Mgiseq 2000 platform

Metagenomic DNA of snow, ice, and seawater samples was extracted from 0.22-μm filters with a PowerSoil DNA Isolation Kit (Qiagen, Germantown, USA) following its protocol. The concentration of DNA was quantified by Qubit DNA Assay Kit with a Qubit 2.0 fluorometer (Invitrogen, Carlsbad, CA, USA), and the quality was checked by gel electrophoresis. The sequencing library was prepared using the KAPA hyper Prep Kit (Roche, Shanghai, China). The shotgun sequencing was performed on Illumina NovaSeq 6000 Platform PE150 (Illumina, San Diego, CA, USA) with 150 bp paired-end reads. Genomic DNA of sediment samples from TY038 and TY042 stations was extracted from ~ 5 g by modified SDS method as descript in a previous study [33 (link)]. DNA of sediment samples from GT04 station was extracted from ~ 0.3 g of sediment per sample using PowerSoil® DNA Isolation Kit (QIAGEN, Germany). The sequence library was built in Beijing Genome Institution (BGI, Shenzhen, China) and sequenced on the BGI MGISEQ-2000 platform, which has been reported can obtain comparable results as Illumina platform [34 (link)].

Total RNA was extracted from SCs‐Exo and LIPUS‐SCs‐Exo, and miRNA sequencing was conducted. The sequence and library preparation were performed by BGI Genomics Co., Ltd. RNA sequencing was conducted using the Illumina MGIseq2000 platform. Total RNA was extracted and isolated from exosomes, and only small RNAs of 18–30 nt were selected for library construction. A computational target prediction algorithm, miRWalk (http://mirwalk.umm.uni‐heidelberg.de/), was used to identify miRNA binding sites to predict the potential target genes of differential miRNAs. However, total RNA extracted from MPG neurons treated with SCs‐Exo and LIPUS‐SCs‐Exo was used for mRNA sequencing. The miRNA targets and mRNA pathways in the Kyoto Encyclopedia of Genes and Genomes (KEGG) were annotated. The predicted exosomal miRNA target and the target gene of neurons were intersected and analyzed using a VENN diagram. Additionally, principal component analysis (PCA), volcano, and heatmap were generated using R‐Studio.

Equal volume of 2× lysis buffer (final concentration of 20 mM EDTA and 0.5% SDS) was added to the SM buffer with cyanophages, which was then incubated with 50 μg/mL proteinase K at 56 °C for 1 h. The phage suspension was sequentially treated with phenol, phenol-chloroform-isoamyl alcohol (25:24:1), and chloroform at a volume ratio of 1:1, respectively. Then, the genomic DNA was precipitated with 1/10 volume of 3 mol/L CH₃COONa, pH 7.5, followed by threefold volume of ethanol at −80 °C for 4 h. The precipitated phage genomic DNA was washed twice with 70% ethanol and then resuspended with sterile ultrapure water at an appropriate volume. Subsequently, the extracted genomic DNA was sequenced by whole-genome shotgun (WGS) strategy to construct a library of different inserts, based on the MGISEQ-2000 platform (BGI-Shenzhen, China) or Illumina NovaSeq platform (Shanghai Personal Biotechnology Co., Ltd., China). For each cyanophage, after removing the adapters and poor-quality reads, all the clean reads were applied for genome assembly with the software SPAdes [43 (link)] to obtain the de novo assembled contigs.