Download Time-resolved footprinting for the study of the structural dynamics of

New Methods for the Study of Protein–Nucleic Acid Interactions
Time-resolved footprinting for the study of the
structural dynamics of DNA–protein interactions
Bianca Sclavi1
LBPA (Laboratoire de Biotechnologies et Pharmacologie génétique Appliquée), UMR 8113 du CNRS, Ecole Normale Supérieure de Cachan, Cachan, France
Abstract
Transcription is often regulated at the level of initiation by the presence of transcription factors or nucleoid
proteins or by changing concentrations of metabolites. These can influence the kinetic properties and/or
structures of the intermediate RNA polymerase–DNA complexes in the pathway. Time-resolved footprinting
techniques combine the high temporal resolution of a stopped-flow apparatus with the specific structural
information obtained by the probing agent. Combined with a careful quantitative analysis of the evolution
of the signals, this approach allows for the identification and kinetic and structural characterization of the
intermediates in the pathway of DNA sequence recognition by a protein, such as a transcription factor or
RNA polymerase. The combination of different probing agents is especially powerful in revealing different
aspects of the conformational changes taking place at the protein–DNA interface. For example, hydroxyl
radical footprinting, owing to their small size, provides a map of the solvent-accessible surface of the DNA
backbone at a single nucleotide resolution; modification of the bases using potassium permanganate can
reveal the accessibility of the bases when the double helix is distorted or melted; cross-linking experiments
report on the formation of specific amino acid–DNA contacts, and DNase I footprinting results in a strong
signal-to-noise ratio from DNA protection at the binding site and hypersensitivity at curved or kinked DNA
sites. Recent developments in protein footprinting allow for the direct characterization of conformational
changes of the proteins in the complex.
Several methods exist for the quantitative study of the
complexes formed between proteins and their specific sites
on DNA. Each method brings a different level of detail and
unique information on the structural signatures of the complex. For a complete characterization of a complex, the use of
a few different complementary techniques is often necessary.
In this paper, I will describe the use of footprinting methods
for the equilibrium and kinetic characterization of DNA–
protein complexes. Footprinting is a general term that refers
to those techniques that rely on the comparative modification
of the polynucleotide in the presence or absence of a binding
molecule, a drug or a protein. Footprinting techniques are also
applied to the study of RNA structure and RNA folding [1,2]
and, more recently, have been applied to the measurement of
solvent-accessible surfaces of proteins [3].
Briefly, a fragment of DNA, for example, is cleaved or
modified by a chemical agent or enzyme at all accessible
sites. When the same reaction is carried out in the presence
of the binding partner, such as a protein, the sites involved
in the interactions will, in general, become protected from
cleavage or modification. The resulting DNA fragments are
separated and identified by denaturing gel electrophoresis and
the extent of the resulting cleavage pattern can be quantified.
The same kind of experiment can be carried out on a
supercoiled plasmid. In this case, primer extension is used to
Key words: DNA–protein interaction, equilibrium experiment, footprinting, kinetic experiment,
structural dynamics, time-resolved experiment.
1
email [email protected]
Biochem. Soc. Trans. (2008) 36, 745–748; doi:10.1042/BST0360745
identify the sites of cleavage or modification [4]. The primer
extension by a thermostable DNA polymerase can be done
in a PCR machine, thus obtaining a linear amplification of
the signal. This is particularly useful for detection of in vivo
footprints both on plasmids and on chromosomal DNA [5,6].
In order to carry out a quantitative analysis, it is necessary
to have identified at least a region of binding of the protein
on the DNA. This is often obtained by a gel shift assay
using different fragments of the putative binding region.
Once a specifically bound fragment is identified, footprinting
techniques can be used to identify the location of the binding
site(s). In the case in which the consensus binding sequence
is known, the fragment containing this sequence is selected.
In addition, it is necessary to obtain a sample of the purified
protein of interest.
Equilibrium experiments
Equilibrium experiments are used to measure the affinities
and co-operativity of binding. In this case, the DNA is
kept at a fixed concentration, while the concentration of the
protein is gradually increased. In order to measure a binding
affinity, the concentration of the DNA should be lower
than the K d (affinity equilibrium constant) and the protein
concentration should be in large excess so that the amount of
free protein does not change significantly during the titration;
this simplifies the subsequent analysis. If the concentration
of DNA is above the K d , one can measure the stoichiometry of
binding.
C The
C 2008 Biochemical Society
Authors Journal compilation 745
746
Biochemical Society Transactions (2008) Volume 36, part 4
Kinetic experiments
Kinetic experiments are used to measure the rates of complex
formation and to identify the presence and the structure
of possible intermediates in the binding pathway. For these
experiments, the protein and DNA are mixed and the complex
is probed after increasing incubation times until equilibrium is
reached. In this case, an excess of protein over DNA equally
ensures small changes in the amount of free protein and
concentrations of protein at least ten times greater than the
K d ensure that the binding site is saturated and the signal is at
its highest level. For this reason kinetic experiments usually
follow an equilibrium characterization.
In order to carry out time-resolved footprinting experiments, the use of a stopped-flow apparatus with a dead time
of 20–50 ms may be necessary if the reaction is completed
within several seconds [7]. The manual mixing and reaction
time is usually limited to 10–30 s and can be used to measure
timescales in the order of several minutes.
Three main methods have been described to produce
hydroxyl radicals [10]. The first is the Fenton reaction, which
requires chelated iron, H2 O2 and often a reducing agent such
as ascorbate to reduce the iron. The second is peroxynitrous
acid, which, however, is very sensitive to the pH of the buffer
[14]. The third is from the radiolysis of water-induced γ - or
X-ray radiation. All of these methods have been used for both
equilibrium and time-resolved experiments [15].
A variant of the Fenton reaction is the binding of the iron
by a BABE (bromoacetamidobenzyl-EDTA) molecule that
has been added to a specific site on the protein of interest.
This approach has been used for both protein–DNA and
protein–protein interactions and it is especially useful to place
a specific protein domain with respect to the DNA-binding
sequence or another protein partner [16,17].
Exposure of live or frozen cells to γ - or X-ray radiation
can be used to obtain in vivo footprints of protein–DNA
complexes or RNA structures [6,18].
UV laser cross-linking
Footprinting techniques
DNase I
DNase I is an enzyme that cleaves the DNA from the minor
groove with a slight sequence and structural preference. For
example, AT-rich regions are cut with decreased efficiency,
whereas sequences that are curved or kinked towards the
major groove are cut with a higher efficiency due to increased
affinity of the enzyme for these structures [8]. This results
in the appearance of hypersensitive cleavage sites. DNase I
experiments result in a high signal-to-noise ratio and report
on the presence of curved or kinked structures; however, the
size of the binding site for the enzyme and the steric clash
between DNase I and the protein bound to the DNA results
in a low spatial resolution. Equilibrium footprinting and
time-resolved quantitative DNase I footprinting have been
used for the study of the affinity, co-operativity and kinetics
of binding of proteins, most often transcription factors, to
DNA [9].
Hydroxyl radical footprinting
Hydroxyl radicals are small and highly reactive. They abstract
a proton from the sugar in the polynucleotide backbone,
resulting in a radical rearrangement and a break in the
DNA or RNA strand in virtually a sequence-independent
reaction [10]. Because of their small size, about the same
as a water molecule, the resulting cleavage pattern shows a
good correlation with the solvent accessibility of the backbone [11]. This is influenced by the sequence-dependent
helical structure, particularly at long stretches of adenines
or thymines resulting in a reduced minor groove width. The
main advantage of hydroxyl radical footprints is thus a high
structural resolution of the protection pattern. This pattern
has often been used to build molecular models of the structure of the DNA–protein complex or of tertiary RNA
structure [12,13].
C The
C 2008 Biochemical Society
Authors Journal compilation UV laser cross-linking is based on the modification of DNA
bases after their absorption of UV light [19]. This can occur
by the formation of thymine dimers or guanosine oxidation.
If the DNA structure were modified by the binding
of a protein, the probability of the formation of these
modifications may be altered; in addition, if an amino acid
were in close contact with a base, a covalent bond may result
from the UV absorption. These modifications are detected
by a primer extension reaction that can be carried out using
a thermostable polymerase in a PCR machine for a linear
amplification of the signal.
In addition, the formation of a covalent bond between
the protein and the DNA allows for the identification of the
peptide or even the amino acid on the protein involved in that
interaction [19]. This may require the use of MS analysis of
the protein sample after digestion of the DNA fragment. By
using a UV laser instead of a common UV lamp, the exposure
time of the sample is significantly reduced from a few
minutes to a few nanoseconds. A further advantage of UV
cross-linking is that it can be carried out on live cells. The
DNA is subsequently extracted and the primer extension can
be used to detect whether a protein was bound to the DNA
at the time of exposure in vivo (see for example [5,20]).
The yield of cross-links on unmodified DNA is usually
very low, approx. 0.5%; however, it can be increased by a
modification of the DNA bases or backbone (see for example
[21]).
DMS and potassium permanganate footprinting
Both DMS and potassium permanganate can be used to measure the conformational changes of individual bases from the
canonical B helix structure of DNA. They are part of a large
family of chemical reagents that can be used to probe different
structural aspects of the bases within polynucleotide chains.
DMS reacts preferably with G>C, depending on the reaction conditions, resulting in a methylation of the base that
can be detected either by primer extension or by reaction
New Methods for the Study of Protein–Nucleic Acid Interactions
with a strong base. Potassium permanganate instead reacts
preferably with T>C. Both of these techniques have been
used to measure the degree of melting of the double helix
by temperature or the presence of specific proteins such
as RNA polymerase (see for example [22–24]). When used
quantitatively, one must be careful to distinguish changes
in the accessibility of the base from changes to its intrinsic
reactivity [25]. In addition, full double-strand melting may
not be necessary to observe an increase in reactivity; a
distortion of the DNA may be sufficient to obtain a signal.
fragment, in the order of a few nanomolar. The use of capillary
sequencers to obtain footprinting data and semi-automated
peak-fitting software had resulted in a considerable decrease
in the time necessary to analyse the samples and have
increased confidence and reproducibility of the results [2].
Finally, a careful quantitative characterization of the
affinities and rates in the formation of DNA–protein
complexes is necessary if one wants to create a model of the
regulatory network in order to describe its dynamics and/or
the mechanisms of regulation [28].
Quantitative analysis
In order to carry out a quantitative analysis of the extent
of modification or cleavage for a footprinting experiments, a
few conditions should be met. First of all, no more than 30%
of the molecule should be cleaved or modified in order to
ensure that the reaction is within single hit regime, meaning
that each molecule is not cleaved or modified more than
once. In addition, as described above, it is better to work
with a concentration of DNA below the K d and an excess of
protein concentration. The main reason for the latter is that
the equations that are used to fit the binding curves are based
on the assumption that the amount of free protein in solution
does not change from one sample to the next (for a review,
see [2]).
An internal standard of an unmodified region is often used
in order to decrease the noise in the data due to losses of
material during sample purification and gel loading. This may
become more difficult in the case of potassium permanganate
footprinting, where it is rare that bases become modified in
the absence of protein or in the case where the whole length
of the DNA becomes protected by the protein. In these cases,
one can sometimes control the amount of material that is
loaded on to the gel or normalize for the total amount of
DNA in the sample. In some cases, one can add a standard
fragment of DNA that migrates in the gel far from the bands
of interest and is not affected by the presence of the protein.
Once a measure of the change in reactivity of the different
sites is obtained, it can be fit to a binding equation, in the case
of equilibrium experiments, or for time-resolved experiments
to a single, double and sometimes even triple exponential
equations. As the number of exponents, and therefore of
parameters, increases, one should use statistical tests in order
to determine whether the increase in the goodness to fit is
significant. The changes in the amplitudes corresponding
to the different phases can be used to measure the relative
amount of each intermediate [26,27].
The quantification of the band intensities on a polyacrylamide gel can be obtained by exposing the gel to a
phosphoimager screen in order to digitize the image and by
using the quantification software available with the imaging
apparatus [2].
More recently, the use of fluorescently labelled DNA has
become more popular as the detection sensitivity of capillary
sequencers has improved. The amounts of DNA used in this
case are now similar to those used for a radioactively labelled
References
1 Sclavi, B., Sullivan, M., Chance, M.R., Brenowitz, M. and Woodson, S.A.
(1998) RNA folding at millisecond intervals by synchrotron hydroxyl
radical footprinting. Science 279, 1940–1943
2 Shcherbakova, I., Mitra, S., Beer, R. and Brenowitz, M. (2008) Following
molecular transitions with single residue spatial and millisecond time
resolution. Methods Cell Biol. 84, 589–615
3 Takamoto, K. and Chance, M.R. (2006) Radiolytic protein footprinting
with mass spectrometry to probe the structure of macromolecular
complexes. Annu. Rev. Biophys. Biomol. Struct. 35, 251–276
4 Gralla, J.D. (1985) Rapid ‘footprinting’ on supercoiled DNA. Proc. Natl.
Acad. Sci. U.S.A. 82, 3078–3081
5 Valls, M., Buckle, M. and De Lorenzo, V. (2001) In vivo UV-laser
footprinting of the Pseudomonas putida s54 Pu promoter reveals that
IHF couples transcriptional activity to growth phase. J. Biol. Chem. 277,
2169–2175
6 Ottinger, L.M. and Tullius, T.D. (2000) High-resolution in vivo footprinting
of a protein–DNA complex using γ -radiation. J. Am. Chem. Soc. 122,
5901–5902
7 Petri, V. and Brenowitz, M. (1997) Quantitative nucleic acids footprinting:
thermodynamic and kinetic approaches. Curr. Opin. Biotechnol. 8,
36–44
8 Lahm, A. and Suck, D. (1991) DNase I-induced DNA conformation: 2 Å
structure of a DNase I–octamer complex. J. Mol. Biol. 222,
645–667
9 Brenowitz, M., Senear, D.F. and Kingston, R.E. (2001) DNase I footprint
analysis of protein–DNA binding. in Current Protocols in Molecular
Biology (Ausubel, F.M., Brent, R., Kingston, R.E., Moore, D.D., Seidman,
J.G., Smith, J.A. and Struhl, K., eds.), chapter 12, unit 12.4, John Wiley and
Sons, New York
10 Tullius, T. and Greenbaum, J. (2005) Mapping nucleic acid structure by
hydroxyl radical cleavage. Curr. Opin. Chem. Biol. 9, 127–134
11 Balasubramanian, B., Pogozelski, W. and Tullius, T. (1998) DNA strand
breaking by the hydroxyl radical is governed by the accessible surface
areas of the hydrogen atoms of the DNA backbone. Proc. Natl. Acad. Sci.
U.S.A. 95, 9738–9743
12 Murakami, K.S., Masuda, S., Campbell, E.A., Muzzin, O. and Darst, S.A.
(2002) Structural basis of transcription initiation: an RNA polymerase
holoenzyme–DNA complex. Science 296, 1285–1290
13 Lehnert, V., Jaeger, L., Michel, F. and Westhof, E. (1996) New loop–loop
tertiary interactions in self-splicing introns of subgroup IC and ID: a
complete 3D model of the Tetrahymena thermophila ribozyme.
Chem. Biol. 3, 993–1009
14 Chaulk, S.G. and MacMillan, A.M. (2000) Characterization of the
Tetrahymena ribozyme folding pathway using the kinetic footprinting
reagent peroxynitrous acid. Biochemistry 39, 2–8
15 Shcherbakova, I., Mitra, S., Beer, R. and Brenowitz, M. (2006) Fast fenton
footprinting: a laboratory-based method for the time-resolved analysis
of DNA, RNA and proteins. Nucleic Acids Res. 34, e48
16 Murakami, K., Kimura, M., Owens, J.T., Meares, C.F. and Ishihama, A.
(1997) The two α subunits of Escherichia coli RNA polymerase
are asymmetrically arranged and contact different halves of the
DNA upstream element. Proc. Natl. Acad. Sci. U.S.A. 94,
1709–1714
17 Traviglia, S.L., Datwyler, S.A., Yan, D., Ishihama, A. and Meares, C.F.
(1999) Targeted protein footprinting: where different
transcription factors bind to RNA polymerase. Biochemistry 38,
15774–15778
C The
C 2008 Biochemical Society
Authors Journal compilation 747
748
Biochemical Society Transactions (2008) Volume 36, part 4
18 Adilakshmi, T., Lease, R. and Woodson, S. (2006) Hydroxyl radical
footprinting in vivo: mapping macromolecular structures with
synchrotron radiation. Nucleic Acids Res. 34, e64
19 Buckle, M., Place, C. and Pemberton, I.K. (2000) Laser UV cross-linking of
nucleoprotein complexes in vitro. in Protein–DNA Interactions: A Practical
Approach (Buckle, M. and Travers, A., eds.), pp. 189–200, Oxford
University Press, Oxford
20 Murtin, C., Engelhorn, M., Geiselmann, J. and Boccard, F. (1998) A
quantitative UV laser footprinting analysis of the interaction of IHF with
specific binding sites: re-evaluation of the effective concentration of IHF
in the cell. J. Mol. Biol. 284, 949–961
21 Naryshkin, N., Revyakin, A., Kim, Y., Mekler, V. and Ebright, R.H. (2000)
Structural organization of the RNA polymerase–promoter open complex.
Cell 101, 601–611
22 Kirkegaard, K., Buc, H., Spassky, A. and Wang, J.C. (1983) Mapping of
single-stranded regions in duplex DNA at the sequence level:
single-strand-specific cytosine methylation in RNA polymerase–promoter
complexes. Proc. Natl. Acad. Sci. U.S.A. 80, 2544–2548
23 Zaychikov, E., Denissova, L., Meier, T., Gotte, M. and Heumann, H. (1997)
Influence of Mg2+ and temperature on formation of the transcription
bubble. J. Biol. Chem. 272, 2259–2267
C The
C 2008 Biochemical Society
Authors Journal compilation 24 Buckle, M. and Buc, H. (1989) Fine mapping of DNA single-stranded
regions using base-specific chemical probes: study of an open complex
formed between RNA polymerase and the lac UV5 promoter.
Biochemistry 28, 4388–4396
25 Tsodikov, O.V., Craig, M.L., Saecker, R.M. and Record, Jr, M.T. (1998)
Quantitative analysis of multiple-hit footprinting studies to characterize
DNA conformational changes in protein–DNA complexes: application to
DNA opening by Eσ 70 RNA polymerase. J. Mol. Biol. 283, 757–769
26 Sclavi, B., Zaychikov, E., Rogozina, A., Walther, F., Buckle, M. and
Heumann, H. (2005) Real-time characterization of intermediates in the
pathway to open complex formation by Escherichia coli RNA polymerase
at the T7A1 promoter. Proc. Natl. Acad. Sci. U.S.A. 102, 4706–4711
27 Dhavan, G.M., Crothers, D.M., Chance, M.R. and Brenowitz, M. (2002)
Concerted binding and bending of DNA by Escherichia coli integration
host factor. J. Mol. Biol. 315, 1027–1037
28 Shea, M. and Ackers, G. (1985) The OR control system of bacteriophage
lambda: a physical–chemical model for gene regulation. J. Mol. Biol.
181, 211–230
Received 21 April 2008
doi:10.1042/BST0360745