Comparative analysis of fecal DNA extraction methods with phylogenetic microarray: Effective recovery of bacterial and archaeal DNA using mechanical cell lysis

https://doi.org/10.1016/j.mimet.2010.02.007Get rights and content

Abstract

Several different protocols are used for fecal DNA extraction, which is an integral step in all phylogenetic and metagenomic approaches to characterize the highly diverse intestinal ecosystem. We compared four widely used methods, and found their DNA yields to vary up to 35-fold. Bacterial, archaeal and human DNA was quantified by real-time PCR, and a compositional analysis of different extracts was carried out using the Human Intestinal Tract Chip, a 16S rRNA gene-based phylogenetic microarray. The overall microbiota composition was highly similar between the methods in contrast to the profound differences between the subjects (Pearson correlations > 0.899 and 0.735, respectively). A detailed comparative analysis of mechanical and enzymatic methods showed that despite their overall similarity, the mechanical cell disruption by repeated bead beating showed the highest bacterial diversity and resulted in significantly improved DNA extraction efficiency of archaea and some bacteria, including Clostridium cluster IV. By applying the mechanical disruption method a high prevalence (67%) of methanogenic archaea was detected in healthy subjects (n = 24), exceeding the typical values reported previously. The assessment of performance differences between different methodologies serves as a concrete step towards the comparison and reliable meta-analysis of the results obtained in different laboratories.

Introduction

Fecal samples provide a non-invasive material for the analysis of the gastrointestinal (GI) microbes. The GI microbiota consists of up to hundred trillion bacteria, which significantly contribute to our health and well-being by food digestion and nutrition intake, neutralizing pathogens, and shaping our immune system (Tappenden and Deutsch, 2007). Hence, extensive characterization of the GI microbiota using the recently developed high-throughput tools for diversity analysis and functional metagenomics is ongoing (Salonen et al., 2009, Zoetendal et al., 2008).

The fecal microbiota is individual-specific and consists of thousands of species-level phylotypes, most of which belong to Firmicutes, Bacteroidetes, Actinobacteria, Proteobacteria and Verrucomicrobia (Dethlefsen et al., 2008, Rajilić-Stojanović et al., 2007). The GI ecosystem is dominated by the Gram-positive bacteria while Bacteroidetes, Proteobacteria and Verrucomicrobia represent the Gram-negative bacteria among the predominant taxa. In contrast to the bacterial diversity, the intestinal archaea consist only of Methanobrevibacter smithii and Methanosphaera stadtmanae (Eckburg et al., 2005, Scanlan et al., 2008). These methanogens operate on the final level of the intestinal food web, disposing hydrogen and some other fermentation end products, and thereby regulate the dynamics of the entire GI ecosystem (Samuel et al., 2007). The documented prevalence of the methanogens varies from 33% to 96% and may reflect the health status of the carrier (Scanlan et al., 2008, Schwiertz et al., 2009, Dridi et al., 2009, Zhang et al., 2009).

The irregular physico-chemical structure and heterogeneous microbiota composition of the fecal specimen challenges the extraction of representative metagenomic DNA (Swidsinski et al., 2008). Different approaches to lyse bacteria in the fecal samples include mechanical, enzymatic and chemical treatments, which can be used alone or in combination. Mechanical disruption methods are typically based on bead beating, although high pressure or ultrasound treatments are also be employed. The advantage of bead beating in fecal DNA extraction is its efficacy and the concurrent sample homogenization. This allows uniform penetration of lysis buffer to the entire fecal sample regardless of its consistency. However, a disadvantage of the mechanical disruption is the shearing of DNA, which limits its use for applications relying on intact chromosomes (de Lipthay et al., 2004). Chemical and enzymatic lysis methods are gentler but may suffer from extraction bias due to the limited spatial access to all target organisms and selectivity for different cell types, as reported for the community DNA extracted from the feces (Morita et al., 2007) and the soil (Carrigg et al., 2007).

Some of the available fecal DNA extraction methods have been compared and evaluated based on their cell lysis efficiency (Morita et al., 2007), DNA yield, and PCR performance (Nechvatal et al., 2008, Yu and Morrison, 2004, Anderson and Lebepe-Mazur, 2003, McOrist et al., 2002). Moreover, comparisons have been based on the bacterial diversity determined by a variety of fingerprinting methods (Li et al., 2007, Scupham et al., 2007, Yu and Morrison, 2004). The studies revealed that the DNA extraction method affected the resultant microbiota composition, although the methodological variation was minor compared to the inter-individual variation (Li et al., 2007, Scupham et al., 2007). However, the techniques applied did not allow identifying the differentially extracted taxa. Among the tested methods, commercial QIAamp Stool Kit by Qiagen has shown high extraction efficiency and PCR-compatibility (Nechvatal et al., 2008, McOrist et al., 2002, Scupham et al., 2007). This method has frequently been employed in the extraction of community DNA from the human feces (Dethlefsen et al., 2008, Li et al., 2007, Zoetendal et al., 2006, Eckburg et al., 2005, Zhang et al., 2009, Khachatryan et al., 2008, Palmer et al., 2007).

Comparison of results from various groups suggests substantial differences in the fecal bacterial composition of healthy individuals. The most striking example is the reported absence or scarcity of normally predominant taxa such as Bacteroidetes spp (Andersson et al., 2008, Gill et al., 2006) or Actinobacteria (Eckburg et al., 2005, Khachatryan et al., 2008, Palmer et al., 2007). Although such results could arise from biological variation between fecal samples, it is also possible that sample preparation methods, such as the DNA extraction step used in each study, could play a role in the observed variation. We selected four commonly used fecal DNA extraction methods representing different principles for the cell lysis and DNA purification to study the effect of the DNA extraction method on the total microbiota composition. We used the Human Intestinal Tract (HIT) Chip, a novel high-density phylogenetic microarray, which enables the profiling and identification of 1033 intestinal bacterial phylotypes and their further analysis on phylotype, genus or phylum level (Rajilić-Stojanović et al., 2009). Quantitative PCR (qPCR) was used to quantify bacteria and methanogenic archaea. After the initial screening, we selected the two best performing methods and analyzed them in detail. The results revealed that while the vast majority of bacteria were similarly recovered between the two methods, the mechanical DNA extraction method provided the highest bacterial diversity and extraction efficiency of archaeal DNA coupled to high technical reproducibility. This work may have important ramifications for interpretation and comparison of the various GI microbiota analyses that are rapidly accumulating.

Section snippets

Fecal samples

Fecal samples from 5 healthy adults were collected in a context of a clinical trial (Kekkonen et al., 2008). Additional samples from 19 healthy individuals were used to study the methanogen carriage. The Ethics Committee of the Hospital District of Helsinki and Uusimaa approved the study protocols, and all the subjects gave their written informed consent. Samples were taken at home and frozen at − 20 °C, and brought within 24 h to the study centre where they were stored at − 80 °C until further

DNA yield and purity

The purpose of this study was to select and statistically validate an optimal method for the fecal DNA extraction to be applied in the intestinal microbiota characterization of clinical samples. Four extraction methods (Table 1) were compared using homogenized fecal samples and their DNA quality and quantity were used as screening criteria to select the best methods for the final validation.

The mean DNA yields g 1 feces, determined with the fluorometric assay, varied up to 35-fold within a

Acknowledgements

This work was funded by the Finnish Funding Agency for Technology and Innovation (TEKES; grant number 40274/06) and was performed in the Centre of Excellence on Microbial Food Safety Research, Academy of Finland. We thank Hans Heilig for excellent technical assistance and Valio Ltd. for the collection of fecal samples.

References (49)

  • J.H. Apajalahti et al.

    Effective recovery of bacterial DNA and percent-guanine-plus-cytosine-based analysis of community structure in the gastrointestinal tract of broiler chickens

    Appl. Environ. Microbiol.

    (1998)
  • T.A. Brettell et al.

    Forensic science

    Anal. Chem.

    (2007)
  • C. Carrigg et al.

    DNA extraction method affects microbial community profiles from soils and sediment

    Appl. Microbiol. Biotechnol.

    (2007)
  • L. Dethlefsen et al.

    The pervasive effects of an antibiotic on the human gut microbiota, as revealed by deep 16 S rRNA sequencing

    PLoS Biol.

    (2008)
  • J. Dicksved et al.

    Molecular analysis of the gut microbiota of identical twins with Crohn's disease

    ISME J.

    (2008)
  • B. Dridi et al.

    High prevalence of Methanobrevibacter smithii and Methanosphaera stadtmanae detected in the human gut using an improved DNA detection protocol

    PLoS ONE.

    (2009)
  • P.B. Eckburg et al.

    Diversity of the human intestinal microbial flora

    Science

    (2005)
  • Faraway, J. J., 2002. Practical Regression and ANOVA using R. Published on the URL:...
  • S.R. Gill et al.

    Metagenomic analysis of the human distal gut microbiome

    Science

    (2006)
  • C.A. Heid et al.

    Real time quantitative PCR

    Genome Res.

    (1996)
  • M. Kaser et al.

    Optimized method for preparation of DNA from pathogenic and environmental mycobacteria

    Appl. Environ. Microbiol.

    (2009)
  • R.A. Kekkonen et al.

    Probiotic intervention has strain-specific anti-inflammatory effects in healthy adults

    World J. Gastroenterol.

    (2008)
  • Z.A. Khachatryan et al.

    Predominant role of host genetics in controlling the composition of gut microbiota

    PLoS ONE.

    (2008)
  • Z. Liu et al.

    Short pyrosequencing reads suffice for accurate microbial community analysis

    Nucleic Acids Res.

    (2007)
  • Cited by (426)

    View all citing articles on Scopus
    View full text