prediction of Bacterial Community by FAPROTAX
Chemoheterotrophy, aerobic chemoheterotrophy and hydrocarbon degradation are important ecological function groups related to the carbon cycle [72]. Abundant chemoheterotrophy and aerobic chemoheterotrophy in soil suggest the bacteria mainly obtain carbon and energy by oxidizing organic compounds for their growth, thus increasing the emission of greenhouse gases [52]. The rise of these functions indicate that PHE and PE probably induced CO 2 emission singly and jointly in our study. Similarly, Gao et al. reported that PE addition (18%) significantly promoted soil CO 2 emissions [23]. It is noteworthy that PE exhibited a significant influence on the nitrogen cycle, which might lead to a significant decrease in AN content (Sections 3.1. and 4.1.) and an increase in carbohydrate/amino acid/lipid metabolism (Sections 3.4.3. and 4.4.3.). Sulfur cycles play an important role in maintaining the stability and health of the soil.

Predictive Metabolic Pathways Using KEGG
Amino acid metabolism and carbohydrate metabolism were the most abundant in all samples, followed by lipid metabolism, which was closely related to energy metabolism. Nucleotide metabolism is directly related to cell homeostasis, contributing to the production of energy suppliers (i.e., ATP and GTP) [73]. Biosynthesis of secondary metabolites could convert toxic substances that accumulate in the later growth phase, subsequently prolonging the survival of bacteria. In addition, secondary metabolites have been proven to act as antioxidants to protect organisms from oxidative stress [74]. PE showed a significantly impact on these metabolic pathways (p < 0.05), potentially suggesting that the bacterial community promotes their metabolic processes to generate energy for microbial life activities. Analogously, Sun et al. found that the carbohydrate/amino acid/lipid metabolism show significantly positive correlations with the microplastic concentrations (PP, PE, PS) in agricultural soil [52]. This may be due to the fact that MPs addition can change the soil nutrient cycling [12]. As shown in Section 3.4.2., the observably enhanced nitrogen cycle after PE addition can confirm this point.
Membrane transport could mediate the uptake of basic substrates and assist in importing/exporting ions, small molecules and macromolecules, facilitating the growth and life activity of bacteria in soil [75,76]. Moreover, the excess metabolites can be excreted in a timely manner through membrane transport and thus prevent them from accumulating to toxic levels in bacterial cells [77]. PHE, PE and PHE-PE induced the functional response in membrane transport, and this may facilitate the bacterial communities to counteract the changing survival environment and then maintain homeostasis in soil. Replication and repair belonging to genetic information processing had higher abundances in treated samples compared to the CK. Replication is essential for the survival of species. However, the physical and chemical agents may attack DNA and lead to mutations. Bacteria possess the capability to repair damages and hold genetic stability. The increased replication and repair function may result in the stabilized bacterial community, which further verified that there was no significant effect on microbial diversity and composition in the presence of PHE, PE and PHE-PE.
In bacteria, the transcription and translation processes are concurrent for gene expression to accommodate the changing conditions [78,79]. Signal transduction is the biological response of microbial cells to the ever-changing environment, altering the bacterial transcriptome to mitigate the influence of contaminant stress [80]. Cell motility might provide a competitive advantage for the bacterial community to fit the changing environment and move towards nutrients or avoid toxic substances as well [81]. The abundance of these pathways was enhanced with the addition of PHE, PE and PHE-PE. This may be conducive to soil bacteria coping with PHE and PE pollution stress. A similar result that PE increased the cell motility of soil bacterial communities was observed by Fei et al. [12].
Previous studies indicated that the xenobiotics biodegradation and metabolism might include the decomposition of plastic polymers and pollutants (e.g., bisphenol, toluene, benzoate, heavy metals and PAHs) [52,53,82,83]. The increase in this metabolic pathway indicated that bacteria in treated soils could use PE or PHE as a carbon source to grow/reproduce and degrade them. Enzyme families are an extremely important class with catalytic activity [84]. Bacteria may contain enzyme families of functionally diverse members that are employed in various metabolic pathways [85], which would drive the cellular processes and result in cell growth and death. These exhibited increased abundances under PHE, PE and PHE-PE treatments. All of this may result in the higher resistance to PHE and PE, consistent with the enhanced function of environmental adaptation

Conclusions
The addition of PHE, PE and PHE-PE did not distinctly affect the soil AP content and neutral phosphatase activity. Soil pH was significantly decreased by PE and PHE-PE. PHE-PE considerably strengthened dehydrogenase activity in contrast to CK or PHE, which might arise from the degradation of PHE by indigenous bacteria. PHE and PHE-PE showed a slightly positive effect on SOM and FDAse activity but decreased AN and urease activity. PE exerted an observably negative effect on the functions of the nitrogen cycle as indicated by increasing the abundance of nitrate denitrification/ureolysis/nitrite respiration. PE also led to a significant increase of amino acid/carbohydrate/lipid/ nucleotide/xenobiotics metabolism. These may cause the decrease of SOM/AN contents and the enhancement of urease/ FDAse activities. Although PHE and PE solely or jointly exhibited insignificant influences on overall community diversity and composition, some bacterial genera were significantly reduced or enhanced in treated samples. However, the changes in soil chemical properties and enzymatic activities explained by bacterial community diversity and composition under all treatments were relatively small. Summarily, (1) there were some changes in soil chemical properties, enzymatic activities and bacterial communities (i.e., diversity, composition and function) with the addition of single or combined pollutants of PE and