Methylation Analysis by DNA Immunoprecipitation (MeDIP)

  • Emily A. Vucic
  • Ian M. Wilson
  • Jennifer M. Campbell
  • Wan L. Lam
Protocol
Part of the Methods in Molecular Biology™ book series (MIMB, volume 556)

Abstract

Alteration in epigenetic regulation of gene expression is a common event in human cancer and developmental disease. CpG island hypermethylation and consequent gene silencing is observed for many genes involved in a diverse range of functions and pathways that become deregulated in the disease state. Comparative profiling of the methylome is therefore useful in disease gene discovery. The ability to identify epigenetic alterations on a global scale is imperative to understanding the patterns of gene silencing that parallel disease progression. Methylated DNA immunoprecipitation (MeDIP) is a technique that isolates methylated DNA fragments by immunoprecipitating with 5′-methylcytosine-specific antibodies. The enriched methylated DNA can then be analyzed in a locus-specific manner using PCR assay or in a genome-wide fashion by comparative genomic hybridization against a sample without MeDIP enrichment. This article describes the detailed protocol for MeDIP and hybridization of MeDIP DNA to a whole-genome tiling path BAC array.

Key words

Epigenetics DNA methylation hypermethylation hypomethylation CpG islands methylated DNA immunoprecipitation epigenetic methods and technologies array-based methylation analysis MeDIP aCGH 

10.1 Introduction

10.1.1 Brief Overview of Methylation

Epigenomics refers to the genome-wide study of heritable yet reversible changes, which do not alter the DNA sequence itself. Aberrant epigenetic changes such as global DNA hypomethylation and focal DNA hypermethylation are found in almost all tumor types and for many developmental diseases (1, 2, 3, 4). Global hypomethylation characteristically increases with age, and is linked to chromosomal instability in cancer and other diseases (5,6). Specific patterns of hypermethylated DNA are associated with the transcriptional silencing of genes correlating with cancer progression, prognosis, and treatment response, therefore representing promising biological markers (6, 7, 8). Drugs that inhibit methylation are used both as a research tool, to assess reactivation of genes silenced in cancer by hypermethylation, and in the treatment of some hematological malignancies (9,10). The identification of epigenetic targets may serve as the basis for new therapeutics aimed at underlying disease biology.

The study of DNA methylation, especially in the integration of epigenetic data with genomic and expression profiles, will improve our ability to identify causal DNA events and their impact on gene expression and disease behavior. Towards this aim, analysis of isolated methylated DNA fragments by a technique described here called methylated DNA immunoprecipitation (MeDIP), previously described by Weber et al., can be accomplished by a number of whole-genome profiling technologies facilitating the survey of differential methylation on a whole-genome scale and locus-specific resolution (11,12).

10.1.2 Brief Outline of Technique

The availability of antibodies specific to 5′-methylcytosine has enabled the enrichment of methylated fragments of DNA by immunoprecipitation (IP). DNA is sonicated and divided into two fractions; one of which will be used for the IP reaction and the other to be used as whole-genome reference material during comparative analysis, called the input (IN) (11). The DNA sample is incubated with anti-5′-methylcytosine antibodies, and immunoprecipitated products are purified and then analyzed for enrichment of methylated DNA relative to the reference DNA. The MeDIP technique is schematically outlined in Fig.10.1. IP DNA can be compared to IN DNA using a number of array CGH (aCGH) platforms including CpG island arrays, whole-genome BAC or oligonucleotide (oligo) arrays (12).
Fig. 10.1.

Methylated DNA immunoprecipitation (MeDIP). A total of 1 μg of DNA is sonicated and divided into two tubes: one which will serve as a whole-genome reference for comparative analysis (IN DNA), the remainder of which will be immunoprecipitated (IP) (1). DNA for IP reaction is denatured (2), cooled, and incubated with primary antibody specific to methylated cytosine (3). A secondary antibody affixed with magnetic beads and specific to the primary antibody is added (4) and methylated DNA is thereby captured by use of a magnetic rack. IP reaction is then purified to remove proteins and buffer (5). IP DNA can then be compared to IN DNA for enrichment of methylation by PCR or array-based analysis.

10.1.3 Validation

The efficiency of IP may be tested using standard real time quantitative PCR (qPCR) techniques. Reliable assessment of MeDIP efficiency can be performed with only a few nanograms of IP DNA per PCR assay. Alternative methods of validation are bisulfite sequencing and methylation-specific PCR (13, 14, 15). To test the enrichment at a given locus by qPCR it is necessary to construct two sets of primers. The control set should amplify a segment of DNA that is devoid of CpG dinucleotides. The other primer set should amplify the locus one wishes to test (e.g., an imprinted region such as H19). Standard rules for qPCR primer design apply. When optimizing the MeDIP protocol, it is possible to use repetitive DNA sequences, imprinted regions, or developmentally silenced genes as test targets to monitor the enrichment efficiency – so long as their methylation status is known (11). PCR assays should be performed for both primer sets on both the IN and IP fraction. Using threshold values for control and test primers, it is possible to calculate the fold enrichment using the ΔΔCt method described by Livak et al. (16).

10.1.4 Analysis of MeDIP DNA by Array CGH

10.1.4.1 BAC Array CGH

Analysis of MeDIP DNA by whole-genome BAC aCGH is a way to analyze DNA enriched for methylation in the sample on a whole-genome scale. Whole-genome methylation profiles can be obtained by differentially labeling IN and IP DNA from a single MeDIP reaction (or a pool of multiple MeDIP reactions for the same sample) with fluorescent nucleotide dyes, and then hybridizing to a whole-genome tiling path array such as the sub-megabase resolution tiling (SMRT) array CGH platform, which consists of ∼27,000 overlapping BAC clones spanning the human genome (11). These steps are outlined in Fig.10.2. Dye ratios correlate to relative quantities of methylation-enriched IP DNA versus reference DNA (IN) bound to each BAC on the microarray.
Fig. 10.2.

Analysis of MeDIP DNA by BAC aCGH. IN and IP DNA are differentially labeled with fluorescent Cy5 and Cy3 dyes, respectively (1). Unincorporated Cy dyes, buffers, and nucleotides are removed by use of exclusion columns (2). IN and IP reactions are combined, Cot-1 is added to bind repetitive DNA sequences and Cy dye incorporations determined by spectroscopy (3). Combined reactions are denatured. Cot-1 is allowed to anneal to labeled repetitive DNA and each sample is then applied to an array for hybridization (4).

10.1.4.2 Application of Other Array-Based Platforms to MeDIP Analysis

Similarly, other aCGH platforms including CpG island and genome-wide arrays can be used in conjunction with BAC aCGH or on their own. Examples include Agilent’s Human Genome CGH and CpG island array (Agilent Technologies), and NimbleGen’s Human Whole-Genome or custom-designed oligo arrays (NimbleGen Systems, Inc) (seeNote 1).

Application of array-based platforms depends primarily on which platform will produce the most meaningful and reliable data in accordance with the experimenter’s primary objective taking into account sample limitations. As previously described, one factor strongly influencing array alteration detection capacity is element size and distribution (17). For methylation analysis, consideration of CpG island coverage and location (promoter or coding region for example) is especially relevant. The use of CpG island arrays in conjunction with whole-genome arrays may be most suitable for discovery-based research, whereas analysis of methylation status of CpG islands located in the promoters of target genes may be accomplished by lower-density CpG island or custom-designed oligo arrays. Sample IN requirement is another important consideration. Oligo platforms typically require 2 μg (Agilent) to 4 μg (NimbleGen) of product per dye channel (4–8 μg total); therefore, amplification of IP and IN DNA may be necessary (seeNote 2). Since amplification may introduce bias, an alternative approach is to pool many MeDIP reactions to obtain the required amount.

10.2 Materials

10.2.1 DNA Preparation

  1. 1.

    Siliconized centrifuge tubes (1.7 ml SafeSeal Microcentrifuge Tubes, Sorenson BioScience)

     
  2. 2.

    Sterilized dH2O

     

10.2.2 DNA Sonication

  1. 1.

    Sonicating device: Biorupter (Diagenode, UCD-200 TM)

     

10.2.3 Immunoprecipitation

  1. 1.

    IP Buffer: 10 mM sodium phosphate pH 7.0, 140 mM NaCl, 0.05% Triton X-100. Stored at room temperature.

     
  2. 2.

    Primary antibody: Anti-5′-methylcytosine Mouse mAb (162 33 D3) (CalBiochem). Stored at −20°C. Avoid freeze–thaw cycles.

     
  3. 3.

    Secondary antibody: Dynabeads M-280 Sheep anti-Mouse IgG (Dynal Biotech, Invitrogen). Stored at 4°C.

     
  4. 4.

    Rotating tube rack

     
  5. 5.

    Magnetic tube rack: magnetic stand (Invitrogen)

     

10.2.4 IP DNA Purification

  1. 1.

    100% ethanol

     
  2. 2.

    70% ethanol stored at −20°C

     
  3. 3.

    3 M sodium acetate (pH 5.2)

     
  4. 4.

    1:1 phenol:chloroform (buffered to pH 7)

     
  5. 5.

    Sterilized dH2O

     

10.2.5 Validation by qPCR

  1. 1.

    MeDIP DNA (IP and IN)

     
  2. 2.

    iQ SYBR Green Supermix (Biorad laboratories)

     
  3. 3.
    Primers: Example of sequences for enrichment testing of immunoprecipitated human DNA:
    1. 3.1

      H19_F 5′-GGCGTAATGGAATGCTTGA

       
    2. 3.2

      H19_R 5′-CCTCGCCTAGTCTGGAAGC

       
    3. 3.3

      Producing a 63 bp product.

       
    4. 3.4

      CTRL_F 5′-GGTTCAGTTTATTGTCCTAAAATCAG

       
    5. 3.5

      CTRL_R 5′-TCAGCCAGACCAAAGCAAAT

       
    6. 3.6

      Producing a 92 bp product.

       
     
  4. 4.

    Real-time qPCR machine

     
  5. 5.

    Real-time PCR strip caps or real-time PCR plates and sealing tape

     

10.2.6 Analysis of MeDIP DNA by BAC aCGH

10.2.6.1 Labeling IN and IP DNA

  1. 1.

    Random primers buffer (PB) (5X Promega Klenow buffer and 7 μg/μl random octamers)

     
  2. 2.

    10X dNTP mix (2 mM each dATP, dGTP, dTTP, 1.2 mM dCTP) (Promega)

     
  3. 3.

    Cyanine 5-dCTP (1 nmol/μl) (Amersham, GE Healthcare)

     
  4. 4.

    Cyanine 3-dCTP (1 nmol/μl) (Amersham, GE Healthcare)

     
  5. 5.

    Klenow fragment (9 U/μl) (Promega)

     

10.2.6.2 Sample Clean-Up

  1. 1.

    Microcon YM-30 column (Millipore)

     
  2. 2.

    Cot-1 DNA (1 μg/μl) (Invitrogen)

     
  3. 3.

    Hybridization buffer: DIG Easy Hyb (Roche Applied Science)

     
  4. 4.

    Sheared herring sperm DNA (20 mg/ml): Optional depending on slide chemistry (seeNote 3)

     

10.2.6.3 Cy Dye Incorporation Calculation

  1. 1.

    ND-1000 Spectrophotometer V3.1.0 (NanoDrop Technologies Inc.)

     

10.2.6.4 Hybridization to BAC Array

  1. 1.

    Coverslips (22 × 60 mm, Fisher Scientific)

     
  2. 2.

    Hybridization cassette (Telechem)

     

10.2.6.5 Array Washing

  1. 1.

    Wash buffer: 0.1X SSC/0.1X SDS (pre warmed to 45°C)

     
  2. 2.

    Rinse buffer: 0.1X SSC (room temperature)

     
  3. 3.

    Coplin jar

     

10.2.6.6 Scanning

GenePix Professional 4200A (Molecular Devices)

10.3 Methods

10.3.1 Preparation of Samples

In one siliconized tube per sample, prepare 1 µg of DNA in 50 μl of sterilized dH2O.

10.3.2 DNA Sonication

DNA sonication and MeDIP protocol must be performed in siliconzied tubes to prevent non-specific binding of proteins to tube walls. Here we describe a method to obtain 300–1,000 bp DNA fragments by sonication with an automated Biorupter (Diagenode, UCD-200 TM) (seeNote 4).
  1. 1.

    Water in Biorupter must be at 4°C with layer of crushed ice on top to water mark.

     
  2. 2.

    Run for 5–7 min on automatic settings (30 s on 30 s off at maximum power).

     
  3. 3.

    Verify fragment size on 1% agarose gel with 100 bp ladder.

     
  4. 4.

    Remove 800 ng (40 μl) of sonicated product and place in siliconized 1.7 ml centrifuge tube for the IP reaction.

     
  5. 5.

    Set aside remainder (200 ng) to serve as IN reference DNA (store at 4°C).

     

10.3.3 Methylated DNA Immunoprecipitation

  1. 1.

    Denature the DNA that will be used for IP reaction (800 ng) at 95°C for 10 min in heat block.

     
  2. 2.

    Cool immediately on ice. Let DNA cool completely (∼5 min on ice) before proceeding with next step.

     
  3. 3.

    Add 5 μg primary (anti-5′-methylcytosine) antibody.

     
  4. 4.

    Add IP buffer to a final volume of 500 μl.

     
  5. 5.

    Incubate a minimum of 2 h at 4°C in rotating tube holder (seeNote 5).

     
  6. 6.

    Just before Step 5 is complete, prepare Dynabeads (coupled with secondary antibody) by washing. First resuspend the beads thoroughly in the vial by vortexing.

     
  7. 7.

    Transfer 30 μl (∼ 2 × 107) of resuspended beads into a new siliconized tube. If performing more than one reaction, remove 30 μl of beads per reaction plus 1 (e.g., if doing eight reactions, remove enough beads for nine, i.e., 270 μl).

     
  8. 8.

    Place the tube on the magnetic rack for 2 min at room temperature.

     
  9. 9.

    Pipette off the supernatant. When pipetting off supernatant avoid touching the beads against inside wall (where the beads attract to the magnet) with the pipette tip.

     
  10. 10.

    Remove the tube from the magnet, and resuspend the beads in an excess volume of IP buffer (750–1,000 μl).

     
  11. 11.

    Repeat the wash once more, and then resuspend the washed beads in IP buffer in the original volume removed in Step 7.

     
  12. 12.

    Add 30 μl of washed Dynabeads to each IP reaction and incubate in a rotating tube holder for 2 h at 4°C.

     
  13. 13.

    After incubation with both antibodies is complete, place the tube on the magnetic rack for 2 min at room temperature.

     
  14. 14.

    Pipette off the supernatant. Avoid touching the inside wall of the tube (where the beads attract to the magnet) with the pipette tip.

     
  15. 15.

    Wash the bound Dynabeads three times in IP buffer, resuspending for the final time in 500 μl IP buffer.

     
  16. 16.

    Treat the reaction with 100 μg of proteinase K for 3 h at 50°C. Spike with 50 μg more proteinase K and continue to digest overnight at 50°C (seeNote 6).

     

10.3.4 IP DNA Purification

  1. 1.

    Add 500 μl 1:1 phenol/chloroform pH 7 and vortex thoroughly.

     
  2. 2.

    Spin at 13,000 g for 10 min at room temperature.

     
  3. 3.

    Remove aqueous (top) fraction to a new tube (seeNote 7).

     
  4. 4.

    Repeat Steps 1–3 one time to ensure complete removal of protein matter.

     
  5. 5.

    Add one-tenth volume 3 M sodium acetate (50 μl) and vortex.

     
  6. 6.

    Add 2 volumes (1,000 μl) of 100% ethanol; place in −20°C freezer for 20 min.

     
  7. 7.

    Spin at max speed at 4°C for 20 min.

     
  8. 8.

    Remove ethanol, pulse spin, and remove residual ethanol.

     
  9. 9.

    Add 500 μl cold 70% ethanol to wash. Vortex briefly, and spin at max speed for 20 min at 4°C.

     
  10. 10.

    Remove 70% ethanol, pulse spin, and remove residual ethanol by pipetting.

     
  11. 11.

    Dry pellets in heat block at 50°C for 10 min with caps open to remove all traces of residual ethanol.

     
  12. 12.

    Resuspend DNA pellet in sterile dH2O (seeNote 8).

     
  13. 13.

    Place tubes at 65°C for 10 min followed by 1 h incubation at 37°C to completely resuspend the DNA before use. Alternatively, after 37°C incubation store at −20°C for future use.

     

10.3.5 Validation by qPCR

  1. 1.

    Using iQ SYBR Green Supermix set up 25 µl PCR reactions with a final primer concentration of 0.6 µM for the forward and 0.6 µM for the reverse (seeNote 9).

     
  2. 2.
    Real-time PCR should be performed using the following cycling parameters, with data collection occurring during the 60°C annealing/extension step:
    1. 2.1

      95°C for 5 min, (95°C for 15 s, 60°C for 45 s) × 40 cycles.

       
     

10.3.6 Analysis of MeDIP DNA by BAC aCGH

It is important to limit exposure of Cyanine dyes to light at all times.

10.3.6.1 Labeling IN and IP DNA

For every sample to be hybridized there will be two labeling reactions: one reaction tube for the reference (IN DNA) and one reaction tube for the IP DNA.
  1. 1.
    In sterilized PCR tubes, for every separate labeling reaction combine:
    1. a.

      DNA (50–400 ng)

       
    2. b.

      5 μl of 5X PB buffer

       
    3. c.

      Dilute to 17 μl total volume with sterile dH2O

       
     
  2. 2.

    Boil for 10 min at 100°C in a PCR machine.

     
  3. 3.

    Transfer immediately to ice for 2 min.

     
  4. 4.

    Vortex 10X dNTP thoroughly and add 4 μl to each labeling reaction.

     
  5. 5.

    Vortex and spin down Cyanine dyes. Add 2 nmol (2 μl of 1 nmol/μl stock) of Cyanine 5-dCTP (Cy5) to IN DNA and 2 nmol (2 μl of 1 nmol/μl stock) of Cyanine 3-dCTP (Cy3) to IP DNA.

     
  6. 6.

    Add 22.5 U Klenow (2.5 μl of 9 U/μl stock) to all reactions and mix gently by pipetting.

     
  7. 7.

    Incubate in the dark at 37°C overnight (∼18 h).

     

10.3.6.2 Sample Clean-Up

It is important to limit the exposure of labeling reactions to light at all times. All steps in sample clean-up are performed at room temperature.

Using a Microcon YM-30 column (seeNote 10):
  1. 1.

    Combine the two reactions (IN and IP) for each sample and add 100 μg Cot-1 DNA (1 μg/μl) to column (do not touch membrane with tip).

     
  2. 2.

    Spin at 13,000 g for 10 min in provided tubes.

     
  3. 3.

    Discard elution and add 200 μl sterile dH2O to membrane and repeat spin in Step 2 to wash.

     
  4. 4.

    Discard collection tube and add 45 μl DIG Easy to column (seeNote 11).

     
  5. 5.

    Leave at room temperature for 2 min then invert Microcon in a new tube and spin at 3,000 g for 3 min.

     

10.3.6.3 Cy Dye Incorporation Calculation

Remove 1.5 μl from the labeling reaction and measure Cy dye incorporation using NanoDrop Spectrophotometer (seeNote 12). Use appropriate blank solution.

10.3.6.4 Hybridization to BAC Array

It is recommended to carry out all hybridization steps in a darkened room (seeNote 13).
  1. 1.

    Denature labeled DNA (probes) at 85°C for 10 min.

     
  2. 2.

    Pulse spin, then place in 45°C incubator or heat block and allow Cot-1 to anneal to labeled repetitive DNA for 1 h.

     
  3. 3.

    Pre-warm a hybridization cassette to ∼45°C.

     
  4. 4.

    Keeping the probe at 45°C in a heat block, pipette 45 μl of the probe solution onto the array slide (seeNote 14).

     
  5. 5.

    Gently lower coverslip onto slide over probe solution avoiding bubbles.

     
  6. 6.

    Place the slide into a pre-warmed hybridization cassette and add 10 μl of H2O in the lower groove (seeNote 15).

     
  7. 7.

    Incubate for 36–40 h at 45°C.

     

10.3.6.5 Array Washing

All wash solutions should be at pH 7.0. All steps should be carried out in a darkened room (seeNote 13).
  1. 1.

    Remove coverslip by gently submerging slide into a wash solution of 0.1X SSC/0.1X SDS until the coverslip falls off (seeNote 16).

     
  2. 2.

    Wash the slides 2–3 times in 0.1X SSC/0.1X SDS pre-warmed to 45°C for 5 min with agitation in Coplin jar.

     
  3. 4.

    Rinse SDS by rinsing the slides 2–3 times in 0.1X SSC for 5 min with agitation until no bubbles appear.

     
  4. 5.

    Dry the slides with an air stream (oil free) or by centrifuging the slides in Falcon tubes at 900g for 3 min. (Blow off dust with air gun)

     
  5. 6.

    Store the slides in the dark (seeNote 17).

     

10.3.6.6 Scanning

We currently use GenePix Professional 4200A (Molecular Devices). Slides are scanned at a 10 μm resolution, on automated optimization. The scanner requires a file with information about spot location to run automatic optimization (seeNote 18).

Before exporting to visualization program, normalization programs are commonly used to correct for systematic variation inherent in experimental or technical processes such as gradients in the array images (18) (seeNote 19).

10.4 Notes

  1. 1.

    Cost may be another consideration. Besides obvious differences in cost of different arrays and services, many arrays have only one spot per unique element and therefore two arrays may be required per sample if reproducibility calculations are desired.

     
  2. 2.

    WGA2 GenomePlex Complete Whole Genome Amplification (WGA) Kit (Sigma-Aldrich), or Bioscore Screening and Amplification Kit (Enzo Life Sciences) are most convenient.

     
  3. 3.

    Traditionally herring sperm is used to prevent non-specific background binding on slides coated with amine reactive groups. However, most current array slide chemistry (including the BAC aCGH protocol described herein) are comprised of non-reactive groups (aldehyde or epoxy) and therefore do not require the use of herring sperm.

     
  4. 4.

    DNA sonication is most easily and consistently accomplished in batch sets with an automated sonicator as described in this protocol. Sonication can also be performed using a conventional sonicating device such as Fisher Scientific Sonic Dismembrator (Fisher Scientific). Optimization of all sonicating devices must be performed for specific DNA quantity, total volume, and type and size of tube sonication will be performed in, prior to IP reaction. After initial optimization, base pair size should be periodically tested on an agarose gel over the course of normal usage.

     
  5. 5.

    Incubation with primary antibody can be left overnight.

     
  6. 6.

    It is normal for beads to clump during protein digestion. Overnight protein digestion is convenient but not necessary.

     
  7. 7.

    Tubes do not need to be siliconized from here on.

     
  8. 8.

    If hybridizing to BAC array then resuspend pellet in 13 μl dH2O.

     
  9. 9.

    It is always necessary to determine primer efficiency separately.  Each primer set will need an efficiency of near 100% for the ΔΔCt method to be valid.

     
  10. 10.

    Microcon YM-30 columns filter species smaller than 50 bp of DNA, including all unincorporated Cy dyes, nucleotides, and buffers. Larger DNA fragments and undigested protein fragments will remain.

     
  11. 11.

    If slide chemistry requires, add an additional 4.5 μl (20 mg/ml) sheared herring sperm DNA. Please seeNote 3 for explanation.

     
  12. 12.

    Incorporations below 3.0 pmol/μl in either channel can produce variable results.

     
  13. 13.

    Preferably carry out all hybridization, washing, and scanning procedures in an ozone-controlled room. Ozone has been show to degrade Cy3 and Cy5 dye signal intensities (19). Dedicating a room to hybridization procedures with an ozone monitoring and filtering device is ideal. Slides are particularly vulnerable when exposed to air, such as during washing, scanning, and storing.

     
  14. 14.

    Alternatively, place 45 μl probe solution onto coverslip and lower array slide onto coverslip.

     
  15. 15.

    This is to control humidity and prevent slide from drying out.

     
  16. 16.

    If the coverslip will not fall off, carefully remove by hand, trying not to scratch the array.

     
  17. 17.

    Humidity, light, and ozone may all contribute to signal degradation during storage.

     
  18. 18.

    Scan resolutions vary depending on array platform. For example, oligo arrays are usually scanned at 5 μm or may potentially require multiple scans per slide.

     
  19. 19.

    Because methylation data is not discrete data (i.e., many CpGs per element on array), segmentation analysis, such as applied to copy number analysis, is not as applicable. Generating lists of clones that are enriched or un-enriched for methylation can be accomplished by filtering normalized Cy3/Cy5 ratios based on thresholds. Ratios imported to software designed to view and analyze microarray data are very useful in comparing multiple samples or across assay types or platforms (20,21).

     

Notes

Acknowledgments

The authors wish to thank Bradley Coe, Chad Malloff, and Spencer Watson for useful discussion and assistance with this manuscript. This work was supported by funds from the Canadian Institutes for Health Research, Canadian Breast Cancer Research Alliance, Genome Canada/British Columbia, and National Institute of Dental and Craniofacial Research (NIDCR) grant R01 DE15965.

References

  1. 1.
    Esteller M, Herman JG. (2002) Cancer as an epigenetic disease: DNA methylation and chromatin alterations in human tumours. J Pathol. 196, 1–7.PubMedCrossRefGoogle Scholar
  2. 2.
    Feinberg AP. (2007) Phenotypic plasticity and the epigenetics of human disease. Nature 447, 433–440.PubMedCrossRefGoogle Scholar
  3. 3.
    Ozanne SE, Constancia M. (2007) Mechanisms of disease: the developmental origins of disease and the role of the epigenotype. Nat Clin Pract Endocrinol Metab. 3, 539–546.PubMedCrossRefGoogle Scholar
  4. 4.
    Shames DS, Minna JD, Gazdar AF. (2007) DNA methylation in health, disease, and cancer. Curr Mol Med. 7, 85–102.PubMedCrossRefGoogle Scholar
  5. 5.
    Jones PA, Baylin SB. (2002) The fundamental role of epigenetic events in cancer. Nat Rev Genet. 3, 415–428.PubMedCrossRefGoogle Scholar
  6. 6.
    Feinberg AP, Ohlsson R, Henikoff S. (2006) The epigenetic progenitor origin of human cancer. Nat Rev Genet. 7, 21–33.PubMedCrossRefGoogle Scholar
  7. 7.
    Glasspool RM, Teodoridis JM, Brown R. (2006) Epigenetics as a mechanism driving polygenic clinical drug resistance. Br J Cancer. 94, 1087–1092.PubMedCrossRefGoogle Scholar
  8. 8.
    Chekhun VF, Lukyanova NY, Kovalchuk O, Tryndyak VP, Pogribny IP. (2007) Epigenetic profiling of multidrug-resistant human MCF-7 breast adenocarcinoma cells reveals novel hyper- and hypomethylated targets. Mol Cancer Ther. 6, 1089–1098.PubMedCrossRefGoogle Scholar
  9. 9.
    Shames DS, Girard L, Gao B, Sato M, Lewis CM, Shivapurkar N, Jiang A, Perou CM, Kim YH, Pollack JR, Fong KM, Lam CL, et al. (2006) A genome-wide screen for promoter methylation in lung cancer identifies novel methylation markers for multiple malignancies. PLoS Med. 3, e486.PubMedCrossRefGoogle Scholar
  10. 10.
    Gryn J, Zeigler ZR, Shadduck RK, Lister J, Raymond JM, Sbeitan I, Srodes C, Meisner D, Evans C. (2002) Treatment of myelodysplastic syndromes with 5-azacytidine. Leuk Res. 26, 893–897.PubMedCrossRefGoogle Scholar
  11. 11.
    Weber M, Davies JJ, Wittig D, Oakeley EJ, Haase M, Lam WL, Schubeler D. (2005) Chromosome-wide and promoter-specific analyses identify sites of differential DNA methylation in normal and transformed human cells. Nat Genet. 37, 853–862.PubMedCrossRefGoogle Scholar
  12. 12.
    Wilson IM, Davies JJ, Weber M, Brown CJ, Alvarez CE, MacAulay C, Schubeler D, Lam WL. (2006) Epigenomics: mapping the methylome. Cell Cycle 5, 155–158.PubMedCrossRefGoogle Scholar
  13. 13.
    Frommer M, McDonald LE, Millar DS, Collis CM, Watt F, Grigg GW, Molloy PL, Paul CL. (1992) A genomic sequencing protocol that yields a positive display of 5-methylcytosine residues in individual DNA strands. Proc Natl Acad Sci U S A 89, 1827–1831.PubMedCrossRefGoogle Scholar
  14. 14.
    Gonzalgo ML, Liang G, Spruck CH, 3rd, Zingg JM, Rideout WM, 3rd, Jones PA. (1997) Identification and characterization of differentially methylated regions of genomic DNA by methylation-sensitive arbitrarily primed PCR. Cancer Res. 57, 594–599.PubMedGoogle Scholar
  15. 15.
    Huang TH, Laux DE, Hamlin BC, Tran P, Tran H, Lubahn DB. (1997) Identification of DNA methylation markers for human breast carcinomas using the methylation-sensitive restriction fingerprinting technique. Cancer Res. 57, 1030–1034.PubMedGoogle Scholar
  16. 16.
    Livak KJ, Schmittgen TD. (2001) Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods 25, 402–408.PubMedCrossRefGoogle Scholar
  17. 17.
    Coe BP, Ylstra B, Carvalho B, Meijer GA, Macaulay C, Lam WL. (2007) Resolving the resolution of array CGH. Genomics 89, 647–653.PubMedCrossRefGoogle Scholar
  18. 18.
    Khojasteh M, Lam WL, Ward RK, MacAulay C. (2005) A stepwise framework for the normalization of array CGH data. BMC Bioinformatics 6, 274.PubMedCrossRefGoogle Scholar
  19. 19.
    Branham WS, Melvin CD, Han T, Desai VG, Moland CL, Scully AT, Fuscoe JC. (2007) Elimination of laboratory ozone leads to a dramatic improvement in the reproducibility of microarray gene expression measurements. BMC Biotechnol. 7, 8.PubMedCrossRefGoogle Scholar
  20. 20.
    Chi B, DeLeeuw RJ, Coe BP, MacAulay C, Lam WL. (2004) SeeGH – a software tool for visualization of whole genome array comparative genomic hybridization data. BMC Bioinformatics. 5, 13.PubMedCrossRefGoogle Scholar
  21. 21.
    Chari R, Lockwood WW, Coe BP, Chu A, Macey D, Thomson A, Davies JJ, MacAulay C, Lam WL. (2006) SIGMA: a system for integrative genomic microarray analysis of cancer genomes. BMC Genomics 7, 324.PubMedCrossRefGoogle Scholar

Copyright information

© Humana Press, a part of Springer Science+Business Media, LLC 2009

Authors and Affiliations

  • Emily A. Vucic
    • 1
  • Ian M. Wilson
    • 1
  • Jennifer M. Campbell
    • 1
  • Wan L. Lam
    • 1
  1. 1.British Columbia Cancer Research CentreVancouverCanada

Personalised recommendations