Moss Tom. DNA-protein interactions: principles and protocols

Подождите немного. Документ загружается.

DNA–Protein Photocrosslinking 327

4. Combine the fractions containing 8-bromo-dATP (8-BrdATP) (main peak of

the elution profile) and dry by lyophilization. 8-BrdATP is obtained as the

triethylammonium salt. The expected yield should be 65% (spectroscopically).

5. Dissolve 0.1 mmol (87.3 mg) of dried 8-BrdATP (triethylammonium salt) in 3 mL

of freshly distilled dimethylformamide (see Notes 3 and 4). Add a dried solution

of 0.8 mmol (34.4 mg) of hydrazoic acid (HN

) in 800 µL of benzene and

0.8 mmol (111.3 µL) of freshly distilled triethylamine. Keep the reaction mixture

in the dark at 75°C for 7 h (the absorption maximum shifts from 262 nm to 280 nm).

6. Evaporate the solvents under vacuum and redissolve the residue in 1 mL of water.

Further purification is achieved by ion-exchange chromatography over DEAE–

Sephadex A-25 as described in step 3. Figure 5 shows the elution profile of the

chromatography (see Notes 4 and 5).

7. Combine the fractions containing 8-N

dATP and dry the solution by lyophiliza-

tion. 8-N

dATP is obtained as the triethylammonium salt. Yield: 30% (spectro-

scopically). 8-N

dATP can be stored at –20°C in the dark freeze-dried (see Notes

6–8) or frozen in aqueous solution, pH 7.0.

3.2. Characterization of 8-N

dATP

1. Thin-layer chromatography (TLC). TLC is carried out on silica gel plates F

254

cellulose plates F. The development is performed in either isobutyric acid/water/

ammonia (66:33:1 v/v) or n-butanol/water/acetic acid (5:3:2 v/v).

Fig. 5. Elution profile (anion-exchange chromatography on DEAE–Sephadex A

25; elution buffer: linear gradient of 1000 mL each of water and 0.7 M triethyl-

ammonium bicarbonate (pH 7.3) of the reaction products of 8-N

dATP synthesis:

front (a), 8-N

dAMP (b), 8-BrdAMP (c), 8-N

dADP (d), 8-BrdADP (e), 8-N

dATP (f),

8-BrdATP (g), and probably a higher phosphorylated 8-azidoadenosine derivative (h).

328 Meffert et al.

2. Ultraviolet absorbance. Record the UV absorbance spectrum of 8-N

dATP. It shows a

maximum at 280 nm. The UV absorbance of 8-N

dATP is pH dependent (see Note 9).

3. Photoreactivity. The photoreactivity of 8-N

dATP is tested by two different meth-

ods. It can either be demonstrated by the spectroscopic observation of the pho-

tolysis (Fig. 6; see Note 10) or by the ability of the photolabel to bind irreversibly

to cellulose on thin-layer plates on UV irradiation (Mineralight handlamp UVSL

25) prior to the development of the chromatogram. After development, most of

the irradiated label is detected at the origin of the chromatogram in contrast to the

nonirradiated control, which has completely migrated.

3.3. Preparation of Azido-Modified DNA

Azido-modified and [

P]-labeled DNA are prepared by nick translation.

For this purpose, the detailed and exact composition of the reaction medium

depends strongly on the size as well as on the amount of the DNA to be modi-

fied. The optimal ratio of DNA, DNase I, and DNA polymerase I (Kornberg

enzyme) should be tested in preliminary experiments (see Notes 11 and 12).

Here, we describe the well-tested reaction conditions for the modification of

plasmid pBR322 (4363 bp). The preparation of azido-modified pWH106 (4970 bp)

can be performed analogously.

1. Add 17.3 pmol of pBR322 to a mixture of 50 nmol of dGTP, 50 nmol of dTTP,

50 nmol of 8-N

dATP, and 500 pmol of dCTP; prepare on ice.

2. Add 370 kBq of [α-

P]-dCTP (110 TBq/mmol) and 20 pg of DNase I (freshly

prepared out of a stock solution of 1 mg of DNase I in 1 mL of 0.15 M NaCl and

50% glycerol).

3. Adjust the reaction medium to an end concentration of 50 mM Tris-HCl, pH 7.2,

10 mM MgSO

, and 50 mg/mL of bovine serum albumin (standard reaction

volume: 100 µL).

4. Start the nick translation reaction by adding 100 U of DNA polymerase I

from E. coli.

5. Incubate for 1 h at 15°C in the dark.

6. Stop the reaction by adding EDTA (final concentration: 20 mM).

7. Separate the unincorporated deoxyribonucleotides from photoreactive

[

P]-labeled pBR322 by gel filtration over Sephadex A-25 column using a

1-mL syringe.

8. Precipitate photoreactive pBR322 by adding two volumes of cold ethanol and

redissolve the precipitate in double-distilled water. Store the aqueous solution at

–20°C in the dark.

Control nonphotoreactive DNA can be prepared analogously replacing the

8-N

dATP by 50 nmol of dATP.

3.4. Photocrosslinking

1. Prepare 20–30 µL aqueous solutions containing the photoreactive DNA (0.5 pmol)

and the protein (1–25 pmol) to be cross-linked (see Notes 13 and 14).

DNA–Protein Photocrosslinking 329

Fig. 6. Change of the optical absorption spectrum of 8-N

dATP on UV irradiation

in Tris-HCl buffer (0.01 M, pH 7.0, 20°C). The irradiation time between two subse-

quent absorption spectra was 2 min initially. It was increased up to 10 min toward the

end of photolysis. The final spectrum was taken after 30 min of irradiation. During the

photolysis, the absorbance at 280 nm decreased; two new absorbance maxima at 248

and 305 nm are formed.

2. Incubate the reaction mixture for 10 min at 37°C in the dark.

3. Expose the sample to UV irradiation (see Notes 15 and 16). The irradiation times

can be chosen in a range from 1 s to 60 min (see Note 17).

4. Keep the solutions in the dark before and after photolysis (see Note 6).

3.5. Analysis of DNA-Protein Adducts

Analysis of DNA–protein adducts can be made, for example, by polyacryla-

mide gel electrophoresis of the irradiated samples followed by autoradio-

330 Meffert et al.

graphy. SDS–polyacrylamide gel electrophoresis should be performed

immediately after photocrosslinking according to Laemmli (24) with some

variations. After the addition of 20 mg/mL of bromophenol blue, the samples

are loaded onto a SDS–polyacrylamide gel of 5% polyacrylamide (separating

gel) with an overlay of 3.5% polyacrylamide (stacking gel) containing 1% SDS.

After the electrophoretic separation, the gels are silver-stained according to

Adams and Sammons (25), dried, and exposed to X-ray film at –70°C. A quan-

titative determination of the DNA–protein adducts is possible by densitometric

measurement of the autoradiogram (26). Figure 7 shows a typical result on

photocrosslinking of EcoRI-digested plasmid pWH106 with a specific inter-

acting protein (Tet repressor).

Another possibility to detect the DNA–protein adducts is the application of

the nitrocellulose filter binding assay according to Braun and Merrick (27), see

also Chapter 1.

4. Notes

1. Experiments to synthesize [α-

P] or [U-

C]-labeled 8-N

dATP by starting the

synthesis with [α-

P] or [U-

C]dATP, respectively, failed. This is most prob-

ably because of the formation of bromine radicals induced by radiation. These

radicals could react unspecifically with the deoxyribonucleotide, suppressing

the very specific electrophilic substitution of the hydrogen in position 8 of the

adenine ring by the bromine ion.

2. Do not stop the reaction of dATP with bromine before 6 h even if the absorption

maximum is near 262 nm after 1 or 2 h, otherwise a significant reduction of the

yield of 8-BrdATP may occur.

3. 8-BrdATP obtained as triethylammonium salt is soluble in dimethylformamide

in contrast to the alkali salts of this nucleotide. This is advantageous for the

following substitution of bromine by the azido group yielding 8-N

dATP.

4. The exchange reaction of bromine by the azido group requires absolute dryness.

However, the formation of 8-N

dAMP and 8-N

dADP is usually observed,

resulting from a limited hydrolytic cleavage of 8-N

dATP (see Fig. 5).

5. Besides the three azidoadenine deoxyribonucleotides, minor amounts of unreacted

8-bromoadenine deoxyribonucleotides are eluted as well (see Fig. 5).

6. Because of the photoreactivity of azido compounds, samples containing

8-N

dATP should always be kept in the dark if possible. However, short

exposure of azido compounds to normal daylight in our laboratory never

falsified the results obtained.

7. 8-N

dATP can be stored frozen at –20°C in aqueous solution, pH 7.0, in the dark

for at least 2 yr without significant loss of photoreactivity, as demonstrated by

subsequent photocrosslinking experiments.

8. Exclude dithiothreitol from any buffers or other solutions that contain 8-N

dATP.

It is well-known that dithiothreitol reduces azido groups to the corresponding

amines (28). In addition, the UV absorbance of dithiothreitol resembles that

DNA–Protein Photocrosslinking 331

Fig. 7. Photocrosslinking of proteins to DNA (pWH106). Autoradiogram of a dena-

turing 5% SDS-polyacrylamide gel electrophoresis showing photocrosslinking of Tet

repressor to azido-activated 187-bp and 3848-bp fragments of pWH106 (radioactive

labeled by

P). Each 187-bp fragment contains two tet operator sequences, the 3848-

bp fragment none. UV irradiation of azidomodified 187-bp fragment in the presence

of Tet repressor results in a reduced migration of the 187-bp fragment because of

covalent crosslinking of the DNA to one or two Tet repressor dimers. In each of lanes

1–7, 0.06 pmol pWH106 (cleaved by EcoRI) and 20 pmol Tet repressor were applied.

Lane 1: photoactive fragments of pWH 106 without protein (30' UV); lane 2:

nonphotoactive fragments of pWH106 with Tet repressor (30' UV); lanes 3–7:

photoactive fragments of pWH106 with Tet repressor (0', 1', 4', 10', 30' UV). Frac-

tions: Origin of sample loading (a); traces of 3848-bp fragment covalently crosslinked

(unspecifically) to Tet repressor (b); 3848-bp fragment (no Tet repressor bound) (c);

187-bp fragment covalently crosslinked to two Tet repressor dimers (d); 187-bp frag-

ment covalently crosslinked to one Tet repressor dimer (e); 187-bp fragment (no Tet

repressor bound) (f).

of 8-N

dATP because of the formation of disulfide bonds by oxidation on stor-

age in aqueous solution. This results in a reduced rate of photocrosslinking by the

UV irradiation.

9. The UV absorption of 8-N

dATP shows a maximum at 280 nm. The absorbance

at 280 nm increases with decreasing pH value (see step 2 of Subheading 3.2.). A

second absorption maximum at 219 nm shifts to 204 nm in acidic solution. Both

effects are a result of the protonation at N

of the purine ring (29). The UV

absorption spectrum of 8-N

dATP resembles that of 8-N

ATP (23).

332 Meffert et al.

10. When testing its photoreactivity, take into account that the photolysis of 8-N

dATP

is pH dependent. Exhaustive irradiation in neutral solution yields new absorption

maxima at 248 and 305 nm, whereas in acidic or basic solution, the destruction of

the purine ring is observed, as indicated by the disappearance of the absorbance

between 240 and 310 nm (data not shown).

11. By analogy with 8-azidoadenine nucleotides, 8-N

dATP should prefer the syn

conformation (Fig. 8) as a result of the bulky substituent in position 8 of the

purine ring (30). This, however, seems to be contradicted by our results indicat-

ing that DNA polymerase I (E. coli) accepts 8-N

dATP in the nick translation

reaction (it has been suggested that this enzyme only interacts with 2-deoxy-

nucleoside triphosphates in the anti conformation [31]). The discrepancy may be

explained in two ways: First, the steric requirements for the binding of 8-N

dATP

by DNA polymerase I are less restrictive than assumed (32) or second, 8-N

dATP

interacts in the anti conformation with the binding site of the enzyme. This

could be demonstrated for the interaction of 8-N

ATP with the F

ATPase

from mitochondria (33).

12. The preparation of azido-modified and [

P]-labeled DNA by nick translation is

critically dependent on the ratio of DNA, DNase I, and DNA polymerase I in the

reaction medium. High concentrations of DNase I, on the one hand, result in a

very efficient incorporation rate of the azido-modified and radioactive labeled

deoxynucleotides, but on the other hand, the degradation of the DNA probes by

DNase I has to be evaluated. Application of too small amounts of DNase I results

in inefficient incorporation of the photoactivatable deoxynucleotides and in an

insufficient photocrosslinking to the interacting proteins during subsequent irra-

diation of the azido-modified DNA.

13. Tris-HCl buffer, 50 mM, pH 7.2, can be used instead of double-distilled water

without any significant effect on the photocrosslinking efficiency.

14. The amount of protein planned to be photocrosslinked to photoreactive DNA can

be varied over a wide range. Too high an excess of proteins, however, should be

Fig. 8. Conformation of adenine nucleotides. (From ref. 23.)

DNA–Protein Photocrosslinking 333

avoided because the absorbance maximum of proteins at 280 nm will lead to

inefficient crosslinking rates.

15. One way to expose the samples to UV light is to deposit the probes (typically 30–

50 µL) in plastic wells (normally used for radioimmunoassay or enzyme-linked

immunosorbent assay tests). The UV lamp is positioned directly above the

samples; thus more than one probe can be irradiated simultaneously.

16. The emitted light of the UV lamp used for photocrosslinking should not contain

light of shorter wavelengths than 300 nm because of the possibility of photo-

damaging DNA or protein. For example, the Mineralight handlamp UVSL 25

(long wave) emits UV light of mainly 366 nm. The small portion of UV light of

wavelengths between 300 and 320 nm emitted allows the photoactivation of the

azido group without any significant photodamage of DNA or protein.

17. Ultraviolet irradiation times for photocrosslinking can be chosen over a wide

range (see step 3 of Subheading 3.4.). Optimal UV irradiation rates (flux per

unit area) must be tested. In our experiments (using the Mineralight handlamp

UVSL 25 fixed in a position resulting in a fluence rate of 4 J/m

/s at the position

of the sample) DNA–protein adducts are first detectable after irradiation times of

10–30 s; irradiation periods longer than 15–20 min do not improve the yield of

photocrosslink products.

Acknowledgments

The authors thank Dr. Marianne Schüz (Wiesbaden) for editing the manu-

script. This work was supported by the Bundesministerium für Forschung

und Technologie (07QV8942) and by the Deutsche Forschungsgemeinschaft

(Scha 344/1-3).

References

1. Bayley, H. and Knowles, J. R. (1977) Photoaffinity labeling. Methods Enzymol.

46, 69–114.

2. Bayley, H. (1983) Photogenerated reagents in biochemistry and molecular biol-

ogy, in Laboratory Techniques in Biochemistry, vol. 12 (Work, T. S. and Burdon,

R. H., eds.), Elsevier, Amsterdam.

3. Schäfer, H.-J. (1987) Photoaffinity labeling and photoaffinity crosslinking of

enzymes, in Chemical Modifications of Enzymes, Active Site Studies (Eyzaguirre,

J., ed.), Ellis Horwood, Chichester, pp. 45–62.

4. Smith, K. C. (1962) Dose dependent decrease in extractability of DNA from bac-

teria (by UV-light). Biochem. Biophys. Res. Commun. 8, 157–163.

5. Shetlar, M. D. (1980) Crosslinking of proteins to nucleic acids by UV-light.

Photochem. Photobiol. Rev. 5, 105–197.

6. Welsh, J. and Cantor, C. R. (1984) Protein–DNA crosslinking. Trends Biochem.

Sci. 9, 505–508.

7. Ekert, B., Giocanti, N., and Sabattier, R. (1986) Study of several factors in RNA–

protein crosslink formation induced by ionizing radiations within 70S ribosomes

of E. coli MRE 600. Int. J. Radiat. Biol. 50, 507–525.

334 Meffert et al.

8. Lesko, S. A., Drocourt, J. L., and Yang, S. U. (1982) DNA–protein and DNA

interstrand crosslinks induced in isolated chromatin by H

. and Fe-EDTA-che-

lates. Biochemistry 21, 5010–5015.

9. Wedrychowsky, A., Ward, W. S., Schmidt, W. N., and Hnilica, L. S. (1985) Chro-

mium-induced crosslinking of nuclear proteins and DNA. J. Biol. Chem. 260,

7150–7155.

10. Summerfield, F. W. and Tappel, A. L. (1984) Crosslinking of DNA in liver and

testis of rats fed 1,3-propanediol. Chem. Biol. Interact. 50, 87–96.

11. Dose, K., Bieger-Dose, A., Martens, K.-D., Meffert, R., Nawroth, T., Risi, S., et al.

(1987) Survival under space vacuum—biochemical aspects. Proc. 3rd Eur. Symp.

Life Sci. Res. in Space (ESA SP–271), pp. 193–195.

12. Lin, S. Y. and Riggs, A. D. (1974) Photochemical attachment of lac repressor to

bromodeoxyuridine-substituted lac operator by UV radiation. Proc. Natl. Acad.

Sci. USA 71, 947–951.

13. Evans, R. K., Johnson, J. D., and Haley, B. E. (1986) 5-Azido-2'-deoxyuridine-5'-

triphosphate: a photoaffinity labeling reagent and tool for the enzymatic synthesis

of photoactive DNA. Proc. Natl. Acad. Sci. USA 83, 5382–5386.

14. Bartholomew, B., Kassavetis, G. A., Braun, B. R., and Geiduschek, E. P. (1990)

The subunit structure of Saccharomyces cerevisiae transcription factor IIIC

probed with a novel photocrosslinking reagent. EMBO J. 9, 2197–2205.

15. Lee, D. K., Evans, R. K., Blanco, J., Gottesfeld, J., and Johnson, J. D. (1991)

Contacts between 5 S DNA and Xenopus TFIIIA identified using 5-azido-2'-

deoxyuridine-substituted DNA. J. Biol. Chem. 266, 16,478–16,484.

16. Blatter, E. E., Ebright, Y. W., and Ebright, R. H. (1992) Identification of an amino

acid-base contact in the GCN4-DNA complex by bromouracil-mediated photo-

crosslinking. Nature 83, 650–652.

17. Willis, M. C., Hicke, B. J., Uhlenbeck, O. C., Cech, T. R., and Koch, T. H. (1993)

Photocrosslinking of 5-iodouracil-substituted RNA and DNA to proteins. Science

262, 1255–1257.

18. Hicke, B. J., Willis, M. C., Koch, T. H., and Cech, T. R. (1994) Telomeric protein–

DNA point contacts identified by photo-cross-linking using 5-bromodeoxyuridine.

Biochemistry 33, 3364–3373.

19. Bartholomew, B., Braun, B. R., Kassavetis, G. A., and Geiduschek, E. P. (1994)

Probing close DNA contacts of RNA–polymerase III transcription complexes

with the photoactive nucleoside 4-thiodeoxythymidine. J. Biol. Chem. 269,

18,090–18,095.

20. Meffert, R. and Dose, K. (1988) UV-induced crosslinking of proteins to plasmid

pBR322 containing 8-azidoadenine 2'-deoxyribonucleotides. FEBS Lett. 239,

190–194.

21. Meffert, R., Rathgeber, G., Schäfer, H.-J., and Dose, K. (1990) UV-induced cross-

linking of Tet repressor to DNA containing tet operator sequences and 8-azido-

adenines. Nucleic Acids Res. 18, 6633–6636.

22. Bartholomew, B., Tinker, R. L., Kassavetis, G. A., and Geiduschek, E. P. (1995)

Photochemical crosslinking assay for DNA tracking by replication proteins. Meth-

ods Enzymol. 262, 476–494.

DNA–Protein Photocrosslinking 335

23. Schäfer, H.-J., Scheurich, P., and Dose, K. (1978) Eine einfache darstellung

von 8-N

ATP: ein Agens zur Photoaffinitätsmarkierung von ATP-bindenden

Proteinen. Liebigs Ann. Chem. 1978, 1749–1753.

24. Laemmli, U. K. (1970) Cleavage of structural proteins during the assembly of the

head of bacteriophage T4. Nature 227, 680.

25. Adams, L. D. and Sammons, D. W. (1981) A unique silver staining procedure for

color characterization of polypeptides. Electrophoresis 2, 155–165.

26. Westermeier, R., Schickle, H., Thesseling, G., and Walter, W. W. (1988)

Densitometrie von gelelektrophoresen. GIT Labor-Medizin 4/88, 194–202.

27. Braun, A. and Merrick, B. (1975) Properties of UV-light-mediated binding of

BSA to DNA. Photochem. Photobiol. 21, 243–247.

28. Staros, J. V., Bayley, H., Standring, D. N., and Knowles, J. R. (1978) Reduction

of aryl azides by thiols: implications for the use of photoaffinity reagents.

Biochem. Biophys. Res. Commun. 80, 568–572.

29. Koberstein, R., Cobianchi, L., and Sund, H. (1976) Interaction of the photoaffinity

label 8-azido-ADP with glutamate dehydrogenase. FEBS Lett. 64, 176–180.

30. Vignais, P. V. and Lunardi, J. (1985) Chemical probes of the mitochondrial ATP

synthesis and translocation. Annu. Rev. Biochem. 54, 977–1014.

31. Czarnecki, J. J. (1978) Ph.D. Thesis, University of Wyoming, Laramie, WY.

32. Englund, P. T., Kelly, R. B., and Kornberg, A. (1969) Enzymatic synthesis of

DNA: binding of DNA to DNA polymerase. J. Biol. Chem. 244, 3045–3052.

33. Garin, J., Vignais, P. V., Gronenborn, A. M., Clore, G. M., Gao, Z., and Bäuerlein,

E. (1988) 1H-NMR studies on nucleotide binding to the catalytic sites of bovine

mitochondrial F

-ATPase. FEBS Lett. 242, 178–182.

Protein–DNA Photocrosslinking 337

337

From:

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

Site-Specific Protein–DNA Photocrosslinking

Analysis of Bacterial Transcription Initiation Complexes

Nikolai Naryshkin, Younggyu Kim, Qianping Dong,

and Richard H. Ebright

1. Introduction

1.1. Site-Specific Protein-DNA Photocrosslinking

In work carried out in collaboration with the laboratory of D. Reinberg

(University of Medicine and Dentistry of New Jersey), we have developed a

site-specific protein–DNA photocrosslinking procedure to define positions of

proteins relative to DNA in protein–DNA and multiprotein–DNA complexes

(1–3). The procedure has four parts (Fig. 1):

1. Chemical (4–6) and enzymatic (7) reactions are used to prepare a DNA fragment

containing a photoactivatible crosslinking agent and an adjacent radiolabel

incorporated at a single, defined DNA phosphate (with a 9.7 Å linker between

the photoreactive atom of the crosslinking agent and the phosphorus atom of the

phosphate, and with an approximately 11 Å maximum “reach” between potential

crosslinking targets and the phosphorus atom of the phosphate).

2. The multiprotein–DNA complex of interest is formed using the site-specifically

derivatized DNA fragment, and the multiprotein–DNA complex is ultraviolet

(UV)-irradiated, initiating covalent crosslinking with proteins in direct physical

proximity to the photoactivatible crosslinking agent.

3. Extensive nuclease digestion is performed, eliminating uncrosslinked DNA and

converting crosslinked DNA to a crosslinked, radiolabeled 3–5 nucleotide “tag.”

4. The “tagged” proteins are identified.

The procedure is performed in systematic fashion, with preparation and

analysis of at least 10 derivatized DNA fragments, each having the photo-

activatible crosslinking agent incorporated at a single, defined DNA phosphate