CaptureSeq: Hybridization-Based Enrichment of cpn60 Gene Fragments Reveals the Community Structures of Synthetic and Natural Microbial Ecosystems

Abstract

Background: The molecular profiling of complex microbial communities has become the basis for examining the relationship between the microbiome composition, structure and metabolic functions of those communities. Microbial community structure can be partially assessed with "universal" PCR targeting taxonomic or functional gene markers. Increasingly, shotgun metagenomic DNA sequencing is providing more quantitative insight into microbiomes. However, both amplicon-based and shotgun sequencing approaches have shortcomings that limit the ability to study microbiome dynamics.

Methods: We present a novel, amplicon-free, hybridization-based method (CaptureSeq) for profiling complex microbial communities using probes based on the chaperonin-60 gene. Molecular profiles of a commercially available synthetic microbial community standard were compared using CaptureSeq, whole metagenome sequencing, and 16S universal target amplification. Profiles were also generated for natural ecosystems including antibiotic-amended soils, manure storage tanks, and an agricultural reservoir.

Results: The CaptureSeq method generated a microbial profile that encompassed all of the bacteria and eukaryotes in the panel with greater reproducibility and more accurate representation of high G/C content microorganisms compared to 16S amplification. In the natural ecosystems, CaptureSeq provided a much greater depth of coverage and sensitivity of detection compared to shotgun sequencing without prior selection. The resulting community profiles provided quantitatively reliable information about all three domains of life (Bacteria, Archaea, and Eukarya) in the different ecosystems. The applications of CaptureSeq will facilitate accurate studies of host-microbiome interactions for environmental, crop, animal and human health.

Conclusions: cpn60-based hybridization enriched for taxonomically informative DNA sequences from complex mixtures. In synthetic and natural microbial ecosystems, CaptureSeq provided sequences from prokaryotes and eukaryotes simultaneously, with quantitatively reliable read abundances. CaptureSeq provides an alternative to PCR amplification of taxonomic markers with deep community coverage while minimizing amplification biases.

Keywords: chaperonin-60; hybridization; microbiome; soil microbiota.

Figure 1

Examination of an artificial microbial community (Zymobiomics Microbial Community Standard) using 16S rRNA-encoding gene amplification and CaptureSeq. The relative abundances of each of the 8 bacterial OTU present in the synthetic community are shown for each of 3 replicates of each method compared to the theoretical composition provided by the manufacturer (dashed line).

Figure 2

CaptureSeq results obtained by reference mapping on an ecologically diverse range of samples including soil (n = 6); manure storage tanks (n = 3); and a freshwater pond (n = 1). The relative abundances of individual phyla are expressed as a proportion of the entire pan-domain microbial communities.

Figure 3

Quantification of an assembled OTU corresponding to Acinetobacter baumanii/A. calcoaceticus in soil samples treated with antibiotics. The number of CaptureSeq reads mapping to this taxonomic cluster in each soil sample is indicated above each bar.

Figure 4

Good’s coverage estimate as a function of the average total sequencing effort for six soil samples each profiled using amplicon (red), CaptureSeq (blue), or shotgun metagenomic (green) approaches.

Figure 5

Alpha diversity metrics for soil samples profiled using cpn60 amplicon (red), CaptureSeq (blue), or shotgun metagenomic (green) approaches. Metrics were calculated using libraries that were down-sampled from 250–2750 reads and were averaged across 100 bootstrapped datasets. The shaded area corresponds to the standard deviation of the three replicate soil plots for each antibiotic treatment condition.

Figure 6

Proportional abundances of taxonomic clusters obtained by reference mapping for type I chaperonins in soil samples profiled using amplicon, CaptureSeq, or shotgun metagenomic approaches. Samples were clustered based on Bray–Curtis distance, and reference clusters composing a minimum of 0.5% of the mapped sequencing reads in any one sample are shown. Samples are coded according to antibiotic treatment: black (10 ppm) or gray (0 ppm).

Figure 7

Principal coordinate analysis of Bray–Curtis dissimilarity between soil samples profiled using CaptureSeq (square), shotgun metagenomics (triangle), or cpn60 amplicon (circle) approaches. OTU frequencies were determined by reference mapping as described in the text. (A) All approaches considered together. (B) CaptureSeq. (C) Shotgun metagenomics. (D) cpn60 amplicon.