al/Water Research 165(0)114979 3 。nt 2.7.Data analysis an initial den (Win.The ue o n Rank n test 10HL PCR the first step Th culating the pairwise Spearman's co 105,65 C for 30smin and and pooled together.Hig and and the was c ed out uein g the Oume et 3.Results and discussion lapping re an omgpastcbofmwsmorehanroctiofhimbr Trimmomatic(version 0.33)( 214)and chimerase number of readsin each san me which cells 2.6.Shotgun metagenomics and data processing oth evident.The d by the au ere RFs)in (n leaf bio ersion 2.6.3)(Hy ass of the three bi es10- roc query co eaf,thu ing the ava tothe s (Ma et al.2016 sition the relea pies of ARGs per y of 16 xamp kholderia ater than 500 bp mplexity of microbial populatio dynamics.Microplastic g gy I ation (NCB)Seq ence Read Archive( quatic environ ent,th for metagenomic analysis) analysis and accession nizerswill be attracted by the and thisinitialgenome DNA and ddH2O to make up a total volume to 50 mL. The PCR procedure conditions were as follows: an initial dena￾turation at 95 C for 5 min; followed by 15 cycles at 95 C for 30 s, 50 C for 30 s and 72 C for 40 s; with a final extension at 72 C for 7 min. The PCR products from the first step PCR were purified through VAHTSTM DNA Clean Beads. A second round PCR was then performed in a 40 mL reaction which contained 20 mL 2 Phmsion HF MM, 8 mL ddH2O, 10 mM of each primer and 10 mL PCR products from the first step. Thermal cycling conditions were as follows: an initial denaturation at 98 C for 30s; followed by 10 cycles at 98 C for 10s, 65 C for 30 s min and 72 C for 30 s; with a final extension at 72 C for 5 min. Finally, all PCR products were quantified by Quant-iT™ dsDNA HS Reagent and pooled together. High￾throughput sequencing analysis was performed on the purified, pooled sample using the Illumina Hiseq 2500 platform (2 250 paired ends). Sequence analysis was carried out using the QIIME pipeline (version 1.8.0) (Caporaso et al., 2010). In brief, the two pair-end sequencing data were merged into one according to the over￾lapping relationship using FLASH (version1.2.7) (Magoc and Salzberg, 2011). After filtering low-quality and short sequences by Trimmomatic (version 0.33) (Bolger et al., 2014) and chimera se￾quences by Uchime (Kozich et al., 2013), we subsampled the reads to obtain the same number of reads in each sample, which was 73,289 clean reads. The sequences were clustered into operational taxonomic units (OTUs) at 97% identity with UCLUST (Caporaso et al., 2010; Edgar, 2010). Taxonomy assignments were conducted by applying SILVA database as the reference (Caporaso et al., 2010). 2.6. Shotgun metagenomics and data processing Paired-end (2 150) metagenomic sequencing was performed on an Illumina HiSeq 2000 platform. The raw reads were der￾eplicated and trimmed by the quality. The clean reads were assembled into scaffolds by using IDBA-UD (version 1.1.1) (Peng et al., 2012). The open reading frames (ORFs) in scaffolds were predicted by Prodigal (version 2.6.3) (Hyatt et al., 2010) and an￾notated using BLASTp by applying the ARGs database (E value  10
5 , sequence identity  80%, query coverage  70%, alignment length  25 amino acids). The abundance of ARGs was calculated by mapping reads to the gene sequences (Ma et al., 2016) and normalized by the abundance of 16S rRNA genes (relative abundance), expressed as copies of ARGs per copy of 16S rRNA gene (Li et al., 2015), consistent with the qPCR results reported in many previous studies (Chen et al., 2017; Feng et al., 2018; Marti et al., 2018). The relative abundance of the ARGs type or subtype were calculated using the following equation (Li et al., 2015): The ORF sequences of the scaffolds carrying ARGs were anno￾tated using RefineM (version 0.0.23) (Parks et al., 2017) and scaf￾folds with length greater than 500 bp were retained. All sequencing data are deposited in the National Center for Biotechnology Information (NCBI) Sequence Read Archive (acces￾sion No. SRP174395 for 16S analysis and accession No. SRP174465 for metagenomic analysis). 2.7. Data analysis Statistical analyses were performed using R (version 3.4.1, R Found Stat Comput, Vienna) and the results were visualized by Origin. The P value of <0.05 was regarded as statistically significant (Wilcoxon Rank Sum test). To understand the inter associations among microorganisms in the whole community, we used network analysis (Barberan et al., 2012). After calculating the pairwise Spearman's correlation co￾efficients (r), a matrix was constructed to explore the potential relationships in the microbial community. The correlation between two nodes were considered as statistically significant for r  0.8 and P value  0.01 and the correlation network was formed. The analysis result was performed using R (version 3.4.1) and visualized by the Gephi (version 0.9.2). 3. Results and discussion 3.1. Biomass of microplastic biofilm was more than rock biofilm but less than leaf biofilm Measured by flow cytometry every two days, the biomass of microplastic biofilm constantly increased and reached a peak of 3.1 108 cells g
1 on day 8 (Fig. 1A, Table S1). The planktonic cells in the surrounding water maintained a relatively steady concentra￾tion of 2.7 107 cells mL
1 . The highest biomass of microplastic was 2.4 times that of the rock biofilm and 0.14 times that of leaf biofilm. Based on the SEM micrographs, the occurrence of the biofilm after a short period of time compared with the pristine surface of the materials was further demonstrated (Fig. 1B, Fig. S1). The outline of bacterial cells in microplastic and leaf biofilm were both evident. The cell surface of microplastic biofilm was smooth while the cells of leaf biofilm were rough and small amount of filiform extracellular polymeric substance (EPS) could be observed. The observed biomass was in the order leaf biofilm > micro￾plastic biofilm > rock biofilm. The leaf biofilm possessed the highest biomass of the three biofilms, which could be the result of the fast breakdown process of the dissolved organic matter (DOM) in the leaf, thus increasing the availability of nutrients and promoting bacterial growth (Gulis and Suberkropp, 2003). During leaf decomposition, the release of a large amount of organic substances attracts colonizers to utilize the nutrients and supports the growth of the thick biofilm. The nature of the microbial community in the leaf biofilm plays an important role in the leaf utilization (McArthur et al., 1985). The varied responses of the bacterial species to com￾ponents during leaf decomposition have been observed (McNamara and Leff, 2004); for example, the population of Bur￾kholderia cepacia increased when DOM concentrations were greatest, while the population of Pseudomonas putida was inhibi￾ted when total DOM concentrations were greatest, which indicates the complexity of microbial population dynamics. Microplastics and rock do not decompose and therefore, after entering the aquatic environment, their surfaces will absorb nutrients from water and form the conditioning film (Siboni et al., 2007). Colo￾nizers will be attracted by the conditioning film and this initial Abundance ¼ Xn 1 NARG
like sequence Lreads. LARG reference sequence N16S sequence Lreads L16S sequence ! X. Wu et al. / Water Research 165 (2019) 114979 3