Moss Tom. DNA-protein interactions: principles and protocols

Подождите немного. Документ загружается.

316 Plyte and Kneale

generated, possessing increased binding affinity, in human replication protein A

(5,6). In addition to its use for the analysis of domain structure, limited proteolytic

fragments from Escherichia coli DNA gyrase B, for example, permitted the suc-

cessful crystallization and structure determination of one of its domains (7).

1.1. Strategy

The strategy adopted for the limited proteolysis of nucleoprotein complexes

can be considered in four parts: optimization of the proteolysis, characteriza-

tion of the proteolysed complex, purification of the DNA-binding domains,

and sequence characterization of the fragment(s).

1.1.1. Proteolysis of Nucleoprotein Complex

The nucleoprotein complex should be digested with various proteases to

establish which conditions are optimal for generating a protease-resistant

domain. We routinely vary two parameters (enzyme/substrate ratio and time of

digestion) when determining the best conditions for limited proteolysis.

However, other parameters, such as temperature, ionic strength, and pH may

also be varied. To determine the appropriate enzyme/substrate ratio for a par-

ticular protease, the nucleoprotein complex is digested at several enzyme/sub-

strate ratios, removing samples at regular time intervals for sodium dodecyl

sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) analysis. The appear-

ance of a discrete domain, resistant to further degradation (even if only tran-

siently), is evidence for the existence of a domain, although not necessarily

one that binds DNA. Choice of protease is often critical (see Table 1). Ini-

tially, it is best to try a relatively nonspecific enzyme (e.g., papain) because

this decreases the likelihood of activity being dependent on primary sequence

rather than tertiary structure.

1.1.2. Preliminary Characterization of DNA-Binding Properties

of the Proteolyzed Nucleoprotein Complex

An initial indication of DNA binding can be found during the proteolysis

experiment by removing two aliquots for gel analysis that can be run on poly-

acrylamide or agarose gels appropriate for the size of complex in the presence

and absence of the denaturant SDS. A retardation in the mobility of the DNA

(seen under ultraviolet [UV] light) in the absence of SDS implies that the frag-

ment is still associated with DNA and constitutes a DNA-binding domain.

However, this does not prove that the proteolyzed fragment is a discrete DNA-

binding domain; it is possible that the nucleoprotein complex has only been

“nicked” by the protease and maintains its native tertiary structure by noncova-

lent interactions. Therefore, it is necessary to purify the domain and fully char-

acterize its DNA-binding properties.

Limited Proteolysis of Protein–Nucleic Acid 317

1.1.3. Purification of the DNA-Binding Domain

Purification of the fragment can make use of the fact that it will still be

associated with DNA. Ultracentrifugation of the proteolyzed nucleoprotein

complex (if large fragments of DNA are used) concentrates the domain and

removes residual protease and small proteolytic fragments. The proteolyzed nucle-

oprotein complex can then be dissociated and the domain further purified if

necessary. Alternatively, the DNA-binding fragment can be purified by affinity

chromatography on DNA agarose. Several techniques are available to deter-

mine whether the purified domain binds DNA (discussed in several chapters)

and include gel retardation assay, a variety of footprinting techniques, fluores-

cence spectroscopy, and circular dichroism.

1.1.4. Determination of the Amino Acid Sequence of the Domain

N-Terminal sequencing and amino acid analysis of the purified DNA-bind-

ing domain should be sufficient to establish the sequence of the domain, if the

Table 1

Useful Enzymes for Limited Proteolysis

Enzyme Substrate specificity Inhibitors

α-Chymotrypsin Preferentially cuts C-terminally Aprotinin, PMSF, DFP,

to aromatic amino acids TPCK, cymostatin

Elastase Cuts C-terminally to aliphatic PMSF, DFP

noncharged amino acids

(e.g., Ala, Val, Leu, Ile, Gly, Ser)

Endoproteinase Arg-C Cuts C-terminally to arginine residues DFP, TLCK

Endoproteinase Lys-C Cuts C-terminally to lysine residues Aprotinin, DFP

Papain Nonspecific protease but shows some PMSF, TPCK, TLCK,

preference for bonds involving leupeptin, heavy

Arg, Lys Gln, His, Gly, and Tyr metal ions

Pepsin Nonspecific protease Pepstatin

Subtilisin Nonspecific protease DFP, PMSF

Trypsin Cuts C-terminally to lysine DFP, PMSF, TLCK

and arginine residues

Endoproteinase Glu-C Cuts C-terminally to glutamic acid DFP

(V8 protease) and or aspartic acid residues

Abbreviations used: DFP, diisopropyl fluorophosphate (extremely toxic!); PMSF, phenyl-

methyl sufonyl fluoride; TPCK, N-tosyl-l-phenylalanine chloromethyl ketone; TLCK, Nα-p-

tosyl-l-lysine chloromethly ketone.

Will cut C-terminally to glutamic acid residues in ammonium bicarbonate, pH 8.0, or ammo-

nium acetate, pH 4.0; will cut C-terminally to glutamic and aspartic acid residues in phosphate

buffer, pH 7.8.

318 Plyte and Kneale

native amino acid sequence is known. Alternatively, N-terminal sequencing

and mass spectroscopy should enable unambiguous identification of the

domain. If certain proteases have been used to generate the domain (e.g.,

trypsin, α-chymotrypsin, endoproteinase Arg-C, and so forth), the C-terminal

amino acid may also be known. If there are still ambiguities, carboxypeptidase

digestion of the fragment can also be used to help identify the C-terminal resi-

dues, although this is not always reliable. If this still does not yield an unam-

biguous result, one must resort to amino acid sequencing of the entire fragment.

2. Materials

1. Spectra-Por dialysis membrane washed thoroughly in double-distilled water.

2. All proteases should be of the highest grade available and treated for contaminat-

ing protease activity, if necessary. A list of useful enzymes and their inhibitors is

given in Table 1.

3. Buffers should be AnalR grade or higher and made up in double-distilled water.

4. SDS–polyacrylamide gel stock solutions:

Solution A: 152 g acrylamide and 4 g bis-acrylamide. Make up to 500 mL.

Solution B: 2 g SDS and 30 g Tris base, pH 8.8. Make up to 500 mL.

Solution C: 2 g SDS, 30 g Tris base, pH 6.8. Make up to 500 mL.

When making up these solutions, they should all be degassed and filtered using a Buchner

filter funnel. They should be stored in lightproof bottles and will keep for many months.

6. 10% ammonium persulfate (APS): dissolve 0.1 mg in 1 mL of dH

7. 15% SDS–polyacrylamide gel: Mix together 8.0 mL of solution A, 4.0 mL of

solution B, and 3.9 mL of dH

O. Add 150 µL of 10% APS and 20 µL of N,N,N',N'-

tetramethylethylene diamine (TEMED). Mix well and then pour between the

plates. Immediately place a layer of dH

0 (or butanol) on top of the gel to create

a smooth interface with the stacking gel. When the resolving gel has set, pour

off the water and prepare the stacking gel. This is done by adding 750 µL of

solution A and 1.25 mL of solution C to 3.0 mL of dH

O. Finally, add 40 µL

of APS and 10 µL of TEMED, pour on the stacking gel, and insert the comb.

Remove the comb as soon as the gel has set to avoid the gel sticking to the comb.

8. 10X SDS running buffer: 10 g SDS, 33.4 g Tris base, and 144 g glycine made up to 1 L.

9. High-methanol protein stain: technical-grade methanol 500 mL, 100 mL glacial

acetic acid and 0.3 g PAGE 83 stain (Coomassie blue), made up to 1 L.

10. Destain solution: 100 mL methanol and 100 mL glacial acetic acid, made up to 1 L.

11. 2X SDS-PAGE loading buffer: 4% (w/v) SDS, 60 mM Tris-HCl, pH 6.8, 20%

glycerol, 0.04% (w/v) bromophenol blue, and 1% (v/v) β-mercaptoethanol.

12. 6X agarose gel loading buffer: 0.25% (w/v) bromophenol blue, 0.25% (w/v)

xylene cyanol, and 30% glycerol.

13. 6X agarose gel loading buffer plus SDS: as in item 12 plus 12% SDS (w/v).

14. TE buffer: 10 mM Tris-HCl, pH 7.5, and 1 mM EDTA.

15. 5 M NaCl or MgCl

(or other concentrated salt solutions for dissociation of the

nucleoprotein complex [e.g., NaSCN]).

Limited Proteolysis of Protein–Nucleic Acid 319

3. Methods

The method given here covers the first three objectives outlined in Sub-

heading 1.1. Experimental details for the determination of the amino acid

sequence of the fragment can be found in any standard text on protein

chemistry. The following protocol was used for the generation of an 11-kDa

DNA-binding domain from the Pf1 gene 5 protein (8). This protein binds

cooperatively to ssDNA to produce a nucleoprotein complex of several million

Daltons. Different nucleoprotein complexes will require different conditions

of digestion and purification, but the basic principles remain the same.

3.1. Limited Proteolysis

1. Dialyze the nucleoprotein complex into the appropriate digestion buffer (see the

manufacturer’s recommendations for the buffer, temperature of reaction, and

inhibitor). We routinely digest the nucleoprotein complex at approx 1 mg/mL,

but the concentration is not too critical.

2. Prepare 40 tubes containing 5 mL of 2X SDS loading buffer plus 1 µL of the

appropriate protease inhibitor. Leave on ice.

3. Pipet 55 µL of the nucleoprotein complex (55 mg) into each of four tubes labeled

1:100, 1:200, 1:500, 1:1000, respectively. Place on ice until needed.

4. Dissolve the protease in digestion buffer to a concentration that will give an

enzyme/substrate ratio of 1:100 (w/w) when 1 µL of the protease is added to

50 µL of nucleoprotein complex (i.e., 0.5 mg/mL).

5. Prepare three dilutions of the protease. In this case, the protease is diluted 1:2,

1:5, and 1:10 with digestion buffer that will result in a final enzyme substrate

ratio of 1:200, 1:500, and 1:1000 (w/w).

6. Remove 5 µL of the nucleoprotein complex from each of the four tubes and add

to 1 of the 40 tubes containing 2X loading buffer (plus inhibitor) and place on

ice. This is the time = 0 tube and should be labeled accordingly.

7. Add 1 µL of the protease to the appropriate nucleoprotein solution (e.g., pro-

tease diluted 1:5 to the nucleoprotein solution marked 1:500) and incubate at

the specified temperature.

8. Remove 5 µL samples every 15 min and add to 2X loading buffer (in the appro-

priately marked tube) and then place on ice.

9. At the end of the experiment, boil the samples and run an SDS polyacrylamide

gel. The presence of a degraded fragment(s), resistant to further proteolysis, is

evidence for a discrete domain (see Note 1).

10. Adjustment of the enzyme/substrate ratios, time course, and choice of enzymes is

often necessary. The optimum conditions must be found by trial and error.

3.2. Purification of the DNA-Binding Domain

1. Digest a large quantity (several milligrams) of the nucleoprotein complex under

the optimized conditions determined in Subheading 3.1. to produce the DNA-

binding domain (see Note 2). Add the appropriate inhibitor and run a sample on

SDS-PAGE to check the digestion.

320 Plyte and Kneale

2. For very large nucleoprotein complexes, the proteolyzed complex can be puri-

fied away from the protease and small proteolytic fragments by ultracentrifuga-

tion. Spin the nucleoprotein complex at 229,000g (Beckman 70.1 Ti rotor)

for 3 h at 4°C (see Note 3). Carefully wash the centrifuge tube with 4 mL of TE

buffer, discard the washings, and resuspend the nucleoprotein complex in 2 mL

of TE buffer on ice. Another ultracentrifugation step can be performed to remove

all traces of the protease. For smaller nucleoprotein complexes, the DNA can

be immobilized on a large resin (e.g., DNA cellulose) prior to interaction with

the DNA-binding protein. Low-speed centrifugation can then be used to purify

the DNA-associated domain. Sometimes, limited proteolysis can generate sev-

eral fragments that bind DNA. These may arise from the same region of the pro-

tein; if so, this can be overcome by allowing the proteolysis to proceed further or

by increasing the amount of protease.

3. Dissociate the proteolyzed nucleoprotein complex by the addition of salt to the

appropriate concentration (see Note 4). The DNA can then be removed by ultra-

centrifugation (if sufficiently large) or nuclease digestion. If the DNA was

originally bound on a solid support, then it can be removed by low-speed centri-

fugation (see Note 5).

4. Remove the high-salt buffer by desalting or dialysis. If the sample contains sev-

eral different domains or a residual undigested protein, it will be necessary to

purify the domains to homogeneity. Various chromatographic techniques are

available to further purify the domains, including chromatofocusing, ion

exchange, affinity, and gel filtration chromatography. These techniques permit

recovery of the domain in a native state for further biochemical analysis. Alterna-

tively, if the fragment is only to be used for sequence analysis, the mixture can be

applied to a C

reverse-phase high-performance liquid chromatography column

and separated in an acetonitrile gradient.

5. If the sequence of the native protein is known, then the sequence of the DNA-

binding domain can be established by N-terminal sequencing and amino acid

analysis. Additionally, the mass of the fragment (determined by mass spectros-

copy) should help locate the sequence of the DNA-binding domain.

4. Notes

1. Often during the experiment, a protease-resistant fragment is only transiently

formed during complete digestion of the protein. If this occurs, vary some of the

parameters (enzyme dilution, temperature, etc.) to try and prolong the lifetime of

the fragment.

2. Scaling up of the digestion is not generally a problem and we routinely digest

several milligrams (>10) of nucleoprotein complex if necessary.

3. The speed and duration of centrifugation will vary depending on the size of the

nucleoprotein complex. For smaller complexes, ultracentrifugation may not be

appropriate.

4. In many cases, a NaCl concentration between 1 M and 2 M is sufficient to disso-

ciate the nucleoprotein complex. However, some nucleoprotein complexes

Limited Proteolysis of Protein–Nucleic Acid 321

remain associated above 2 M NaCl and require 1 M MgCl

or 1 M NaSCN for

dissociation (8). The appropriate salt concentration can be determined by SDS-

PAGE analysis of the pellet and supernatant after ultracentrifugation at different

ionic strengths.

5. The DNA can also be removed by DNase digestion followed by gel fitration (i.e.,

desalting column) or by extensive dialysis against TE buffer.

References

1. Vita, C., Dalzoppo, D., and Fontana, A (1987) Limited proteolysis of globular

proteins: molecular aspects deduced from studies on thermolysin, in Macromo-

lecular Biorecognition (Chaiken, I., Chaiancone, E., Fontana, A., and Veri, P.,

eds.), Humana Press, Clifton, NJ.

2. Merril, B., Stone, K., Cobianchi, F., Wilson, S., and Williams, K. (1988)

Phenylalanines that are conserved among several RNA-binding proteins form part

of a nucleic acid binding pocket in the heterogeneous nuclear ribonucleoprotein.

J. Biol. Chem. 263, 3307–3313.

3. Bandziulis, R., Swanson, M., and Dreyfuss, G. (1989) RNA binding proteins as

developmental regulators. Genes Dev. 4, 431–437.

4. Plyte, S.E. and Kneale, G.G. (1993) Characterization of the DNA-binding domain

of the Pf1 gene 5 protein. Biochemistry 32, 3623–3628

5. Huet, J., Conesa, C., Carles, C., and Sentenac, A. (1997) A cryptic DNA-binding

domain at th COOH terminus of TFIIIB70 affects formation, stability and func-

tion of perinitiation complexes. J. Biol. Chem. 272, 18,341–18,349.

6. Bochareva, E., Frappier, L., Edwards, A., and Bochareva, A. (1998) The RPA32

subunit of human replication protein A contains a single stranded DNA-binding

domain. J. Biol. Chem., 273, 3932–3936.

7. Wigley, D., Davies, G., Dodson, E., Maxwell, A., and Dodson, G. (1991) Crystal

structure of an N-terminal fragment of the DNA gyrase B protein. Nature 351,

624–629.

8. Kneale, G.G. (1983) Dissociation of the Pf1 nucleoprotein assembly complex and

characterization of the DNA-binding protein. Biochem. Biophys. Acta 739, 216–224.

DNA–Protein Photocrosslinking 323

323

From:

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

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

Ultraviolet Crosslinking of DNA–Protein

Complexes via 8-Azidoadenine

Rainer Meffert, Klaus Dose, Gabriele Rathgeber,

and Hans-Jochen Schäfer

1. Introduction

In biological systems, photoreactive derivatives have been widely applied

to study specific interactions of receptor molecules with their ligands by

photoaffinity labeling (1–3). While the receptors are generally proteins (e.g.,

enzymes, immunoglobulins, or hormone receptors), the ligands differ widely

in their molecular structure (e.g., sugars, amino acids, nucleotides, or oligo-

mers of these compounds).

The advantage of photoaffinity labeling compared with affinity labeling, or

chemical modification with group-specific reagents is that photoactivatable

nonreactive precursors can be activated at will by irradiation (Fig. 1). These

reagents do not bind covalently to the protein unless activated. On irradiation

of the precursors, highly reactive intermediates are formed that react indis-

criminately with all surrounding groups. Therefore, after activation, a

photoaffinity label, interacting at the specific binding site, can label all the

different amino acid residues of the binding area. Today, aromatic azido com-

pounds are mostly used as photoactivatable ligand analogs. They form highly

reactive nitrenes upon irradiation because of the electron sextet in the outer

electron shell of these intermediates (Fig. 2).

In addition to the azido derivatives, photoreactive precursors forming

radicals or carbenes on irradiation can be used as photoaffinity labels. All of

these intermediates (nitrenes, e.g.) vigorously try to complete an electron octet

(Fig. 3).

To produce covalent crosslinks between proteins and DNA, various meth-

ods have been applied (4–11): ultraviolet (UV) irradiation, γ-irradiation, chemi-

324 Meffert et al.

Fig. 1. Photoaffinity labeling of receptor proteins (e.g., enzymes) by photo-

activatable ligand analogs (e.g., substrate analog/product analog). In the dark (upper

line), the biological interactions of the protein with the ligand analog can be studied.

On irradiation (lower line), the protein (enzyme) is labeled and inactivated by the

substrate analog/product analog.

Fig. 2. Highly reactive photogenerated intermediates: radical (A), carbene (B), and

nitrene (C).

cal methods, and even vacuum or extreme dryness. Besides these methods,

photoaffinity labeling and photoaffinity crosslinking are helpful tools for the

study of specific interactions between proteins and deoxyribonucleic acids. To

date, many successful attempts have been made to photocrosslink proteins

to nucleic acids using different photoactivatable deoxynucleotides. 5-bromo-,

5-iodo-, 5-azido-, and 5-[N-(p-azidobenzoyl)-3-aminoallyl]-2'-deoxyuridine-

5'-monophosphate (12–18), 4-thio-2'-deoxythymidine-5'-monophosphate (19),

and 8-azido-2'-deoxyadenosine-5'-monophosphate (20,21) have been incorpo-

rated into deoxyribonucleic acids to bind DNA covalently to adjacent proteins

(for a review see ref. 22).

Here, we describe the synthesis of 8-azido-dATP (8-N

dATP), its incorpo-

ration into DNA by nick translation, and the procedure to photocrosslink azido-

modified DNA to proteins (20,21).

DNA–Protein Photocrosslinking 325

2. Materials

2.1. Synthesis of 8-N

dATP

1. dATP (disodium salt, Boehringer Mannheim, Mannheim, Germany).

2. Potassium acetate buffer: 1 M, pH 3.9.

3. Bromine.

4. Sodium disulfite (Na

5. Ethanol.

6. DEAE–Sephadex A-25.

7. Triethylammonium bicarbonate buffer: 0.7 M, pH 7.3.

8. Dimethylformamide.

9. Hydrazoic acid: 1 M in benzene.

10. Triethylamine.

2.2. Characterization of 8-N

dATP

1. Silica gel plates F

254

(Merck, Darmstadt, Germany).

2. Cellulose plates F (Merck).

3. Isobutyric acid/water/ammonia (66:33:1 v/v).

4. n-Butanol/water/acetic acid (5:3:2 v/v).

2.3. Preparation of Azido-Modified DNA

1. DNA (e.g., pBR 322 or pWH 106).

2. Deoxyribonucleotides (dATP, dGTP, dCTP, dTTP, [α-

P]-dCTP).

Fig. 3. Reactions of nitrenes. Cycloaddition to multiple bonds forming three-mem-

bered cyclic imines (1), addition to nucleophiles (2), direct insertion into C-H bonds

yielding secondary amines (3), and hydrogen atom abstraction followed by coupling

of the formed radicals to a secondary amine (4a, 4b).

326 Meffert et al.

3. DNase I (Escherichia coli, 2000 U/mg, Boehringer Mannheim) in 0.15 M NaCl

and 50% glycerol.

4. 50 mM Tris-HCl, pH 7.2.

5. Magnesium sulfate (MgSO

6. Bovine serum albumin.

7. DNA polymerase I (E. coli, Boehringer Mannheim, No. 104493, purchased con-

taining definite amounts of DNase I).

8. Ethylenediaminetetraacetic acid disodium salt (EDTA).

9. Sephadex A-25.

2.4. Photocrosslinking

An ultraviolet lamp (e.g., Mineralight handlamp UVSL 25 at position “long

wave”) emitting UV light at wavelengths of 300 nm and longer.

3. Methods

3.1. Synthesis of 8-N

dATP

The synthesis of 8-N

dATP (Fig. 4) is performed principally by analogy to

the synthesis of 8-N

ATP (23) (see Note 1). In the first step, bromine exchanges

the hydrogen at position 8 of the adenine ring. Then, the bromine is substituted

by the azido group.

1. Dissolve 0.2 mmol (117.8 mg) of dATP in 1.6 mL of potassium acetate buffer

(1 M, pH 3.9) and add 0.29 mmol (15 µL) of bromine. Keep the reaction mixture

in the dark at room temperature for 6 h (the absorption maximum shifts from 256 nm

to 262 nm; see Note 2).

2. Reduce excessive bromine by addition of traces of (approx 5 mg) Na

until the

reaction mixture looks colorless or pale yellow. Pour the reaction mixture into 20 mL

of cold ethanol (–20°C) and allow to stand for at least 30 min at –20°C in the dark.

3. Collect the precipitated deoxynucleotide by centrifugation and redissolve the resi-

due in 0.5 mL of double-distilled water. Further purification is achieved by ion-

exchange chromatography over DEAE–Sephadex A-25 column (50 × 2 cm) with

a linear gradient of 1000 mL each of water and triethylammonium bicarbonate

(0.7 M, pH 7.3).

Fig. 4. Synthesis of 8-N

dATP.