Moss Tom. DNA-protein interactions: principles and protocols

Подождите немного. Документ загружается.

Footprinting with Exonuclease III 45

nondenaturing acrylamide gel, as described in Subheading 3.1., provided the

complexes show different gel mobilities and have half-lives long enough to

survive the electrophoresis. Such a purification step requires the use of 10 times

more radioactively labeled DNA.

The procedure is as follows:

1. Form the complex.

2. Subject the complex to digestion with Exo III according to Subheading 3.3.,

step 2.

3. Dialyze the complex by drop dialysis against a low salt buffer (e.g., 10 mM

Tris-HCl, pH 7.9) in order to avoid salt effects during electrophoresis (Subhead-

ing 3.1., step 5).

4. Apply the complex to a nondenaturing gel as described in Subheading 3.1., steps

6 and 7. The half-life of the complex is in most cases not changed by the Exo III

digestion of the DNA.

5. Expose the gel at –70°C to X-ray film using an intensifying screen. A 1-h expo-

sure should be enough to recognize the complexed bands. If not, the recovered

DNA will be insufficient for the subsequent sequencing gel analysis.

6. Before removing the film for development mark exactly its position on the gel.

7. Excise the complex bands of interest with a spatula. The band representing the

free DNA will be visible as a smear after Exo III digestion.

8. To elute the complexed DNA, put the excised gel slice in 600 µL of bidistilled

water. Heat the complex to 90°C for 3 min and shake overnight at room tempera-

ture. The effectiveness of the elution can be easily monitored by comparing the

radioactivity of the eluate with the radioactivity of the gel slice.

9. Vacuum-dry the eluate in a SpeedVac concentrator.

10. Dissolve the pellet in 10 µL formamide buffer and spin down the gel residue.

11. Transfer the supernatant to a new Eppendorf tube and apply the sample to a

sequencing gel as described in Subheading 3.2., steps 7–9.

4. Notes

1. The gel concentration has to be adjusted according to the molecular weight of the

protein–DNA complex. Here we describe the conditions established for the study

of the E. coli RNA polymerase (MW 455,000) and a DNA fragment of 130 bp

carrying a promoter (9). For some applications, another widely used non-

denaturing gel system may be appropriate: 1X TBE buffer, 4% acrylamide, and

0.1% bis-acrylamide. Recirculation of the buffer is not necessary here.

2. It has been observed that many batches of commercially available Exo III contain an

activity that removes the 5' label. A 5'-phosphatase or a 5'–3' exonuclease activ-

ity could account for this phenomenon. Filling in the 5' protruding ends using α-

thio-dNTPs as described by some authors (10,11) may eliminate the problem.

Addition of E. coli tRNA can reduce the effect, but will not completely avoid it.

46 Metzger and Heumann

3. Investigation of the complexes of specific binding of proteins and DNA in crude

extracts using Exo III requires additional precautions in order to avoid problems

caused by endogenous nuclease activities during Exo III exposure. To avoid this

problem, sodium-phosphate, tRNA, deoxyoligonucleotides and fragmented

phage DNA (e.g., 2 mM sodium phosphate, 1 µg of FX 174 DNA cut with HaeIII,

10 µg of yeast tRNA, and 1 µg mixed p[dN]

) should be added to the assay (5).

We find this suppresses nuclease and possibly phosphatase activities contained

in the crude extracts (see also Note 2).

4. Testing different concentrations of Exo III and different incubation periods can pro-

vide additional information about the nature of the protein–DNA complex under

study. If Exo III is able to “nibble” into a protected area with increasing exposure

time, this indicates differences in the strength of protein–DNA interaction (10).

5. Different binding sites for one or more proteins may be detected as distinct stop

points for Exo III, as shown in refs. 12–14. This applies as much when working

with crude extracts as when using purified factors. It is necessary, however, that

the ratio of DNA to binding proteins be >1.

6. Heparin, which is often used as a DNA competitor for E. coli RNA–polymerase

and other DNA-binding proteins, also interacts with Exo III and reduces its activ-

ity markedly.

References

1. Rogers, S. G. and Weiss, B. (1980) Exonuclease III of Escherichia coli K-12, an

AP endonuclease. Methods Enzymol. 65, 201–211.

2. Shalloway, D., Kleinberger, T., and Livingston, D. M. (1980) Mapping of SV 40

DNA replication origin region binding sites for the SV 40 DNA replication anti-

gen by protection against Exonuclease III digestion, Cell 20, 411–422.

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

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

4. Pavco, P. A. and Steege, D. A. (1990) Elongation by Escherichia coli RNA poly-

merase is blocked in vitro by a site specific DNA binding protein. J. Biol. Chem.

265, 9960–9969.

5. Wu, C. (1985) An exonuclease protection assay reveals heat-shock element and

TATA box binding proteins in crude nuclear extracts. Nature 317, 84–87.

6. Loh, T. P., Sievert, L. L., and Scott, R. W. (1990) Evidence for a stem cell-spe-

cific repressor of Moloney murine leukemia virus expression in embryonic carci-

noma cells. Mol. Cell. Biol. 10, 4045–4057.

7. Carnevali, F., La Porta, C., Ilardi, V., and Beccari, E. (1989) Nuclear factors spe-

cifically bind to upstream sequences of a Xenopus laevis ribosomal protein gene

promoter. Nucleic Acids Res. 17, 8171–8184.

8. Fried, M. and Crothers, D. M. (1981) Equilibria and kinetics of lac repressor-

operator interactions by polyacrylamide gel electrophoresis. Nucleic Acids Res. 9,

6505–6525.

9. Heumann, H., Metzger, W., and Niehörster, M. (1986) Visualization of interme-

diary transcription states in the complex between Escherichia coli DNA-depen-

Footprinting with Exonuclease III 47

dent RNA polymerases and a promoter-carrying DNA fragment using the gel

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

10. Straney, D. C. and Crothers, D. M. (1987) A stressed intermediate in the formation

of stably initiated RNA chains at the Escherichia coli lac UV 5 promoter. J. Mol.

Biol. 193, 267–278.

11. Straney, D. C. and Crothers, D. M. (1987) Comparison of the open complexes

formed by RNA polymerase at the Escherichia coli lac UV 5 promoter. J. Mol.

Biol. 193, 279–292.

12. Gaur, N. K., Oppenheim, J., and Smith, I. (1991) The Bacillus subtilis sin gene, a

regulator of alternate developmental processes, codes for a DNA-binding protein.

J. Bact. 173, 678–686.

13. Owen, R. D., Bortner, D. M., and Ostrowski, M. C. (1990) ras Oncogene activa-

tion of a VL30 transcriptional element is linked to transformation. Mol. Cell. Biol.

10, 1–9.

14. Wilkison, W. O., Min, H. Y., Claffey, K. P., Satterberg, B. L., and Spiegelman, B.

M. (1990) Control of the adipisin gene in adipocyte differentiation. J. Biol. Chem.

265, 477–482.

Further Reading

Kow, Y. W. (1989) Mechanism of action of Escherichia coli Exonuclease III.

Biochemistry 28, 3280–3287.

Hydroxyl Radical Footprinting 49

From:

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

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

Hydroxyl Radical Footprinting

Evgeny Zaychikov, Peter Schickor, Ludmilla Denissova,

and Hermann Heumann

1. Introduction

The basic principle of the DNA footprinting technique is the measurement

of accessibility of the DNA using a probe. The probe can be any enzyme or a

chemical reagent that is able to cut the DNA backbone. When the target DNA

is a fragment containing a signal sequence for a sequence-specific binding

protein, sites on the DNA that interact with the protein are inaccessible to the

probe. After electrophoretic separation based on molecular weight, these inac-

cessible sites appear as blanks in an otherwise regular DNA cleavage pattern,

thus revealing the characteristic interaction footprint for the binding protein.

Although the footprinting pattern is a characteristic of the protein–DNA

interaction, it is also greatly affected by the type of probe used. Hydroxyl radi-

cals provide DNA footprinting probes, which are very convenient to handle

and are distinguished by a number of distinct advantages:

1. Hydroxyl radicals cut the DNA with almost no sequence dependence.

2. Because the probe is very small, the resolution of the footprint is very high (1 bp).

3. The cleavage reaction is effective over a wide range of buffer compositions, salt

concentrations, pHs, and temperatures. Only glycerol, a radical scavenger, inter-

feres with the cutting when present at concentrations higher than 0.5%.

4. All chemicals needed are easily available and uncomplicated in their handling.

1.1. Generation and Action of Hydroxyl Radicals

Hydroxyl radicals are generated according to the Fenton reaction by reduc-

tion of iron(II) with hydrogen peroxide as follows:

ascorbate or

diothiothreited

(EDTA

4–

) + H

→

(EDTA

4–

) + OH·

50 Zaychikov et al.

The resulting iron(III) is reduced by ascorbate or dithiothreitol back to

iron(II), which can start a new cycle. The use of a negatively charged

[Fe(EDTA)]

2–

complex prevents the iron from interacting electrostatically with

DNA, so the only reactant interacting with DNA is the hydroxyl radical gener-

ated in solution (1).

An alternative source of hydroxyl radicals is potassium peroxonitrite (2).

The radicals are generated via its conjugate acid (ONOOH) when adding a

stable alkaline solution of ONOOK in samples buffered at neutral pH:

ONOOH → NO

· + OH·

2NO

· → N

+ H

O → NO

–

+ NO

–

+ 2H

The exact manner in which hydroxyl radicals act on DNA is still not known.

The radicals are thought to abstract an H-atom from the sugar moiety of the

DNA backbone, and secondary reactions of the resulting sugar radical cause

the backbone to break, leaving a gap in one strand of the double helix with

the phosphate groups on either side (3).

1.2. Principle of the Procedure

After formation of the complex of a sequence-specific binding protein and a

DNA fragment carrying the binding sequence (see Fig. 1), the complex is sub-

Fig. 1. A putative DNA-binding protein interacts with the DNA over three helical

turns. The major portion of the protein interacts with only one side of the DNA over

two helical turns. A minor portion of the protein wraps fully around the DNA. The

asterisk indicates the position of the radioactive label.

Hydroxyl Radical Footprinting 51

jected to hydroxyl radical treatment. Hydroxyl radicals introduce single-base

deletions randomly distributed in the DNA. The concentration of the hydroxyl

radicals is adjusted so that the yield of deletions is less than one per DNA such

that approx 10% of the DNA fragments are affected. Cutting of the DNA is

prevented at those sites on the DNA where the protein is bound. This partially

cut DNA is applied to a sequencing gel. If the DNA is detected by a unique

terminal radioactive label, a DNA ladder is produced that is similar to that

obtained by sequence analysis. Blanks within this regular ladder indicate the

sites where the protein is bound. This footprint becomes more evident if a ref-

erence DNA is included that has been subjected to the same procedure but

without previous protein binding. If complex formation is incomplete (i.e., the

assay contains free DNA), the footprint will be masked by apparent cleavage

within the protein-binding site. This can be avoided by separating the hydroxyl-

radical-treated protein DNA complex from free DNA by nondenaturing gel

electrophoresis (Fig. 2) or by nitrocellulose filter binding before application to

a sequencing gel. Figure 3 shows schematically the footprinting pattern of the

protein DNA complex depicted in Fig. 1.

1.3. Interpretation of the Footprinting Pattern

Figure 3 shows that a footprinting analysis of both DNA strands includes

six DNA ladders (i.e., two DNA sequencing reaction length standards, two

free DNAs as reference, and two complexed DNAs). Blanks in the DNA lad-

der indicate exclusion of radical attack of the DNA because of the presence of

the bound protein. These blanks can be assigned to specific sequence positions

via the length standards.

Fig. 2. Protein-DNA complexes are separated from free DNA by nondenaturing gel

electrophoresis. The two lanes represent the same complex with a single label at one

end of the DNA, at the 3' end, or at the 5' end. Hydroxyl radicals create gaps in the

DNA, which cause a significant retardation of the modified fragments within the gel.

This effect is visible only in the free DNA and is indicated by the shaded shoulder on

the lower band.

52 Zaychikov et al.

The following information can be extracted from the hydroxyl radical

footprinting pattern:

1. The total size of the DNA sequence interacting with the protein can be deter-

mined from the extent of the uncleaved or blank positions.

2. A variation of the intensity of the bands within the interacting sequence reflects

differences in the modes of interaction. By a comparison of the footprints on both

strands, the modes of interaction can often be interpreted:

a. If both strands show a blank at the same region, it indicates that the protein

wraps around the DNA.

b. A blank on only one strand indicates single-strand formation, with one strand

protected by interaction with the protein.

Fig. 3. The bands of the nondenaturing gel in Fig. 2 containing the complex and the

free DNA are eluted and applied (under denaturing conditions) to a sequencing gel.

Lanes 1 and 4 show the free DNA labeled respectively at the 3' and the 5' ends. Lanes

2 and 3 show the DNA recovered from the complex. The complex depicted schemati-

cally in Fig. 1 would result in the footprint displayed in lanes 2 and 3. Lanes G contain

the length standards obtained by a G-specific Maxam–Gilbert sequencing reaction.

Hydroxyl Radical Footprinting 53

c. Modulation of the intensity of the bands with a regular phasing according to

the helix repeat (e.g., 10.3 bp for B-DNA) indicates binding of the protein to

one side of the DNA. This interpretation is supported if the complementary

strand shows the same pattern but with an offset of two or three bases. This

offset is a consequence of the double-helical nature of the DNA, as shown

schematically in Fig. 1.

1.4. Examples of the Application

of Hydroxyl Radicals as Footprinting Probes

1.4.1. Protein DNA Complexes

Numerous Fe

-dependent hydroxyl radical footprinting studies were per-

formed on protein–DNA complexes. Most useful were those studies of protein–

DNA contacts within the transcription machinery.

1. Hydroxyl radicals were used to follow the formation of the transcriptionally

active complex between the DNA-dependent RNA polymerase of E. coli and

T7A1 promoter (4). Temperature-dependent footprinting studies showed that the

transcription initiation complex undergoes three different conformations charac-

terized by a specific “footprint” until a transcription competent complex is

formed. These conformations could be attributed to the so-called closed,

intermediate, and open complex. A bent conformation of DNA in complex with

DNA was concluded from the OH radical probing of the lambda P

promoter

complex (5).

2. Site-specific cleavage of the DNA of a RNA polymerase binary complex by both

free and EDTA-chelated Fe

was detected in absence of Mg

ions (6,7). A phe-

nomenon in FeEDTA-dependent OH radical footprints is hypercleavage of DNA

which can be observed in footprints of Mg

-dependent proteins such as E. coli

RNA polymerase (6) or HIV reverse transcriptase (8). The E. coli RNA poly-

merase footprint contains a hyperreactive cleavage spot within the protected

region. This spot could be attributed to a site-specific OH radical cleavage mecha-

nism. This view was supported by footprinting studies using peroxonitrite as an

alternative reagent for generating OH radicals. Using this latter reagent, no

hyperreactive spot was visible in the OH radical footprint of E. coli RNA poly-

merase, indicating that specifically bound FeEDTA probe was responsible for

the hyperreactive cleavage. Free Fe

ion can replace the catalytically active Mg

at the polymerization site of RNA polymerase, being chelated with aspartates of

the active site (7). The chelated Fe

generates OH radicals probably according

to a mechanism that is analogous to the Fenton reaction and causes a strong local

cleavage of both DNA and protein. This hyperreactivity was also observed in

-dependent OH radical footprints of reverse transcriptase (RT) of the human

immunodeficiency virus (HIV-RT). The catalytically active Mg

of the RNaseH

active site of HIV-RT was replaced by Fe

, leading to site-specific OH radical

cleavage of the DNA. It is interesting to note that no hyperreactive cleavage was

observed at the other Mg

-carrying active site of HIV-RT (i.e., at the polymer-

54 Zaychikov et al.

ization site). These different effects of Mg

/Fe

substitution in the polymeriza-

tion sites of HIV-RT and E. coli RNA polymerase probably reflect a variation of

the redox potentials of the two sites.

3. The movement of E. coli RNA polymerase during mRNA synthesis was followed by prob-

ing a series of specifically arrested transcribing complexes with hydroxyl radicals (9–11).

1.4.2. Antibiotic DNA Complexes

Mithramycin, a small antitumor antibiotic drug, was shown to bind to the

minor groove of GC-rich DNA sequences, thereby protecting only three bases

from hydroxyl radical attack (12).

1.4.3. DNA Structures

The accessibility of bent DNA was studied using hydroxyl radicals. The

bend was induced by A tracts repeated in phase with the helical repeat (13).

Hydroxyl radicals can also be used to measure the number of base pairs per

helical turn along any DNA molecule. The DNA is adsorbed onto crystalline

calcium phosphate before being subjected to radical treatment. From the varia-

tion of the intensity of the bands, the helical periodicity of the DNA can be

directly obtained (14).

1.4.4. RNA Protein Complexes

Splicing-specific ribonucleoprotein complexes were analyzed by hydroxyl

radical treatment. These studies revealed that several regions of the 3'-splice

site of mRNA precursors are not accessible for hydroxyl radicals, for example,

the 3'-intron/exon junction, the polypyrimidine tract, and the site of branch

formation were found to be all inaccessible (15).

1.4.5. RNA Structures

By using hydroxyl radicals, Celander and Cech (16) demonstrated that at least

three magnesium ions are necessary for formation of a catalytically active ribozyme

RNA molecule. By using both FeEDTA and ONOOK-generated hydroxyl radi-

cals, information on higher-order structures of tRNA was obtained (2).

2. Materials

2.1. The Cutting Reaction

Prepare the following solutions separately (see Note 1):

1. 0.1 M dithiotreitol (DTT).

2. 1% Hydrogen peroxide.

3. Iron(II)–EDTA mix: Mix equal volumes of 2 mM ammonium iron(II) sulfate

hexahydrate ((NH

)

Fe(SO

)

·6H

O) and 4 mM EDTA.

4. Stop mix: 4% glycerol, 0.6 M sodium acetate, 0.1 mg/mL carrier DNA.

Hydroxyl Radical Footprinting 55

2.2. The Sequencing Gel

1. Urea (ultra pure).

2. 20X TBE: 1 M Tris-base, 1 M boric acid, and 20 mM EDTA.

3. Acrylamide solution: 40% acrylamide and 0.66% bis-acrylamide (see Note 2).

4. 10% Sodium persulfate (see Note 3).

5. 10% N,N,N',N'-tetramethylethylene diame (TEMED).

6. Sequencing gel (8%): 21 g urea, 2.5 mL of 20X TBE, and 10 mL of the 40% acrylamide

solution are made up to 50 mL with bidistilled H

O and stirred under mild heating until

urea is dissolved. The solution is filtered (filter pore size: 0.2 µm) and degassed for

5 min. Immediately before pouring the solution between the glass plates, add 0.3 mL

of 10% sodium persulfate and 0.3 mL of 10% TEMED.

7. Loading buffer for the sequencing gel (stock solution): 100 mL forma-

mide (deionized), 30 mg xylenecyanol FF, 30 mg bromophenol blue, and

750 mg EDTA.

8. Electrophoresis buffer: 1X TBE.

2.3. The Nondenaturing Gel for DNA Isolation

and Electrophoretic Mobility Shift Assay

1. 20X TBE.

2. Acrylamide solution: 30% acrylamide, and 0.8% bis-acrylamide (see Note 2).

3. 10% Sodium persulfate (see Note 3)

4. 10% TEMED (aqueous solution).

5. 3% Nondenaturing gel: 1.5 mL of 20X TBE, 3 mL of the acrylamide solution,

and 25.5 mL of bidistilled water are mixed and degassed for 5 min. Before pour-

ing the solution between the glass plates add 300 µL of 10% ammonium persulfate

and 300 mL of 10% TEMED.

6. Loading buffer for the nondenaturing gel (stock solution): 50% glycerol and 0.1%

bromophenol blue.

7. Electrophoresis buffer: 1X TBE.

2.4. Other Items

1. Sequencing gel apparatus.

2. Apparatus for nondenaturing gel electrophoresis.

3. Filters for drop dialysis, VS 0.025 µm (Millipore, Bedford, MA).

4. Filtration device (optional).

5. Nitrocellulose filters (BA85, Schleicher & Schüll) (optional).

6. Peristaltic pump (optional).

3. Method

3.1. Establishing the Conditions for Obtaining Optimum Yield

of Specific Protein–DNA Complexes

The method of establishing the conditions for complex formation using the

band-shift assay is described in Chapter 4, see also Chapter 2.