Abstract
Within the past decade, microbiologists have moved from detecting single antibiotic resistance genes (ARGs) to detecting all known resistance genes within a sample due to advances in next generation sequencing. This has provided a wealth of data on the variation and relative abundances of ARGs present in a total bacterial population. However, to use these data in terms of therapy or risk to patients, they must be analyzed in the context of the background microbiome. Using a quantitative PCR ARG chip and 16S rRNA amplicon sequencing, we have sought to identify the ARGs and bacteria present in a fecal sample of a healthy adult using genomic tools. Of the 42 ARGs detected, 12 fitted into the ResCon1 category of ARGs: cfxA, cphA, bacA, sul3, aadE, blaTEM, aphA1, aphA3, aph(2′)-Id, aacA/aphd, catA1, and vanC. Therefore, we describe these 12 genes as the core resistome of this person's fecal microbiome and the remaining 30 ARGs as descriptors of the microbial population within the fecal microbiome. The dominant phyla and genera agree with those previously detected in the greatest abundances in fecal samples of healthy humans. The majority of the ARGs detected were associated with the presence of specific bacterial taxa, which were confirmed using microbiome analysis. We acknowledge the limitations of the data in the context of the limited sample set. However, the principle of combining qPCR and microbiome analysis was shown to be helpful to identify the association of the ARGs with specific taxa.
Introduction
T
The aims of this study were to identify and measure the relative abundances of ARGs and bacteria present in a fecal sample of a healthy adult using molecular biology tools. As we move further toward genomic analysis of ARGs and bacteria, we must generate guidelines for the interpretation of the data generated. To do this, we must understand which bacteria and ARGs are present in the healthy human and then what constitutes a risk to the treatment of a patient in terms of likelihood of transfer to pathogenic bacteria.
Materials and Methods
Sample preparation and DNA extraction
A fecal sample was collected from a healthy adult who had taken no antibiotics in the previous 2 years. It was immediately homogenized and 0.6 g was added directly to the MoBio™ Power Soil® DNA isolation kit tube in step one of the protocol. The DNA was extracted using the kit protocol.
The NUI Maynooth Biomedical and Life Sciences Research Ethics Sub-Committee approved this study and experimental protocols (application reference number BRSERC-2014-007). All methods were performed in accordance with ethical guidelines and regulations. Written informed consent was obtained from the subject.
16S rRNA amplicon sequencing
The V3 and V4 regions of the 16S rRNA genes were amplified and sequenced using the Illumina MiSeq® primers and protocol. 9 A 2 × 150 paired-end configuration was used for sequencing. The image analysis and base calling were processed using MiSeq Control Software. The data were quality control checked and trimmed and analyzed initially using the Illumina BaseSpace app. 10 The sequencing data were processed using Quantitative Insights Into Microbial Ecology version 1.5.0. 11 Shannon diversity, collectors curve, and Chao-1 were used to determine the evenness of the 16S rRNA results. 12
ARG relative abundance (qPCR)
DNA extracted from the fecal samples was used to analyze the relative abundances of the ARGs using a highly parallel quantitative PCR (qPCR) platform (Wafergen Smartchip). 13 The samples analyzed comprised three biological replicates for each sample and three technical replicates of each biological replicate. The samples were analyzed for the relative abundance of 384 primer sets targeting known ARGs and mobile genetic elements (Supplementary Table S1; Supplementary Data are available online at www.liebertpub.com/mdr). The Ct values of each primer pair were normalized using the 16S rRNA gene values (deltaCt = CtARG−Ct16SrRNA). Results with a Ct value of >28 were removed. The deltaCt values and fold change were calculated according to Relative Gene Expression Data Using Real-Time Quantitative PCR and the deltadeltaCt method. 14 The SmartChip has been validated by comparison with metagenomics. 15
Results
ARG resistome
Forty-two different ARGs (plus six mobile elements and two repressor genes) were detected in the fecal sample (Table 1). Using the definitions of ARGs provided by Martinez et al. would exclude efflux genes, ampC genes, the erythromycin resistance genes, and the tetracycline resistance genes as these are not considered ARGs from an ecological viewpoint. 16 The erythromycin and tetracycline resistance genes are frequently identified on the chromosomes of anaerobes commonly found in the human gut microbiome and are not considered to confer resistance in these bacteria. Their presence identified the bacteria harboring these genes, rather than a resistance reservoir. Of the 42 ARGs detected, 12 fitted into the ResCon1 category of ARGs as defined by Martinez et al. 16 These were ARGs cfxA, cphA, bacA, sul3, aadE, aphA1, aphA3, aph(2′)-Id, aacA/aphd, blaTEM, catA1, and vanC. Therefore, we describe these 12 genes as the core resistome of this person's fecal microbiome and the remaining 30 ARGs as descriptors of the microbial population within the fecal microbiome.
Three technical replicates were averaged to measure the Ct values for each primer set.
MDR, multidrug resistance; MGE, mobile genetic element.
Relative abundances of genes
The relative abundances of each ARG were determined by comparison with the 16S rRNA gene abundance. This ensured that variations in DNA quantities were not responsible for variations in the ARG abundances in the sample (Table 1). The genes with the highest relative abundances comprised mobile genetic elements (insertion sequences and transposases), tetracycline and erythromycin resistance genes associated with the anaerobes present in the human gut microbiome, and the cfxA beta-lactamase gene. The genes detected in the lowest relative abundances comprised efflux genes (tetC, ceoA, and tetB), a tetracycline resistance gene regulator (tetR), vanC, mobile genetic elements (IncN and pNI105), beta-lactamase (blaTEM), aminoglycoside (aacA/aphd), and chloramphenicol (catA1) resistance genes.
Bacterial community analysis using 16S rRNA gene amplicon sequencing
Microbial composition
The phyla, which constituted the taxa at >1% within the 89303 OTU sequence reads, that passed the quality control filtering comprised Firmicutes (66%), Bacteroidetes (28%), Proteobacteria (2.5%), unclassified (1.5%), and Actinobacteria (1.2%). The dominant phyla are consistent with previous findings. 16 Within these phyla, the taxa were spread across 32 classes, with 6 of those representing >1% relative abundances: Clostridia (65%), Bacteroidia (20%), Flavobacteria (7.3%), unclassified (2%), Actinobacteria (1.2%), and Bacilli (1%). The classes were subdivided into 68 orders with Clostridiales (64%), Bacteroidales (20%), Flavobactriales (7.3%), and unclassified (2.3%), representing those at >1%. The families within the microbiome (n = 143) at >1% were represented by Lachnospiraceae (32%), Ruminococcaceae (25%), Bacteroidaceae (11%), Flavobacteriaceae (7.3%), unclassified (4.6%), Clostridiaceae (3.8%), Paraprevotellaceae (3.7%), Odoribacteraceae (2.7%), Porphyromonadaceae (2.2%), and Eubacteriaceae (1.6%). A total of 263 different genera were represented in the fecal microbiome and those with >1% relative abundance comprised 19 different genera (including unclassified), which represented 90% of the total microbiome composition (Table 2). Dominant genera also agree with those previously detected in the greatest abundances in fecal samples of a healthy human. 17 Although correlations between bacterial phylogenies and antibiotic resistomes have been reported by Pehrsson et al., this was using diverse habitats. 18
Discussion
In a mixed bacterial population, such as human feces, bacteria are present that contain ARGs either on their chromosomes or on mobile elements. These bacteria are maintained within the fecal population due to their roles, regardless of the ARG or selective antibiotic pressures. With advances in next generation sequencing (NGS) came studies measuring the influences of antibiotics on complex bacterial populations and their total antibiotic resistomes, such as the human gut microbiome.1–8 However, to identify the changes occurring within a population of bacteria, we must first identify the ARGs present in the natural fecal bacterial population, independent of selective pressure. With reduced cost and increased capacity, NGS has become a potential tool for the identification of ARGs and pathogens directly from patients. In order for such technology to function, one must understand the difference between carriage and selection of bacterial species and ARGs. This study aimed to integrate NGS and antibiotic resistance qPCR chip technology to describe the fecal population of bacteria and the relative abundance of their resistome in a healthy human under no antibiotic selective pressure.
The published data on the human gut microbiome alterations due to antibiotics are highly variable. To date, there have been no definite conclusions on the bacterial genera that proliferate or decrease after specific antibiotic administration. There is no consensus on whether any microbiome changes occur at all or whether intersubject variability was greater than the effects of the antibiotics administered. The first studies of the human gut microbiome tried to identify and describe the core set of bacterial taxa responsible for health and disease. However, such studies among healthy individuals revealed wide variation in the taxonomic composition of the microbiome, which prevented the discovery or identification of a core microbiome.19,20
The majority of the ARGs detected may have been associated with the presence of specific bacterial taxa. There was a high proportion of mobile genetic elements detected, which suggests high genetic mobility within the fecal microbiome and the detection of several ARGs, which are not associated with intrinsic resistance of common gut microbiome bacteria. The conclusion from this study is that although many ARGs can be detected and their abundances measured using DNA-based tools, we must put these genes in the context of the bacterial composition of the sample, in this case feces from a healthy human, to identify the genes, which pose a risk to the treatment of pathogenic infections. This study highlights the need to put DNA analysis in context and has listed several ARGs that may be present on the chromosome of the natural fecal microbiome. The conclusions may not be further extrapolated due to the small sample size.
Footnotes
Acknowledgment
Dr. Walsh received funding from Maynooth University for this study.
Disclosure Statement
The authors have no competing financial interests. The funding bodies did not play a role in the experiments or their interpretation.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
