Sensitive DNA Methylation Detection Using Droplet Digital PCR

DNA Methylation Detection Using Droplet Digital" PCR Hamutal Bonen, PhD; Jeremiah McDole, PhD; Eddy van Collenburg Bio-Rad Laboratories, Inc., Digital Biology Center, 5731 W Las Positas Blvd, Pleasanton, CA 94588 Abstract The year 2023 marks 12 years since the advent of Droplet Digital PCR (ddPCR™). In that time, ddPCR technology has revolutionized the absolute quantification of nucleic acids. With a host of technical advances and enhanced automation in recent years, Droplet Digital PCR is moving from a specialized research technique to a robust tool in clinical and translation research applications. This white paper reviews the use of Droplet Digital PCR for detection of DNA methylation. We review ddPCR technology, compare different workflow strategies, and explain critical parameters for optimal assay design to get unprecedented sensitivity, linearity, and robustness when detecting DNA methylation by Droplet Digital PCR. Introduction DNA methylation plays a critical role in many normal developmental and sustaining physiological processes. However, aberrant methylation can lead to a variety of disease states, including several types of cancer. For instance, global hypomethylation can contribute to the activation of oncogenes while hypermethylation within tumor suppressor genes can result in silencing. Mounting evidence demonstrates the utility of analyzing methylation status as an indicator of early-stage cancer development and as a method for monitoring tumor response to therapy. Detecting methylation changes in cell-free DNA (cfDNA) is especially attractive given the noninvasive nature of blood draws. Unfortunately, sensitive, robust, and cost-effective methods to perform methylation analysis on cfDNA samples remain an unmet need. Here we demonstrate how common methylation detection methods combine with Droplet Digital PCR to enhance analysis of these precious samples. DNA Digestion Using Methylation-Sensitive Restriction Enzymes (MSRE) By far, the fastest and easiest method of DNA methylation quantification incorporates MSRE digestion. Further, this method does not have the drawbacks of bisulfite conversion, such as extensive DNA degradation and high starting material input requirements. As shown in the schematic below, methylation-sensitive restriction enzymes do not cleave restriction sites in the presence of a methylated cytosine, leading to PCR amplification (Figure 1), while cleavage of an unmethylated site inhibits amplification. Unmethylated Restriction Site Cleavage No Amplification E Amplification Methylated Restriction Site M No Cleavage Fig. 1. Schematic overview of methylation-sensitive restriction enzyme digestion. M, methyl; MSRE, methylation-sensitive restriction enzyme. There are a variety of commercially available MSREs available (Table 1). Specific primers can be designed to amplify a specific sequence of interest that contains an MSRE. See the Primer Design section for recommendations on how to design primers. MSRE PCR Restriction Site CGICG GC+GC CĮCGC GGC1G CĮCGG GGC1C GCGĮC C1GCG To detect the methylation status at a specific locus, the sample is digested by a methylation-sensitive restriction enzyme and analyzed with Droplet Digital PCR (Figure 2). When using probe- based detection methods, multiplexing several methylation sites or targets is possible. For best results, a duplex reaction with a reference target (no restriction sites) is advised (e.g., RPP30). The reference assay is used to normalize and correct for small input differences caused by pipetting or samples with unstable copy number. Methylation percentages are calculated with high precision via Droplet Digital PCR by measuring DNA concentrations in samples with (MSRE+) and without (MSRE-) enzyme. The percentage of methylation, or methylation fraction, is calculated using the precise ddPCR concentrations measured for samples either digested (MSRE+) or undigested (MSRE-), as follows: MSRE+ ratio = target concentration/reference concentration MSRE-ratio = target concentration/reference concentration % methylation = MSRE+ ratio/MSRE- ratio Note that copy number results are available from the undigested (MSRE-) samples and can be used to detect differences in samples (assuming the use of a stable reference gene). DNA Methylation Detection Using Droplet Digital PCR Table 1. Overview of MSREs and restriction site sequences. Restriction Enzyme (all 4-base cutters) Accll (BstUl, Bsh12361) Acil (Ssil) Hpall (Hapll) Hhal (Cfol) Temperature, C 95 95 55 98 12 Droplet reading can be performed on a QX200T or QX600TM Droplet Reader. All the data shown in this white paper were generated using a QX200 Droplet Reader. Cycling Step Enzyme activation Denaturation Annealing/extension Enzyme deactivation Hold Time 10 min 30 sec 1 min 10 min <48 hr Samples are prepared as follows (incubations can be done in a thermal cycler such as the T100 Thermal Cycler): . MSRE+ samples - 50-100 ng of DNA is incubated for 1 hr at 60°C with 2-10 U of BstUl and 0.5 ul 10x CutSmart Buffer (New England Biolabs, Inc., discontinued; rCutSmart Buffer [#B6004S] is a suitable alternative) in a total volume of 5.0 pl. . MSRE-samples - 50-100 ng of DNA is incubated for 1 hr at 60°C with 0.5 ul 10x CutSmart Buffer in a total volume of 5.0 ul. In a 22 ul experiment, 1-100 ng of incubated DNA samples can be analyzed using 11 ul ddPCR Supermix for Probes (No dUTP) (Bio-Rad Laboratories, Inc., catalog #1863024) and primers and probes in a final concentration of 900 and 250 nM, respectively. PCR mixtures are partitioned into ~20,000 droplets using the Automated Droplet Generator (Bio-Rad, #1864101). Subsequent PCR is performed in a thermal cycler using the the protocol outlined in Table 2. Table 2. Thermal cycling protocol. Ramp rate set to 2C/sec for all steps. Number of Cycles 1 40 1 Concen Methylated MSRE- Methylated MSRE+ Unmethylated MSRE- MSRE+ Ratio MSRE-Ratio Methylation, % 2.50 Unmethylated MSRE+ Fig. 3. PTGER4 promotor (FAM) and RPP30 (HEX). One-dimensional plots show the positive and negative droplet clusters from two targets run in duplex, PTGER4 (FAM, AssaylD dHsaEXD82842910) and RPP30 (HEX, AssaylD dHsaCP2500350). Two-dimensional plot shows the four clusters from the same reaction. Hpall is used for digestion. Concentrations (copies/ul) are shown in the bottom graph and used for calculating the percentage of methylation from each sample (see calculations). MSRE, methylation-sensitive restriction enzyme. PTGER4 Promoter (FAM) and RPP30 (HEX) The percentage of PTGER4 promoter methylation was calculated as shown in Table 3. Table 3. PTGER4 promoter methylation calculations. Methylated 175/185 = 0.946 179/183 = 0.978 0.946/0.978 = 97% 2023 Bio-Rad Laboratories, Inc. Unmethylated 2.39/249 = 0.0096 272/252 = 1.079 0.0096/1.079 = 0.9% MSRE+ Ratio MSRE-Ratio Methylation, % Methylated MSRE- Methylated MSRE+ Fig. 4. SEPT9 CGI3 promotor (FAM) and RPP30 (HEX). One-dimensional plots show the positive and negative droplet clusters from two targets run in duplex, SEPT9 CGI3 (FAM, AssaylD dHsaEXD18126781) and RPP30 (HEX, AssayID dHsaCP2500350). Two-dimensional plot shows the four clusters from the same reaction. BstUl is used for digestion. Concentrations (copies/ul) are shown in the bottom graph and used for calculating the percentage of methylation from each sample (see calculations). MSRE, methylation-sensitive restriction enzyme. The percentage of SEPT9 promoter methylation was calculated as shown in Table 4. Table 4. SEPT9 promoter methylation calculations. Methylated 335/208 = 1.611 350/217 = 1.613 1.611/1.613 = 100% Unmethylated MSRE- Unmethylated MSRE+ Unmethylated 2.33/266 = 0.0088 271/272 = 0.996 0.0088/0.996 = 0.9% Bulletin 3585 [Buner) Fig. 5. Probe Mix Triplex WIF1 (FAM), NPY (FAM), and RPP30 (0.4x FAM, 0.25x HEX). One-dimensional plots show the positive and negative droplet clusters from three targets run in triplex. WIF1 (FAM, AssaylD dHsaEXD31159278), NPY (FAM, AssaylD dHsaEXD18238370), and RPP30 (0.4x FAM, AssaylD dHsaCP2500313, 0.25x HEX, AssayID dHsaCP2500350). Two-dimensional plot shows up to eight clusters from the same reaction. A mixture of Acil and Hhal was used for digestion. Two replicate reactions of an unmethylated control sample were used to compare the digestion in sample (buffer: digestion overnight) with the digestion in master mix (sample: 45 min at room temperature). Concentrations (copies/ul) are shown in the bottom graph and used for calculating the percentage of methylation from each sample (see calculations). MSRE, methylation-sensitive restriction enzyme. WIF1 MSRE+ Ratio MSRE-Ratio Methylation, % NPY MSRE+ Ratio MSRE-Ratio Methylation, % WIF1 and NPY methylation was calculated as shown in Table 5. Table 5. WIF1 and NPY methylation calculations. Overnight Digestion 0.345/56 = 0.0062 53.7/57.9 = 0.927 0.0062/0.927 = 0.7% 5.48/56 = 0.0979 58.3/57.9 = 1.007 0.0979/1.007 = 9.7% ddPCR Assay Design Since analysis is based on restriction enzyme cleavage, the amplicon sequence must include cleavage sites. This approach provides high specificity, however only specific restriction sites can be analyzed, which can be a limitation. To reliably measure DNA methylation, we recommend including at least two but not more than four restriction sites in the amplicon. If a single MSRE cannot cut at least two sites in an amplicon, combining multiple MSREs that cut within the amplicon is a good alternative. For an extended overview on how to design specific ddPCR assays, please refer to the Assay Design for Droplet Digital PCR section in bulletin 6407. Consider the following when designing primers: . Check for restriction sites within the area of your target sequence . Look for restriction enzymes that have 2 or more restriction sites close together . Design primers outside these restriction sites to produce an amplicon with 2 or more restriction sites A restriction site can be included in the primers but only in the last 4 nucleotides of the 3' end of each of the primers. Restriction sites outside the primers are preferred Short amplicons (<100 nucleotides) enable fragmented sample analysis (e.g., formalin-fixed paraffin-embedded [FFPE] or circulating tumor DNA [ctDNA]) . Combining different restriction enzymes is possible and used to perform double digestions in the same reaction Example Assay Design Figure 6 shows an example assay design for the transcription factor one cut homeobox 2 (ONECUT2). Amplicon: 64 bp, location hg19|chr18:55103715-55103778:+ Sequence: AAATGCTCAGCCCCAACTTCGACGCGCACCACACT GCCATGCTGACCCGCGGTGAGCAACACC Fig. 6. Example assay design for the transcription factor ONECUT2 (BstUl restriction sites). Bio-Rad PrimePCR AssaylD: dHsaEXD31605634. BstUl sites are bolded, primer sites are underlined, and the probe site is highlighted yellow. Chr, chromosome; hg, human genome. ddPCR Master Mix Digestion 0.0559/53.7 = 0.00104 53.7/57.9 = 0.927 0.00104/0.927 = 0.1% 5.16/53.7 = 0.0961 58.3/57.9 = 1.007 0.0961/1.007 = 9.5% Cytosine Deamination While MSRE digestion protocols are rapid and reliable, sodium bisulfite conversion is the current gold standard of methylation analysis. In this method, DNA is denatured and treated with sodium bisulfite. This process deaminates unmethylated cytosine in DNA to uracil and leaves methylated cytosines unchanged. Amplifying the treated DNA with PCR converts uracils further to thymines while the methylated cytosines remain cytosines. An example of cytosine deamination is shown in Figure 7. Although bisulfite treatment is relatively fast and straightforward, it requires elevated temperatures and high pH that can damage DNA, resulting in fragmentation and depyrimidination. Unmethylated cytosines are especially sensitive to degradation with this method, resulting in challenges amplifying GC-rich areas. A less damaging method for cytosine deamination was recently developed by New England Biolabs, Inc. (NEB). In this method, APOBEC enzyme is used to convert cytosine to uracil, resulting in far less DNA degradation. However, APOBEC can also deaminate 5-methylcytosine (5mC) and 5-hydroxymethylcytosine (5hmC). To protect the methylated cytosine from deamination, NEB uses TET2 enzyme and Oxidation Enhancer (#E7129) to modify 5mC and 5hmC to forms that are not substrates for APOBEC. When deciding which method to utilize, the required starting material should be considered. Sodium bisulfite protocols typically require between 0.1 and 2 g of DNA (depending on the manufacturer) while the enzymatic approach can be performed with much less. Since both methods rely on the conversion of unmethylated DNA to thymine, primers and probe designs are the same for both. Primer Design There are two main design strategies available to specifically amplify converted (unmethylated) and unconverted (methylated) sequences with Droplet Digital PCR: methylation-specific primers and methylation-independent primers (Figure 8). Primer design considerations for methylation-specific primers: . Include as many CpG sites as possible, especially at the 3' end of the primer . Consider the same CpG sites in the primer sequence for methylated DNA and unmethylated DNA Maximize specificity by using a detection probe including CpG sites Primer design considerations for methylation-independent primers: . No CpG sites within the primer sequence Use an adequate number of C (no CpG) Spanning a maximum number of CpG sites in the amplicon Maximize specificity by using a detection probe including CpG sites