Moss Tom. DNA-protein interactions: principles and protocols

Подождите немного. Документ загружается.

56 Zaychikov et al.

3.2. Establishing the Conditions for Cutting the DNA

by Hydroxyl Radicals

1. End label an aliquot of the DNA fragment of interest under standard conditions

(17) at the 5'-position, using T

polynucleotide kinase and [γ-

P] ATP and end

label a second aliquot at the 3'-position, using the Klenow fragment of DNA

polymerase I and the appropriate [α-

P] dNTP. In each case, remove one label

end by asymmetric cleavage of the DNA fragment with an appropriate restriction

endonuclease. Purify your uniquely end-labeled DNA fragment by a non-

denaturing gel electrophoresis (see Subheading 2.3.). The total amount of radio-

activity in one assay should be approx 10,000–15,000 cpm. The total amount of

DNA in an assay of 20 µL should not be below 100 ng. The optimum length of

the DNA fragment is between 100 and 200 base pairs (bp) (see Note 4).

2. For the cutting reaction, use the buffer conditions and temperature that are opti-

mum for formation of the protein–DNA complex, (see Note 5). Add to the 20 µL

incubation assay 2 µL of each of the previously prepared solutions of DTT,

hydrogen peroxide, and the iron(II)–EDTA and mix rapidly. This can be done by

placing the individual 2-µL drops separated on the inner wall of the tube and then

rapidly mixing them before combining them with the sample using a micropipet.

3. Incubate for variable times (1–5 min is recommended) at the appropriate

temperature.

4. Add 25 µL stop mix and 150 µL of ice-cold 100% ethanol to precipitate the DNA.

Keep the solution at –70°C for 30 min.

5. Spin down the precipitate in a microcentrifuge for 30 min. Wash the pellet with

ice-cold 80% ethanol, dry under vacuum and dissolve the pellet in formamide

buffer. Adjust the amount of radioactivity and volume of each sample to approx

5000–6000 cpm in 4–5 µL.

6. Heat the sample for not longer than 2 min at 90°C and place on ice (see Note 6).

Apply the sample onto a 6–10% sequencing gel (for the analysis of fragments in

the range of 50–150 bases, a gel consisting of 8% acrylamide is adequate). Use as

length standards a Maxam–Gilbert sequencing reaction of the 5'- or 3'-labeled

DNA fragment.

7. Run the gel at 50 W at a temperature of 60°C for 1.5–2 h. The gel is ready when

the xylenecyanol dye marker is about 3–5 cm above the bottom of the gel.

8. After electrophoresis, expose the gel to an X-ray film using an intensifying screen

at –70°C overnight. For subsequent experiments, choose the time of hydroxy

radical cleavage that provides an even distribution of bands and leaves around

90% of the DNA uncleaved.

3.3. Footprinting of Protein-DNA Complexes

3.3.1 Preparation of Complexes

and Performing Footprinting Reaction

1. Prepare two 15-µL samples of the complexes, one with the DNA labeled at the 3'

end and another with the label at the 5' end. Use the conditions established for

Hydroxyl Radical Footprinting 57

optimal complex formation from Subheading 3.1. The total amount of radioac-

tivity in one assay should be approx 60,000–80,000 cpm.

2. Pour 30–40 mL of the dialysis buffer containing 8 mM Tris–HCl, pH 7.9, into a

Petri dish (see Note 5). Place a Millipore filter (VS 0.025 µm) on the surface of

the buffer, shiny side (hydrophobic side) up. Put the samples containing the com-

plexes onto the filter for 1 h in order to remove glycerol and salt (see Note 7).

Remove the samples from the filter using a micropipet and transfer them to a

fresh 1.5-mL Eppendorf tube.

3. Subject the two samples to hydroxyl radical treatment as described in Subhead-

ing 3.2., steps 2 and 3 only.

3.3.2. Separation of the Complex and the Free DNA

by Nondenaturing Acrylamide Gel Electrophoresis

The conditions for studying a high-molecular-weight complex are described

below (see Note 8). Use the conditions for preparing the nondenaturing gel as

described in Subheading 2.3. These low-concentration gels are difficult to

handle, therefore, the glass plates must be subjected to a special treatment by which

the gel is bound to one of the plates. Wash the glass plates (20 × 20 cm) with

ethanol. Treat one plate with γ-methacryl-oxypropyl-trimethoxy-silane.

Wash this plate carefully four times with ethanol to avoid sticking of the

other plate to the gel. Treat the second plate with dichlorodimethylsilane.

Use spacers 1–1.5 mm thick. Use combs that allow you to apply amounts of

the 50-µL sample.

1. Add loading buffer to the sample (1/10 of the sample volume). Apply the sample onto

the nondenaturing acrylamide gel (20 × 20 cm). Run the gel at 20 V/cm for about

2 h. The gel is ready when the dye marker is approx 3–5 cm above the end of the gel.

2. Remove one of the plates by lifting it carefully with a spatula and cover the gel

(which sticks to the other glass plate) with plastic wrap. Place an X-ray film on

the gel. Put the gel and the film into a cassette and expose for 1–2 h at 4°C. Mark

the bands containing the complex and, if visible, the free DNA. This is possible

by replacing the film after developing on the gel in exactly the same position.

The use of fluorescent marker tapes (e.g., from Stratagene) during exposure of

the gel may be very helpful for exact repositioning of the film.

3. Remove the plastic wrap from the gel, cut out the marked bands with a scalpel or

spatula, and cut them into small pieces. Put the slices into a eppendorf tube along

with 300 µL of elution buffer (0.5% SDS, 1 mM EDTA, 0.1 mg/mL carrier DNA)

and shake for several hours at room temperature.

4. Spin the tube for a few seconds in a microcentrifuge and transfer the liquid to a

new tube. Avoid transferring gel pieces to the new tube. Add 200 µL of water-

saturated neutralized phenol, shake 1 min, and centrifuge for 5 min at room tem-

perature. Remove aqueous layer (upper), transfer it to a fresh Eppendorf tube,

and add 30 µL of 3 M sodium acetate and 1 mL of ice-cold 100% ethanol. Shake

the tube for a few seconds and put into –70°C for at least 30 min.

58 Zaychikov et al.

5. Spin the sample in a microcentrifuge for 30 min and remove the supernatant.

Wash once with 1 mL of ice-cold 80% ethanol. Dry the sample under vacuum.

Dissolve the pellet in a small volume of formamide buffer (usually 5–10 µL,

depending on the amount of radioactivity).

3.3.3. Separation of the Complex

and the Free DNA Filtration Through Nitrocellulose Filter

(Alternative to Subheading 3.3.2.,

see

Note 9)

The reader may also find it useful to refer to Chapter 1, which discusses the

uses of the filter binding assay, and to Subheading 3.2.3. of that chapter, which

deals with the recovery of DNA from filter-retained complexes.

1. Stop the hydroxyl radical reaction by dilution (at least twofold) with binding

buffer containing 2% glycerol. (Alternatively, the sample may be poured

directly into the filtration device after completion of the reaction and immedi-

ately washed).

2. Mount a nitrocellulose filter (BA85, Schleicher & Schüll) into the filtration

device and connect it to a peristaltic pump. (See Note 10.)

3. Pore the solution containing the complex treated with radicals onto the filter.

Switch on the pump and filter off the solution at the speed of 1–2 mL/min. Wash

the filter twice with binding buffer at the same speed. Switch off the pump.

4. Disassemble the filtration device, take off the filter and cut it into small pieces.

Place the pieces into an Eppendorf tube.

5. Add 300 µL of solution containing 1% SDS, 0.3 M sodium acetate, and 10 µg/mL

carrier DNA to the tube and shake for 15 min at room temperature.

6. Carefully transfer the liquid into fresh tube (avoid transferring the nitrocellulose)

and add 1 mL of ethanol. Shake the tube for a few seconds and put into –70°C for

at least 30 min.

7. Spin the sample in a microcentrifuge for 20 min and remove the supernatant.

Wash once with 1 mL of ice-cold 80% ethanol. Dry the sample under vacuum.

Dissolve the pellet in a small volume of formamide buffer (usually 5–10 µL,

depending on the amount of radioactivity).

3.3.4. Analysis of DNA by Denaturing Gel Electrophoresis

Heat the samples at 90°C for 1–2 min. Proceed as described for free DNA in

Subheading 3.2., steps 7–10. For both 3' and 5' end-labeled DNA, load the

following samples on the gel: the DNA recovered from the complex, the free

DNA (either recovered from the gel and/or prepared separately as in Subhead-

ing 3.2.) and the Maxam–Gilbert reaction ladders as length standard.

4. Notes

1. The iron(II), iron–EDTA mix, and the H

solutions should be freshly made

before use. The solutions of DTT (0.1 M), EDTA (4 mM), H

(as a 30% stock

solution), and the stop mix are stable for months stored at –20°C.

Hydroxyl Radical Footprinting 59

2. The acrylamide solutions are stable for months if protected from light and

kept at 4°C.

3. Sodium persulfate has an advantage over routinely used ammonium persulfate of

being much more stable in aqueous solution. The 10% sodium persulfate solution

may be kept at least 1 mo at +4°C without loss of activity.

4. It is strongly recommended to check the quality of the labeled DNA before use

on a sequencing gel. Nicks in the double strand, which could be derived from

DNase activities during preparation, will appear as additional bands in the

sequencing gel. This admixture of bands will spoil the whole footprint, even when

present in only small amounts. Furthermore, it is recommended not to store the

pure, labeled DNA longer than 2 wk at –70°C, because the radiation of the label

also creates nicks in the DNA.

6. Longer heating or boiling creates additional cuts in the DNA.

5. The buffer conditions can be varied (e.g., the pH), but the ionic strength should

not be too high (maximum 50 mM NaCl) in order to obtain sharp bands during

the nondenaturing electrophoresis to separate complexed and uncomplexed DNA.

Many protein–DNA complexes are very stable at low ionic strength (e.g., com-

plexes between RNA polymerase and promoters [18]). Therefore, in most cases

the stability of the pH in the following electrophoresis is the only limitation to

lowering the ionic strength.

7. The purpose of the dialysis is twofold: Removal of glycerol which interferes

with the cutting reaction and removal of salt which lowers the quality of the

electrophoresis pattern. As a rough approximation, one can remove up to

80–90% of the glycerol and salt present in the sample within 1 h using drop

dialysis as described.

8. The gel concentration should be adjusted according to the molecular weight

of the protein–DNA complex. Here we have described the conditions established

for the study of the E. coli RNA polymerase (molecular weight 490,000) and a

DNA fragment of 130 bp carrying a promoter (3).

9. Filtration via nitrocellulose is a simpler and faster way to remove unbound

DNA, but it does not resolve complexes having different stoichiometries of

protein and DNA.

10. A variety of filtration devices may be used allowing handling of relatively small

volumes (below 1 mL). Mild vacuum may be used instead of a peristaltic pump.

References

1. Tullius, Th. D., Dombroski, B. A., Churchill, M. E. A., and Kam, L. (1987)

Hydroxyl radical footprinting: a high resolution method for mapping protein–DNA

contacts. Methods Enzymol. 155, 537–558.

2. Götte, M., Marquet, R., Isel, C., Anderson, V. E., Keith, G., Gross, H., Ehresmann, C.,

et al. (1996) Probing the higher order structure of RNA with peroxonitrous acid. FEBS

Lett. 390, 226–228.

3. Shafer, G. E., Price, M. A., and Tullius, Th. D. (1989) Use of the hydroxyl radical

and gel electrophoresis to study DNA structure. Electrophoresis 10, 397–404.

60 Zaychikov et al.

4. Schickor, P., Metzger, W., Werel, W., Lederer, H., and Heumann, H. (1990)

Topography of intermediates in transcription initiation of E. coli. EMBO J. 9,

2215–2220.

5. Craig, M. L., Suh, W. C., and Record T. M. (1995) HO

and DNase I probing

of Eσ70 RNA polymerase–λP

promoter open complexes: Mg

binding and

its structural consequences at the transcription start site. Biochemistry 34,

15,624–15,632.

6. Zaychikov, E., Denissova L., Meier, T., Götte, M., and Heumann H. (1997) Influ-

ence of Mg

and temperature on formation of the transcription bubble J. Biol.

Chem. 272, 2259–2267.

7. Zaychikov, E., Martin, E., Denissova, L., Kozlov, M., Markovtsov, V., Kashlev,

M., et al. (1996) Mapping of catalytic residues in the RNA polymerase active

center. Science 273, 107–109.

8. Götte, M., Maier, G., Gross, H., and Heumann, H. (1998) Localization of the

active site of HIV-1 reverse transcriptase–associated RNase H domain on a DNA

template using site-specific generated hydroxyl radicals. J. Biol. Chem. 273,

10,139–10,146.

9. Metzger, W., Schickor, P., and Heumann, H. (1989) A cinematographic view of

Escherichia coli RNA polymerase translocation. EMBO J. 8, 2745–2754.

10. Zaychikov, E., Denissova, L., and Heumann, H. (1995) Translocation of the

Escherichia coli transcription complex observed in the registers 11 to 20: “Jumping”

of RNA polymerase and asymmetric expansion and contraction of the “transcription

bubble”. Proc. Natl. Acad. Sci. USA 92, 1739–1743.

11. Heumann, H., Zaychikov, E., Denissova, L., and Hermann, T. (1997) Transloca-

tion of DNA-dependent E. coli RNA polymerase during RNA synthesis, in

Nucleic Acids and Molecular Biology, vol. 11, (Eckstein, F. and Lilley, D., eds.),

Springer-Verlag, Heidelberg, Germany, pp. 151–177.

12. Cons, B. M. G. and Fox, K. R. (1989) High resolution hydroxyl radical

footprinting of the binding of mithramycin and related antibiotics to DNA. Nucleic

Acids Res. 17, 5447–5459.

13. Burkhoff, A. M. and Tullius, Th. D. (1988) Structural details of an adenine tract

that does not cause DNA to bend. Nature 331, 455–457.

14. Tullius, Th. D. and Dombroski, B. A. (1985) Iron(II) EDTA used to measure the

helical twist along any DNA molecule. Science 230, 679–681.

15. Wang, X. and Padgett, R. A. (1989) Hydroxyl radical “footprinting” of RNA: appli-

cation to pre-mRNA splicing complexes. Proc. Natl. Acad. Sci. USA 86, 7795–7799.

16. Celander, D. W. and Cech, Th. R. (1991) Visualizing the higher order folding of a

catalytic RNA molecule. Science 251, 401–407.

17. Maniatis, T., Fritsch, E. F., and Sambrock, J. (1982) Molecular Cloning, a Labo-

ratory Manual, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY.

18. Heumann, H., Metzger, W., and Niehörster, M. (1986) Visualization of intermedi-

ary transcription states in the complex between Escherichia coli DNA-dependent

RNA polymerase and a promoter-carrying DNA fragment using the gel retardation

method. Eur. J. Biochem. 158, 575–579.

Hydroxyl Radical Footprinting 61

Further Reading

Tullius, Th. D. (1989) Physical studies of protein–DNA complexes by footprinting.

Annu. Rev. Biophys. Biophys. Chem. 18, 213–237.

Tullius, Th. D. (1989) Structural studies of DNA through cleavage by the hydroxyl

radical, in Nucleic Acids and Molecular Biology, vol. 3, (Eckstein, F. and Lilley,

D., eds.), Springer-Verlag, Heildelberg, Germany, pp. 1–12.

Use of DEPC and Potassium Permanganate as Probes 63

From:

Methods in Molecular Biology, vol. 148: DNA–Protein Interactions: Principles and Protocols, 2nd ed.

Edited by: T. Moss © Humana Press Inc., Totowa, NJ

The Use of Diethyl Pyrocarbonate and Potassium

Permanganate as Probes for Strand Separation

and Structural Distortions in DNA

Brenda F. Kahl and Marvin R. Paule

1. Introduction

In the search for methods to explore the interaction between proteins and

DNA, a plethora of footprinting techniques have been developed, many of

which are discussed elsewhere in the present work. Most footprinting tech-

niques are based on the simple premise of specific DNA regions being pro-

tected from the reagent by the bound protein or molecule of interest. However,

a number of studies over the past decade have revealed remarkable distortion

of the DNA molecule, including bending and strand separation, in response to

the bound protein. These distortions are often within the classical footprint but

are rarely detected by these classical techniques. Thus, their detection requires

alternative approaches. Unlike enzymatic methods, the chemical probes diethyl

pyrocarbonate (DEPC) and potassium permanganate can access and react with

the entire sequence of the DNA, distinguishing DNA distortions “under the

foot” of a typical footprinting experiment. Combining the information obtained

from these chemical probing techniques with the spatial information afforded

by traditional footprinting gives an in-depth account of the various ways pro-

teins and other molecules interact with DNA.

Diethyl pyrocarbonate and potassium permanganate are useful probes

because of their preferential reactivity with single-stranded vs double-stranded

DNA. In addition, unlike typical footprinting techniques where near saturation

of the DNA with protein is necessary to observe a clean footprint, potassium

permanganate and DEPC probing does not require most of the DNA to be in

complex with the protein or molecule of interest. Because of the sensitivity of

64 Kahl and Paule

these chemical probes, it is possible to observe reactive bases even when the

population of single-stranded DNA is very small.

The mechanism by which DEPC modifies bases has been investigated, but it

is still not clearly understood (1–3). DEPC predominantly reacts with purine

residues, but it may react weakly with cytosine residues as well. DEPC modi-

fies DNA by an out-of-plane attack on several of the nucleophilic centers in

purines, leading to the scission of the glycosidic bond. Double-stranded B-form

DNA does not undergo modification by DEPC because the close stacking of

neighboring bases occludes access to the out-of-plane surfaces. However, under

conditions in which the conformation deviates from B form (such as strand

separation or bending), purines in the sequence become more accessible to

modification by DEPC. Carbethoxylation of the imidazole ring N-7 produces

strand scission under alkaline conditions. Thus, DEPC is commonly used to

detect purines that are present in melted or distorted DNA sequences. Although

DEPC can react with both purines, it shows a marked preference for adenines

vs guanines in most instances.

Potassium permanganate reacts with double bonds, oxidizing them to vici-

nal diols. In nucleic acids, the base thymine is oxidized most vigorously,

whereas reaction with C, G, and A is minimal. The mechanism behind this

preferential reactivity is believed to arise from an out-of-plane attack on the

5,6 double bond of the thymine ring (4–9). Although the ring is still intact, the

loss of aromaticity resulting from insertion of hydroxyl groups on the 5 and 6

carbons leads to a reduction in hypochromicity. Treatment of the vicinal diol

with strong base leads to ring opening and cleavage of the phosphodiester back-

bone. As with DEPC, stereochemical hindrance from base stacking prohibits

reactivity of double-stranded B-form DNA. In DNA, which is denatured or is

altered from the B form, the thymine ring becomes susceptible to modification

by potassium permanganate. Because of their base preferences, the combined

use of DEPC and potassium permanganate allows complete analysis of both

GC– and AT–containing sequences in DNA.

Diethyl pyrocarbonate and potassium permanganate modification have been

used to detect a number of distorted DNA structures in both prokaryotic and

eukaryotic cells, including open complex formation during transcription (10–14a),

steps in promoter clearance (14a–17), elongation (17–19), and termination

(20,21), RNA–DNA hybrid structures in transcription elongation complexes

(22,23), drug binding to DNA (24–26), chromatin positioning (27,28), recom-

bination events (29–31), and single-stranded binding protein binding domains

(32–34). On DNA alone, these reagents can reveal sequence-dependent distor-

tions (35–37), negatively supercoiled DNA (38–40), cruciform DNA structures

(41,42), DNA hairpins such as those found in triplet expansion diseases like

fragile X-syndrome (43,44), and Z-DNA, H-DNA, or triplex DNA (45–48).

Use of DEPC and Potassium Permanganate as Probes 65

Figures 1 and 2 give examples of the data which can be obtained. Figure 1

shows one of the more common uses of potassium permanganate, the analysis

of open promoter complex formation. Figure 2 shows results for both DEPC

and potassium permanganate acting on a stalled transcription elongation

complex, showing both the melted transcription bubble and the unreactive

RNA–DNA hybrid.

Virtually any sequence that deviates from B-form DNA is a candidate for

probing with DEPC and potassium permanganate. Because DEPC and

potassium permanganate only modify the susceptible bases without cleaving

the phosphodiester backbone, further steps need to be taken to visualize the

positions of the modified bases. There are three different methods commonly

used: The first, which can only be used for experiments performed in vitro,

utilizes 5' or 3' end-labeled DNA fragments. After treatment with DEPC or

potassium permanganate, the DNA is treated with piperidine to cleave the

Fig. 1. Potassium permanganate sensitive sites when an open promoter complex is

formed on the coding strand by RNA polymerase I from Acanthamoeba castellanii.

Lane M contains a G + A Maxam–Gilbert sequencing ladder, lane 1 contains DNA

exposed to KMnO

treatment in the absence of any proteins; lane 3 contains DNA

with the addition of proteins necessary for melting of DNA. Numerical designa-

tions refer to the transcription start site. Hypersensitive sites in bold denote regions of

strand separation.

66 Kahl and Paule

phosphate backbone on the 3' side of the modified nucleotide. This procedure

is useful when comparing the results from multiple footprinting techniques,

because the same fragment of labeled DNA may be utilized with each type of

experiment. This method also works well when examining short tracts of DNA

or DNA that does not amplify well in a thermocycler. The other two methods,

primer extension and thermocycle amplification, are used when working in

vivo (potassium permanganate only) or with circular pieces of DNA in vitro.

Thermocycle amplification is particularly beneficial when working with

limited amounts of DNA (in the low nanogram range) or when the ratio of

protein–DNA complex to DNA is low. The following materials and methods

Fig. 2. Potassium permanganate and DEPC probing of stalled transcription complex.

(A) Potassium permanganate probing of the coding strand when RNA polymerase I is

paused at +31 from the transcription start site. Lane M contains G + A Maxam-Gilbert

sequencing ladder, lane 1 contains DNA alone with KMnO

treatment; lane 2 displays

the permanganate sensitive sites when RNA polymerase from Acanthamoeba

castellanii is paused at +31 from the transcription start site. The hypersensitive sites in

bold at positions +21, +22, and +32, +33 define the transcription bubble on the coding

strand. Hyposensitive thymidines in the transcription bubble depict protection due to

the formation of an RNA-DNA hybrid. (B) DEPC probing on the noncoding strand

when RNA polymerase I is paused at +31 from the transcription start site. Lanes are the

same as in (A). DEPC-sensitive sites in bold define the leading edge of the transcription

bubble on the noncoding strand. Unreactive As in the region may be the result of

protein interference or the slightly less reactive nature of DEPC compared to KMnO