Evaluation of methods to purify virus-like particles for metagenomic sequencing of intestinal viromes

Manuel Kleiner, Lora V. Hooper, Breck A. Duerkop

Research output: Contribution to journalArticle

52 Citations (Scopus)

Abstract

Background: Viruses are a significant component of the intestinal microbiota in mammals. In recent years, advances in sequencing technologies and data analysis techniques have enabled detailed metagenomic studies investigating intestinal viromes (collections of bacteriophage and eukaryotic viral nucleic acids) and their potential contributions to the ecology of the microbiota. An important component of virome studies is the isolation and purification of virus-like particles (VLPs) from intestinal contents or feces. Several methods have been applied to isolate VLPs from intestinal samples, yet to our knowledge, the efficiency and reproducibility between methods have not been explored. A rigorous evaluation of methods for VLP purification is critical as many studies begin to move from descriptive analyses of virus diversity to studies striving to quantitatively compare viral abundances across many samples. Therefore, reproducible VLP purification methods which allow for high sample throughput are needed. Here we compared and evaluated four methods for VLP purification using artificial intestinal microbiota samples of known bacterial and viral composition. Results: We compared the following four methods of VLP purification from fecal samples: (i) filtration + DNase, (ii) dithiothreitol treatment + filtration + DNase, (iii) filtration + DNase + PEG precipitation and (iv) filtration + DNase + CsCl density gradient centrifugation. Three of the four tested methods worked well for VLP purification. We observed several differences between methods related to the removal efficiency of bacterial and host DNAs and biases against specific phages. In particular the CsCl density gradient centrifugation method, which is frequently used for VLP purification, was most efficient in removing host derived DNA, but also showed strong discrimination against specific phages and showed a lower reproducibility of quantitative results. Conclusions: Based on our data we recommend the use of methods (i) or (ii) for large scale studies when quantitative comparison of viral abundances across samples is required. The CsCl density gradient centrifugation method, while being excellently suited to achieve highly purified samples, in our opinion, should be used with caution when performing quantitative studies.

Original languageEnglish (US)
Article number7
JournalBMC Genomics
Volume16
Issue number1
DOIs
StatePublished - Jan 22 2015

Fingerprint

Metagenomics
Virion
Deoxyribonucleases
Density Gradient Centrifugation
Bacteriophages
Viruses
Bacterial DNA
Gastrointestinal Contents
Dithiothreitol
Microbiota
Ecology
Reproducibility of Results
Feces
Nucleic Acids
Mammals

Keywords

  • Bacteriophage
  • CsCl density gradient
  • Microbiome
  • Viral metagenomes
  • Virus metagenomics
  • Virus-like particles

ASJC Scopus subject areas

  • Biotechnology
  • Genetics

Cite this

Evaluation of methods to purify virus-like particles for metagenomic sequencing of intestinal viromes. / Kleiner, Manuel; Hooper, Lora V.; Duerkop, Breck A.

In: BMC Genomics, Vol. 16, No. 1, 7, 22.01.2015.

Research output: Contribution to journalArticle

@article{1dd51fd1957a4f87a2068ac90cdd1f33,
title = "Evaluation of methods to purify virus-like particles for metagenomic sequencing of intestinal viromes",
abstract = "Background: Viruses are a significant component of the intestinal microbiota in mammals. In recent years, advances in sequencing technologies and data analysis techniques have enabled detailed metagenomic studies investigating intestinal viromes (collections of bacteriophage and eukaryotic viral nucleic acids) and their potential contributions to the ecology of the microbiota. An important component of virome studies is the isolation and purification of virus-like particles (VLPs) from intestinal contents or feces. Several methods have been applied to isolate VLPs from intestinal samples, yet to our knowledge, the efficiency and reproducibility between methods have not been explored. A rigorous evaluation of methods for VLP purification is critical as many studies begin to move from descriptive analyses of virus diversity to studies striving to quantitatively compare viral abundances across many samples. Therefore, reproducible VLP purification methods which allow for high sample throughput are needed. Here we compared and evaluated four methods for VLP purification using artificial intestinal microbiota samples of known bacterial and viral composition. Results: We compared the following four methods of VLP purification from fecal samples: (i) filtration + DNase, (ii) dithiothreitol treatment + filtration + DNase, (iii) filtration + DNase + PEG precipitation and (iv) filtration + DNase + CsCl density gradient centrifugation. Three of the four tested methods worked well for VLP purification. We observed several differences between methods related to the removal efficiency of bacterial and host DNAs and biases against specific phages. In particular the CsCl density gradient centrifugation method, which is frequently used for VLP purification, was most efficient in removing host derived DNA, but also showed strong discrimination against specific phages and showed a lower reproducibility of quantitative results. Conclusions: Based on our data we recommend the use of methods (i) or (ii) for large scale studies when quantitative comparison of viral abundances across samples is required. The CsCl density gradient centrifugation method, while being excellently suited to achieve highly purified samples, in our opinion, should be used with caution when performing quantitative studies.",
keywords = "Bacteriophage, CsCl density gradient, Microbiome, Viral metagenomes, Virus metagenomics, Virus-like particles",
author = "Manuel Kleiner and Hooper, {Lora V.} and Duerkop, {Breck A.}",
year = "2015",
month = "1",
day = "22",
doi = "10.1186/s12864-014-1207-4",
language = "English (US)",
volume = "16",
journal = "BMC Genomics",
issn = "1471-2164",
publisher = "BioMed Central",
number = "1",

}

TY - JOUR

T1 - Evaluation of methods to purify virus-like particles for metagenomic sequencing of intestinal viromes

AU - Kleiner, Manuel

AU - Hooper, Lora V.

AU - Duerkop, Breck A.

PY - 2015/1/22

Y1 - 2015/1/22

N2 - Background: Viruses are a significant component of the intestinal microbiota in mammals. In recent years, advances in sequencing technologies and data analysis techniques have enabled detailed metagenomic studies investigating intestinal viromes (collections of bacteriophage and eukaryotic viral nucleic acids) and their potential contributions to the ecology of the microbiota. An important component of virome studies is the isolation and purification of virus-like particles (VLPs) from intestinal contents or feces. Several methods have been applied to isolate VLPs from intestinal samples, yet to our knowledge, the efficiency and reproducibility between methods have not been explored. A rigorous evaluation of methods for VLP purification is critical as many studies begin to move from descriptive analyses of virus diversity to studies striving to quantitatively compare viral abundances across many samples. Therefore, reproducible VLP purification methods which allow for high sample throughput are needed. Here we compared and evaluated four methods for VLP purification using artificial intestinal microbiota samples of known bacterial and viral composition. Results: We compared the following four methods of VLP purification from fecal samples: (i) filtration + DNase, (ii) dithiothreitol treatment + filtration + DNase, (iii) filtration + DNase + PEG precipitation and (iv) filtration + DNase + CsCl density gradient centrifugation. Three of the four tested methods worked well for VLP purification. We observed several differences between methods related to the removal efficiency of bacterial and host DNAs and biases against specific phages. In particular the CsCl density gradient centrifugation method, which is frequently used for VLP purification, was most efficient in removing host derived DNA, but also showed strong discrimination against specific phages and showed a lower reproducibility of quantitative results. Conclusions: Based on our data we recommend the use of methods (i) or (ii) for large scale studies when quantitative comparison of viral abundances across samples is required. The CsCl density gradient centrifugation method, while being excellently suited to achieve highly purified samples, in our opinion, should be used with caution when performing quantitative studies.

AB - Background: Viruses are a significant component of the intestinal microbiota in mammals. In recent years, advances in sequencing technologies and data analysis techniques have enabled detailed metagenomic studies investigating intestinal viromes (collections of bacteriophage and eukaryotic viral nucleic acids) and their potential contributions to the ecology of the microbiota. An important component of virome studies is the isolation and purification of virus-like particles (VLPs) from intestinal contents or feces. Several methods have been applied to isolate VLPs from intestinal samples, yet to our knowledge, the efficiency and reproducibility between methods have not been explored. A rigorous evaluation of methods for VLP purification is critical as many studies begin to move from descriptive analyses of virus diversity to studies striving to quantitatively compare viral abundances across many samples. Therefore, reproducible VLP purification methods which allow for high sample throughput are needed. Here we compared and evaluated four methods for VLP purification using artificial intestinal microbiota samples of known bacterial and viral composition. Results: We compared the following four methods of VLP purification from fecal samples: (i) filtration + DNase, (ii) dithiothreitol treatment + filtration + DNase, (iii) filtration + DNase + PEG precipitation and (iv) filtration + DNase + CsCl density gradient centrifugation. Three of the four tested methods worked well for VLP purification. We observed several differences between methods related to the removal efficiency of bacterial and host DNAs and biases against specific phages. In particular the CsCl density gradient centrifugation method, which is frequently used for VLP purification, was most efficient in removing host derived DNA, but also showed strong discrimination against specific phages and showed a lower reproducibility of quantitative results. Conclusions: Based on our data we recommend the use of methods (i) or (ii) for large scale studies when quantitative comparison of viral abundances across samples is required. The CsCl density gradient centrifugation method, while being excellently suited to achieve highly purified samples, in our opinion, should be used with caution when performing quantitative studies.

KW - Bacteriophage

KW - CsCl density gradient

KW - Microbiome

KW - Viral metagenomes

KW - Virus metagenomics

KW - Virus-like particles

UR - http://www.scopus.com/inward/record.url?scp=84924029843&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84924029843&partnerID=8YFLogxK

U2 - 10.1186/s12864-014-1207-4

DO - 10.1186/s12864-014-1207-4

M3 - Article

VL - 16

JO - BMC Genomics

JF - BMC Genomics

SN - 1471-2164

IS - 1

M1 - 7

ER -