Candiloro Ida L M, Mikeska Thomas, Dobrovic Alexander
Peter MacCallum Cancer Centre, Melbourne, Victoria 3000 Australia.
Department of Pathology, The University of Melbourne, Parkville, Victoria 3010 Australia.
Clin Epigenetics. 2017 Apr 4;9:31. doi: 10.1186/s13148-017-0328-4. eCollection 2017.
Determining the role of DNA methylation in various biological processes is dependent on the accurate representation of often highly complex patterns. Accurate representation is dependent on unbiased PCR amplification post bisulfite modification, regardless of methylation status of any given epiallele. This is highly dependent on primer design. Particular difficulties are raised by the analysis of CpG-rich regions, which are the usual regions of interest. Here, it is often difficult or impossible to avoid placing primers in CpG-free regions, particularly if one wants to target a specific part of a CpG-rich region. This can cause biased amplification of methylated sequences if the C is placed at those positions or to unmethylated sequences if a T is placed at those positions.
We examined the effect of various base substitutions at the cytosine position of primer CpGs on the representational amplification of templates and also examined the role of the annealing temperature during PCR. These were evaluated using methylation-sensitive high-resolution melting and Pyrosequencing.
For a mixture of fully methylated and unmethylated templates, amplification using the C-, C/T (Y-) and inosine-containing primers was biased towards amplification of methylated DNA. The bias towards methylated sequences increased with annealing temperature. Amplification using primers with an A/C/G/T (N) degeneracy at the cytosine positions was not biased at the lowest temperature used but became increasingly biased towards methylated DNA with increased annealing temperature. Using primers matching neither C nor T was in the main unbiased but at the cost of poor PCR amplification efficiency. Primers with abasic sites were also unbiased but could only amplify DNA for one out of the two assays tested. However, with heterogeneous methylation, it appeared that both the primer type and stringency used have a minimal influence on PCR bias.
This is the first comprehensive analysis of base substitutions at CpG sites in primers and their effect on PCR bias for the analysis of DNA methylation. Our findings are relevant to the appropriate design of a wide range of assays, including amplicon-based next-generation sequencing approaches that need to measure DNA methylation.
确定DNA甲基化在各种生物过程中的作用取决于对通常高度复杂模式的准确呈现。准确呈现取决于亚硫酸氢盐修饰后无偏差的PCR扩增，而与任何给定表观等位基因的甲基化状态无关。这高度依赖于引物设计。富含CpG的区域是通常感兴趣的区域，对其分析会带来特别的困难。在此，通常很难或不可能避免将引物置于无CpG的区域，特别是如果想要靶向富含CpG区域的特定部分时。如果在这些位置放置C，这可能导致甲基化序列的偏差扩增；如果放置T，则会导致未甲基化序列的偏差扩增。
我们研究了引物CpG位点胞嘧啶位置的各种碱基替换对模板代表性扩增的影响，并研究了PCR过程中退火温度的作用。使用甲基化敏感的高分辨率熔解分析和焦磷酸测序对这些进行评估。
对于完全甲基化和未甲基化模板的混合物，使用含C、C/T（Y）和含次黄嘌呤的引物进行扩增偏向于甲基化DNA的扩增。对甲基化序列的偏向随着退火温度的升高而增加。在胞嘧啶位置具有A/C/G/T（N）简并性的引物在使用的最低温度下无偏差，但随着退火温度升高，对甲基化DNA的偏向性越来越大。使用与C和T均不匹配的引物在主要方面无偏差，但代价是PCR扩增效率低下。具有无碱基位点的引物也无偏差，但在测试的两种分析中只能扩增其中一种的DNA。然而，对于异质性甲基化，似乎所用的引物类型和严格性对PCR偏差的影响最小。
这是对引物中CpG位点碱基替换及其对用于DNA甲基化分析的PCR偏差影响的首次全面分析。我们的发现与广泛分析方法的适当设计相关，包括需要测量DNA甲基化的基于扩增子的下一代测序方法。
