2.6 | Measuring methylation rate
The 20 µl of 300 ng extracted DNA samples (section 2.3) and purified standard DNA (section 2.4) were subjected to bisulfite conversion using the EZ DNA Methylation-Gold Kit (Zymo Research, Irvine, CA, USA). This converts unmethylated cytosines to uracil. The differences between the bisulfite-converted DNA samples were measured using MS-HRM using a LightCycler 480 Instrument II (Roche Molecular Systems, Branchburg, NJ, USA) equipped with the Gene Scanning Software in a 25 µl total volume containing 1x EpiTect HRM PCR Master Mix (EpiTect HRM PCR kit: QIAGEN GmbH, Hilden, Germany), 750 nM of each primer, and 2–4 µl of template DNA (bisulfite-converted DNA). The chosen detection format was set to SYBR Green I/HRM Dye. First, the initial PCR activation step was performed at 95 ˚C for five mins to activate the polymerase, then the three-step PCR was performed (PCR conditions are presented in Table 1). Finally, MS-HRM analysis was initiated by cooling the samples to 65 ˚C for 1 sec and then heated to 95 ˚C at a ramp rate of 0.02 ˚C/s. Gene Scanning Software was used to normalize the raw melt curves, allowing the comparison of samples. After MS-HRM, the normalizing process was performed, the pre-melt temperature region was set to 72–74 ˚C and post-melt temperature region to 82-84 ˚C for both gene.
We defined the minimum absolute value of the relative signal difference from the difference plots as the Df value. For the analysis, we used the average Df value, which was calculated by measuring each sample twice. To construct a standard curve that correlates the Df value with the methylation rate, we used the 100% and 0% methylated standard DNA samples described in section 2.2. These standard DNA samples were mixed in various ratios, resulting in a set of standard DNA samples with different methylation percentages (0, 5, 10, 15, 20, 50, and 100%). The Df values of the standard curve were obtained from the MS-HRM using the standard DNA samples for both genetic regions. The plotted standard curve followed a non-linear regression model designed by Wameckeet al. (1997ed by Hamano et al . (2016) as follows:
\begin{equation} \frac{a*M}{b-M}=\ \frac{\text{Df}}{\text{Df}_{\max}-\ Df}\nonumber \\ \end{equation}
Where a: coefficient, b: The maximum methylation rate used for the standard curve construction, M: methylation rate,\(\text{Df}_{\max}\): Df value of the maximum methylated standard samples (GRIA2: 0.2, CDKN2A: 1.0).
Due to the low range of methylation percentages observed in GRIA2during subsequent analysis, the standard curve was redrawn using the standard DNA samples with the methylation percentages of 0, 5, 10, 15, and 20%. All subsequent analyses of GRIA2 utilized this recalibrated calibration curve to quantify the methylation percentage of the samples.