DNA Methylation Analysis by DHPLC | Genetics | Biology

In this article we will discuss about SIRPH – A Rapid, Quantitative SNuPE-IP RP HPLC Method for Simultaneous Methylation Analysis at Multiple CpG Sites.

Introduction:

Bisulfite Based DNA Methylation Analysis:

The most widely used protocols for detailed DNA methylation analyses are based on the use of bisulfite converted genomic DNA. This methodology was introduced by M. Frommer in the early 1990’s providing the first highly sen­sitive and sequence independent approach to detect CpG (mammals) and CpG and CpNpG (plants) methylation, independent of the sequence context. The method uses the ability of sodium bisulfite to selectively interact with cytosines at their carbon-6 position to form sulphonated cytosine intermediates.

These intermediates are then converted to uracil by pH-dependent deamination and desulphonation steps 5-methyl cytosine remains un-reactive under such con­ditions since the presence of the methyl group inhibits the formation of sulphonation at the carbon-6 position.

The uracil in the bisulfite convert­ed DNA is replaced by thymine in the subsequent PCR reaction. After cloning and sequencing of such bisulfite generated PCR products methylated cytosine will read as cytosine while un-methylated cytosine will read as thymine. Several modifications and improvements of the original protocol have been introduced. A detailed protocol allowing the use of smallest amounts of starting material was published recently.

Overview of Methods for the Analysis of PCR Products Derived from Bisulfite-Treated Genomic DNA:

Rapid and reliable quantification of methylated cytosines in a given genom­ic region is the main goal in DNA methylation analysis. The traditional method of cloning and sequencing of PCR products derived from bisulfite-treated genomic DNA provides the most detailed information. This approach, however, is very time consuming, laborious and expensive. For many applications the analysis of methylation at selected CpG sites will be sufficient.

Since there is a great demand for inexpensive, rapid, and quantitative methods for performing high throughput screening of DNA methylation for diagnostic purposes, several alternative methods have been developed in recent years. One such method is COBRA, which is based on restriction enzyme digestion of PCR products derived from bisulfite treated genomic DNA.

The use of this method, however, is lim­ited since it allows only the analysis of CpG methylation within (newly gener­ated) restriction sites of the PCR products derived from bisulfite-treated genom­ic DNA. Recently, a more flexible method based on differential hybridisation of PCR fragments from bisulfite treated genomic DNA, using oligonucleotide-containing chips, was introduced.

While this method allows for a high through­put screening, it requires high technological laboratory standards and sophisti­cated and laborious chip-design and analysis tools.

More flexible methods are based on single nucleotide primer extension techniques (SNuPE techniques) such as:

(1) MethyLight, a real time PCR based SNuPE method, which is quan­titative and highly sensitive, but which requires special fluorescently labelled primers;

(2) Ms-SNuPE which uses the incorporation of radioactive nucleotides that are quantified by autoradiography, requiring separation of extension prod­ucts on acrylamide gels.

Recently, we have developed an inexpensive, non­radioactive variation of such a SNuPE protocol using ion pair reversed phase high performance liquid chromatography (SIRPH: SNuPE-IP RP HPLC) for methylation detection and quantification.

SIRPH – Single Nucleotide Primer Extension with Ion Pair Reversed Phase High Performance Liquid Chromatography:

PCR products, derived from bisulfite treated DNA, of a region of interest are obtained using standard protocols. We recommend using agarose imbedded DNA material to obtain optimal results. The PCR products are then purified (see protocol below) to remove residual PCR primers and dNTPs. Subsequently primer(s) immediately 5′ to a CpG site(s) are hybridised to the denatured sin­gle-stranded PCR product.

Temperature cycling using Thermo Sequenase in the presence of both ddCTP and ddTTP extends the annealed primers; ddTTP (un- methylated CpG) or ddCTP (methylated CpG, see Figure 8-1) extended products are subsequently analysed by DHPLC directly (WAVE System, Transgenomic).

Due to the incorporation of the more hydrophobic ddTTP the retention time of such an extended product is longer compared to products containing ddCTP (Figure 8-1). The amount of the ddTTP and ddCTP extended products can then be quantified by either integrating peak areas or measuring peak heights and calculation of percentage ratios using WAVEMAKER Software version 4.1 – see analysis section below.

Materials and Methods:

i. Purification of PCR Products Derived from Bisulfite-Treated Genomic DNA:

PCR products can be purified either by gel extraction, e.g.: QIAquick Gel Extraction Kit, or by treatment with a mixture of exonuclease I and shrimp alkaline phosphatase, e.g.: ExoSap-IT (Amersham), which will degrade excess primer and unincorporated dNTPs.

ii. SNuPE Reaction:

Thermo Sequenase DNA Polymerase, 2′,3′-dideoxythymidine 5′-triphosphate (ddTTP), and 2′,3′-dideoxycytidine 5′-triphosphate (ddCTP) were purchased from Amersham Biosciences.

iii. IP RP HPLC:

All DHPLC analyses were performed on a WAVE Nucleic Acid Fragment Analysis System (Transgenomic) equipped with a DNASep Cartridge (Transgenomic). The standard dual buffer system was used with buffer A being 0.1 M aqueous triethylammonium acetate (TEAA) and buffer B being 0.1 M aqueous TEAA with 25% (v/v) acetonitrile. Acetonitrile was obtained from ROTH and HPLC grade water was purchased from Merck.

iv. Protocol Outline:

The SIRPH protocol for quantitative methylation analysis at specific CpG sites can be divided into three steps. The first step involves the generation of PCR products derived from bisulfite-treated genomic DNA.

The second step com­prises the SNuPE reaction, while the third step involves the separation and quantification of primer extension products with the WAVE System. Recently, we published a detailed description of the first step. In the following we describe in detail the latter two steps.

v. Detailed Protocol for the Analysis of Methylated DNA:

Treatment of isolated genomic DNA or cells and PCR is performed as described in the protocol by Hajkova et al. The PCR product is purified by either of two methods. (1) Run the PCR product on 1% agarose gel until opti­mum separation, excise the specific band of interest and recover the product by using a standard PCR-gel extraction kit.

In our experience the QIAquick Gel Extraction Kit from Qiagen yields highly pure products with high rate of recov­ery. (2) Add 2µl of exonuclease I (10 units/µl) and shrimp alkaline phosphatase (2 units/µl) to 5µl of PCR product, heat at 37°C for 15 min followed by 15 min at 80°C to deactivate the enzyme mixture.

Enzymatic treatment has the advantage that it is rapid and easier to perform, however, it is relatively more expensive. The gel extraction on the other hand is more laborious, but it has the advantage of concentrating low abun­dance PCR products.

In cases were non-specific or spurious PCR products are present, these may be removed by this method. However, we found that the enzymatic treatment is very convenient and easy to perform, especially when analysing large numbers of samples.

SNuPE-reaction sample setup (20µl total reaction volume):

The SNuPE PCR program is performed as follows:

Step 1: 94°C for 2 min;

Step 2: 92°C for 10 s;

Step 3: X°C for 30 s;

Step 4: 60°C for 1 min; repeat steps 2 to 4 50 times.

X = the optimum annealing temperature for the primer(s) used.

See below Analysis of results:

vi. HPLC Analysis:

Aliquots (10µl) of the above primer extension reactions are loaded direct­ly onto the WAVE Nucleic Acid Fragment Analysis System (Transgenomic). The oven temperature is set to 50°C. The elution gradient used is shown in Table 8-1 (also see Figure 8-2).

The values of B1 and B2 were defined empirically for each set of SNuPE primers. As a first step, the set of primers used is run with a wide gradient of 10% (B1) to 60% (B2) buffer B. Most short primers of 10 to 20 bp should be eluted using this gradient.

At a later stage, depending on the retention times of the primers, the gradient can be narrowed down/shortened by either increasing the starting percentage of buffer B, lowering the final per­centage of buffer B, or both. Resolution can be adjusted by changing the slope of the gradient.

A decrease in slope, the reduction of the change of buffer com­position in percent over a fixed time interval, generally translates into better resolution, while increasing gradient slope generally leads to decreased resolu­tion.

vii. Design Guidelines for SNuPE Primers:

The most important step of the SIRPH method is the selection and design of suitable sets of primers.

Some recommendations for designing appropriate primers are listed below:

(a) The 3′-end of the SNuPE primer has to be just 5′ (flanking) to the spe­cific CpG site to be studied.

(b) Avoid placing primers (if possible) in T-rich regions since this could increase mispriming and lead to inaccuracy in the methylation measurements.

(c) Primers must not include a CpG site, as this will bias measurements.

(d) Avoid designing excessively short primers, as these would have a higher chance for mispriming. In our experience primers as short as 10 bases can give accurate measurements. However, for routine uses our primers are 15 to 18 bases long.

(e) Primers that will be used in multiplex SNuPE reactions should be matched for their melting temperature. However, primers that differ by up to 10 °C in their optimum annealing temperature can be used without apparent decrease in the accuracy of measurements.

(f) For multiplexed SNuPE reactions that are to be analyzed by DHPLC, the retention times of the individual primers and their elongation products must exhibit differences. A general guideline for a rough estimation of relative reten­tion times for two primers is given in Table 8-2 (see also Figure 8-3). If for practical reason two primers have to be designed that give similar retention times, we recommend to extend one of the primers by the addition of thymidins (T) to its 5′-end.

In our experience, this addition has no functional effect on either the annealing temperature to the template, or on the SNuPE reaction itself. The number of T to be added has to be determined empirically. Each additional T leads to an incremental, stepwise increase of the retention time observed.

Analysis of Results:

The percentage of the methylated portion of genomic DNA before bisulfite treatment can be calculated from the chromatograms of the SNuPE products according to the formula: M = (AC/(AC+AT)) 100, were AC and AT are the peak areas or peak heights of ddCTP and ddTTP extended primers, respectively.

Use of peak heights yields more precise results than integrated peak areas. The latter, however, still yields a very good estimate of the percentage of methylation present. WAVEMAKER software automatically calculates the AC and AT peak areas and heights.

Advantages and Drawbacks of SIRPH Analyses:

The methylation analysis presented in this article has the advantages of being quantitative, rapid, non-radioactive, and suitable for automation. Moreover, a complete picture of the methylation states at a given CpG site in the PCR product under investigation is obtained.

However, some drawbacks of the method have to be considered:

(a) Detailed sequence information of the region to be studied is required

(b) In contrast to the cloning and sequencing approach, methylation analy­sis of only specific CpG sites can be performed with each primer.

(c) Mosaic patterns of methylation, were at given CpG sites random pat­terns of methylation occur on the same chromosome, cannot be detected.

(d) Imprinted patterns of methylation, with about half of the PCR product being representative of completely methylated genomic DNA and the other half of completely un-methylated DNA, cannot be revealed directly, but have to be deduced from the level of methylation at only few sites.

(e) In a CpG island where CpG sites are very close to each other’s, i.e.: between 0 and 10 bases, some CpG sites cannot be analyzed because of length restrictions on the primers used in SNuPE reactions.

Conclusion:

We described a detailed protocol for the SIRPH method for quantitative methylation analysis at specific CpG sites. The WAVE Nucleic Acid Fragment Analysis System (Transgenomic) proved to be adequate for this analysis.

The SIRPH assay can be optimized to analyze more than one CpG sites simultane­ously and is highly reproducible. The assay is highly beneficial in comparing large numbers of samples in a relatively short time.