[TOC]

芝麻菜中肠杆菌科主导核心微生物组并贡献抗生素抗性组

Enterobacteriaceae dominate the core microbiome and contribute to the resistome of arugula (Eruca sativa Mill.)

Article, 2019-01-29

Microbiome, [9.133]

https://doi.org/10.1186/s40168-019-0624-7

第一作者
Tomislav Cernava,  Armin Erlacher

通讯作者
Gabriele Berg

其他作者
Jung Soh,  Christoph W Sensen,  Martin Grube

关键字：宏基因组、耐药组、抗生素耐药基因、CARD

日报

https://www.mr-gut.cn/papers/read/1088891609

叶表面的微生物组和抗性组提示：不宜生吃芝麻菜

创作：刘永鑫 审核：高春辉 02月28日

原标题：芝麻菜中肠杆菌科主导核心微生物组并贡献抗生素抗性组

1. 芝麻菜是一种传统的药用植物，也是西餐中流行的生吃绿叶菜，是一种潜在的食源性疾病传播途径；
2. 采用宏基因组、扩增子测序和细菌培养技术，全面揭示芝麻菜微生物组的特异性和可食用植物部分的抗生素抗性组；
3. 叶际土著细菌群落以肠杆菌科为主，基因组中含有多种耐药基因，包括外排泵和氟喹诺酮、氯霉素、苯尼考、大环内酯、氨基香豆素的抗性基因等；
4. 此抗性组多抗药用抗生素，与自然环境中抗性组明显不同，预示着独特的生态学模式。

主编评语：本研究发现芝麻菜表面不仅含有很多可能致病的肠杆菌，肠杆菌中还富集了一些非常不受欢迎的抗性基因。貌似以后是不能生吃了。不过小编觉得虽然不能直接生吃，但是炝拌一下放点酱油陈醋也未尝不可吧！

文章思路总结

方法：16S + 宏基因组 + 培养组

结果：物种组成 + 肠杆菌组成 + 肠杆菌网络 + 宏基因组抗生素抗性组成 + 抗生素抗生分类

摘要

背景：当前芝麻菜是一种传统的药用植物和受欢迎的绿叶菜。它主要在西餐中生吃，并含有各种生物活性次级代谢产物。然而，芝麻菜也与引起严重食源性人类疾病的暴发有关。采用多维方法：整合来自宏基因组学、扩增子测序和来源于芝麻菜细菌培养的数据，以了解土著微生物组的特异性和可食用植物部分的抗性。

结果：我们的研究结果表明，芝麻菜定植一个多样的，植物栖息地特有的微生物组。土著叶际细菌群落以肠杆菌科为主，具有多种抗生素抗性。出乎意料的是，特异流行的性抗性机制为靶向治疗性抗生素，如外排泵、氟喹诺酮、氯霉素、苯尼考、大环内酯、氨基香豆素。

结论：肠杆菌是芝麻菜的核心微生物群成员，在整个耐药基因组中具有重要意义。对芝麻菜相关微生物抗生素耐药性自然存在情况的调查表明，该植物是独特防御机制的研究热点。微生物在这个不寻常的生态系统中的特殊功能提供了一个独特的模型来研究在生态环境中的抗生素抗性。

Background: Arugula is a traditional medicinal plant and popular leafy green today. It is mainly consumed raw in the Western cuisine and known to contain various bioactive secondary metabolites. However, arugula has been also associated with high-profile outbreaks causing severe food-borne human diseases. A multiphasic approach integrating data from metagenomics, amplicon sequencing, and arugula-derived bacterial cultures was employed to understand the specificity of the indigenous microbiome and resistome of the edible plant parts.

Results: Our results indicate that arugula is colonized by a diverse, plant habitat-specific microbiota. The indigenous phyllosphere bacterial community was shown to be dominated by Enterobacteriaceae, which are well-equipped with various antibiotic resistances. Unexpectedly, the prevalence of specific resistance mechanisms targeting therapeutic antibiotics (fluoroquinolone, chloramphenicol, phenicol, macrolide, aminocoumarin) was only surpassed by efflux pump assignments.

Conclusions: Enterobacteria, being core microbiome members of arugula, have a substantial implication in the overall resistome. Detailed insights into the natural occurrence of antibiotic resistances in arugula-associated microorganisms showed that the plant is a hotspot for distinctive defense mechanisms. The specific functioning of microorganisms in this unusual ecosystem provides a unique model to study antibiotic resistances in an ecological context.

主要结果

图1. 三类样本的细菌组成

图中用物种层级树展示了组间各分类级别的相对关系。圆圈代表叶际、根际和土壤宏基因组物种组成平方根标准化的相对丰度，分别用绿色、蓝色和灰色表示。图中仅展示测序数量1000以上的18个目(界、门、纲、目，因为目水平即不多也不少，到科水平数量较多，只能用圈图360度展示才美观)，此图用MEGAN 5.7版本绘制。

Fig. 1 Composition of the bacterial biota in the three analyzed samples. Circles represent the square root scaled taxonomic structure of the assembled metagenomes of the phyllosphere (green), rhizosphere (blue) and bulk soil (gray). Only abundant taxa (n > 1000 hits) were plotted using MEGAN (v.5.7)

图2. 宏基因组中肠杆菌群体结构和丰度

挑选肠杆菌群体进一步细划。此处展示了在一个样品中丰度大于100个序列的菌种类。堆叠柱状图展示肠杆菌科中鉴定的种系。相对丰度为基于RefSeq注释序列中所占的比例。

Fig. 2 Structure and abundance of enterobacterial population in the metagenomes. Only those taxa with an assigned read number higher than 100 in at least one metagenome are shown. The chart illustrates the distribution and abundance of identified lineages of Enterobacteriaceae. Relative abundances within the bacterial fraction are based on non-assembled reads with taxonomic assignments in NCBI’s RefSeq database (ncbi.nlm.nih.gov/refseq)

图3. 叶际和根际中肠杆菌科的核心微生物组

100%阈值OTU的相关网络。采用Cytoscape v3.1绘制网络，Centiscape v2.2计算中心值。OTUs大于定义的阈值时标记为蓝色。点的大小与序列数量相关；三个结点大小图例位于右下角。物种分类表见附表4。

Fig. 3 Enterobacterial core microbiome in the arugula phyllo- and rhizosphere. The OTU-based network correlates OTUs at a 100% cut-off level. Cytoscape v.3.1.0 was used for network rendering and Centiscape v.2.2 to calculate centroid values. OTUs that are above the defined centrality threshold are labeled in blue. The node size correlates with the number of assigned reads; three reference node sizes are visualized in the legend. Taxonomic assignments of the predominant OTUs are provided in Additional file 1: Table S4

图4. 芝麻菜抗性组评估

基于宏基因组挖掘菌株资源的抗性分析来评估芝麻菜的抗性组。采用CARD分析基因以确定抗生素抗性。

a 每个样品的重叠群抽样至89,462条，再采用blastx比对确定抗生素抗性组。

b 其中5,494个属于肠杆菌片段的注释

c 180个培养菌株抗生素抗性检测

Fig. 4 The arugula resistome assessed with metagenome mining and resistance analyses of a strain collection. The CARD-based [45] analysis targets known genes conferring antibiotic resistance. a The general analysis of antibiotic resistance in bacteria is based on blastx assignments for 89,462 subsampled contigs from each sample and b 5494 respective query contigs for the Enterobacteriales fraction. c A total of 180 cultivable isolates obtained from arugula plants was subjected to antibiotic susceptibility testing

图5. 可食用植物部分细菌的主要抗生素抗性

CARD展示主要抗性素抗性的组成。采用CARD鉴定了11个主要的抗生素抗性并展现在圈的外环，不连续的环形。各部分的面积与CARD分类重叠群的数量相关。

Fig. 5 Schematic profile of predominant antibiotic resistances of bacteria in edible plant parts. The eleven most abundant resistance mechanisms identified within the CARD-based [45] analysis are displayed in the outer, discontinuous circle. Areas of the segments correlate with the respective number of contigs that were assigned to distinct CARD categories

结论

结合培养和非培养测序分析微生物组，允许对芝麻菜的肠杆菌群进行全面评估。核心微生物组中各种肠杆菌科成员的出现表明它们很好地适应了这种特殊的寄主植物和部分天然生物群。此外，它们在地上植物部分的富集程度是它们在有限营养环境中存在竞争力的一个指标，而限制性环境是由植物的生物活性代谢物加剧导致的。宏基因组分析与分离株筛选相结合，为微生物群中各种抗生素耐药机制的发生提供了证据。这些防御机制主要针对治疗性抗生素。到目前为止，在自然生境中还没有观察到这种抗生素抗性(AR)谱。因此，芝麻菜系统提供了一个独特的模式来研究在生态环境中的AR。本研究中发现的抗药性携带者对人类健康的影响，以及这些微生物与周围环境和入侵病原体的相互作用，仍有待阐明。

The conducted study allowed a holistic assessment of the enterobacterial population in the Eruca sativa Mill. microbiome by combining different cultivationdependent and cultivation-independent analyses. The occurrence of various Enterobacteriaceae members in the core microbiome indicated that they are well adapted to this particular host plant and part of the natural biota. Moreover, their enrichment in above-ground plant parts is an indicator for their competitiveness in a restrictive environment, which is aggravated by the plant’s bioactive metabolites. Metagenomic analyses in combination with isolates screenings provided evidence for the occurrence of various antibiotic resistance mechanisms in the microbiome. These defense mechanisms were mainly against therapeutic antibiotics. So far, such an AR spectrum was not observed in a natural habitat. Therefore, the arugula system provides a unique model to study AR in an ecological context. The implications of the resistance carriers found in this study for human health as well as the interactions of such microorganisms with the surrounding environment and intruding pathogens remain to be elucidated.

材料方法

看见上面的结果，是不是又简单，又简洁，还发了高分文章。那你一定关注此文是如何分析的，这些结果是如何做出来的，那我们就一起来学习一下方法吧！

样品取材

Sampling of arugula plants

芝麻菜植物生长在格拉茨（奥地利，大约北纬47°4′13′，东经15°28′19〃）郊区的花园土壤（以下简称非根际土或土壤）中。所用的样品汇总见附加文件1：表S1中。食用植物部分由长叶和短茎组成，在本研究中，当提及微生物栖息地时，称为叶际。在两个连续采样中，分别提取了芝麻菜的叶际和根际微生物组。在第一次采样期间，采集了芝麻菜植株，通过16S rRNA基因片段的扩增子测序来评估肠杆菌微生物组。植物在7月份叶片发育的最后生长阶段收获。在后续采样过程中，同一个花园的样本是在11月初，即植物已经形成花和种子体的最后生长阶段获得的。它们生长在大陆性或北半球气候温暖的夏季，记录的气候数据（附加文件1：图S1）在采样前2周开始，显示相对温暖的温度、温和但有波动的降水、阳光和大气压力水平、总体低风水平和平均空气湿度。除了叶际和根际样本外，还对土壤进行宏基因组测序。所有样品都储存在冰上，到达附近的实验室后立即进行处理。

Arugula plants were grown in garden soil (hereafter referred to as bulk soil) in a suburban region of Graz (Austria; approx. 47°4′ 13′′ N 15° 28′ 19″ E). All samples used are summarized in Additional file 1: Table S1. Edible plant parts consist of elongated leafs with short stalks and are termed as phyllosphere throughout this study when referring to the microbial habitat. The microbial fraction of the Eruca sativa system was extracted separately for both the phyllo- and rhizosphere in two consecutive samplings. During the first sampling, arugula plants were collected to assess the enterobacterial microbiome by amplicon sequencing of 16S rRNA gene fragments. Plants were harvested in their final growth stage of leaf development in July. During the follow-up sampling, samples from the same garden were obtained in early November at their final growth stage when the plants had already formed flowers and seed bodies. They were grown in a warm summer with continental or hemiboreal climate, and the recorded climate profiles (Additional file 1: Figure S1), which started 2 weeks before sampling, show relatively warm temperatures, moderate but fluctuating precipitation, sunshine and atmospheric pressure levels, overall low wind levels, and average air humidity. In addition to the phyllo- and rhizosphere samples, bulk soil was sampled for metagenome sequencing. All samples were stored on ice and immediately processed after arrival at a nearby laboratory.

扩增子测序与肠杆菌16S基因片段处理

Amplicon sequencing and processing of enterobacterial 16S rRNA gene fragments

样品准备

Sample preparation

分别提取了根际和叶际的微生物组，并分别收集和保存每个生境3个独立的生物学重复，每个样本由15-20个叶或根组成。对于样品的均质化，用无菌研钵和研钵物理破坏5 g植物材料，重新悬浮在10 ml 0.85%NaCl中，转移到2 ml eppendorf管中，然后离心（16500 g，20 min，4°C）。所获得的颗粒用土壤FastDNA®提取试剂盒（MP Biomedicals，美国俄亥俄州索伦市）分离总DNA。在FastPrep®FP120仪器（Qbiogene，Bio101，Carlsbad，CA，USA）中以5.0 m s^-1的速度将细胞均匀化两次30 s以实现DNA打段，并根据制造商的方案进行处理。

The microbial fractions were extracted separately for the rhizo- and phyllospere, and three independent single replicates per habitat, consisting of 15–20 leaves or roots, respectively, were collected and stored separately. For the homogenization of the samples, 5 g of plant material per replicate was physically disrupted with a sterile pestle and mortar, re-suspended in 10 ml of 0.85% NaCl, transferred in two 2 ml Eppendorf tubes, and subsequently centrifuged (16,500 g, 20 min, 4 °C). The pellet obtained was used for isolation of the total-community DNA with the FastDNA® SPIN Kit for Soil (MP Biomedicals, Solon, OH, USA). For mechanical lysis, the cells were homogenized twice in a FastPrep® FP120 Instrument (Qbiogene, BIO101, Carlsbad, CA, USA) for 30 s at a speed of 5.0 m s−1 and treated according to the manufacturer’s protocol.

16S扩增与测序

Amplification and sequencing of 16S rRNA fragments

采用两轮PCR扩增，使用标签标记的引物对肠杆菌科16S rRNA基因片段进行了扩增。根据Binh等人描述的方法，使用肠杆菌-F234和肠杆菌-R1423（v3-v8）特定的引物进行PCR[10]。按照Heuer[33]等人的描述进行巢式PCR，使用根据规范修改用于多样品混合的454测序的引物对F984和R1378。使用Wizard SV凝胶PCR纯化系统（PROMEGA，麦迪逊，美国）混合和纯化两步独立PCR反应的产物。纯化的pcr产物混合（每个200 ng）并在罗氏FLX+ 454 Titanium平台（韩国首尔千年基因Macrogen Korea）上进行测序。

The 16S rRNA gene fragments of the Enterobacteriaceae were amplified in a dual-phase nested PCR approach using multiplex identifier (MID) tagged primers. The PCR was conducted with specific primers for enterics, using Entero-F234 and Entero-R1423 (V3-V8), according to the method described by Binh et al. [10]. The nested PCR was carried out as described by Heuer et al. [33] using the primer pair F984 and R1378 modified for multiplex 454 sequencing according to the specification. The products of two independent PCR reactions per sample were pooled and purified using the Wizard SV Gel and PCR Clean-Up System (Promega, Madison, USA). Purified PCR products were pooled (200 ng each) and sequenced on a Roche GS FLX+ 454 Titanium platform (Macrogen Korea, Seoul, South Korea).

生物信息方法处理序列

Bioinformatics sequence processing

使用QIIME软件版本1.8.0[16]分析序列。去除标签序列、引物序列和接头序列，序列质量（最低得分50，编码方式不同，类似于Q20，99%准确度）和长度过滤（最小原始片段长度：E1≥350；E2和E3≥430），然后进行去噪步骤（使用去噪流程和去噪脚本；均包含在QIIME分析流程中）。剩余序列去除非特异扩增、质体和线粒体起源的序列和嵌合体。OTU表采用UCLUST[25]在100%的相似度水平上创建。使用RDP分类器[63]v2.5和NCBI RefSeq数据库对唯一序列进行分类。数据集在naїve贝叶斯分类器[63]上被抽平为每个实验中的最小测序数据量，以计算α和β多样性指数。利用Cytoscape V.3.1.0[57]构建了一个基于OTU的网络，该网络基于100%相似水平的OTU。Centiscape V.2.2[55]用于计算网络每个节点的质心值，并对其进行相应标记。

Sequences were analyzed with the QIIME software version 1.8.0 [16]. MID, primer, and adapter sequences were removed, and the sequences were quality- (minimal score 50) and length filtered (minimal raw fragment length: E1 ≥ 350; E2 and E3 ≥ 430), followed by a denoising step (using the denoise wrapper and denoiser script; both contained within the QIIME pipeline). Chimeras and remaining sequences of non-target, plastidal, and mitochondrial origin were removed. OTU tables were created with UCLUST [25] at a 100% cut-off level. Unique reads were classified with the RDP classifier [63] v2.5 and the NCBI RefSeq database. The datasets were rarified on a naїve Bayesian classifier [63] to the number of least reads within each experiment, to compute alpha and beta diversity indices. An OTU-based network correlating the OTUs at 100% cut-off level was constructed with Cytoscape v.3.1.0 [57]. Centiscape v.2.2 [55] was used to calculate centroid values for each node of the network and label them accordingly.

宏基因组测序和分析

Metagenome shotgun sequencing and bioinformatic processing

样品准备

Sample preparation

称取5克每种样品（叶际、根际和土壤），与10毫升0.85%氯化钠一起放入无菌塑料袋中，并在一台袋式搅拌机（Interscience，St.Nom，France）中机械处理两次，处理时间为210秒。将样品置于冰上5分钟。再将均质细胞悬浮液进一步转移到S34管中，以10000转/分的速度离心20分钟。每种方法丢弃上清液，并将细胞颗粒储存在−20°C。我们总共处理了90袋（细胞颗粒堆积三次，上清液每轮离心之后丢弃以增加总产量），从35袋中取出8×s34管（35 ml/s34）根际样品，从35袋中取出6×s34管（35 ml/s34）土样。使用fastdna®土壤DNA提取试剂盒（MP Biomedicals，Solon，OH，USA）从每个栖息地的六个植物样品（叶、根际、土壤）中提取DNA，使用300 mg的每个粗细胞颗粒，并按照制造商的方案进行处理。DNA样品在100μL超纯水溶液中洗脱，并使用纳米滴光度计检查质量和数量。然后，将每一个生物学重复的2μg汇集起来，得到每个生境总共12μg的宏基因组DNA。

Five grams of each sample (phyllo-, rhizosphere, and bulk soil) were weighed, transferred into sterile plastic bags together with 10 ml 0.85% NaCl, and mechanically processed twice for 210 s in a Bagmixer (Interscience, St. Nom, France). Samples were placed at interims of 5 min on ice. Homogenized cell suspensions were further transferred into S34 tubes, and centrifuged at 10,000 rpm for 20 min. The supernatant was discarded for each approach and the cell pellets were stored at − 20 °C. In total, we processed 24 × S34 tubes (35 ml/S34) foliage samples from a total 90 bags (cell pellets were stacked three times and the supernatant discarded between each centrifugation round in order to increase the total yield), 8 × S34 tubes (35 ml/S34) rhizosphere samples from a total of 35 bags, and 6 × S34 tubes (35 ml/ S34) soil samples from a total of 35 bags, respectively. DNA was extracted using the FastDNA® SPIN Kit for Soil (MP Biomedicals, Solon, OH, USA) from six pellets per habitat (foliage, rhizosphere, soil) using 300 mg each of the crude cell pellets and processing as stated in the manufacturer’s protocol. The DNA samples were eluted in 100 μL H2Oultra pure and checked for quality and quantity using NanoDrop Photometer. Then, 2 μg of each replicate were pooled for a total 12 μg of metagenomic DNA per habitat.

宏基因组测序

Shotgun sequencing of total community DNA

在Illumina-Hiseq2000系统（2×150 bp）上进行了宏基因组测序，方法采用Eurofins MWG Operon的操作指南。

Metagenomic sequencing was performed on an Illumina HiSeq2000 system (2 × 150 bp) by Eurofins MWG Operon (Ebersberg, Germany) following the Eurofins MWG Operon protocol.

序列质量控制

Sequence quality filtering

原始读取数据上传到Galaxy主服务器（http://usegalaxy.org）并使用FastQ Joiner、FastQ Groomer、FastQ Quality Trimmer和Filter FastQ（最小长度=75，最小质量=20，允许超出质量范围=2）进行处理。对于每个数据集，只保留高质量的序列进行进一步处理，其中：叶际：41,867,724（79.8%），根际：35,463,395（74.7%），土壤：27,085,866（79.3%）。

The raw read data were uploaded to the Galaxy Main server (http://usegalaxy.org) and processed using the FASTQ Joiner, FASTQ Groomer, FASTQ Quality Trimmer, and Filter FASTQ (min. length = 75, min. quality = 20, allowed outside quality range = 2). For each of the datasets, only high-quality reads were retained for further processing, resulting in the following read numbers to process further: phyllosphere: 41,867,724 (79.8%), rhizosphere: 35,463,395 (74.7%), and soil: 27,085,866 (79.3%).

比对参考序列

Reference alignment

为了减少植物来源序列引起的遗传偏差，对油菜Brassica oleracea、油菜brapa-rapa(brapa-1.0）和萝卜raphanus sativus(rs1.0)的参考基因组进行了序列比对。为了将过滤后的高质量序列与参考基因组比对，使用了bowtie 2[39]和默认比对参数。附加文件1：表S2显示了比对结果。以上研究表明，与三个基因组中的至少一个的基因可比对将被排除。因此，只有那些与三个参考基因组中的任何一个不一致的序列才被用作下一步程序的输入。在这三个样本中，叶际和土壤样本的可比对序列分别是最大和最小的，这与预期的一致。

Reads were aligned to the reference genomes of Brassica oleracea (BOL; ncbi.nlm.nih.gov/genome), Brassica rapa (Brapa_1.0), and Raphanus sativus (Rs1.0), respectively, with the goal of discarding aligned reads in order to reduce plant-derived genetic bias. To align the quality-filtered reads to the reference genomes, Bowtie 2 [39] was used, with the default alignment parameters. Additional file 1: Table S2 shows the alignment results. Those reads that aligned to at least one of the three genomes were excluded from further processing. Thus, only those reads that were not aligned to any of the three reference genomes were used as input to the assembly program. The phyllosphere and soil samples had the largest and smallest numbers of aligned reads among the three samples, respectively, which was to be expected.

序列组装

Sequence assembly

使用Velvet（版本1.2.10；[67]）从头组装测序结果为重叠群(Contigs)。由于k-mer长度的选择至关重要，因此对每个样品进行了多个不同K值的组装，以确定最佳k-mer长度。在所有程序集中，Velvet会自动找到预期的覆盖范围值，并且成对的末端读取的插入长度设置为350。附加文件1：表S3显示了最终组件的关键统计数据，其中k是所用k-mer的长度。只有那些覆盖率中位数至少为90%且长度至少为k值两倍的k值被选作后续分析。

Assembly of the reads into contigs was performed using the Velvet (version 1.2.10; [67]) de novo assembly software. Since the choice of the k-mer length is crucial, multiple assemblies with different k values were performed on each sample in order to determine the best k-mer length. In all assemblies, the expected coverage value was automatically found by Velvet and the insert length for paired-end reads was set at 350. Additional file 1: Table S3 shows the key statistics of the final assemblies, where k is the length of the k-mer used. Only those contigs that had at least 90% of the median coverage and were at least as long as twice the k value were selected for subsequent analysis.

物种和功能分析

Taxonomic and functional analyses

对于每个样本，使用Megan程序（版本5.11.3；[36]）执行分类和功能分析。对于分类分析，BLASTN程序在TimeLogic DeCypher boards（www.timelogic.com）上对所有过滤后的序列上运行，与2016年2月12日从NCBI下载的34,646,553个核苷酸序列的“NT”数据库相对应。为了创建功能分析的输入，BLASTX（核苷酸-顶蛋白）程序在时间逻辑密码板上的所有过滤后的序列上运行，与“nr”数据库相对应，2016年2月12日从NCBI下载了81622391个蛋白质序列。将每个样本的blastn结果导入MEGAN中，对样本进行分类分析，其中每个contig被分配到尽可能低的细菌谱系，并具有足够的置信度。从每个样本中随机抽取相同数量的分配给细菌的菌落（叶际、根际和土壤样本中的最小菌落数），并用于比较随后三个样本中的细菌分类和功能分析结果。因此，细菌的分类和功能分析分别基于BLASTN和BLASTX结果，从每个样本中提取89462个亚样本。同样，肠杆菌科的分类和功能分析分别基于叶际和根际样本中5494个亚样本的BLASTN和BLASTX结果。从土壤样本中提取的极低数量（356）的重叠群被分配到肠杆菌科；因此，我们无法将它们包括在有意义的比较中。

For each sample, taxonomy and functional analyses were performed using the MEGAN program (version 5.11.3; [36]). For taxonomic analyses, the blastn program was run on all the filtered contigs on the TimeLogic DeCypher boards (www.timelogic.com), against the “nt” database with 34,646,553 nucleotide sequences downloaded from NCBI on February 12, 2016. In order to create input for functional analyses, the blastx (nucleotide-toprotein) program was run on all the filtered contigs on the TimeLogic DeCypher boards, against the “nr” database with 81,622,391 protein sequences downloaded from NCBI on February 12, 2016. The blastn results from each sample were imported into MEGAN to perform the taxonomy analysis for the sample, where each contig is assigned to a bacterial lineage at the lowest rank possible with sufficient confidence. An identical number of contigs assigned to bacteria (the smallest such number among the phyllosphere, rhizosphere and bulk soil samples) were randomly extracted from each sample and used to compare subsequent bacterial taxonomic and functional analyses results among the three samples. Therefore, the taxonomy and functional analyses of the bacteria are based on the blastn and blastx results, respectively, for the 89,462 subsampled contigs from each sample. Similarly, the taxonomy and functional analyses of Enterobacteriaceae are based on the blastn and blastx results, respectively, for the 5494 subsampled contigs from the phyllosphere and rhizosphere samples, respectively. A very low number (356) of contigs from the soil sample was assigned to the Enterobacteriaceae; therefore, we were not able to include them in a meaningful comparison.

抗生素抗性分析

Antibiotic resistance analysis

使用综合抗生素耐药性数据库（CARD；[45]）分析微生物抗生素耐药性概况。从CARD网站（card.mcmaster.ca）下载用于宏基因组数据的抗生素抗性蛋白质序列fasta文件，作为每个样本的装配序列的核苷酸到蛋白质比对（blastx）分析的数据库。该分析仅使用最优比对结果，一致率至少为50%。如果蛋白质比对到多个类别，那么对抗生素抗性蛋白的一次比对被多个抗生素抗性类别计数。

Microbial antibiotic resistance profiles were analyzed using the Comprehensive Antibiotic Resistance Database (CARD; [45]). The antibiotic resistance protein sequences for metagenomic data were downloaded from the CARD website (card.mcmaster.ca) as a FASTA file and used as the target sequences for the nucleotide-to-protein Blast (blastx) analysis of the assembled contigs for each sample. The analysis was performed by using only the top blast hits, with a percentage identity of at least 50%. A single hit of a contig to an antibiotic resistance protein was counted as multiple antibiotic resistance category hits, if the protein mapped to multiple categories.

MG-RAST分析末组装序列

MG-RAST analysis of non-assembled reads

基于非组装的宏基因组序列，采用单独的生物信息学方法分析了肠道细菌部分的丰度。宏基因组（叶际、根际和土壤）上传到宏基因组MG-RAST服务器[47]上，并使用默认参数进行初始处理。处理过程包括去除人工PCR扩增重复序列[29]、低质量序列[21]、短序列和含有不明确碱基的序列。从RefSeq数据库获得系统发育注释，最大E值为10-5，最小相似度为60%，最小比对长度为15 bp。

Abundances within the enterobacterial fraction were analyzed in a separate bioinformatics approach based on non-assembled metagenome sequences. The metagenomes (phyllosphere, rhizosphere, and bulk soil) were uploaded on the Metagenomic RAST (MG-RAST) server [47] and initially processed using the default parameters. The processing included the removal of artificial replicate sequences [29], low-quality sequences [21], short sequences, and sequences containing ambiguous bases. Phylogenetic annotations were obtained from the RefSeq database, with a maximum e value of 10−5, minimum identity of 60%, and minimum alignment length of 15 bp for RNA databases.

菌株分离和抗性检测

Isolation of a strain collection and antibiotic susceptibility tests

为了分离肠杆菌，将第二次取样的均质细胞悬浮液置于培养基上。从芝麻菜叶际、根际和土壤中共分离到180株好氧菌。不同的培养基，包括PDA、Mac-Conkey、Kingsb、R2A（均来自德国卡尔斯鲁厄的Carl Roth GmbH）、MIS（根据[61]编制）和NAII（德国柏林的Sifin），允许广谱覆盖以创建一个代表性菌株库。为了避免选择性条件引起的分离偏倚，最初分离所用的培养基均未添加抗生素。采用纸片扩散抗菌药敏试验方法[35]检测对一组特定抗生素的敏感性。用NAII琼脂代替MH琼脂作为试验板。分别检测其对氨苄西林（10μg）、氯霉素（30μg）、红霉素（15μg）、庆大霉素（10μg）、青霉素（10单位；6μg）、多粘菌素B（300单位；30μg）、链霉素（10μg）和四环素（30μg）的敏感性。

For the isolation of enterobacteria, homogenized cell suspensions from the second sampling were plated on growth media. In total, 180 aerobic bacteria were isolated from the arugula phyllosphere and rhizosphere as well as bulk soil. Different media including PDA, Mac- Conkey, KingsB, R2A (all obtained from Carl Roth GmbH, Karlsruhe, Germany), MIS (prepared according to [61]), and NAII (Sifin, Berlin, Germany) allowed a broad-spectrum coverage to create a representative strain library. None of the media used for the initial isolation was supplemented with antibiotics, in order to avoid an isolation bias caused by selective conditions. A disk diffusion antimicrobial susceptibility test method [35] was used to test for susceptibility towards a defined set of antibiotics. Instead of MH agar, NAII agar was used for the test plates. Unique morphotypes were tested for their sensitivity against Ampicillin (10 μg), Chloramphenicol (30 μg), Erythromycin (15 μg), Gentamicin (10 μg), Penicillin G (10 Units; 6 μg), Polymyxin B (300 Units; 30 μg), Streptomycin (10 μg), and Tetracycline (30 μg).

猜你喜欢

• 10000+: 菌群分析
  宝宝与猫狗 提DNA发Nature 实验分析谁对结果影响大  Cell微生物专刊 肠道指挥大脑
• 系列教程：微生物组入门 Biostar 微生物组  宏基因组
• 专业技能：生信宝典 学术图表 高分文章 不可或缺的人
• 一文读懂：宏基因组 寄生虫益处 进化树
• 必备技能：提问 搜索  Endnote
• 文献阅读 热心肠 SemanticScholar Geenmedical
• 扩增子分析：图表解读 分析流程 统计绘图
• 16S功能预测   PICRUSt  FAPROTAX  Bugbase Tax4Fun
• 在线工具：16S预测培养基 生信绘图
• 科研经验：云笔记  云协作 公众号
• 编程模板: Shell  R Perl
• 生物科普: 肠道细菌 人体上的生命 生命大跃进  细胞暗战 人体奥秘  

写在后面

为鼓励读者交流、快速解决科研困难，我们建立了“宏基因组”专业讨论群，目前己有国内外5000+ 一线科研人员加入。参与讨论，获得专业解答，欢迎分享此文至朋友圈，并扫码加主编好友带你入群，务必备注“姓名-单位-研究方向-职称/年级”。技术问题寻求帮助，首先阅读《如何优雅的提问》学习解决问题思路，仍末解决群内讨论，问题不私聊，帮助同行。

学习扩增子、宏基因组科研思路和分析实战，关注“宏基因组”

点击阅读原文，跳转最新文章目录阅读
https://mp.weixin.qq.com/s/5jQspEvH5_4Xmart22gjMA