Protocol

Terminal Transferase-Dependent PCR (TDPCR) for In Vivo UV Photofootprinting of Vertebrate Cells

See allHide authors and affiliations

Science's STKE  10 Apr 2001:
Vol. 2001, Issue 77, pp. pl1
DOI: 10.1126/stke.2001.77.pl1

Abstract

Terminal transferase-dependent PCR (TDPCR) is a versatile, sensitive method for detecting DNA lesions such as those generated by the footprinting agents commonly used to detect in vivo protein-DNA interactions. Data similar to those obtained by ligation-mediated PCR (LMPCR) are obtained, but one advantage of TDPCR is that no special enzymes are needed other than terminal deoxynucleotide transferase, T4 DNA ligase, and thermostable DNA polymerases. A detailed TDPCR protocol is given for using UV photofootprinting to detect in vivo footprints and chromatin fine structure in vertebrate cells. One version of the protocol makes use of nonradioactive labeling by near-infrared fluorochromes and detection by a LI-COR DNA sequencing instrument. Sensitivity similar to that of 32P-labeling is obtained, but with superior band resolution and quantitation.

Introduction

Gene function depends on chromatin structure and protein-DNA interactions, which are among the most fundamental forms of macromolecular recognition that take place in a living cell regardless of its state of growth, division, differentiation, quiescence, or senescence. Specific and nonspecific DNA-protein interactions underlie the packaging of DNA into chromatin, and many signals from the cellular environment terminate at sequence-specific trans-acting factors. These, in turn, often initiate chromatin-remodeling events leading to changes in gene expression. The elucidation of which trans-acting factors bind to their cognate recognition sequences and how they change chromatin structure is therefore of utmost importance for our understanding of the differential regulation of gene expression. One of the few methods available to study in vivo chromatin and transcription factor binding is in vivo footprinting. For this purpose, several in vivo footprinting agents, such as dimethylsulfate (DMS), KMnO4, DNase I, micrococcal nuclease (MNase), irradiation with ultraviolet light at 254 nm (UVC), and H2O2 have been reported (1-4) for use with several genomic sequencing techniques (5-8). Among these techniques, ligation-mediated PCR (LMPCR) has been the most sensitive and widely used method for quantitatively detecting DNA lesions and DNA-protein interactions in mammalian cells at single nucleotide resolution (9-11). UV irradiation is emerging as the simplest and least perturbing method to examine changes in DNA structure generated by proteins in vivo (12-15). UVC irradiation generates predominantly UV dimers at dipyrimidines. The frequency with which this occurs is a function of DNA structure, which is often affected by bound proteins. To make these lesions detectable by LMPCR, the DNA must contain single-strand breaks with phosphate groups at their 5′ termini. Hence, to reveal UV dimers by use of LMPCR, the DNA isolated after UVC irradiation has to be further modified to generate breaks.

As an alternative method by which to reveal DNA lesions such as those generated by UV, Komura and Riggs (16) introduced a terminal transferase-dependent PCR (TDPCR) method (Fig. 1), which can be used for the detection of DNA adducts and in vivo footprints in eukaryotic cells (16, 17, 24-28). The protocol given here is a modified version of TDPCR used to determine changes in transcription factor and chromatin fine structure profiles during myeloid cell differentiation (17). Noteworthy in the protocol given here and in Kontaraki et al. (17) is the use of streptavidin-coupled magnetic beads (1, 26) and the optional use of nonradioactive labeling with near-infrared fluorochromes and a LI-COR DNA sequencing system, which provide detection sensitivity close to that of 32P-labeling (4, 17, 18). TDPCR has become a relatively quick, versatile, and sensitive method for the detection of in vivo protein-DNA interactions (4, 17, 28), as well as protein-RNA interactions (19-22) in mammalian cells.

Fig. 1.

Schematic outline of the TDPCR procedure.

Figure 1 outlines the TDPCR procedure. Any DNA containing single-strand breaks or adducts that pause or stop DNA polymerase can be used as the template for multiple cycles of primer extension using a gene-specific, biotinylated primer. The biotin-containing, primer-extended molecules are captured on streptavidin-coupled magnetic beads, and the rest of the primer extension components, including the original DNA template molecules, are removed by washing with buffer after denaturation by NaOH. The captured DNA molecules are 3′-tailed using terminal deoxynucleotidyl transferase (TdT) under conditions that result in the addition, on average, of three riboguanosine triphosphates (riboGTPs). The tailed molecules are then ligated to a double-stranded linker containing a three-deoxycytidine overhang at the 3′ end of the longer oligonucleotide. All fragments are PCR amplified using a second, nested, gene-specific primer and a linker-specific primer, which is complementary to the lower primer of the linker. The amplified DNA fragments can be visualized either by hybridization with a single-stranded probe (1, 16) or by primer extension after the PCR step (1, 4, 17).

Figure 2 shows two examples of UV photofootprinting. Figure 2A shows a UV photofootpinting study of the well-characterized human phosphoglycerate kinase (PGK) promoter using the procedures described here. All of the characteristic footprints previously reported (10, 11, 13) were clearly reproduced, even when the ligation time was reduced to two hours, which allows the procedure to be done in one day. Figure 2B is an example of a photofootprinting experiment examining the chicken lysozyme -3.9 kb enhancer element by TDPCR (17). TDPCR products were visualized with a fluorescent primer and electrophoresis on a LI-COR DNA sequencing system. The method clearly revealed a number of subtle changes in UV dimer formation in lysozyme-expressing cells around the transcription factor binding sites (17, 23). About 140 bp are displayed in both Figure 2A and Figure 2B. Using standard gel electrophoresis and 32P autoradiography for detection, about 200 nucleotides (nt) of good quality data are usually obtained. Using the LI-COR system with fluorescent detection, at least 300 nt of good quality data are usually obtained. The most important advantage of using the LI-COR gel system is its equally spaced, high-quality data output, allowing accurate quantification of each band, and thus, generation of high-resolution chromatin fine structure profiles (17).

Fig. 2.

UV Photofootprinting. (A) Human PGK promoter. Results obtained with two ligation times are shown. The enzymes used were Vent (exo-) for primer extension and AmpliTaq for PCR. Control, in vivo, and in vitro UVC-treated DNAs were isolated from the Y162-11C cell line (10, 11, 13). Lanes 1 to 9 represent results with 2 hours of ligation and lanes 10 to 18 show results with overnight ligation. Lanes 1 to 3 and 10 to 12 are control DNAs (c) ; lanes 4 to 6 and 13 to 15 are in vitro UVC-treated DNAs (t); lanes 7 to 9 and 16 to 18 are in vivo UVC-treated DNA (v). The positions relative to the transcriptional start site in the human PGK promoter are indicated at the right. (B) In vivo UV footprinting of the -3.9 kb enhancer region of the chicken lysozyme locus. The chicken cell lines used, HD37, MEP (HD50 MEP), and HD11, are shown on the top. Naked DNA (gDNA, lane 1) was compared with DNA from the chicken cell lines (lanes 2-7). TDPCR was done with primer set D376 (17), which amplifies the upper strand. (+) Cells grown in the presence of 100 nM Trichostatin (TSA; Sigma). (0) Diminished UV dimer formation as compared to naked DNA; (·) enhanced reactivity, with position relative to the transcriptional start site indicated at the right.

Materials

Reagents and Chemicals

32P-γATP and 32P-αdCTP (NEN Life Science Products, Boston, MA)

Phenol (Sigma, St. Louis, MO)

Chloroform (Mallinckrodt Baker, Phillipsburg, NJ)

Isopropanol (Mallinckrodt Baker)

Ethanol

Mineral oil, white, light (Mallinckrodt Baker, Phillipsburg, NJ)

10% sodium dodecyl sulfate (SDS) (BRL, Grand Island, NY)

Dynabeads M-280 streptavidin (Dynal, Lake Success, NY)

SequaGel solutions (National Diagnostics, Atlanta, GA)

100 mM GTP (Riboguanosine triphosphate), 100 mM ATP (Riboadenosine triphosphate), and 100 mM dNTP set (BMB/Roche, Indianapolis, IN)

Dithiothreitol (DTT) (Sigma)

TEMED (Sigma)

Ammonium persulfate (BMB/Roche)

Xylene cyanol (Sigma)

Bromophenol blue (Sigma)

LI-COR IR2 stop solution and LI-COR molecular weight STR markers (50 bp to 350 bp, and 50 bp to 700 bp, either IRD-700 or IRD-800 labeled) (LI-COR, Lincoln, NE)

25-bp and 50-bp DNA ladders (BRL)

AG 501-X8 (D) resin, 20 to 50 mesh (Bio-Rad, Hercules, CA)

PBS (Irvine Scientific, Irvine, CA)

Betaine (Sigma)

BSA (10 mg/ml) (New England Biolabs)

Disposables

150-mm culture plates (Corning, Corning, NY)

50-ml centrifuge tubes (Falcon, Franklin Lakes, NY)

Standard media (BRL or other sources such as Irvine Scientific) suitable to your cell lines

0.65 ml and 1.7 ml siliconized, DNase-free, RNase-free thin-walled microcentrifuge tubes (Sorenson Bioscience, Salt Lake City, UT)

Flat pipette tips (multiFlex, 0.4 mm, Sorenson Bioscience)

G-25 columns (pre-packed, Eppendorf, Boulder, CO)

3 mm Whatman paper

Saran wrap

Thermostable DNA Polymerases

Vent (exo-) (recombinant exonuclease minus) DNA polymerase (2 units/µl) with included 10× buffer and 100 mM MgSO4 solution (New England Biolabs, Beverly, MA)

Vent DNA polymerase (2 units/µl) with included 10× buffer and 100 mM MgSO4 solution (New England Biolabs)

AmpliTaq (5 units/µl) with included 10× buffer (Perkin Elmer/Applied Biosystems, Foster City, CA)

Pfu DNA polymerase with included 10× buffer (Stratagene, San Diego, CA)

Pfu Turbo (2 units/µl) with included 10× buffer (Stratagene)

Expand Long PCR System (3.5 units/µl) with 3 kinds of 10× buffer (BMB/Roche)

Other Enzymes

DNase-free RNase A (100 mg/ml) (Qiagen, Valencia, CA)

Proteinase K (BMB/Roche, Indianapolis, IN)

Terminal deoxynucleotidyl transferase (TdT) (15 units/µl) with included 5× TdT buffer (BRL)

T4 DNA ligase (3 units/µl) (Promega, Madison, WI)

T4 DNA kinase (10 units/µl) with included 10× kinase buffer (New England Biolabs)

Escherichia coli Exonuclease I (10 units/µl) (Amersham, Arlington Heights, IL)

T4 DNA polymerase (3 units/µl) with included 10× buffer (New England Biolabs or BRL)

Equipment

Magnetic Particle Concentrator (MPC, Dynal, Lake Success, NY)

UV Stratalinker 2400 Oven with 254 nm UV (UVC) Bulbs (Stratagene San Diego, CA)

Hamilton 8-channel syringe (Hamilton Company, Reno, NV)

PhosphorImager 425S (Molecular Dynamics, Sunnyvale, CA)

Thermocyclers

Any good thermocycler can be used. We use:

PTC-100 Programmable Thermal Controller (MJ Research, San Francisco, CA)

Robocycler Gradient 96 Temperature Cycler with Hot Top (Stratagene, San Diego, CA)

Sequencing Gel Apparatus and Power Supplies

For 32P-ATP labeled samples, we use an electrophoresis Model SA system (BRL) and a power supply (6000 volts) suitable for sequencing gels.

For near-infrared fluorescent IRD-700 or IRD-800 labeled samples, we use the IR2 Long Ranger 4200 system (LI-COR, Lincoln, NE), which has its own power supply.

Software

Gene ImagIR (Scanalytics, Billerica, MA)

Adobe PhotoShop (Adobe Systems, San Jose, CA)

Oligo 4 or Oligo 5.1 software (National Biosciences, Plymouth, MN)

Recipes

Recipe 1: Stock Solutions
5 M NaCl
5 M NaOH
1 M Tris-HCl, pH 8.0
1 M Tris-HCl, pH 7.5
1 M MgCl2
1 M DTT (store at -20°C)
0.5 M EDTA, pH 8.0
Each of the above solutions should be prepared in dH2O and can be stored at room temperature, unless indicated otherwise.
Recipe 2: Cell Lysis Buffer B
Stock Amount Final Concentration
5 M NaCl 30 ml 150 mM NaCl
0.5 M EDTA (pH 8.0) 10 ml 5 mM EDTA
Add dH2O to 1 L and store at 4°C.
Recipe 3: Cell Lysis Buffer C
Stock Amount Final Concentration
1.0 M Tris-HCl, pH 8.0 20 ml 20 mM
5 M NaCl 4 ml 20 mM
0.5 M EDTA, pH 8.0 40 ml 20 mM
Add dH2O to 1 L and store at 4°C.
Take an appropriate amount and add SDS to a final concentration of 1% and Proteinase K to a final concentration of 600 µg/ml just before use.
Recipe 4: Dynal Bead Binding and Wash Buffer (2× BW)
Stock Amount Final Concentration
NaCl (solid) 58.5 gm 2 M
1 M Tris-HCl, pH 7.5 5 ml 10 mM
0.5 M EDTA, pH8.0 1 ml 1 mM
Add dH20 to 500 ml.
Recipe 5: 10× TBE Buffer
Amount per liter of dH2O
Tris base 121.1 g
Boric acid 55.8 g
EDTA (disodium salt) 7.2 g
pH should be at least 8.3.
Recipe 6: TE
TE, pH 7.5: Prepare a 10-mM Tris and 1-mM EDTA solution using 1 M Tris-HCl (pH 7.5) and 0.5 M EDTA (pH 8.0).
TE, pH 8.0: Prepare a 10-mM Tris and 1-mM EDTA solution using 1 M Tris-HCl (pH 8.0) and 0.5 M EDTA (pH 8.0).
Recipe 7: 0.1× TE
Prepare a 0.1× TE, pH 8.0 solution and a 0.1× TE, pH 7.5 solution by diluting the proper TE solutions (Recipe 6) with dH2O.
Recipe 8: Formamide Loading Solution
Formamide 95% (v/v)
EDTA 20 mM, pH 8.0
Xylene cyanol 0.05% (w/v)
Bromphenol blue 0.05% (w/v)
Store at -20°C.
Recipe 9: Kinase Mix for TDPCR Linkers
Component Volume (µl) per sample
H2O 292
10× kinase buffer 40
100 mM ATP 4
200 µM lower primer (LP24*C5) 44
T4 DNA kinase (10 units/µl) 20
Final volume is 400 µl.
Recipe 10: Extension Mix
Component Volume (µl) per sample
H2O X
10× Vent buffer 3
100 mM MgSO4 1.2
25 mM dNTP mix (mix 1 part each of 100 mM dNTPs) 0.3
20 µM Bio-P1 (custom-made) 0.1
Mix thoroughly at this point before adding the enzyme.
Vent (exo-) (2 units/µl) 1.0
Total volume 20 µl
Note: 10× Vent Buffer gives a final [Mg+2] of 2 mM. The optimal Mg+2 concentration for Vent and Vent (exo-) is often 6 mM. Performing a [Mg+2] titration curve is recommended. In some cases, Vent (exo-) can be substituted with a Vent (exo-):Vent mix (the ratio should be determined experimentally). Normally, 16:1 to 19:1 Vent (exo-):Vent works well. Vent can also be replaced with Pfu DNA polymerase and Pfu Turbo from Stratagene and Expand Long from BMB; in this case, the buffer and amount of each component should be changed according to each manufacturer's recommendation. Always check the total volume with a pipet after mixing gently and performing a pulse at maximum speed in a microcentrifuge to bring the liquid to the bottom of the tube.
Recipe 11: TdT Mix
Component Volume (µl) per sample
H2O 4.93
5× TdT buffer 4.0
100 mM riboguanosine triphosphate 0.4
Mix thoroughly at this point before adding the enzyme.
TdT (15 units/µl) 0.67
Total volume 10 µl
Recipe 12: Ligation Mix
Component Volume (µl) per sample
H2O 7.95
1 M Tris-HCl, pH 7.5 1.5
1 M MgCl2 0.3
1 M DTT 0.3
100 mM ATP 0.3
10 mg/ml BSA 0.15
20 µM TDPCR annealed linker 3.0
Mix thoroughly at this point before adding the enzyme.
T4 DNA ligase (3 units/µl)1.5
Total volume 15 µl
Recipe 13: PCR Mix
Component Volume (µl) per sample
H2O appropriate volume to make a total volume of 40 µl
5× Taq buffer 10
25 mM MgCl2 3 - 4
25 mM dNTP mix (mix 1 part each of 100 mM dNTPs) 0.5
20 µM primer 2 (P2) 0.5
20 µM LP25 0.5
Mix thoroughly at this point before adding the enzyme.
AmpliTaq (5 units/µl) 1.0
Total volume 40 µl
Note: The 5× Taq Buffer we make consists of 200 mM NaCl, 50 mM Tris-HCl, pH 8.9, and 0.05% (w/v) gelatin. 10× Taq buffer from Perkin Elmer is also suitable, but our homemade 5× Taq buffer with pH at 8.9 works better in our hands. To optimize the PCR reaction, the final Mg+2 concentration should be between 1.5 to 2.0 mM, depending on the primer and template. The addition of betaine to a final concentration of 1.5 M from a 5-M stock has proved advantageous for resolving high G + C regions with apparent secondary structure problems.
Recipe 14: Kinase Mix
Component Volume (µl) per sample Volume (µl) for 5 samples
H2O 5.9 29.5
10× kinase buffer 1.0 5.0
20 µM P3 1.1 5.5
T4 DNA kinase (10 units/µl) 1.0 5.0
Total volume 10 µl 50 µl

Instructions

Preparation of Genomic DNA Samples

A few general comments that apply to all of the genomic samples follow. The purity of the DNA should be ascertained by measuring the UV ratio (260 nm/280 nm). This ratio should be 1.8 or greater. All of the genomic DNA preparations should be checked not only by a regular agarose gel, but also by a 1.5% alkaline gel to ensure that there are not excessive strand breaks. DNA prepared by the procedure given here usually has a single-strand size of >20 kb. Genomic DNAs should be stored at 4°C, because repeated freezing and thawing might cause shear breakage of the DNA strands.

In vivo UVC-treated genomic DNA

1. Culture adherent cells in 150-mm plates as a monolayer to near confluence.

Note: For suspension cells, spin at 4000 rpm (table top centrifuge) for 2 min to pellet, wash with PBS, resuspend the cells in 5 ml PBS, and spread on 150-mm plates for UV treatment.

2. Aspirate the medium from the monolayer cell culture and wash two times with 20 ml of PBS, aspirating after each wash.

3. Pre-warm the UV Stratalinker oven equipped with 254-nm UV bulbs for 2 min.

4. Place the washed cell plates without lids inside the pre-warmed UV oven and set the energy level at 1000 J per m2 or 1500 J per m2, and start the UV treatment.

Note: The optimal dose of UVC may vary due to a number of factors, so we recommend performing a dosage curve from 500 J per m2 to 2000 J per m2.

5. Add 5 ml of Cell Lysis Buffer B (Recipe 2) and 5 ml of Cell Lysis Buffer C (containing 1% SDS and 600 µg/ml Proteinase K, Recipe 3) per plate.

6. Cover the plates with the lids and gently swirl to allow the lysis buffer to mix and completely cover the cells.

7. Seal the plates with parafilm and incubate at 37°C for 3 hours.

8. Transfer the lysed cell solution from the plates to a 50-ml centrifuge tube.

9. Add DNase-free RNase A (100 mg/ml) to each tube to a final concentration of 100 µg/ml and continue incubating at 37°C for 1 hour.

10. Add NaCl to a final concentration of 0.29 M per tube from the 5 M NaCl stock solution.

11. Perform phenol-chloroform extractions. Extract two times with an equal volume of phenol, once with an equal volume of a mixture of phenol and chloroform (1:1, v/v), and once with an equal volume of chloroform.

12. Add an equal volume of isopropanol to the extracted aqueous phase, invert the tube to mix, and spool the DNA using an end-sealed glass pipet.

Note: End-sealed glass pipets can be prepared by using a Bunsen burner to melt the ends of glass Pasteur pipets.

13. Dip the tip containing the DNA in 75% ethanol to wash.

14. Air dry for 3 min.

15. Resuspend the DNA by dipping the tip in 1 ml of TE, pH 8.0. Leave at 4°C overnight or long enough to completely resolubilize the DNA. This is the in vivo UV-treated genomic DNA sample.

Control and in vitro UVC-treated genomic DNAs

1. Isolate genomic DNA from cultured cells as described above, except do not perform the UV treatment (steps 3 and 4 above). This is the control genomic DNA sample.

2. Dilute the control genomic DNA in TE, pH 8.0 (Recipe 6) to 0.5 µg/µl.

3. Pre-warm the UV Stratalinker oven equipped with 254-nm UV bulbs for 2 min.

4. Deposit 5-µl droplets (50 to 100) onto a piece of Parafilm inside the pre-warmed UV Stratalinker oven and irradiate to the same or higher (1) dose as the in vivo-treated DNA was irradiated. Combine the droplets and transfer the DNA to a microcentrifuge tube. This is the in vitro UV-treated genomic DNA sample.

Primers and Linkers

Although several software programs give good predictions of Tm, it is best to check the effective Tm in the TDPCR reactions (or conventional PCR). This can be done conveniently by running a temperature gradient on a gradient-capable thermocycler, such as a Robocycler (Stratagene).

Gene-specific primers

Our laboratory uses either the Oligo 4 or the Oligo 5.1 software to design the primers. The first gene-specific primer (Bio-P1) is 5′-biotinylated and should be designed to have a melting temperature (Tm) of approximately 60°C; the second primer (P2) should have a Tm of approximately 63°C; and the third primer (P3) should have a Tm of approximately 65°C. Ideally, for optimal direct labeling, the gene-specific primers should be designed so that their Tms are in ascending order. Because the P2 is used together with LP25 (the linker primer) at the PCR step, designing a P2 with a Tm similar to that of LP25 is optimal.

Universal linker primer

The universal linker primer (LP25) is a 25-mer (1) of sequence:

5′-GCGGTGACCCGGGAGATCTGAATTC-3′.

TDPCR linker primers

The sequences of these primers are shown below.

Upper primer (LP27): 5′-GCGGTGACCCGGGAGATCTGAATTCCC-3′, a 27-mer (16). Lower primer (LP24*C5): 5′-AATTCAGATCTCCCGGGTCACCGC-APNH2-3′, a 24-mer with an aminopentyl blocking group at its 3′ terminus (16).

Kinasing of the lower TDPCR linker primer

1. Prepare the Kinase Mix for TDPCR Linkers (Recipe 9) in a 1.7-ml microcentrifuge tube. (Usually 10 tubes are done at the same time.)

2. Incubate at 37°C for 1.5 to 2 hours.

3. Incubate at 65°C for 20 min.

4. Place samples on ice.

Annealing the TDPCR linker primers

1. Add 44 µl of a 200-µM solution of the upper primer directly to the kinased samples from the steps above and mix. (The final concentration of the linkers is 20 µM.)

2. Pulse the samples at maximum speed in a microcentrifuge.

3. Secure tubes with a lid lock and denature at 95°C for 5 min in a heating block.

4. Turn off the power to the heating block, but leave tubes in the block, and allow the samples to gradually cool to room temperature in the block.

5. Leave at 4°C overnight and then store at -20°C.

Note: Always thaw the annealed linkers on ice before preparing the Ligation Mix (Recipe 12).

TDPCR Procedure

The following is a Vent (exo-)-AmpliTaq 2-day TDPCR protocol used regularly in our laboratory (Fig. 1). It is possible to change from a 2-day to a 1-day protocol by performing only a 2-hour ligation instead of an overnight ligation (Fig. 2A). However, the efficiency of the short ligation is dependent on the activity of the ligase, so overnight ligation is likely to be more reproducible.

Vent (exo-), Vent, Pfu, Pfu Turbo, and Expand Long enzymes have all worked well for the primer extension step, as well as for the PCR step; however, we have not tested AmpliTaq in the primer extension step. Any combination of thermostable polymerases should be tested in each lab.

Vent, Pfu, Pfu Turbo, and Expand Long enzymes all have 3′ exonuclease activity, which will degrade single-stranded DNA such as primers. Usually, under the conditions described here, these enzymes nevertheless work well. However, the reason Vent (exo-) is commonly used is that it does not have 3′ exonuclease activity.

Primer extension (day 1)

1. Prepare the proper amount of Extension Mix (Recipe 10) according to the number of samples, allowing enough for two extra per experiment.

2. Add 10 µl of DNA sample (0.5 µg to 1.0 µg) and 20 µl of the Extension Mix (Recipe 10) to a siliconized 0.65-ml microcentrifuge tube on ice.

3. Mix by pipetting, and pulse at maximum speed in a microcentrifuge to bring the liquid to the bottom of the tube.

4. Overlay with 20 µl of mineral oil.

Note: Mineral oil can be omitted if the thermocycler has a hot bonnet; in this case, remember to prestart the program and set at pause to allow the hot bonnet to equilibrate to 95°C (5 to 10 min).

5. Perform the extension reaction using the following parameters to produce the Bio-P1 extended DNA.

1 cycle: 95°C 3 min

10 cycles: 95°C 45 s

Tm of primer 1 (Bio-P1) (or up to Tm + 5°C) 2 min

72°C 3 min

Note: More cycles can be run, but the background level may increase.

Binding to streptavidin-coupled magnetic beads (day 1)

1. Gently swirl the bottle containing the beads from the manufacturer to completely resuspend the beads and then place an amount sufficient for all of the samples (usually 10 µl to 20 µl per sample) into a microcentrifuge tube.

2. Remove supernatant using the MPC.

3. Wash beads with a 2× volume of 2× BW (Recipe 4) (see the manufacturer's instructions). Remove supernatant using the MPC.

4. Resuspend the beads in 2× BW (Recipe 4), allowing 30 µl per sample.

5. Transfer 30 µl of beads to each sample after primer extension and mix by pipetting gently.

6. Immobilize the Bio-P1 extended DNA molecules to the beads by rotating the mixture at room temperature for 15 to 60 min (see the manufacturer's instructions).

7. Remove supernatant using the MPC.

8. Wash twice with 50 µl of 2× BW (Recipe 4) and remove the supernatant using the MPC.

9. Separate the DNA strands by incubating the beads with 50 µl of freshly prepared 0.15 M NaOH (prepare just before use by mixing 30 µl of 5M NaOH with 970 µl of dH2O) at 37°C for 5 to 10 min. Remove supernatant using the MPC.

10. Wash once with 50 µl of freshly prepared 0.15 M NaOH. Remove the supernatant using the MPC.

11. Wash twice with 100 µl of TE, pH 7.5 (Recipe 6). Mix thoroughly and perform a 1-s to 3-s pulse (low speed) in a microcentrifuge to bring down any residual NaOH. Remove the supernatant using the MPC.

12. Resuspend each sample in 10 µl of 0.1× TE, pH 7.5 (Recipe 7).

RiboGTP tailing using terminal deoxynucleotidyl transferase (TdT) (day 1)

1. Prepare the amount of TdT Mix (Recipe 11) needed for the number of samples, allowing enough for two extra per experiment.

2. Mix the solution gently by pipetting, pulse in a microcentrifuge to bring to the bottom of the tube, and check the total volume with a micropipette.

3. Add 10 µl of TdT Mix (Recipe 11) to each sample from step 13 above, and incubate at 37°C for 15 min.

4. Remove the supernatant using the MPC.

5. Wash twice with 100 µl of TE, pH 7.5 (Recipe 6) and remove the supernatant using the MPC.

6. Resuspend each sample in 15 µl of 0.1× TE, pH 7.5 (Recipe 7).

Overnight ligation (day 1)

1. Prepare the amount of Ligation Mix (Recipe 12) needed for the number of samples, allowing enough for two extra per experiment.

2. Mix the solution gently by pipetting, pulse in a microcentrifuge to bring to the bottom of the tube, and check the total volume with a micropipette.

3. Add 15 µl of Ligation Mix (Recipe 12) to each sample, mix well by pipetting, and incubate overnight at 17°C using a thermocycler.

Note: Close each tube with a lid lock if using a water bath for overnight ligation. The overnight ligation can be substituted by a two-hour ligation (Fig. 2A).

PCR (day 2)

1. Pulse the samples for 1 to 3 sec (low speed) in a microcentrifuge and remove supernatant using the MPC.

2. Wash the beads twice with 100 µl of TE, pH 8.0 (Recipe 6), remove supernatant using the MPC, and resuspend each bead sample in 10 µl to 30 µl of 0.1× TE, pH 8.0 (Recipe 7).

Note: If the beads are resuspended in 20 µl o 30 µl of 0.1× TE, pH 8.0, store extra unused sample at 4°C. Do not freeze samples containing the magnetic beads!

3. Prepare the PCR Mix (Recipe 13) needed for the number of samples (40 µl per sample), allowing enough for two extra per experiment.

Note: We have also used the Expand Long PCR System (BMB) and obtained longer readouts when using the LI-COR Long Ranger gel electrophoresis system. We follow the instructions from BMB to prepare PCR mixtures.

4. Add 40 µl of PCR Mix (Recipe 13) to 10 µl of each resuspended sample. Mix well by pipetting. The total volume should be 50 µl.

5. Overlay the PCR reactions with 30 µl of mineral oil.

Note: The oil can be omitted if the thermocycler has a hot bonnet; in this case, remember to prestart the program and set at pause to allow the hot bonnet to equilibrate to temperature.

6. Perform a PCR reaction using the following parameters:

1 cycle: 95°C 3 min

20 cycles: 95°C 45 s

Tm of primer 2 (P2) 2 min

72°C 3 min

7. Store the PCR products at 4°C (do not freeze!) for use in future labeling reactions.

Note: Although most of the thermostable DNA polymerases we have used remained active for months if the PCR mixtures were stored properly at 4°C, we add 0.1 µl of additional polymerase into the final labeling reaction.

Escherichia coli Exonuclease I treatment (optional) (day 2)

1. Dilute Exo I from a stock of 10 units/µl to 1 unit/µl with dH2O.

2. Transfer 9 µl of each PCR product to a new tube using the MPC and add 1 µl of the diluted Exo I.

3. Incubate at 37°C for 30 min to eliminate unincorporated primers (P1, P2, and LP25) left from the previous steps.

4. Inactivate Exo I by incubating at 72°C for 15 min.

Detection by 32P-Labeling and Autoradiography (Day 2)

These are the instructions to follow if the samples are to be analyzed by autoradiography following detection with a 32P-labeled fragment. If samples are to be detected with fluorescence infrared dyes, please see the next section.

Direct labeling of primer 3 (P3) using 32P

1. Prepare Kinase Mix (Recipe 14). Mix well and place on ice. Pulse the sample at maximum in a microcentrifuge and return to ice.

Note: The following steps must be performed taking the proper precautions for use of radioactivity and should be performed in the designated radioactivity work area in the laboratory.

2. Add 1 µl (1×) or 5 µl (5×) 32P-γATP, mix well, and pulse at maximum in a microcentrifuge to bring liquid to the bottom of the tube.

3. Incubate at 37°C for 1 to 1.5 hours.

4. Incubate at 65°C for 15 min.

5. Purify the labeled primer using a G-25 spin column.

6. Count 0.5 to 1.0 µl of the reaction in a scintillation counter. Counts per min per µl should be at least 1 × 106 to 5 × 106 cpm per µl. (Counts achieved will vary with the primer.)

7. Store at -20°C or use directly.

Primer extension

1. Remove 10 µl from each PCR reaction and add the appropriate volume (we recommend 1 × 106 to 5 × 106 counts per min per reaction) of 32P-ATP labeled P3.

2. Pulse in a microcentrifuge to bring liquid to the bottom of the tube.

3. Overlay with 10 µl of mineral oil.

4. Set the thermal cycler to pause at 95°C and warm the tubes in the preheated thermal cycler for no longer than 1 min.

5. Perform a primer extension reaction using a thermal cycler with the following parameters:

1 cycle: 95°C 2 min

3 to 9 cycles: 95°C 45 s

Tm of P3 2 min

72°C 3 min

1 cycle 72°C 10 min

Note: If Exo I treatment was performed, add the labeled P3 directly to the Exo I-treated sample and perform the primer extension reaction.

6. Prepare the samples for gel loading by adding an equal volume of Formamide Loading Solution (Recipe 8) to each sample and denaturing the samples at 95°C for 2 min.

7. Place samples on ice before loading onto a sequencing gel (usually 6 to 8% acrylamide).

8. Store labeled samples at -20°C.

Preparation of labeled DNA markers

1. Label a 25-bp or 50-bp DNA ladder with 32P-αdCTP using T4 DNA polymerase according to instructions from BRL.

2. Stop the reaction with 2.5 µl of 0.5 M EDTA.

Note: This is the amount of EDTA required if BRL's instructions are followed and the reaction's total volume is 50 µl.

3. Purify the labeled DNA ladder on a G-25 column.

4. Add a 2× volume of Formamide Loading Dye (Recipe 8) and denature for 2 min at 95°C before loading alongside of TDPCR samples on a sequencing gel.

Note: When the 32P markers are fresh, 0.5 to 1.0 µl is sufficient for an overnight exposure. Markers can be used for up to 3 months, compensating for radioactive decay by increasing the amount loaded.

Detection with the LI-COR System (Day 2)

These are the instructions to follow if the samples are to be analyzed by labeling with the fluorescence infrared dyes IRD-700 or IRD-800. Both IRD-700 and IRD-800 give similar sensitivity. Both dyes can be detected simultaneously, allowing two differently labeled samples to be run together in the same lane.

Direct labeling using oligonucleotide primers 5′-labeled with LI-COR dyes IRD-700 or IRD-800

1. Order primer 3 (LI-COR P3) from LI-COR labeled with either IRD-700 or IRD-800.

Note: The LI-COR primers are light sensitive and should be handled under dimmed or yellow light.

2. Resuspend the primer, which arrives as a dry pellet from LI-COR, according to LI-COR's instructions, to obtain a 1-µM final concentration.

Primer extension

1. Remove 9 µl from each PCR reaction and add 1 µl of 1 µM LI-COR P3.

2. Pulse in a microcentrifuge to bring liquid to the bottom of the tube.

3. Overlay with 10 µl of mineral oil.

4. Set the thermal cycler to pause at 95°C and place the tubes in the preheated thermal cycler.

5. Perform the primer extension reaction using a thermal cycler with the following parameters:

1 cycle: 95°C 2 min

3 to 9 cycles: 95°C 45 s

Tm of LI-COR P3 2 min

72°C 3 min

1 cycle 72°C 10 min

Note: If Exo I treatment was performed, add the LI-COR P3 directly to the Exo I-treated sample and perform the primer extension reaction.

6. Prepare the samples for gel loading by adding 3 µl of LI-COR gel loading dye (IR2 Stop Solution, LI-COR) to each sample and denaturing the samples at 95°C for 2 min

7. Cool the samples on ice and load onto a 5 or 6% LI-COR sequencing gel.

8. Store labeled samples in the dark at -20°C.

Gel electrophoresis

32 P-γATP-labeled samples

Any DNA sequencing electrophoresis apparatus and power supply (6000 volts) are suitable.

1. Prepare a 6 to 8% acrylamide sequencing gel.

Note: For better resolution during gel electrophoresis, the SequaGel solutions are routinely deionized using AG 501-X8 (D) Resin before casting the gels.

2. Load the samples and the labeled DNA markers.

Note: An 8-channel syringe or pipettor is recommended for equal loading of samples onto the sequencing gels.

3. Run the gel until the xylene cyanol reaches the bottom of the gel.

4. Transfer the gel to 3 mm Whatman paper.

5. Cover the gel with a piece of Saran wrap.

6. Dry under vacuum, as is typically done for standard sequencing gels.

7. Expose the gel overnight using a PhosphorImager cassette, and scan next morning (Fig. 2A).

IRD-700 or IRD-800 labeled samples

To separate LI-COR IRD-primer labeled samples, we use a IR2 Long Ranger electrophoresis system from LI-COR, which was designed for DNA sequence determination, along with LI-COR DNA markers, which can be purchased from LI-COR. The data are automatically saved as a TIFF file during the run and the file can be opened directly by Adobe PhotoShop (Fig. 2B) or analyzed quantitatively by other programs that accept TIFF files.

Both PhosphorImager and LI-COR gel files can be read and quantitated using Gene ImagIR [an upgraded version of restriction fragment length polymorphism (RFLP) analysis software (17, 18)] from Scanalytics.

Notes And Remarks

We now prefer the LI-COR nonradioactive procedure. However, the LI-COR instrument is expensive and may not be available, so the conventional, radioactive detection system is also described.

As controls to aid sequence identification and check efficiency, we usually include DNA samples made using Maxam-Gilbert reactions. Maxam-Gilbert-treated DNA is an excellent substrate for both LMPCR and TDPCR. In fact, sequence can be determined by genomic sequencing (18). It is important to have 32P-labeled molecular weight markers or LI-COR molecular weight markers included in the gels, because they help to locate the footprints and read sequence in the Maxam-Gilbert lanes. The markers are especially useful when the Maxam-Gilbert sequence ladders are poor.

To ensure the reproducibility of footprints, performing experiments in duplicate or triplicate (see Fig. 2A) is highly recommended. In general, thin-wall tubes or microtiter plates, such as those used for the Robocycler, have given better results in our hands.

Issues related to the amount of DNA sample required are discussed here. In theory, around 0.5 µg to 1.0 µg of vertebrate genomic DNA will provide about 150,000 to 300,000 template molecules for the primer extension reaction of LMPCR and TDPCR. Each band seen in the final DNA fragment ladder requires a template molecule. Up to one thousand bands are produced (the LI-COR gel often shows bands to about this size, though good-quality data usually do not go beyond 500 bases), so template molecules are only in about 150- to 300-fold excess. This is adequate if all the reactions are efficient, but usually overall efficiency of molecule usage seems to be about 10%. Thus, one must be careful when using less than 0.5 to 1 µg of genomic DNA to ensure that the primers and all enzyme components are working efficiently. Inefficient reactions often give adequate signal and useful information, but bands may be missing or reduced in intensity because of statistical sampling fluctuation (16).

TDPCR theoretically and experimentally (16) requires less DNA than LMPCR, because each template molecule can be sampled more than once by repeated primer extensions. However, if the efficiencies of tailing and linker ligation are not high, this advantage is reduced. If the starting genomic DNA is limited such that less than 0.1 µg DNA is used per reaction, one is advised to include carrier DNA and to make sure that all reactions are as efficient as possible by adjusting cycling temperatures and concentrations of Mg2+, dNTP, primers, and other reagents.

Troubleshooting

No Bands or Weak Bands

Check the primer sequences and the DNA samples (for example, the region of interest might be polymorphic) by performing a standard PCR using the primers that cover the region of interest. It is tempting to just increase the number of PCR cycles to enhance weak or nonexistent signal, but this should be done with caution because for footprinting, one must stay in the linear, quantitative range of PCR, as is almost always the case for the first 20 cycles. If more than 20 cycles are needed, it is likely that at least one of the primers is poor or at least one step is inefficient. If this is the case, the patterns seen may not be reproducible.

Extra Bands

TDPCR generally shows at least a weak band at every position. This is normal and due to the polymerase pausing, as well as variability in riboG-tailing by TdT. However, the UV-generated band pattern seen for TDPCR is similar to that obtained by LMPCR (16). Occasionally when Taq DNA polymerase is used for PCR, every LMPCR or TDPCR band has a shadow band. This shadowing is due to variability in the addition of an extra base by Taq and indicates that the enzyme was no longer adequately active in the last cycle of PCR. Note also that the size purity of the linker and labeling primers is important, because the final labeled DNA fragments need to be separated with single nucleotide resolution.

Lack of Reproducibility

In vivo footprinting is a procedure that involves several steps, numerous pipettings, and a PCR reaction. Make sure that all components of the reaction mixtures are at the correct concentrations and are active, especially the enzymes. As a rough rule of thumb, differences in band intensities ≥ fold can be reliably detected by TDPCR, as is the case for LMPCR, where, with care, the coefficient of variation for band intensities can be as low as 10% for strong bands (18).

References

  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.
  21. 21.
  22. 22.
  23. 23.
  24. 24.
  25. 25.
  26. 26.
  27. 27.
  28. 28.
  29. 29.
View Abstract

Navigate This Article