Winnowing DNA for rare sequences: highly specific sequence and methylation based enrichment

Abstract

Rare mutations in cell populations are known to be hallmarks of many diseases and cancers. Similarly, differential DNA methylation patterns arise in rare cell populations with diagnostic potential such as fetal cells circulating in maternal blood. Unfortunately, the frequency of alleles with diagnostic potential, relative to wild-type background sequence, is often well below the frequency of errors in currently available methods for sequence analysis, including very high throughput DNA sequencing. We demonstrate a DNA preparation and purification method that through non-linear electrophoretic separation in media containing oligonucleotide probes, achieves 10,000 fold enrichment of target DNA with single nucleotide specificity, and 100 fold enrichment of unmodified methylated DNA differing from the background by the methylation of a single cytosine residue.

Conflict of interest statement

Competing Interests: The authors have the following competing interest: Jason D. Thompson and Andre Marziali are inventors on, and entitled to royalties arising from, a patent application on methods described in this manuscript. Jason D. Thompson and Andre Marziali own shares in Boreal Genomics, a company holding an exclusive license to the technology described in this manuscript, and currently developing a product based on the methods described in this manuscript. This does not alter the authors' adherence to all the PLoS ONE policies on sharing data and materials, as detailed online in the guide for authors.

Figures

Figure 1. Measurement of temperature dependence of DNA target mobility through a gel containing immobilized complementary oligonucleotide probes.

The upper limit of these curves were cut off because target DNA had migrated off the end of the gel used for taking these measurements. Both sets of data points were fit to equation (2) using the non-linear least squares fitting tool in the Origin 7.5 software package (OriginLab Corporation, Northampton MA). Mfold calculated values were used for the enthalpy and entropy terms; μ0 and α were determined by the fit. For the mismatch curve μo = 1.34e-9+/−0.05e-9 m2/Vs and α = 1.12e-6+/−0.07e-6. For the perfect complement μ0 = 1.28e-9+/−0.05e-9 m2/Vs and α = 2.01e-7+/−0.13e-7.

Fig<?Pub Caret?>ure 2. Time series of ssSCODA focusing under bias. PC DNA is tagged with 6-FAM (green) and snMM DNA is tagged with Cy5 (red).

Images were taken at 2 min intervals with the first image taken immediately following injection. The camera gain was reduced from 32 to 16 on the green channel after the first image was taken. DNA was injected from a chamber adjacent to the right side of the gel. After injection, focusing plus bias fields are applied. The PC target (green) experiences a convergent drift velocity and focuses to the centre of the gel. The more weakly focusing snMM target (red) is washed completely from the gel by the bias field.

Figure 3. Demonstration of length independent focusing.

Left: Focus location under bias for 250 bp (green) and 1000 bp (red) fragments. Right: Image of the gel at the end of the run. Green and red channels have been superimposed.

Figure 4. Rejection ratio of snMM DNA. Four different ratios of snMM:PC were injected into a gel and focused under bias to remove excess snMM.

The PC DNA was tagged with 6-FAM and the snMM DNA was tagged with Cy5. Top: fluorescence signal from the final focus spot after excess snMM DNA had been washed from the gel. The fluorescence signals are normalized to the fluorescence measured on an initial calibration run where a 1∶1 ratio of PC-FAM:PC-Cy5 DNA was injected and focused to the centre of the gel. Bottom: rejection ratios calculated by dividing the initial ratio of snMM:PC by the final ratio after washing.

Figure 5. Enrichment of EZH2 Y641N mutation from a 1∶1 mixture of wild type (green) and mutant (red) amplicons.

30 ng of each 470 bp target was diluted into a 250 µl solution of 0.9 mM tris, 0.9 mM boric acid and 2 mM NaCl. The sample was placed in a boiling water bath for 5 min to denature the double stranded DNA prior to injection. The targets were injected from a chamber adjacent to the lower right corner of the gel. After injection, focusing and bias fields were applied to simultaneously concentrate the mutant amplicon while washing the wild type amplicon from the gel.

Figure 6. Measurement of mobility versus temperature for methylated and unmethylated targets.

Data points were fit to equation (3). The upper limit of these curves were cut off because target DNA had migrated off the end of the gel used for taking these measurements. Both sets of data points were fit to equation (2) using the non-linear least squares fitting tool in the Origin 7.5 software package (OriginLab Corporation, Northampton MA). For the unmethylated curve Mfold calculated values were used for the enthalpy and entropy terms; μ0 and α were determined by the fit. Using ΔH = 144.4 kcal/mol and ΔS = 0.3988 kcal/mol K the unmethylated curve fit resulted in values of μo = 1.34e-9+/−0.05e-9 m2/Vs and α = 1.12e-6+/−0.07e-6. For the methylated curve, it was assumed that the parameters μ0, α and ΔS were unaffected by the addition of the methyl group and the parameters obtained in the unmethylated fit were used to obtain a value for ΔH = 144.62+/−0.04 kcal/mol. Inset: Separation of methylated (6-FAM, green) and unmethylated (Cy5, red) targets by focusing with an applied DC bias at 69°C.

Figure 7. Washing of unmethylated DNA from the gel.

Top two images were taken after an initial focus but before attempts to wash. The bottom two images were taken after washing the unmethylated target from the gel. All images were taken with the same gain and shutter settings which resulted in sensor saturation and some ghost images from reflections off of lenses in the top left image. Note that the dyes have been swapped compared to the image in Figure 7.

Figure 8. Time series of a demonstration of enrichment and extraction of target DNA fragments from a mixture of 20 fmol of 100 nt target (shown in red) and 1 µg of non-target 460 bp dsDNA (shown in green).

The original images in the green and red channels are in gray scale; a–c. The sample mixture is electrophoretically injected from the sample chamber on the bottom left towards to top right. d–g. A focusing field and a time-multiplexed DC bias field is applied to focus the target fragments (red) while washing the background (green) back towards the bottom left. The extraction well in the center is still empty, though condensation on the walls of the well can be observed. h. The extraction well is filled with 12 µl of buffer and only the focusing field is applied. The target fragments enter the buffer in the extraction well. Once focusing is complete, the output buffer can be collected with a pipettor.