Elsevier

Methods

Volume 52, Issue 3, November 2010, Pages 242-247
Methods

Review Article
Quantitative detection of DNA methylation states in minute amounts of DNA from body fluids

https://doi.org/10.1016/j.ymeth.2010.03.008Get rights and content

Abstract

Quantitative and reliable estimation of DNA methylation levels for multiple genomic regions pose a major challenge where starting DNA is available in very low quantity. Here we review major advances in the development of techniques for quantitative detection of DNA methylation in minute amount of DNA and describe a detailed protocol for quantitative Methylation Analysis of Minute DNA amounts after whole Bisulfitome Amplification (qMAMBA), a combination of techniques that allows quantitative and sensitive detection of DNA methylation at multiple CpG sites and for multiple gene assays. Recently we successfully used this technique to quantitatively detect DNA methylation for a set of cancer-related genes in lung cancer patient plasma samples [18]. This method involves genome-wide amplification of bisulfite-modified DNA template followed by quantitative methylation detection using pyrosequencing. This allows a precise assessment of DNA methylation at CpG sites and could be adapted for high-throughput settings. It can also be applied in conjunction with studies involving single-cells or laser capture microdissected samples. Thus, this method should facilitate DNA methylation studies aiming to discover epigenetic biomarkers, and should prove particularly valuable in profiling a large sample series of body fluids from molecular epidemiology studies.

Introduction

DNA methylation, the covalent addition of a methyl group to the cytosine base in DNA, is a central epigenetic modification and has essential roles in cellular processes including genome regulation, development and disease. DNA hypermethylation and hypomethylation are two distinct forms of the aberrant DNA methylation, both of which have been associated with a large number of human malignancies, other non-neoplastic diseases and aging. Interestingly, aberrant DNA methylation changes are disease specific and their extents may reflect the stage of the disease progression state. Thus, aberrant DNA methylation patterns can serve as a highly specific biomarker [1], [2].

The cell-free DNA present in different body fluids is an attractive target for biomarker discovery. In addition to genetic changes (mutations), cell-free circulating DNA from individuals with tumors has been shown to harbor alterations in DNA methylation at CpG sites in the promoter regions of a wide range of tumor suppressor genes and other cancer-associated genes [3], [4], [5], [6]. Notably, due to sampling ease the cell-free DNA present in different body fluids holds potential to serve as a promising noninvasive biomarker with diagnostic and prognostic value.

For successful and reliable application of DNA methylation-based biomarkers in diagnostic routine, the following criteria need to be satisfied:

  • (a)

    Prevention of false positives resulting from incomplete conversion of unmethylated cytosines. DNA methylation analysis is relying on bisulfite DNA modification as an initial step towards detection of the presence of methylated cytosines. It involves treatment of DNA with sodium bisulfite resulting in chemical conversion of the unmethylated cytosines into uracil while methylated cytosines remain unchanged. Incomplete or inefficient modification reaction results in unconverted cytosines, which will indicate the presence of methyl-cytosine, resulting in a false positive detection of DNA methylation. Hence, it is essential that the technique used has an in-built measure to check for completeness of the bisulfite conversion reaction.

  • (b)

    Confirmation of the PCR amplified product used for methylation determination. To detect the methylation status of cancer-associated gene(s), the bisulfite-modified DNA is used as a template to amplify the region of interest by PCR. PCR amplifications heavily depend on the robustness of the assay conditions and primer-specificity, thus remaining prone to non-specific amplifications. Validation of PCR amplified products thus becomes necessary before making any final conclusions.

  • (c)

    Quantitative detection of methylation at multiple CpG sites for a given gene. Regulation of gene expression by promoter DNA methylation depends on the frequency of methylation in the CpG sites surrounding the transcription start sites [2], [7]. Different CpG sites within a genomic region may be methylated at different levels, thus it is desirable to obtain a quantitative estimate of the extent of methylation and its distribution over several neighboring CpG sites in the genomic region of interest. In this case gene expression levels may be evaluated in relation to the extent of DNA methylation.

  • (d)

    Possibility to carry out several assays to confirm the cancer signature in individual samples. Successful utilization of DNA methylation-based biomarkers for tumor diagnosis often involves investigation of methylation status for a set of marker genes to confirm the cancer specific signature. In such a situation both development and routine application of DNA methylation-based biomarkers, the use of minute amounts of patient DNA samples will require an approach allowing multiple assays for individual patient samples.

Many efforts have been made to develop a reliable technique that would allow sensitive detection of DNA methylation in minute amounts of nucleic acids (Table 1). Methylation-Specific PCR (MSP), quantitative Methylation-Specific PCR (qMSP) and Methylationsensitive Single-Nucleotide Primer Extension are relatively inexpensive and highly sensitive techniques that have been widely used to characterize DNA methylation changes in a wide variety of biological samples, including tumor tissues and different body fluids [8], [9], [10], [11], [12], [13]. However, the major disadvantages of these techniques are the lack of an in-built measure for checking the completeness of bisulfite conversion (conversion control) and limited coverage of CpG sites (typically a single CpG site is analyzed). Moreover, being solely based on PCR amplification these techniques are prone to mispriming. These problems are further aggravated in situations of low concentrations of starting DNA templates (typical for DNA samples isolated from body fluids) and when high numbers of PCR cycles are used. Although problems with mispriming can be remedied by subsequent confirmation steps such as re-analysis using bisulfite sequencing [14] or methylation sensitive restriction enzymes [15], these additional validation steps represent significant logistical disadvantages by reducing both time-and cost-effectiveness. To a certain extent these limitations could be overcome by more advanced and modified version of the same techniques, wherein a sequence specific fluorescent probe is added to validate the PCR amplification (e.g., MethyLight [12]), and an additional fluorescent probe directed against unconverted DNA added in each qPCR reaction to detect presence of any unconverted DNA (e.g., conversion-specific detection of DNA methylation using real-time polymerase chain reaction; ConLight-MSP) [16]. Most of these techniques still remain limited by the small amounts of starting DNA materials and can provide methylation information about only one or two CpG sites (Table 1).

To overcome the problems associated with false positives, Liloglou and colleagues developed the Methylation Enrichment Pyrosequencing (MEP) method [17], which benefits from high sensitivity of MSP and specificity of the Pyrosequencing Methylation Assay (PMA) that serves as a confirmatory step. While being a pyrosequencing-based technique, MEP has important advantages over the above described methylation detection techniques. Its intrinsic weakness lies in the fact that it cannot be used to monitor quantitative levels of the methylated alleles in samples with minute amounts of DNA.

First, because the primers of the first step (MSP) are designed to specifically amplify methylated DNA, and exclude unmethylated DNA, this technique cannot provide a quantitative measure of different amounts of methylated and unmethylated DNA in samples. Since the fraction of circulating DNA containing DNA methylation changes varies between samples and typically represents only a tiny fraction of total circulating DNA, an ideal method should be able to provide a quantitative measure of methylated alleles in total circulating DNA. Second, application of this method remains limited by the quantity of the available starting DNA material, and hence such a method cannot allow methylation analysis of a large number of genes in samples with low amounts of DNA.

Section snippets

Quantitative Methylation Analysis of Minute DNA amounts after whole Bisulfitome Amplification, qMAMBA

We sought to develop a method that would satisfy the key criteria for their applicability in quantitative analysis of DNA methylation in samples with low concentration of DNA. To this end, we have recently described the qMAMBA (quantitative Methylation Analysis of Minute DNA amounts after whole Bisulfitome Amplification) method, which takes advantage of a combination of techniques allowing for quantitative and sensitive detection of DNA methylation in minute amounts of starting DNA [18].

This

Concluding remarks

In summary, qMAMBA should facilitate studies aiming to discover and validate new DNA methylation markers and studies that are limited by the tiny amounts of starting DNA material. Additionally it should prove particularly valuable in profiling large sample series of body fluids from molecular epidemiology studies and clinical studies for early diagnostics and prognostic purposes as well as in monitoring the efficacy of epigenetics-based cancer therapies and preventive strategies.

Acknowledgements

T. Vaissière is supported by a Ph.D. fellowship from la Ligue National (Française) Contre le Cancer. The work in Epigenetics Group at IARC is supported by grants from the National Institutes of Health/National Cancer Institute (NIH/NCI), United States; L’Association pour la Recherche sur le Cancer (ARC), France, la Ligue Nationale (Française) Contre le Cancer, France, and Agence Nationale de Recherhe Contre le Sida et Hépatites Virales (ANRS, France), and the Swiss Bridge Award (to Z.H.).

References (30)

  • C. Goessl et al.

    Eur. Urol.

    (2002)
  • Z. Herceg et al.

    Mol. Oncol.

    (2007)
  • M.L. Gonzalgo et al.

    Methods

    (2002)
  • D.S. Shames et al.

    Cancer Lett.

    (2007)
  • K. Rand et al.

    Methods

    (2002)
  • Z. Liu et al.

    Anal. Biochem.

    (2009)
  • T. Vaissiere et al.

    Cancer Res.

    (2009)
  • Z. Herceg

    Mutagenesis

    (2007)
  • J.P. Issa

    Clin. Cancer Res.

    (2007)
  • P.W. Laird

    Nat. Rev. Cancer

    (2003)
  • J.P. Issa

    Nat. Rev. Cancer

    (2004)
  • S.A. Belinsky et al.

    Cancer Res.

    (2002)
  • M. Sanchez-Cespedes et al.

    Cancer Res.

    (2000)
  • D.J. Rossi et al.

    Nature

    (2007)
  • C.A. Eads et al.

    Nucleic Acids Res.

    (2000)
  • Cited by (0)

    View full text