Quantification of DNA methylation in electrofluidics chips (Bio-COBRA).

Abstract

Alterations of normal gene expression patterns are a hallmark of human cancers. It is now clear that the dysregulation of epigenetic modifications of the DNA and surrounding histones contributes to aberrant gene silencing, thus being major participants not only in the progression but also the initiation of the disease phenotype. The best-studied epigenetic modification is DNA methylation, which converts cytosine to 5-methylcytosine. Aberrant hypermethylation of the promoter is frequently observed in cancer and is generally associated with gene silencing. Currently, accurate and reproducible quantification of DNA methylation remains challenging. Here, we describe Bio-COBRA, a modified protocol for Combined Bisulfite Restriction Analysis (COBRA), that incorporates an electrophoresis step in microfluidics chips. Microfluidics technology involves the handling of small amounts of liquid in miniaturized systems. In the life sciences, microfluidics usually entails the scaling down of at least one application, such as electrophoresis, to chip format, which often results in increased efficiency and reliability. Bio-COBRA provides a platform for the rapid and quantitative assessment of DNA methylation patterns in large sample sets. Its sensitivity and reproducibility also makes it a tool for the analysis of DNA methylation in clinical samples. The Bio-COBRA assay can be performed on 12 samples in less than 1 h. If the protocol is started at the DNA isolation step, however, approximately 48 h would be required to complete the entire procedure.

Full text

Page 1

52 | VOL.1 NO.1 | 2006 | NATURE PROTOCOLS
PROTOCOL
Quantification of DNA methylation in electrofluidics
chips (Bio-COBRA)
Romulo M Brena1, Herbert Auer2, Karl Kornacker3 & Christoph Plass4
1Department of Molecular Genetics, The Ohio State University, 420 West 12th Ave, Room 435, Columbus, Ohio 43210, USA. 2Columbus Children’s Research Institute,
Columbus Children’s Hospital, 700 Children’s Drive, Columbus, Ohio 43205, USA. 3Division of Sensory Biophysics, The Ohio State University, 484 West 12th Ave, Room
105, Columbus, Ohio 43210, USA. 4Department of Molecular Virology, Immunology and Medical Genetics, The Ohio State University, 420 West 12th Ave, Room 464A,
Columbus, Ohio 43210, USA. Correspondence should be addressed to C.P. (christoph.plass@osumc.edu).
Published online 27 June 2006; doi:10.1038/nprot.2006.8
Alterations of normal gene expression patterns are a hallmark of human cancers. It is now clear that the dysregulation of epigenetic
modifications of the DNA and surrounding histones contributes to aberrant gene silencing, thus being major participants not only in
the progression but also the initiation of the disease phenotype. The best-studied epigenetic modification is DNA methylation, which
converts cytosine to 5-methylcytosine. Aberrant hypermethylation of the promoter is frequently observed in cancer and is generally
associated with gene silencing. Currently, accurate and reproducible quantification of DNA methylation remains challenging. Here, we
describe Bio-COBRA, a modified protocol for Combined Bisulfite Restriction Analysis (COBRA), that incorporates an electrophoresis
step in microfluidics chips. Microfluidics technology involves the handling of small amounts of liquid in miniaturized systems. In the
life sciences, microfluidics usually entails the scaling down of at least one application, such as electrophoresis, to chip format, which
often results in increased efficiency and reliability. Bio-COBRA provides a platform for the rapid and quantitative assessment of DNA
methylation patterns in large sample sets. Its sensitivity and reproducibility also makes it a tool for the analysis of DNA methylation
in clinical samples. The Bio-COBRA assay can be performed on 12 samples in less than 1 h. If the protocol is started at the DNA
isolation step, however, approximately 48 h would be required to complete the entire procedure.
MATERIALS
REAGENTS
• Genomic DNA isolated from individual patients, tissue samples or cell lines,
prepared as described in REAGENT SETUP
• Genomic DNA isolated from peripheral blood lymphocytes (PBLs), prepared
as described in REAGENT SETUP
• SssI methylase (20 U µl–1) (New England Biolabs, cat. no. M0226L)
INTRODUCTION
The development and progression of human cancers is character-
ized by profound changes in cellular function, which stem from
altered patterns of normal gene expression1,2. Over the past decade,
it has become evident that genetic mechanisms account for only a
subset of the alterations that typify the cancer genome3. It has also
become apparent that epigenetic modifications of the genome,
which are heritable modifications to the DNA without the altera-
tion of the primary DNA sequence, are largely dysregulated in the
cancer cell2,4–6. DNA methylation is the best-studied epigenetic
modification, and it takes place predominantly in the context of
5′-CpG-3′ dinucleotides7–9.
There are currently several approaches for evaluating DNA
methylation. Most of these approaches, however, only assess
DNA methylation in a qualitative manner10. Quantitative analysis
of DNA methylation can be achieved using bisulfite DNA sequenc-
ing11. The main drawbacks of this technique are that it is costly and
time consuming. Also, depending on the difference in the frequen-
cy of DNA methylation of the samples to be compared, a very large
number of clones for each sample might have to be sequenced in
order to attain statistically significant results.
We recently described the coupling of Combined Bisulfite
Restriction Analysis (COBRA)12, a well-established method
widely used in DNA-methylation laboratories, with an electro-
phoresis step in microfluidics chips (Agilent 2100 Bioanalyzer,
http://www.agilent.com) for a rapid, accurate and cost-efficient
quantification of methylation patterns in DNA samples of any
origin13. The main strength of our assay is that the DNA meth-
ylation status of all DNA molecules in a PCR product is assessed
for each sample, which circumvents the need for sequencing indi-
vidual clones. Bio-COBRA results are calibrated to a standard
curve in order to extract true DNA methylation percentages. Our
method, however, can be performed without a standard curve,
if the analysis desired is a relative comparison of DNA methyla-
tion levels in a sample set. It should be noted, however, that Bio-
COBRA analysis is limited to DNA sequences that possess at least
one restriction-enzyme site with at least one CpG dinucleotide
in its recognition sequence. Our method also provides a plat-
form for the screening of DNA methylation in genomic regions
where aberrant DNA methylation is known to occur, especially
in genomic areas suspected of having diagnostic or prognostic
value. Furthermore, Bio-COBRA can be used as a discovery tool
to detect novel aberrant DNA methylation sites. Conversely, it is
important to note that only the methylation status of the CpG
dinucleotide or dinucleotides within the recognition sequence of
the restriction enzyme used in the assay are interrogated for their
methylation status.
It should be noted that Agilent Technologies is currently pro-
ducing the Agilent 5100 Automated Lab-on-a-Chip Platform. This
instrument allows investigators to perform Bio-COBRA assays
using 96- or 384-well formats, thus affording Bio-COBRA the
high-throughput capacity needed for the simultaneous screen of
large sample sets.
© 2006 Nature Publishing Group http://www.nature.com/natureprotocols

Page 2

NATURE PROTOCOLS | VOL.1 NO.1 | 2006 | 53
PROTOCOL
PROCEDURE
1| Fragment the genomic DNA using a sonicator at 70% power for at least 2 min. DNA fragments should not be larger than 2 kb.
2| To methylate genomic DNA in vitro, prepare the reaction mix as follows:
Fragmented genomic DNA
15 µg
160 U
3.2 µl
40 µl
to 400 µl
SssI
20 mM SAM
10× enzyme buffer
Water
Incubate the reaction at 37 °C for 4 h. Two hours into the incubation, add an extra 2 µl of SAM and incubate for the remaining 2 h.
3| Recover the DNA from the reaction using QIAquick gel extraction kit columns. Mix the 400-µl reaction with 1,200 µl QG
buffer (all buffers called for in this step are from kit) and 400 µl isopropanol. Vacuum the mix through the column. Wash twice
with 800 µl PB buffer and spin the column at 16,100g for 2 min. Eluate the DNA 200 µl EB buffer.
4| Repeat Steps 2 and 3 once more.
▲ CRITICAL STEP Perform the in vitro methylation reaction on the same genomic DNA twice to ensure complete conversion of
cytosine to 5-methylcytosine. DNA recovery from the columns can be maximized by performing the elution step four times, each
with 50 µl elution buffer.
■ PAUSE POINT The DNA can be stored for up to 6 months at –20 °C.
5| Prepare a 12-point DNA methylation standard as shown in Table 1. Adjust the concentration of the 100% in vitro–
methylated DNA to 20 ng µl–1. Sonicate and adjust the concentration of PBL genomic DNA to 20 ng µl–1. Mix the 100%
in vitro–methylated DNA with the PBL genomic DNA in ratios to obtain the following DNA methylation gradient: 0%, 1.6%,
3.2%, 6.4%, 12.5%, 25%, 50%, 75%, 87.5%, 93.6%, 96.8%, 100%.
▲ CRITICAL STEP The accuracy of the gradient rests first on the precise determination of the DNA concentration of both
the 100% in vitro–methylated DNA and the sheared PBL DNA samples. Using a NanoDrop spectrophotometer can help obtain
accurate optical density readings while minimizing sample loss. A second important consideration is that it is crucial to
carefully pipette the indicated volumes so as to maintain the appropriate ratio of the two reagents.
• 20 mM S-adenosylmethionine (SAM) (New England Biolabs, cat. no. B9003S)
• QIAquick gel extraction kit (Qiagen, cat. no. 28706)
• Platinum Taq DNA polymerase (Invitrogen, cat. no. 10966-034)
• 10 mM dNTP mix (Invitrogen, cat. no. 10297-018)
• Sodium bisulfite (Fisher Scientific, cat. no. S654-500)
• Hydroquinone (Sigma, cat. no. H 9003) ! CAUTION Toxic
• NaOH pellets (Fisher Scientific, cat. no. BP359-212)
• Sodium acetate (Fisher Scientific, cat. no. S93352)
• 10× PCR buffer: 166 mM ammonium sulfate, 670 mM Tris pH 8.8, 67 mM
MgCl2, 100 mM 2-mercaptoethanol
• DNase- and RNase-free water (Invitrogen, cat. no. 10977-015)
• Oligonucleotide primers designed against the target sequence of interest
(10 pmol) (see REAGENT SETUP)
• Restriction enzymes: for example, BstUI (cat. no. R0518L), HpyCH4 IV (cat.
no. R0619L), HhaI (cat. no. R0139L) (New England Biolabs) ▲ CRITICAL Each
restriction enzyme must have a CpG dinucleotide in its recognition sequence
suitable for the analysis of the sequence of interest.
• 29% acrylamide (wt/vol) mixed with 1% N, N’-methylene-bisacrylamide
(wt/vol) solution ! CAUTION Toxic; avoid contact with the skin. Use a respirator
when handling acrylamide and/or bisacrylamide in powder form.
• Ethidium bromide, 0.1% wt/vol (Sigma, cat. no. E 8751) ! CAUTION
Carcinogen
• 10× Tris/borate/EDTA (TBE): 0.90 M Tris, 0.90 M boric acid, 10 mM EDTA
• DNA 500 LabChip kit (Agilent Technologies, cat. no. 5064-8284)
EQUIPMENT
• GeneAmp PCR System 9700 (Applied Biosystems) or any thermal cycler
• ND 1000 NanoDrop spectrophotometer (NanoDrop Technologies)
or any spectrophotometer Agilent 2100 Bioanalyzer (Agilent Technologies)
• Concentrator 5301 (Eppendorf) or any vacuum dryer
• Sonicator
• Hybridization oven
• Water bath
• Electrophoresis power source
• Electrophoresis chamber
• Alphaimager UV transilluminator (Alpha Innotech) or any gel-imaging system
REAGENT SETUP
DNA isolation Freeze in liquid nitrogen ~100 mg human tissue with 1,000 µl
lysis buffer (10 mM Tris pH 8.0, 25 mM EDTA pH 8.0, 100 mM NaCl, 0.5%
wt/vol SDS). Crush the frozen tissue to a fine powder with a mortar and pestle.
Add 10 µl proteinase K (10 mg ml–1) to the crushed tissue. Incubate the crushed
tissue at 60 °C for at least 1 h. Extract the DNA using a mixture of phenol and
chloroform. Dialyze the DNA sample against 4 liters 10 mM Tris, pH 8.0, to
remove all traces of phenol. Precipitate the DNA using cold (–20 °C)100%
ethanol. As the genomic DNA samples isolated for Bio-COBRA are treated with
bisulfite, the isolation of high-molecular-weight DNA is not required. Thus,
commercially available kits, such as DNeasy (Qiagen) can be used to rapidly
obtain low-molecular-weight DNA.
Oligonucleotide primers Design oligonucleotide primers to amplify the target
sequence of interest. ▲ CRITICAL To reduce the PCR bias toward the preferential
amplification of either methylated or unmethylated sequences, primer binding
sites should not contain CpG dinucleotides. Free online software to aid in
the design of primers for DNA methylation analysis is available (http://www.
urogene.org/methprimer/index1.html14 and http://bisearch.enzim.hu)15.
© 2006 Nature Publishing Group http://www.nature.com/natureprotocols

Page 3

54 | VOL.1 NO.1 | 2006 | NATURE PROTOCOLS
PROTOCOL
? TROUBLESHOOTING
■ PAUSE POINT The DNA can be stored for up to 6 months at –20 °C.
6| Prepare a 3 M NaOH solution by dissolving 1.20 g of NaOH pellets in 10 ml of double-distilled water. To bisulfite convert
and purify the DNA methylation standard, first add 5 µl 3 M NaOH to each sample of the DNA methylation standard (50 µl
volume). Incubate the samples at 42 °C for 30 min. After the incubation, add the following to each sample:
Denatured DNA standard sample
55 µl
515 µl
30 µl
600 µl
3 M sodium bisulfite, pH 5.0
10 mM hydroquinone
Final volume
Mix all reagents, wrap the samples in foil (to exclude light) and incubate at 50 °C for 16 h in a hybridization oven.
7| Purify samples using a QIAquick gel extraction kit. To do so, add the following to each sample:
Bisulfite-treated sample
600 µl
1,800 µl
600 µl
3,000 µl
QG buffer
Isopropanol
Final volume
Mix all reagents well and vacuum through a spin column. Wash twice with 800 µl PB buffer and spin the column at 16,100g for
2 min. Eluate the DNA 50 µl EB buffer.
8| To complete the bisulfite treatment, add 5 µl 3 M NaOH to each sample (50 µl volume) and incubate at 37 °C for 30 min.
After the incubation, add 10 µl 5 M sodium acetate to each sample.
9| Purify the samples using a QIAquick gel extraction kit by adding the following:
Bisulfite-treated sample
65 µl
195 µl
65 µl
325 µl
QG buffer
Isopropanol
Final volume
Vacuum the mix through the spin column. Wash twice with 800 µl PB buffer and spin the column at 16,100g for 2 min. Eluate
the DNA 300 µl EB buffer.
▲ CRITICAL STEP All reagent solutions needed for bisulfite treating DNA (3 M sodium bisulfite, pH 5.0, and 10 mM
hydroquinone) should be prepared fresh every time they are needed. Prevent their exposure to light by wrapping them in foil.
Add 100–200 µl 10 M NaOH per 10 ml 3 M sodium bisulfite solution to raise its pH to 5.
■ PAUSE POINT The bisulfite-treated DNA can be stored for up to 1 year at –20 °C.
10| Amplify by PCR the bisulfite-treated DNA methylation standard and samples of interest according to the reaction
mix below:
TABLE 1 | Example of a 12-point DNA methylation standard created by mixing 0% and 100% methylated DNA.
DNA methylation percent
In vitro treated DNA (µl)
Sonicated PBL DNA (µl)
Total volume (µl)
DNA amount (µg)
0 1.6 3.2 6.412.5 2550 75 87.5 93.6 96.8100
— 0.8 1.63.26.2512.5 25 37.543.7546.8 48.450
5049.248.446.843.7537.5 2512.56.25 3.2 1.6
—
5050 5050 5050 505050 5050 50
1.01.01.0 1.01.0 1.0 1.01.01.01.0 1.01.0
© 2006 Nature Publishing Group http://www.nature.com/natureprotocols

Page 4

NATURE PROTOCOLS | VOL.1 NO.1 | 2006 | 55
PROTOCOL
Bisulfite-treated DNA
10× PCR buffer
10 mM dNTP mix
10 µl
5 µl
1 µl
1 U
2 µl
Platinum Taq polymerase
Oligonucleotide mix (10 pmol each)
DNase- and RNase-free water
up to 50 µl
Place the PCR reactions in a thermal cycler and activate the polymerase by incubating at 95 °C for 10 min. Carry out
the amplification reaction for 35 cycles, using 96 °C as the denaturing temperature in each cycle: e.g., 95 °C for 10 min
(96 °C for 30 s, 60 °C for 30 s, 72 °C for 30 s) × 35 cycles, 72 °C for 10 min.
▲ CRITICAL STEP Dimethylsulfoxide (DMSO) may be added to the PCR reactions to help amplify the target sequence. However,
high DMSO concentrations (>5%) can inhibit the DNA polymerase. Thus, consider adding more units of enzyme if high DMSO
concentrations are needed for your specific PCR reaction to work.
? TROUBLESHOOTING
■ PAUSE POINT The PCR products can be stored at 4 °C for up to 1 week.
11| Check the PCR amplification. Run 5 µl of each reaction in an 8% acrylamide gel. Run the gel for at least 1 h at 250V.
Visualize the PCR products by staining the gel with a 0.001% ethidium bromide solution.
12| Clean the remaining PCR product (45 µl) using a QIAquick gel extraction kit by adding the following to each sample:
PCR product 45 µl
QG buffer 135 µl
Isopropanol45 µl
Final volume225 µl
Vacuum the mix through the spin column. Wash twice with 800 µl PB buffer and spin the column at 16,100g for 2 min. Eluate
the DNA 50 µl EB buffer.
■ PAUSE POINT The purified PCR products can be stored at
4 °C for up to 1 week.
13| Concentrate the eluted PCR product to a final volume
of 7 µl. Place the samples in a Concentrator 5301 heated to
60 °C. It takes approximately 15 min for the samples to reach
the desired volume.
▲ CRITICAL STEP Concentrating the PCR samples is required
in order to achieve a DNA concentration within the dynamic
range of the Agilent DNA 500 LabChip chemistry (5–50
ng µl–1). This step is necessary if the expected restriction
pattern (Step 10 and on) is made of multiple (four or more)
restriction fragments.
14| Digest the PCR products with the appropriate restriction
enzyme, which is selected based on the restriction sites
present in the target sequence. Mix all reagents indicated
below well and incubate at the appropriate temperature for at
least 2 h.
PCR product 7 µl
Enzyme
10× buffer
Water
10 U
1 µl
up to 10 µl
Figure 1 | Typical electropherogram generated by plotting Bioanalyzer
raw data (CSV files) into Excel. The axis in the plot indicates fluorescence
intensity (y) and migration time (x). Each peak in the electropherogram
represents a DNA fragment. As in conventional DNA electrophoresis, DNA
fragments in this system migrate according to their molecular weight.
Because smaller fragments are detected first, their fluorescence intensities
are recorded the earliest in the plot (39 s in the example provided). Peak
A is the result of the signal generated by the undigested portion of the
PCR product. Peaks B, C and D each represent a DNA fragment product
of the restriction digestion carried out on the PCR product. The percent
methylation of the sample shown is calculated by adding the fluorescence
values of peaks B, C and D, divided by the addition of the fluorescence
values of peaks A, B, C and D: [(B + C + D) / (A + B + C + D)]. The
fluorescence units for each peak are calculated by measuring the height of
each peak (examples A and B).
0
15
30
45
60
75
90
39 424548 51 54 57
A: 91.32
B
A
Fluorescence units (FU)
Time (s)
B: 30.02C
D
© 2006 Nature Publishing Group http://www.nature.com/natureprotocols

Page 5

56 | VOL.1 NO.1 | 2006 | NATURE PROTOCOLS
PROTOCOL
■ PAUSE POINT The digested PCR products can be stored at
4 °C for up to 1 week.
15| Check the restriction digestion. Run 5 µl of each reaction
in an 8% acrylamide gel. Run the gel for at least 1.5 h at
250 V. Consider using a loading dye, such as YellowSub, that
will not migrate over the expected restriction fragments.
Visualize the restriction digestion by staining the gel with a
0.001% ethidium bromide solution. Check to ensure complete
digestion was obtained in the 100% methylated sample, and
that no restriction fragments are observed in the 0% sample
(PBL control).
▲ CRITICAL STEP Incomplete digestion in the 100%
methylated sample may be attributed to either incomplete in vitro methylation or incomplete enzymatic digestion. Consider
checking the in vitro methylation efficiency by PCR amplifying and digesting an aliquot of the SssI treated DNA before mixing
the DNA methylation standard. Restriction digestion in the 0% sample could result from either basal DNA methylation in PBL
DNA or incomplete bisulfite conversion.
? TROUBLESHOOTING
16| Follow exactly the protocol provided by Agilent for the DNA 500 LabChip. Load 1 µl of the restriction digestion into a DNA
500 LabChip well. Perform the electrophoresis step by running the Agilent 2100 Bioanalyzer
▲ CRITICAL STEP Gel and dye must be equilibrated to room temperature before mixing and loading. Reset plunger of Chip
Priming Station to 1 ml before opening station. This avoids spills to the seal, which could cause leakage of the station.
17| To maximize the efficiency of the assay, it is recommended to work with fully loaded chips (12 samples per run).
? TROUBLESHOOTING
18| Export the raw data (CSV file) for each of the samples and plot the raw data from each sample (fluorescence and migration
time) in Microsoft Excel. The Agilent 2100 Bioanalyzer includes a software package that can be used to determine the peak area
of each of the peaks generated during the electrophoretic run for each of the samples. However, we strongly encourage the use
of peak height, instead of peak area, for calculating DNA methylation percentages10.
? TROUBLESHOOTING
Figure 2 | Eight percent acrylamide gel, virtual gel and electropherogram
visualization of a restriction-digested, 12.5% methylated sample.
(a) Restriction digestion of a 12.5% methylated sample separated by
electrophoresis in an 8% acrylamide gel and subsequently stained with
ethidium bromide (left lane). Virtual gel for the same sample automatically
generated by the Bioanalyzer during the sample run (right lane) and the
corresponding electropherogram (plot). It should be noted that low-
intensity bands are visible in the acrylamide gel but are not present in the
corresponding virtual gel. Nevertheless, all visible bands in the acrylamide
gel can be matched to a peak in the electropherogram. The numbers to
the left and the right of the gel lanes correspond to the peaks in the
electropherogram. It should also be noted that in this example, a 17-bp
fragment (indicated with the number 8) comigrates with the front marker
(FM). The FM is a 15-bp DNA fragment included in all Bioanalyzer runs to
calibrate the system. The overlapping of the 17-bp fragment with the front
marker makes the 17-bp fragment unquantifiable. However, the elimination
of this fragment from the quantification process would affect all samples
equally, thus reducing the bias created by the elimination of one digestion
product. Also, the low fluorescence intrinsically emitted from such a short
fragment is, in most cases, only a small fraction of the total fluorescence
generated by all the restriction fragments in the sample. The reason that
the front marker peak measures 100 fluorescence units is because 2.55 ×
1011 15-bp DNA molecules comprise that peak. (b) Same set up as in a, but
of a 50% methylated sample. (c) Same set up as in a and b, but of a 96.5%
methylated sample.
0
20
40
60
80
100
4348 5358 636873
1
2
3
4
5
6
7
8 & FM
0
20
40
60
80
100
42 475257 626772
1
2
3
4
5
6
7
8 & FM
0
20
40
60
80
100
120
4045 5055 60 6570
3
4
5
6
7
8 & FM
1
2
3
4
5
6
7
8
1
4
6
7
8 & FM
1
2
3
4
5
6
7
8
1
4
6
7
8 & FM
3
4
5
6
7
8
4
6
7
8 & FM
a
b
c
© 2006 Nature Publishing Group http://www.nature.com/natureprotocols

Page 6

NATURE PROTOCOLS | VOL.1 NO.1 | 2006 | 57
PROTOCOL
19| Measure the peak height for each of the restriction fragments and calculate the total DNA methylation level by using the
following formula: fluorescence of methylated products/(fluorescence of methylated products + fluorescence of unmethylated
product); see Figure 1. To help determine which peak corresponds to which restriction fragment, it is useful to generate a plot
of the DNA size marker that is included in the chip with each experiment. The fragment size for each of the DNA markers is
available in the Agilent DNA 500 LabChip chemistry product sheet. Truncated peaks, or lack of high-molecular-weight peaks,
indicate the early titration of the fluorochrome used in the DNA 500 chemistry, probably as a result of DNA concentrations
well above the dynamic range of the electrophoresis system (tested up to 65 ng µl–1). If these occur, consider performing the
restriction reaction in a larger volume.
? TROUBLESHOOTING
20| Compare the DNA methylation percentages obtained for each sample with the gel image generated in Step 17.
21| Plot the 12 points of the DNA methylation standard in Microsoft Excel and generate a model. Use the equation from that
model to derive the true methylation percentages for the experimental samples. Up to 12 Bio-COBRA samples can be processed
in the 2100 Agilent Bioanalyzer in less than 1 h. However, if the procedure were to be started from the DNA isolation step, it
would take approximately 2 d to complete the entire protocol.
? TROUBLESHOOTING
See Table 2.
ANTICIPATED RESULTS
Electrophoresis of digested PCR products generated from PCR reactions with average efficiency (~10–30 ng µl–1) yield
fluorescence peaks well above background. The function derived from the 12 points in the DNA methylation standard is usually
linear or logarithmic, depending on the difference of PCR amplification efficiency between the methylated and unmethylated
TABLE 2 | Troubleshooting table.
PROBLEM SOLUTION
Step 5
PBL DNA exhibits low-
level methylation in the
sequence of interest
One of the main advantages of Bio-COBRA is that DNA methylation percentages can be generated for each
sample. If PBL DNA shows partial methylation at the sequence of interest, the analysis of a PBL sample will
provide the baseline DNA methylation for that sequence. Thus, subtracting the background methylation level
of PBL DNA from the unknown samples would yield the true increase in DNA methylation in those samples.
Alternatively, DNA extracted from a different tissue in which no background methylation is present in the
sequence of interest can be used instead of PBL DNA.
Try using DMSO at various concentrations (1–5%) with concurrent changes in annealing temperature. It might
also be useful to reduce the time in the extension step of the PCR cycle to 10–15 s.
Step 11
Multiple bands in PCR
amplification
Step 11
The PCR reaction
generates primer dimers
Reduce the original amount of primer used per reaction by 50%. If primer dimers cannot be eliminated,
accurate quantification of DNA methylation levels may still be possible. Calculate the size of all expected
digestion products. If the primer dimer size does not overlap with any of the expected digestion fragments, the
primer dimer fluorescence peak is simply ignored in the electropherogram analysis.
Treat 100 ng SssI-treated DNA with bisulfite, amplify by PCR the CpG island of a housekeeping gene and
directly sequence the PCR product. If C and T peaks overlap at any CpG dinucleotides, the in vitro methylation
reaction was not complete. If only C peaks are detected at all CpG locations, the enzymatic digestion was not
complete.
The DNA 500 Lab-on-a-Chip can accommodate 12 samples per run. Each chip can only be used once; thus, it is
recommended that samples are processed, whenever possible, in multiples of 12 as a way to maximize the cost
efficiency of the assay.
For the purpose of the assay, accurate quantification of DNA methylation levels can be best achieved by
measuring the peak height of each of the signals in the electropherogram13. To calculate the peak height of
a signal, simply place the mouse arrow at the top of the peak in the electropherogram (graphed in Microsoft
Excel).
During separation of nucleic acids by electrophoresis in a DNA 500 Lab-on-a-Chip, DNA fragments migrate
according to size. Thus small DNA fragments are detected before larger fragments. In the electropherogram,
the x-axis represents time in seconds, increasing from left to right. Thus, peaks toward the left of the
electropherogram are generated by the smaller DNA fragments that resulted from the restriction digestion.
Conversely, larger fragments migrate further to the right.
Step 16
Incomplete digestion
of the 100% in vitro-
methylated standard
Step 17
Inefficient processing of
samples in the Bioanalyzer
Step 18
Peak height versus peak
areas to calculate DNA
methylation percentages
Step 19
Interpreting the
electropherogram
generated by the
Bioanalyzer
© 2006 Nature Publishing Group http://www.nature.com/natureprotocols

Page 7

58 | VOL.1 NO.1 | 2006 | NATURE PROTOCOLS
PROTOCOL
COMPETING INTERESTS STATEMENT The authors declare that they have no
competing financial interests.
Published online at http://www.natureprotocols.com
Reprints and permissions information is available online at http://npg.nature.
com/reprintsandpermissions
1. Baylin, S.B. et al. Abnormal patterns of DNA methylation in human neoplasia:
potential consequences for tumor progression. Cancer Cells 3, 383–390 (1991).
2. Esteller, M. Cancer epigenetics: DNA methylation and chromatin alterations in
human cancer. Adv. Exp. Med. Biol. 532, 39–49 (2003).
3. Costello, J.F. et al. Aberrant CpG-island methylation has non-random and
tumour-type-specific patterns. Nat. Genet. 24, 132–138 (2000).
4. Jones, P.A. & Laird, P.W. Cancer epigenetics comes of age. Nat. Genet. 21,
163–167 (1999).
5. Jones, P.A. & Baylin, S.B. The fundamental role of epigenetic events in
cancer. Nat. Rev. Genet. 3, 415–428 (2002).
6. Fischle, W., Wang, Y. & Allis, C.D. Histone and chromatin cross-talk. Curr.
Opin. Cell Biol. 15, 172–183 (2003).
7. Clark, S.J., Harrison, J. & Frommer, M. CpNpG methylation in mammalian
cells. Nat. Genet. 10, 20–27 (1995).
8. Franchina, M. & Kay, P.H. Evidence that cytosine residues within 5’-CCTGG-3’
pentanucleotides can be methylated in human DNA independently of the
methylating system that modifies 5’-CG-3’ dinucleotides. DNA Cell Biol. 19,
521–526 (2000).
9. Jabbari, K. & Bernardi, G. Cytosine methylation and CpG, TpG (CpA) and TpA
frequencies. Gene 333, 143–149 (2004).
10. Brena, R.M., Huang, T.H. & Plass, C. Quantitative assessment of DNA
methylation: potential applications for disease diagnosis, classification, and
prognosis in clinical settings. J. Mol. Med. 84, 1–13 (2006).
11. Clark, S.J., Harrison, J., Paul, C.L. & Frommer, M. High sensitivity mapping of
methylated cytosines. Nucleic Acids Res. 22, 2990–2997 (1994).
12. Xiong, Z. & Laird, P.W. COBRA: a sensitive and quantitative DNA methylation
assay. Nucleic Acids Res. 25, 2532–2534 (1997).
13. Brena, R.M. et al. Accurate quantification of DNA methylation using
combined bisulfite restriction analysis coupled with the Agilent 2100
Bioanalyzer platform. Nucleic Acids Res. 34, e17 (2006).
14. Li, L.C. & Dahiya, R. MethPrimer: designing primers for methylation PCRs.
Bioinformatics 18, 1427–1431 (2002).
15. Tusnady, G.E., Simon, I., Varadi, A. & Aranyi, T. BiSearch: primer-design and
search tool for PCR on bisulfite-treated genomes. Nucleic Acids Res. 33, e9
(2005).
molecules. A comparison between an 8% acrylamide gel, a virtual gel generated by the Bioanalyzer and the corresponding
electropherogram is shown in Figure 2.
© 2006 Nature Publishing Group http://www.nature.com/natureprotocols