Interactions of erythromycin and an antibiotic-mixture to the gut microbiome of juvenile rainbow trout

Erythromycin (ERY) is a commonly used antibiotic that can be found in wastewater effluents globally. Due to the mechanisms by which they kill and prevent bacterial growth, antibiotics can have significant unwanted impacts on the fish gut microbiome. The overall objective of this project was to assess the effects of erythromycin and an antibiotic mixture on fish gut microbiomes. The project was split into two experiments to assess gut microbiome in response to exposure with ERY alone or in mixture with other common antibiotics. The objectives of experiment 1 were to understand uptake and depuration of ERY in juvenile rainbow trout (RBT) over a 7 d uptake followed by a 7 d depuration period using three concentrations of ERY. Furthermore, throughout the study changes in gut microbiome response were assessed. In experiment 2, a follow-up study was conducted using an identical experimental design to assess the impacts of an antibiotic-mixture (ERY, ampicillin, metronidazole, and ciprofloxacin at 100 µg/g each). Here, three matrices were analyzed, with gut collected for 16s metabarcoding, plasma for untargeted metabolomics, and brain for mRNA-seq analysis. ERY was depurated from the fish relatively quickly and gut microbiome dysbiosis was observed at 7 d after exposure, with a slight recovery after the 7 d depuration period. A greater number of plasma metabolites was dysregulated at 14 d compared to 7 d revealing temporality compared to gut microbiome dysbiosis. Furthermore, several transformation products of antibiotics and biomarker metabolites were observed in plasma due to antibiotic exposure. Brain transcriptome revealed only slight alterations due to antibiotic exposure. The results of these studies will help inform aquaculture practitioners and risk assessors when assessing the potential impacts of antibiotics in fish feed and the environment, with implications for host health. Five RBT were exposed in each of six replicate 20-L tanks for each treatment in a dilutor system to regulate temperature (12°C) and light (14 h: 10 h light: dark). To maintain optimal water quality and consistency of the water microbiome, fish were housed under flow-through conditions with three complete renewals of cold, filtered facility water every 24 h. During experiments, water quality was assessed every 3 days. RBT were exposed via diet to two treatments of antibiotics, including 1,000 µg ERY/g and a mixture of antibiotics (ant-mix) containing nominal concentrations of 100 µg/g each of ampicillin, metronidazole, ciprofloxacin, and ERY, and one solvent control (MeOH). The 7-d exposure was followed by a 7-d depuration, during which food containing antibiotics was replaced with solvent control food. Brain mRNAseq was ran to assess effects of antibiotics on the transcriptome of the brain. Briefly, flash-frozen brain tissue was extracted for high-quality RNA using RNeasy Plus Universal Mini Kit with QIAzol Lysis Reagent (Qiagen, Germany). Two samples of whole brain collected from each tank after 7 d, were extracted and pooled at equimolar amounts with measurement using Qubit RNA HS Assay Kit (Thermo Fisher, USA). Extracted RNA was assessed for RNA integrity number (RIN) using Agilent RNA 6000 Nano kit on an Agilent 2100 Bioanalyzer (Agilent Technologies, Germany) with all values being ≥ 6.5 with a mean ± standard deviation of 7.84 ± 0.875. Samples were then shipped on dry ice to Genome Quebec Innovation Centre for paired-end sequencing (2x100), using an Illumina NovaSeq 6000. Quality control of resulting sequence output was conducted using FastQC, and reads were assessed for quality, then trimmed to a minimum Phred score of 20 and a minimum length of 35 bases using Trimmomatic (Bolger et al., 2014). The abundance of transcripts was estimated using Salmon (version 1.10.2) with the reference RBT transcriptome USDA_OmykA_1.1 (GenBank Acc. no. GCA_013265735.3) (Patro et al., 2017) and mapped to genes using GenomicFeatures (version 1.52.1) (Lawrence et al., 2013) and corresponding gtf file (Oncorhynchus_mykiss.USDA_OmykA_1.1.110.gtf).红霉素（Erythromycin, ERY）是一种常用抗生素，广泛存在于全球各类废水处理厂的出水中。由于其通过杀菌与抑制细菌增殖发挥作用的机制，抗生素会对鱼类肠道微生物组（gut microbiome）产生显著的不良影响。 本项目的总体目标为评估红霉素及抗生素混合物对鱼类肠道微生物组的影响。研究分为两个实验，分别评估单独暴露于ERY或与其他常见抗生素联合暴露时，鱼类肠道微生物组的响应变化。 实验1的目标为在三种ERY浓度处理下，探究幼年虹鳟（juvenile rainbow trout, RBT）体内ERY的吸收与清除过程：具体为7天暴露吸收阶段后接续7天的清除净化阶段。此外，本研究全程监测肠道微生物组的响应变化。 实验2为后续跟进研究，采用与实验1一致的实验设计，评估抗生素混合物（ERY、氨苄西林（ampicillin）、甲硝唑（metronidazole）、环丙沙星（ciprofloxacin），各组分浓度均为100 µg/g）的影响。本实验共分析三类样本：用于16S扩增子测序（16S metabarcoding）的肠道组织、用于非靶向代谢组学分析的血浆样本，以及用于mRNA测序（mRNA-seq）的脑组织。 ERY可在鱼体内快速被清除，暴露7天后可观察到肠道微生物组失调，经过7天净化期后出现轻微恢复。与暴露7天时相比，暴露14天时血浆中失调代谢物的数量更多，表明其代谢紊乱的时间进程与肠道微生物组失调存在差异。此外，抗生素暴露后血浆中可检测到多种抗生素转化产物及生物标志物代谢物。脑组织转录组仅因抗生素暴露出现轻微改变。 本研究结果可为水产养殖从业者与风险评估人员评估抗生素在鱼类饲料及环境中的潜在影响提供参考，同时对宿主健康研究具有借鉴意义。 本研究采用稀释系统调控水温（12℃）与光照周期（14小时光照:10小时黑暗），每个处理组设置6个重复的20-L水族箱，每个箱内投放5尾虹鳟。为维持最优水质与水体微生物组稳定性，鱼类采用流水养殖模式，每日更换三次经过过滤的低温设施用水。实验期间每3天监测一次水质。 鱼类通过日粮暴露于两种抗生素处理组：①1000 µg ERY/g日粮；②抗生素混合物组（ant-mix），各组分标称浓度为氨苄西林、甲硝唑、环丙沙星、ERY各100 µg/g日粮；同时设置溶剂对照组（甲醇, MeOH）。7天暴露阶段结束后接续7天净化期，此时将含抗生素的日粮替换为含溶剂对照的日粮。 脑组织mRNA测序用于评估抗生素对脑部转录组的影响。简要而言，采用RNeasy Plus Universal Mini Kit结合QIAzol裂解试剂（Qiagen，德国）从快速冷冻的脑组织中提取高质量RNA。实验7天后从每个水族箱中采集2尾全脑组织样本，提取RNA后以等摩尔量混合，使用Qubit RNA HS检测试剂盒（Thermo Fisher，美国）定量。采用Agilent 2100生物分析仪搭配Agilent RNA 6000 Nano试剂盒测定RNA完整性数（RNA integrity number, RIN），所有样本的RIN值均≥6.5，平均值±标准差为7.84±0.875。随后将样本置于干冰中寄送至魁北克基因组创新中心（Genome Quebec Innovation Centre），采用Illumina NovaSeq 6000平台进行双端测序（2×100 bp）。 测序数据的质控采用FastQC完成，对测序reads进行质量评估，随后使用Trimmomatic（Bolger等，2014）将reads修剪至最低Phred质量值20、最小序列长度35 bp。转录本丰度估算采用Salmon（版本1.10.2），参考转录组为虹鳟USDA_OmykA_1.1（GenBank登录号GCA_013265735.3）（Patro等，2017），并通过GenomicFeatures（版本1.52.1）（Lawrence等，2013）及对应的gtf文件（Oncorhynchus_mykiss.USDA_OmykA_1.1.110.gtf）将转录本映射至基因水平。