Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2020 Aug 27.
Published in final edited form as: Methods Mol Biol. 2018;1708:497–513. doi: 10.1007/978-1-4939-7481-8_25

MethyLight and Digital MethyLight

Mihaela Campan 1, Daniel J Weisenberger 2, Binh Trinh 3, Peter W Laird 4
PMCID: PMC7450695  NIHMSID: NIHMS1619604  PMID: 29224160

Abstract

MethyLight is a quantitative, fluorescence-based, real-time PCR method to sensitively detect and quantify DNA methylation of candidate regions of the genome. MethyLight is uniquely suited for detecting low-frequency methylated DNA regions against a high background of unmethylated DNA, as it combines methylation-specific priming with methylation-specific fluorescent probing. The quantitative accuracy of real-time PCR and the ability to design bisulfite-dependent, DNA methylation-independent control reactions together allow for a quantitative assessment of these low frequency methylation events. Here we describe the experimental steps of MethyLight analysis in detail. Furthermore, we present principles and design examples for three types of quality control reactions. QC-1 reactions are methylation-independent reactions to monitor sample quantity and integrity. QC-2 reactions are bisulfite-independent reactions to monitor recovery efficiencies of the bisulfite conversion methodology used. QC-3 reactions are bisulfite-independently primed reactions with variable bisulfite-dependent probing to monitor completeness of the sodium bisulfite treatment. We show that these control reactions perform as expected in a time course experiment interrupting sodium bisulfite conversion at various timepoints. Finally, we describe Digital MethyLight, in which MethyLight is combined with Digital PCR, for the highly sensitive detection of individual methylated molecules, with use in disease detection and screening.

Keywords: DNA Methylation, Real-time PCR, TaqMan, Bisulfite, Epigenetics, Cancer, Quantitative, Methylation-specific PCR, Digital PCR

1. Introduction

MethyLight is a sodium-bisulfite-dependent, quantitative, fluorescence-based, real-time PCR method to sensitively detect and quantify DNA methylation in genomic DNA [1–5]. MethyLight relies on methylation-specific priming [6], combined with methylation-specific fluorescent probing [1–5]. This combination of methylation-specific detection principles results in a highly methylation-specific detection technology, with an accompanying ability to sensitively detect very low frequencies of hypermethylated alleles. The high sensitivity and specificity of MethyLight make it uniquely suited for the detection of low-frequency DNA methylation biomarkers as evidence of disease [7]. At the same time, the quantitative accuracy of real-time PCR and the flexibility to design bisulfite-dependent, methylation-independent control reactions [5] allows for a quantitative assessment of these low frequency methyla- tion events.

In addition to discussing in detail how to perform the experimental steps of MethyLight analysis, we present here how template, primer and probe design flexibility can be used to develop quality control reactions. The major principles are presented in Fig. 1. Methylation-independent, bisulfite-dependent reactions can be used as quality controls of sample quantity and integrity, as illustrated in Fig. 1: QC-1 [5]. We have developed a series of reactions to monitor recovery and completeness of the sodium bisulfite conversion step. One of the challenges in monitoring recovery during the conversion step is that the sequence changes as a result of the conversion. We therefore selected a region of the genome that does not contain any cytosine residues on one DNA strand over a short stretch (Fig. 1: QC-2), and thus remains unaffected by treatment with sodium bisulfite. Therefore, we can use a reaction targeting this locus (C-LESS) to monitor DNA recovery at any step during the sodium bisulfite conversion. Figure 2, right panel, shows a time course of a sodium bisulfite conversion reaction. It is evident that the C-LESS reaction is relatively impervious to the effects of sodium bisulfite. We also designed reactions to monitor the efficacy and completeness of sodium bisulfite conversion for a given sample (Fig. 1: QC-3). For this purpose, we selected a locus in the human genome for which the primer locations do not cover any cytosines in one strand of the DNA. Thus, the amplification of this strand would be bisulfite-conversion-independent. The region covered by the probe contains four cytosine residues. We designed and tested all 16 permutations of the probe, assuming either conversion to uracil, or lack of conversion at each cytosine (Fig. 2). Experimental results with the probes indicated by arrows on the left are shown for the bisulfite conversion time course on the right. We recommend using these probes to monitor completeness of the reaction. A threshold for bisulfite conversion quality control can be implemented simply as a delta cycle threshold (Δ-C(t)) for each of these reactions, compared to the methylation-independent ALU-C4 QC-1 quantity control.

Fig. 1.

Fig. 1

Open in a new tab

TaqMan-based MethyLight experimental and quality control PCR reactions used for quantitative methylation analysis of bisulfite converted DNA. Several types of TaqMan-based PCR reactions are used for the quantitative DNA methylation analysis of bisulfite converted DNA. Following bisulfite conversion, methylated cytosines remain unchanged, while unmethylated cytosines are deaminated to uracils. During subsequent PCR amplification of the bisulfite-converted DNA, thymine is incorporated in the place of uracil. The two strands of the bisulfite-converted DNA are no longer complementary, such that separate PCR reactions can be designed to amplify either the top or the bottom DNA strands. Each horizontal line represents one DNA strand. The tick marks represent cytosines not in the context of CpG dinucleotides, while the lollipops represent cytosines in the context of CpG dinucleotides. The methylated cytosines are depicted as solid black while the unmethylated ones are open white. The bottom strand does not participate in the further analysis after bisulfite conversion, and is depicted in gray. The experimental MethyLight reactions are specific for unmethylated or methylated DNA sequences. These reactions (the first 2 panels) are designed to include cytosines within CpG sequences (methylation-specific) as well as cytosines located outside the CpG context (bisulfite conversion-specific). Reactions towards unmethylated DNA sequences are designed to amplify TG-containing sequences, while the reactions toward methylated DNA sequences are designed to amplify CG-containing sequences. Three methylation-independent reactions are used as quality controls (QC) to monitor the sample quantity and integrity (QC-1), as well as bisulfite conversion recovery (QC-2) and bisulfite conversion completeness (QC-3). The QC-1 reaction is designed towards a CpG-less sequence that still contains cytosines outside the CpG context (bisulfite conversion-specific). QC-2 is a bisulfite-independent reaction in which both primers and the probes are designed towards a DNA region that does not contain any cytosines on one of the DNA strands (C-LESS). QC-3 reactions comprise a panel of 16 different reactions designed towards a single DNA sequence that have the same primer sequences but distinct probes. The DNA sequence covered by the primers lacks cytosine residues on one of the strands, while the DNA sequence covered by the probes contains four cytosines outside the CpG context (Fig. 2)

Fig. 2.

Fig. 2

Open in a new tab

Description of the QC-3 reactions and their performance on bisulfite converted DNA. We designed 16 distinct bisulfite conversion control reactions that have common forward and reverse primers complementary to a DNA strand lacking cytosine residues at the positions of the primers, but have cytosine-containing unique probes that differ in their abilities to recognize various percentages of bisulfite converted DNA (0%, 25%, 50%, 75%, and 100% conversion). The genomic DNA sequence covered by these probes contains four cytosines that are normally modified to uracils after bisulfite conversion and then to thymines after a subsequent PCR amplification. The degree of conversion of these residues can be monitored by these probes, since they contain 16 different permutations of these residues to thymine reflecting possible changes that could occur in case of complete or incomplete bisulfite conversion. We tested the performance of these reactions in a time course experiment where DNA was either left unconverted or was bisulfite converted for different periods of time (0.25 h, 0.5 h, 1 h, 2 h, 4 h, 8 h and 16 h). The ability of the 0% and 100% conversion reaction as well as the best 25%, 50%, and 75% conversion reactions to detect various degrees of bisulfite converted DNA is presented in the right panel of this figure along with the performance of the C-LESS reaction that is not affected by the bisulfite conversion process since it is designed towards a DNA sequence that contains no cytosines on one strand

2. Materials

2.1. M.SssI Modification

  1. M.SssI enzyme supplied with 10× buffer and 32 mM S-adenosyl methionine (SAM) (e.g., New England Biolabs).

  2. Peripheral blood leukocyte (PBL) DNA.

  3. DNA extraction kit for circulating nucleic acids (e.g., Qiagen QIAamp Circulating Nucleic Acid Kit or similar).

2.2. Bisulfite Conversion and Recovery

  1. EZ DNA Methylation kit (Zymo Research) or similar.

2.3. MethyLight PCR

  1. AmpliTaq Gold DNA Polymerase with Buffer II and MgCl2 (Thermo Fisher Scientific). The kit contains the AmpliTaq Gold Enzyme, 10× reaction buffer II (100 mM Tris–HCl, pH 8.3, 500 mM KCl), and the 25 mM MgCl2 stock.

  2. Deoxynucleotide triphosphates (dNTPs) are combined and diluted to a stock concentration of 10 mM for each nucleotide.

  3. Primers and Black-Hole Quencher containing probes are obtained from Biosearch Technologies Inc. (Novato, CA). The primers and probes are prepared as 300 μM and 100 μM solutions, respectively, in H2O. The probes containing the Minor Groove Binder Non Fluorescent Quencer (MGBNFQ) are obtained from Thermo Fisher Scientific, and are prepared as 100 μM solution in H2O.

3. Methods

3.1. DNA Isolation

Highly purified DNA is not a requirement for MethyLight analysis. Crude DNA extraction protocols involving lysis of the cells or tissues followed by DNA precipitation, or just crude lysates alone, can be used in conjunction with MethyLight analysis. These approaches are desirable when limited quantities of DNA are available, such as in tissues embedded in paraffin slides, tissues from biopsies, or in body fluids such as blood (plasma/serum) that contain small amounts of free circulating DNA. After biopsy tissues or microdissected cells from paraffin slides are lysed, an aliquot of this lysis solution can be directly used in bisulfite conversion. For the sensitive detection of DNA methylation in plasma or serum, the plasma/serum DNA needs to be concentrated from a larger initial volume. This can be achieved by using commercially available kits for DNA extraction from blood or various biological fluids or by precipitation of the DNA.

3.2. M.SssI Modification

M.SssI is a CpG methylase and therefore each CpG dinucleotide is a target of the enzyme, which uses S-adenosyl methionine (SAM) as a methyl donor. M.SssI treated DNA is used as a universally methylated reference sample in most MethyLight implementations. PBL DNA is used as a substrate in this protocol. A dilution of bisulfite-converted M.SssI-DNA will be used for normalization and is the basis for the ALU-C4 standard curves.

  1. Carry out the M.SssI treatment overnight at 37 °C in a solution containing H2O, 0.05 μg/μL of PBL DNA, 0.16 mM of SAM, 1× reaction buffer, and 0.05 units/μL of M.SssI enzyme.

  2. On the next day, add an extra boost of M. SssI enzyme and SAM (1/3 of the original volume) for both components together with H2O in a total volume representing 1/50 of the initial treatment volume.

  3. In order to achieve complete methylation at all the genomic CpG sites, multiple rounds of M.SssI treatment can be performed.

  4. Store M.SssI treated DNA at +4 °C, and use 20 μL (~1 μg) for each bisulfite conversion.

3.3. Bisulfite Conversion and Recovery

  1. Before starting, prepare the CT Conversion Reagent and M-Wash Buffer included in the Zymo EZ DNA Methylation kit. Add 750 μL of water and 210 μL of M-Dilution Buffer to one tube of CT Conversion Reagent and mix by vortexing every 1–2 min for a total of 10 min. Each tube of CT Conversion Reagent is designed to treat 10 DNA samples. For best results the prepared CT Conversion Reagent should be used immediately. Add 24 mL of 100% ethanol to the M-Wash Buffer Concentrate.

  2. Start the bisulfite conversion protocol by adding 5 μL of the M-Dilution Buffer to the DNA sample and adjust the total volume to 50 μL with sterile H2O. Mix the sample by flicking or pipetting up and down.

  3. Incubate the sample at 37 °C for 15 min.

  4. Add 100 μL of the prepared CT Conversion Reagent to each sample and vortex gently.

  5. Incubate the sample in the dark at 50 °C for 12–16 h.

  6. Incubate the sample on ice for 10 min.

  7. Add 400 μL of M-Binding buffer to the sample and mix by pipetting up and down.

  8. Load the sample onto a Zymo-Spin I column and place the column into a 2 mL collection tube.

  9. Centrifuge at full speed (>10,000 × g) for 30 s. Discard the flow-through.

  10. Add 200 μL of M-Wash Buffer to the column. Spin at full speed for 30 s.

  11. Add 200 μL of M-Desulfonation Buffer to the column and let column stand at room temperature for 15 min. After the incubation, spin at full speed for 30 s.

  12. Add 200 μL of M-Wash Buffer to the column. Spin at full speed for 30 s.

  13. Add another 200 μL of M-Wash Buffer and spin at top speed for 2 min.

  14. Add 10 μL of M-Elution Buffer directly to the column matrix. Place into a 1.5 mL tube. Spin at top speed for 1 min to elute the DNA.

  15. Bisulfite converted DNA is stored at −20 °C.

3.4. MethyLight Assay Design Guidelines

  1. The size of the MethyLight PCR amplicon should not be larger than 130 base pairs (bp) for optimal performance (shorter amplicons of 80–100 bp are preferred).

  2. Each primer and probe oligomer should contain at least two CpGs.

    Exceptions: MethyLight reactions containing only one CpG in each primer and probe oligomer can be used when there is no alternative, such as when the sequence is located in a CpG poor region. In this situation, the CpG should be positioned at the 3′ end in each primer and towards the middle of the probe oligomer in order to maximize discrimination between methylated and unmethylated loci. Additional specificity can be obtained by using shorter oligomers with minor groove binding (MGB) annealing enhancing modified nucleotides. This approach increases the relative contribution of each individual CpG. Further methylation specificity can be obtained by the use of blocking oligonucleotides, an approach referred to as HeavyMethyl [8].

  3. In general, CpGs should be located at or near the 3′ end of the primer sequences, however, the final five bases at the 3′ end of the primer oligomer should not contain more than three guanines and/or cytosines.

  4. In general, CpGs should be in the middle part of the probe oligomer sequence in order to maximize the melting point differences between methylated and unmethylated template molecules.

  5. Avoid mononucleotide repeats (e.g., AAAAAAA) in the primer and probe sequences, as these may decrease priming specificity. With respect to cytosine and guanines mononucleotide repeats longer than three bases should be avoided in assay design.

  6. The melting temperature (Tm) of the probe oligomer should be 10° higher than that of the primers. The Tm for each primer should be at 59 °C ± 2 °C, and should not be more than one degree different between each primer for an assay.

  7. Avoid using probe sequences in which the 5′ end begins with a guanine. Moreover, probe sequences should contain fewer guanines than cytosines. This can be achieved by using the complementary strand for probe design.

  8. Probe oligomers should not be greater than 30 bp in length.

3.5. TaqMan PCR Reaction Setup for MethyLight Analysis

The MethyLight assay makes use of the TaqMan PCR principle, which requires forward and reverse primers as well as a nonextend- able oligomeric probe which emits fluorescence only after it is degraded by the 5′ → 3′ exonuclease activity of the Taq polymerase.

  1. Each PCR reaction uses the same basic reaction setup. The choice of primer/probe sets is the only variable in these reactions.

  2. Each individual PCR reaction contains 10 μL DNA, 15.4 μL PreMix Solution, 4.5 μL OligoMix Solution (1.5 μL of each primer and probe) and 0.1 μL Taq polymerase in a 30 μL total PCR volume.

  3. The PreMix Solution contains all the TaqMan components except Taq polymerase. These components and their final concentration in a PCR reaction are: MgCl2 (3.5 mM), 1× TaqMan Buffer II (10 mM Tris–HCl, pH 8.3, 50 mM KCl), and 0. 2 mM of each dNTP. Each AmpliTaq Gold DNA Polymerase Kit is sufficient for 350 MethyLight reactions. Therefore, to prepare a PreMix Solution for 350 reactions, mix 1.47 mL of the 25 mM MgCl2 stock with 1.05 mL of 10× TaqMan Buffer, 210 μL of 10 mM combined dNTPs and 2.66 mL of H2O. Small aliquots are stored at +4 °C. For each PCR reaction use 15.4 μL of the PreMix Solution in a 30 μL total PCR volume.

  4. The OligoMix Solution is specific for each MethyLight and quality control reactions and represents a mixture of both primers and the probe. From the working stock of primers (300 μM) and probe (100 μM), prepare an OligoMix Solution by combining both primers and the probe in one tube. The concentrations of each the forward and reverse primers in the OligoMix Solution are 2 μM, and the probe concentration is 0.67 μM. For each PCR reaction use 4.5 μL of the OligoMix Solution in a 30 μL total PCR volume.

  5. The combined PreMix Solution, OligoMix Solution and Taq Gold polymerase for each reaction is referred to as the Master-Mix Solution (see Note 1). Load 10 μL of bisulfite converted DNA and 20 μL of the MasterMix Solution in each PCR well.

  6. For example, to determine the DNA methylation status of a specific gene of interest, such as MLH1, first prepare an MLH1 OligoMix Solution by combining 2 μL of the MLH1 forward primer (300 μM), 2 μL of the MLH1 reverse primer (300 μM) and 2 μL of the MLH1 probe (100 μM) with 294 μL water. The MethyLight primers and probe sequences for MLH1 have previously been published [9]. In each individual MLH1 PCR reaction, combine 4.5 μL of this MLH1 OligoMix Solution with 15.4 μL of PreMix Solution, 0.1 μL Taq polymerase and 10 μL of the bisulfite converted DNA sample to be analyzed.

  7. Individual OligoMix Solutions are prepared for any other gene investigated by MethyLight or any other Quality Control reactions used in the analysis, and 4.5 μL aliquots are then combined with the PreMix Solution, Taq polymerase and bisulfite converted DNA as described above.

  8. Each MethyLight-based data point is the result of the combined analysis of a methylation-dependent PCR reaction (Experimental MethyLight reaction, see Fig. 1) and methylation-independent PCR reaction (CpG-less sequence, see Fig. 1) on reference (M.SssI-treated DNA) and experimental DNA samples. The MethyLight assay setup is described in Subheading 3.7.

  9. All PCR reactions are carried out as follows: one cycle at 95 °C for 10 min followed by 50 cycles at 95 °C for 15s and 60 °C for 1 min.

3.6. Initial Quality Control

  1. QC-1: Sample Quantity and Integrity. Samples vary in the initial template quantity and integrity. The most reliable measure of amplifiable DNA quantities after bisulfite conversion is a bisulfite-dependent, methylation-independent reaction for a multi-copy number sequence well distributed throughout the genome. For this purpose, use the ALU-C4 bisulfite control reaction (5) (see Table 1 for primer and probe sequences) to perform a preliminary TaqMan PCR test to check the C (t) value of each sample. Following the purification of bisulfite treated DNA using the Zymo kit, the DNA is contained in a 10 μL volume. Dilute the sample 1:10 (final volume 100 μL) and test 2 μL by PCR using the ALU-C4 bisulfite control reaction and the PCR conditions described in Subheading 3.4. The The C(t) value generated from this 1:5 dilution will give an indication of the amount of bisulfite converted DNA available for further analysis (see Note 2).

  2. QC-2: Sample Recovery. If problems are regularly encountered with recovery of samples after bisulfite conversion, then C-less bisulfite-independent reactions can be used to monitor recovery of each step. Primer and probe sequences of the C-less reaction are given in Table 1, and the cycling conditions are described in Subheading 3.4.

  3. QC-3: Bisulfite Conversion Completeness. The efficacy and completeness of the bisulfite conversion of the sample can be assessed using a panel of bisulfite conversion reactions (Fig. 2) (see Table 1 for the primer and probe sequences of these reactions and Subheading 3.4 for cycling PCR conditions). These reactions are specific for unconverted DNA (0% conversion), fully converted DNA (100% conversion) or DNA with various degrees of conversion (25%, 50%, and 75% conversion) (see Note 3).

Table 1.

Primer and probes sequences for the quality control PCR reactions used for the analysis of bisulfite treated DNA

HB-Number Reaction ID Description Forward Primer Sequence (5′ to 3′) Reverse Primer Sequence (5′ to 3′)  Probe Sequence (5′ to 3′)
HB-313 ALU-C4 QC-1 GGT TAG GTA TAG TGG TTT ATA TTT GTA ATT TTA GTA ATT AAC TAA ACT AAT CTT AAA CTC CTA ACC TCA CCT ACC TTA ACC TCC C
HB-344 C-LESS-C1 QC-2 TTG TAT GTA TGT GAG TGT GGG AGA GA TTT CTT CCA CCC CTT CTC TTC C CTC CCC CTC TAA CTC TAT
HB-365 CONV-C1 QC-3: 100% AAA TTG GAG ATG AGG GAT GGG T TTA TCC TCC ACT CAT TCC CCA A TCT TAC AAA CTA ATC CTT AAC TTT
HB-368 CONV-C2 QC-3: 0% AAA TTG GAG ATG AGG GAT GGG T TTA TCC TCC ACT CAT TCC CCA A AAC TGG TCC TTG GCT TT
HB-369 CONV-C3 QC-3: 75% AAA TTG GAG ATG AGG GAT GGG T TTA TCC TCC ACT CAT TCC CCA A TTA CAA ACT GAT CCT TAA CTT T
HB-370 CONY-C4 QC-3: 50% AAA TTG GAG ATG AGG GAT GGG T TTA TCC TCC ACT CAT TCC CCA A TTA CAA ACT GAT CCT TAG CTT T
HB-372 CONV-C5 QC-3: 75% AAA TTG GAG ATG AGG GAT GGG T TTA TCC TCC ACT CAT TCC CCA A TTA CAA ACT AGT CCT TAA CTT T
HB-376 CONV-C9 QC-3: 50% AAA TTG GAG ATG AGG GAT GGG T TTA TCC TCC ACT CAT TCC CCA A TTA CAA ACT GGT CCT TAA CTT T
HB-377 CONV-C10 QC-3: 50% AAA TTG GAG ATG AGG GAT GGG T TTA TCC TCC ACT CAT TCC CCA A TTA CAA ACT AAT CCT TGG CTT T
HB-378 CONV-C11 QC-3: 25% AAA TTG GAG ATG AGG GAT GGGT TTA TCC TCC ACT CAT TCC CCA A TTA CAA ACT AGT CCT TGG CTT T
HB-379 CONV-C12 QC-3: 25% AAA TTG GAG ATG AGG GAT GGGT TTA TCC TCC ACT CAT TCC CCA A TTA CAA ACT GAT CCT TGG CTT T
HB-380 CONV-C13 QC-3: 25% AAA TTG GAG ATG AGG GAT GGGT TTA TCC TCC ACT CAT TCC CCA A TTA CAA ACT GGT CCT TAG CTT T
HB-381 CONY-C14 QC-3: 25% AAA TTG GAG ATG AGG GAT GGGT TTA TCC TCC ACT CAT TCC CCA A TTA CAA ACT GGT CCT TGA CTT T
HB-382 CONV-C16 QC-3: 50% AAA TTG GAG ATG AGG GAT GGGT TTA TCC TCC ACT CAT TCC CCA A TTA CAA ACT AGT CCT TAG CTT T
HB-383 CONV-C17 QC-3: 50% AAA TTG GAG ATG AGG GAT GGGT TTA TCC TCC ACT CAT TCC CCA A TTA CAA ACT AGT CCT TGA CTT T
HB-399 CONV-C15 QC-3: 50% AAA TTG GAG ATG AGG GAT GGGT TTA TCC TCC ACT CAT TCC CCA A TTA CAA ACT GAT CCT TGA CTT T
HB-411 CONV-C20 QC-3: 75% AAA TTG GAG ATG AGG GAT GGGT TTA TCC TCC ACT CAT TCC CCA A TTA CAA ACT AAT CCT TGA CTT T
HB-412 CONV-C21 QC-3: 75% AAA TTG GAG ATG AGG GAT GGGT TTA TCC TCC ACT CAT TCC CCA A TTA CAA ACT AAT CCT TAG CTT T
Open in a new tab

All probes contain a 6FAM fluorophore at the 5′ end and a minor groove binder (MGB) nonfluorescent quencher at the 3′ terminus. The genomic coordinates for the QC-2 reaction are; chromosome 20, 19199387-19199455, and for the QC-3 reactions; chromosome 11, 17649485-17649620 obtained from the NCBI Build 36.2. The QC-1 reaction is based on an ALU consensus sequence and therefore does not have precise genomic coordinates

3.7. MethyLight Reactions

Two types of reactions are used in the MethyLight protocol that both use the bisulfite converted DNA as a substrate: methylation-dependent reactions (CpG-based) specific for methylated DNA and methylation-independent control reactions (CpG-less), as described for the QC-1 quality control reaction above.

  1. The methylation-dependent reactions are both bisulfite- and methylation-specific reactions, i.e., they cover CpGs as well as Cs not in a CpG context in their sequence (methylated CpGs will remain CpGs, other Cs and unmethylated CpGs will become Ts or TpGs, respectively, after bisulfite conversion and PCR).

  2. The methylation independent control reaction (CpG-less) (ALU-C4) is used to normalize for differing quantities and quality of DNA samples (see Note 4). This reaction is not methylation-specific since there are no CpGs in the primers/probe sequences, but is specific for bisulfite converted DNA since it covers Cs not in a CpG context.

3.8. MethyLight Assay Setup

In order to determine the methylation status of a specific gene using the MethyLight assay, four PCR reactions are required. Two types of samples are needed: the bisulfite converted DNA of the sample of interest and the M.SssI-converted DNA as a reference sample. For each of these DNA samples, we perform a PCR reaction for the gene of interest (Experimental MethyLight measurement, see Fig. 1) and one control PCR reaction to measure the amount of amplifiable DNA sample (QC-1, ALU-C4 reaction, see Fig. 1). The use of M.SssI-converted DNA as a reference helps to control for variations in reagent batches, including primers and probes, reaction efficiency, machine performance, and various other sources of batch effects (see Note 5).

  1. Dilute the bisulfite converted M.SssI-DNA (1:100) and use 10 μL of this sample for each PCR reaction. Use 10 μL of the experimental sample (diluted based on the ALU-C4 C(t) value from the 1:5 dilution test, see Subheading 3.5, step 1). Perform each MethyLight reaction as well as each control reaction in duplicate.

  2. Two independent four-point standard curves using the ALU-C4 control reaction and bisulfite converted M.SssI-modified DNA (diluted 1:100) are required for quantification. From this initial stock of bisulfite converted M.SssI-modified DNA, perform 1:25 serial dilutions. A volume of 10 μL of each dilution should be used for each serially diluted sample.

3.9. MethyLight Percentage of Methylated Reference (PMR) Calculations

  1. The formula to calculate PMR values represents the quotient of two ratios (multiplied by 100). Thus, the formula is: 100 × [(GENE-X mean value)sample/(ALU-C4 mean value)sample]/[(GENE-X mean value)M.SssI/(ALU-C4 mean value)M.SssI].

  2. Once the real-time PCR program is finished, the C(t) values are converted to mean values/copy numbers using the standard curve for each plate (see Note 6).

  3. One PMR value per sample will be calculated based on the mean values derived from each of the two standard curves. The two PMRs obtained will be averaged at the end of the procedure.

  4. Using the data generated with the first standard curve, divide the mean/copy value for the methylation reaction of the sample of interest by the mean/copy value of the ALU-C4 reaction for the very same sample.

  5. Divide the mean/copy value for M.SssI sample for the same methylation reaction as in step 4 by the mean/copy value for the ALU-C4 reaction of the M.SssI sample. Average this quotient for duplicate reactions.

  6. Divide the value from step 4 by the value from step 5 and multiply that value by 100. This is the first PMR value.

  7. Calculate the second PMR value by the same procedure using the data generated based on the second standard curve. This can be achieved by simply reassigning the second ALU-C4 standard curve wells as standards. Then redetermine the values from steps 4 and 5. The PMR values from each quantitation can then be averaged to generate the final PMR value for each sample.

3.10. Digital MethyLight

We have also applied MethyLight technology to digital PCR applications in developing Digital MethyLight, an ultra-sensitive method for detecting and quantitating individual methylated DNA regions in biological fluids, such as plasma or serum [10]. Digital MethyLight involves distributing a MethyLight reaction across a 96- or 384-well plate or higher in a microfluidic device, such that the mean initial template DNA concentration is less than one molecule per reaction compartment. Amplification of methylated DNA molecules occurs in a small minority of PCR wells, and therefore represents a digital readout of the original number of template molecules in each sample (Fig. 3).

  1. Isolate genomic DNA from a specific volume of plasma, serum, etc., using the QIAamp Circulating Nucleic Acid as described by the manufacturer.

  2. Convert genomic DNA from each sample with sodium bisulfite using the EZ DNA methylation kit as described by the manufacturer.

  3. Mix bisulfite-converted DNAs, representing a specific volume of plasma or serum, with 200 μM dNTPs, 0.3 μM forward and reverse MethyLight primers, 0.1 μM MethyLight probe, 3.5 mM MgCl2 and 50 units of AmpliTaq GOLD polymerase in a 2.88 mL volume and distribute in 30 μL aliquots in each well of a 96-well plate. Since the amount of DNA in these samples is very small, and in order to minimize costs, we recommend scaling down the components of the assay by 1/3 and distribute the total volume of0.996 mL (2.88 mL/3) in 10 μL aliquots in each well of a 96-well plate.

  4. Perform digital MethyLight PCRs using the exact same PCR program as with Classic Digital MethyLight. Each well that shows a PCR amplification is scored as positive, and the total number of amplifications per 96-well plate is later normalized for the volume of input material.

Fig. 3.

Fig. 3

Open in a new tab

Digital MethyLight Technology Overview. Genomic DNA is isolated from a specific volume (usually 100–1000 μL) of plasma or serum. Genomic DNA is then bisulfite converted, mixed with MethyLight assay reagents and distributed across a multi-well PCR plate such that each well contains less than one methylated DNA molecule. Sample distribution also minimizes the amount and influence of unmethylated DNA and/or background contaminants that may affect the performance of classic MethyLight assays. MethyLight is performed on the entire multi-well plate, and the PCR-amplified reaction wells are counted. Each sample/assay is scored as the number of amplifications (number of methylated DNA molecules) per volume of input material

3.11. Multiplexing in Digital MethyLight Assays

Digital MethyLight assays can be performed with multiplexed markers. Multiplexing provides the most cost-effective way of analyzing multiple markers, while maximizing the simultaneous use of precious samples such as plasma or serum from patients. Presently, we have succeeded in combining up to three sets (triplex) of markers. Each probe is labeled with one of three different fluorophores (Fam, Hex and Quasar), so that three candidate gene regions can be concurrently and simultaneously interrogated. In order to develop multiplexed reactions with minimal impact on the performance of each individual reaction, combinations of the three primer/probe sets need to be evaluated to determine the impact of spurious primer annealing events by noncognate primers for each labeled probe.

  1. Set up each MethyLight reaction containing OligoMix Solutions with combinations of the three sets of primers and probes. Each PCR reaction uses the same reaction setup (discussed in Subheading 3.4).

  2. Prepare OligoMix Solutions for Reactions 1, 2, and 3 independently.

  3. To check for any interference of Probe 1 by any of the Primer Sets, prepare one OligoMix Solution with Primers for reactions 1 and 2 with the Probe for reaction 1, and another OligoMix solution with Primers for reactions 1 and 3 with the Probe for reaction 1.

  4. To check for any interference of Probe 2 by any of the Primer Sets, prepare one OligoMix Solution with Primers 1 and 2 with the Probe for reaction 2, and another OligoMix solution with Primers 2 and 3 with the Probe for reaction 2.

  5. To check for any interference of Probe 3 by any of the Primer Sets, prepare one OligoMix Solution with Primers for reactions 1 and 3 with Probe 3, and a second OligoMix solution with Primers 2 and 3 with Probe for reaction 3.

  6. Finally, prepare: (1) OligoMix Solution with all three primers sets and the Probe for reaction 1; (2) an OligoMix solution with all three primers sets and Probe for reaction 2, and (3) an OligoMix solution with all three primers sets and the Probe for reaction 3.

  7. Prepare all of the above OligoMix Solution from a diluted (1:3) working stock of primers (100 μM) instead of 300 μM. The probe working stock concentration of the probe will remain 100 μM. The concentrations of each the forward and reverse primers as well as the probe in the OligoMix Solution is 0. 67 μM. For each PCR reaction use 4.5 μL of the OligoMix Solution in a total PCR volume of 30 μL.

  8. Set up PCR reactions (in triplicate) with these 12 different OligoMix Solutions and the C(t) values of these primer/probes combinations will be compared. Any differences in C(t) values between the individual reactions and the Probes in combination with other primers will be indicative that the multiplex it is not possible for that particular Probe.

  9. The interference of any of the Probes by the other Probes can only be determined after each Probe is labeled with one of two different fluorophores (Hex or Quasar). Similar OligoMix Solutions can be made this time using combinations of two primers with two probes and three primers and all three probes. The C(t) values of these reactions can be compared with those of the individual reactions to determine if the Probes are compatible with each other.

Footnotes

1.

Uracil DNA glycosylase (AMPerase) should not be included in MethyLight reactions. This is a component used in some Taq- Man reaction kits, and this poses a complication in MethyLight reactions as uracil is a product following bisulfite conversion.

2.

It should be noted that low C(t) values are always preferred to achieve the best possible data. An ALU-C4 C(t)≤17 is usually desirable, but data can also be generated from samples with C (t) ≤ 22 when the samples are precious.

3.

The lack of complete decline of the mean values obtained for the 75% and 50% reactions is likely due to some degree of crosshybridization with fully converted sequence. Nevertheless, these two reactions are likely to be the most sensitive probes for detecting incomplete bisulfite conversion of a sample. It should be noted that these reactions assess completeness of bisulfite conversion only at this one locus. To the extent that sequences differ in their resistance to denaturing or bisulfite conversion, this locus may not be representative for other parts of the genome.

4.

Cancer genomic DNA samples can contain copy number alterations, which can affect the quantitation of both the locus of interest as well as the methylation-independent reaction. Ideally, one would design a CpG-less methylation-inde- pendent reaction as close to the MethyLight reaction as possible to adjust for such events. This significantly increases the cost of reaction design and experimental implementation. As a next-best solution, we avoid the influence of copy-number alterations of single loci for the methylation-independent QC-1 control reaction by using a repetitive element. Even in samples with aneuploidy, this control reaction will yield a reasonable approximation of total DNA quantities due to the distributed nature of the target sequence.

5.

We describe here the standard procedure in our laboratory to calculate the PMR value as a universal measure for the percentage of fully methylated alleles of a DNA sample, regardless of origin or locus being assessed. We find this a useful measure for most instances. Although our implementation is based on a comparison to a reference sample, it utilizes the absolute method of quantitation for real-time PCR, which is based on mean values derived from a standard curve of defined initial template quantities. By comparing to a control reaction and to a reference sample, we turn this absolute method into a relative method. We also sometimes implement relative methods of analysis, including the calculation of Δ-C(t) values. We use this most frequently, when we are interested in a separate within-sample comparison, such as the comparison of different bisulfite conversion control reactions. We prefer the PMR method as a general measure of DNA methylation, since it controls for many other sample-independent sources of experimental variation and error.

6.

Under usual real-time PCR conditions, the standard curve is based on dilutions of known absolute quantities of template. Although this could be implemented for each reaction, using synthetic or cloned template, we prefer to avoid this, in part to limit sources of high-concentration PCR contamination. Since the PMR calculation is a relative measure, it is sufficient to use unknown quantities of standard DNA, but with precisely defined dilutions. This will yield mean values that do not have any absolute meaning, but which can be used to derive the ratios in the PMR calculation. In other words, the unknown quantity of template DNA in the standards is divided away.

References

  • 1.Eads CA, Danenberg KD, Kawakami K et al. (1999) CpG island hypermethylation in human colorectal tumors is not associated with DNA methyltransferase overexpression. Cancer Res 59:2302–2306 [PubMed] [Google Scholar]
  • 2.Eads CA, Danenberg KD, Kawakami K et al. (2000) MethyLight: a high-throughput assay to measure DNA methylation. Nucleic Acids Res 28:e32. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Trinh BN, Long TI, Laird PW (2001) DNA methylation analysis by MethyLight technology. Methods 25:456–462 [DOI] [PubMed] [Google Scholar]
  • 4.Ogino S, Kawasaki T, Brahmandam M et al. (2006) Precision and performance characteristics of bisulfite conversion and real-time PCR (MethyLight) for quantitative DNA methyla- tion analysis. J Mol Diagn 8:209–217 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Weisenberger DJ, Campan M, Long TI et al. (2005) Analysis of repetitive element DNA methylation by MethyLight. Nucleic Acids Res 33:6823–6836 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Herman JG, Graff JR, Myohanen S et al. (1996) Methylation-specific PCR: a novel PCR assay for methylation status of CpG islands. Proc Natl Acad Sci U S A 93:9821–9826 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Laird PW (2003) The power and the promise of DNA methylation markers. Nat Rev Cancer 3:253–266 [DOI] [PubMed] [Google Scholar]
  • 8.Cottrell SE, Distler J, Goodman NS et al. (2004) A real-time PCR assay for DNA-methylation using methylation-specific blockers. Nucleic Acids Res 32:e10. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Fiegl H, Gattringer C, Widschwendter A et al. (2004) Methylated DNA collected by tampons-a new tool to detect endometrial cancer. Cancer Epidemiol Biomarkers Prev 13:882–888 [PubMed] [Google Scholar]
  • 10.Weisenberger DJ, Trinh BN, Campan M et al. (2008) DNA methylation analysis by digital bisulfite genomic sequencing and digital MethyLight. Nucleic Acids Res 36:4689–4698 [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES