MSP, methylation-specific PCR.
DHPLC analyses were conducted using a DNASep® HT column (Transgenomic) under nondenaturing conditions at a temperature of 50°C with a 1.5 mL/min flow rate and UV detection. Under these conditions, PCR product separation is based on size not sequence. Resolution is limited to approximately 1% of the PCR product size. DHPLC analysis conditions for stages 1 and 2 nested three-gene MSP assays are listed in (Table 2). PCRs were loaded onto the WAVE autosampler, and 5-µL injections were used for each analysis. DHPLC analysis of both stage 1 and 2 products required <6 min/sample. Typical DHPLC detection of stages 1 and 2 unmethylated MSP products are shown in (Figure 1). It should be noted that given the heterogeneity of biological samples, verification of target gene amplification is normally accomplished by the presence of a stage 1 PCR product. Stage 2 unmethylated primers are typically used only when analyzing extensively degraded DNA as found in some formalin-fixed paraffin-embedded tissue samples.Figure 1.
DHPLC, denaturing high-performanc liquid chromato raphy. Time is in minu es. A, % uffer A 0.1 M trieth l mmon M triethylammonium acetate); %B, % buffer B (0.1 M triethylammonium acetate, 25% acetonitrile).
One-hundred percent concordance was observed in the detection of methylated stage 2 MSP products between the nested single-gene MSP/agarose gel electrophoresis method and the nested three-gene MSP/DHPLC method. An example of MSP product separation using each method is shown in (Figure 2), A–F. DHPLC-based separation has better resolution and increased sensitivity for detecting methylated PCR products compared with agarose gel electrophoresis ((Figure 2)). For example, sputum samples 5 and 8 are methylated for both p16 and DAPK. Resolving the 8-bp size difference between stage 2 PCR products is difficult using agarose gel electrophoresis, while this difference is clearly discerned by DHPLC. The increased resolution of DHPLC makes it easier to separate and identify methylated MSP products. This is essential in accurately determining the promoter methylation status of individual genes in a multiplex reaction. Additionally, as demonstrated in (Figure 2), E and F, DHPLC analysis of MSP products obtained using a formalin-fixed paraffin-embedded DNA template do not present added difficulties. Compared with conventional nested single-gene MSP, a nested multigene MSP/DHPLC method requires less time, reagents, and sample DNA to analyze the gene promoter methylation status of clinical samples from a variety of tissues. The ability to identify methylated MSP products using UV detection also eliminates the cost associated with fluorescence probes required for other nongel-based MSP methods such as TaqMan®. Savings realized are proportional to the number of genes present in the nested multigene MSP and/or the number of gene products analyzed per DHPLC chromatogram. To further increase throughput and savings, we are modifying the multigene MSP/DHPLC method to simultaneously detect promoter methylation of four genes (Pax5α, Pax5β, GATA4, and GATA5) ((Figure 3)). The Pax5 and GATA gene families encode transcription factors implicated in carcinogenesis (10,11). All four genes are silenced by aberrant promoter methylation in lung cancer (10,12). The conditions for the nested four-gene MSP/DHPLC method are available in the supplementary material available online at www.BioTechniques.com .Figure 2.
We have described a nested multigene MSP/DHPLC method to simultaneously examine gene promoter hypermethylation of three genes. Although this method was developed to detect multiple gene methylation events in complex biological samples, quantitation of methylation levels can be accomplished by determining the ratio of methylated to unmethylated DNA as reflected by individual peak heights expressed in millivolts ((Figure 1)B and (2)C). This method will aid in the rapid and economical assembly and validation of gene panels involved in cancer initiation, progression, and malignant transformation. Once validated, biomarkers based on these gene panels can be used in the early detection, diagnosis, and prognosis of multiple types of cancer.