Physiol Genomics 46: 441–447, 2014.
First published May 6, 2014; doi:10.1152/physiolgenomics.00041.2014.
Review
Advanced methods for the analysis of chromatin-associated proteins
Hector Guillen-Ahlers,1 Michael R. Shortreed,2 Lloyd M. Smith,2 and Michael Olivier1
1
Department of Genetics, Texas Biomedical Research Institute, San Antonio, Texas; and 2Department of Chemistry,
University of Wisconsin, Madison, Wisconsin
Submitted 27 March 2014; accepted in final form 30 April 2014
DNA-protein interactions; chromatin; mass spectrometry; chromatin enrichment;
affinity purification
THE COMPLETION OF THE HUMAN genome project has provided the
scientific community with invaluable information about our
genetic makeup. With the actual nucleotide sequence in hand,
the key question in genomics and human biology is now how
this genetic information is used and regulated. How is the
expression of genes controlled on the molecular level? What
molecules regulate and modulate the levels of expression of
individual genes, and how do these regulatory mechanisms
allow a cell and an entire organism to effectively respond to
physiological stimuli and stressors? In an initial effort to
address these questions and to identify parts of the genome that
are essential for this regulatory control of the expression of all
genes in the genome, the Encyclopedia of DNA Elements
(ENCODE) consortium (9) has begun to identify and catalog
the functional elements encoded in the genome. Investigators
focused on three interactions that are important for gene
regulation: the binding of specific transcription factors (TF)
and other DNA-binding proteins to sequence elements in the
genome, the characterization of histone modifications and
nucleosome positioning as a local regulator of chromatin structure, and the analysis of long-range structural interactions
between distal regions of the genome, believed to affect accessibility of chromatin regions for transcription.
With this large-scale effort, significant progress has been
made in understanding how individual proteins that are part of
the transcription machinery mediate the expression of genes by
binding to specific elements in the DNA (For review see Ref.
Address for reprint requests and other correspondence: M. Olivier, Dept. of
Genetics, Texas Biomedical Research Inst., PO Box 760549, San Antonio, TX
78245 (e-mail: [email protected]).
17). Chromatin immunoprecipitation (ChIP) has been a crucial
tool for identifying regions of the genome where TF and other
DNA-binding proteins bind (13, 35). These experimental efforts have been complemented by computational efforts to
predict TF binding sites (6), taking advantage of the data being
obtained by ChIP analyses (5). Additional efforts have used
oligonucleotides as affinity probes to capture candidate binding
proteins interacting with putative functional DNA elements
followed by mass spectrometry to identify those proteins (20,
39). Analogous technologies using arrays containing every
possible DNA sequence permutation have been used in in vitro
assays to map TF binding sites of individual proteins (37).
Histones and their characteristic amino acid modifications
have also been investigated by ChIP approaches. Modifications
such as acetylation or phosphorylation regulate chromatin
conformation by altering the charge on histones (2), whereas
methylation of lysine can serve as point of recognition for
transcriptional regulation via recruitment of reader proteins
(38). When silent or repressed, the chromatin is arranged in a
highly condensed conformation, known as heterochromatin,
that greatly limits accessibility. Conversely, in active states, the
chromatin opens up into a more relaxed conformation, known
as euchromatin, which enables regulators and transcription
machineries to bind more readily (4, 12). Complex mechanisms affect gene activation and repression through structural
conformation changes in chromatin, and through recruitment
of interacting proteins.
Finally, methods like Hi-C (31) and ChIA-PET (15, 40) have
been developed to enable study of three-dimensional chromosome folding through analysis of genome-wide interactions.
These studies of long-range chromatin interactions showed that
1094-8341/14 Copyright © 2014 the American Physiological Society
441
Downloaded from http://physiolgenomics.physiology.org/ by 10.220.33.6 on June 16, 2017
Guillen-Ahlers H, Shortreed MR, Smith LM, Olivier M. Advanced methods
for the analysis of chromatin-associated proteins. Physiol Genomics 46: 441– 447,
2014. First published May 6, 2014; doi:10.1152/physiolgenomics.00041.2014.—
DNA-protein interactions are central to gene expression and chromatin regulation
and have become one of the main focus areas of the ENCODE consortium.
Advances in mass spectrometry and associated technologies have facilitated studies
of these interactions, revealing many novel DNA-interacting proteins and histone
posttranslational modifications. Proteins interacting at a single locus or at multiple
loci have been targeted in these recent studies, each requiring a separate analytical
strategy for isolation and analysis of DNA-protein interactions. The enrichment of
target chromatin fractions occurs via a number of methods including immunoprecipitation, affinity purification, and hybridization, with the shared goal of using
proteomics approaches as the final readout. The result of this is a number of
exciting new tools, with distinct strengths and limitations that can enable highly
robust and novel chromatin studies when applied appropriately. The present review
compares and contrasts these methods to help the reader distinguish the advantages
of each approach.
Review
442
METHODS FOR DNA-PROTEIN INTERACTIONS
ANALYSIS OF DNA-PROTEIN INTERACTIONS
The two most powerful methods for discovery and analysis
of in vivo DNA-protein interaction employ chromatin immunoprecipitation followed by analysis of the captured DNA on
microarrays (ChIP-chip) or by next-generation sequencing
(ChIP-Seq). ChIP-Seq offers higher resolution, less noise, and
greater coverage than ChIP-chip (21). The two methods focus
on one TF, DNA-binding protein, or histone modification at a
time, with the purpose of discovering all of the sequences
throughout the genome where it is bound. Systematic sequential profiling of a cell with ChIP of multiple proteins (TFs,
histones, etc.) can reveal the interesting interplay between
chromatin structure and transcription. A great advantage to
these methods is that they may be performed with relatively
small samples because the DNA captured in the process is
amplified prior to readout by array or by sequencing. Crucial
limitations of this approach are, however, that the protein has
to be known and that antibodies against it need to be available.
Technologies expanding these ChIP-based analysis capabilities are beginning to emerge wherein a locus (or even multiple
loci) is selected and captured followed by mass spectrometric
detection of bound proteins. There are currently two major
ways to approach this. The first is to use a TF, histone, or DNA
binding protein as bait to isolate a specific chromatin region
and analyze the proteome associated with that particular chromatin region (ChIP-like methods). This approach is considered
genome-wide since all genomic fragments bound to the target
protein are enriched. While additional binding proteins that are
part of a complex can be identified, the method does not
provide information for an individual target site. Alternatively,
the second way uses a specific genomic DNA sequence to
extract a target chromatin segment for analysis of the entire
associated proteome and is considered a locus-specific method.
Here, all identified proteins bind to the specific DNA locus
targeted for analysis. Both general approaches to studying
DNA-protein interactions are far more challenging than either
ChIP-chip or ChIP-Seq since captured proteins cannot be
amplified prior to analysis. Each single copy locus will commonly deliver at most two copies of a binding protein per cell
(assuming a diploid organism and full protein occupancy, and
the protein binding as a monomer). Nevertheless, these locusspecific approaches should provide more detailed information
about all of the proteins associated with any given locus and
should reveal many new and exciting DNA-protein interactions.
As part of this review, we will describe methods reported for
both of these approaches and discuss their utility in detail
below.
CHIP-LIKE METHODS
Modified Chromatin Immunopurification
Meant as a tool to elucidate chromatin-associated protein
networks, a method termed modified chromatin immunopurification (mChIP) was developed and then tested in yeast (19).
Tandem affinity purification (TAP)-tagged histones are affinity
purified with associated chromatin-bound protein networks
prior to mass spectrometric analysis. The addition of the
TAP-tag to the targeted histones does not result in overexpression of the protein but is simply a way to improve immunoprecipitation efficiency and also a way to help standardize the
method for other targets since the affinity purification steps
should be relatively consistent throughout. mChIP follows
sample preparation procedures closely resembling established
ChIP protocols, but with a number of modifications, which
include no treatment with formaldehyde or any kind of crosslinker. mChIP requires a mild sonication and centrifugation
prior to the immunoprecipitation step. As is usually the case
with ChIP, the chromatin is sheared mechanically during the
sonication step producing chromatin fragments that range between 0.5 and 2 kb. A great feature of this method is the fact
that the DNA length of the chromatin fragments to be enriched
can be regulated depending on additional fragmentation strategies. Sonication, micrococcal nuclease S7 digestion, or
DNase I digestion produce large, medium, or small fragments,
respectively, without significantly affecting the immunoprecipitation (IP) recovery efficiency. Depending on the purpose
of the experiment, this feature provides great flexibility by
giving the option to work with chromatin fragments that will
contain a few nucleosomes, one nucleosome with some adjacent DNA, or a single nucleosome without any exposed DNA
(Fig. 1A). As would be expected, the amount of protein
recovered by these shearing methods varies with DNA length
and allows differentiating some of the protein-protein interactions from those occurring on adjacent DNA regions. Utilization of TAP-tagged strains improves the efficiency of the IP
step and makes this method feasible for target proteins without
commercially available antibodies. Protein-chromatin complexes are resolved in a SDS-PAGE gel, and the proteins are
analyzed with standard liquid chromatography tandem mass
spectrometry (LC-MS/MS). This methodology allows for the
detection of other chromatin-associated proteins present at the
same locus as the protein being interrogated. The mChIP
Physiol Genomics • doi:10.1152/physiolgenomics.00041.2014 • www.physiolgenomics.org
Downloaded from http://physiolgenomics.physiology.org/ by 10.220.33.6 on June 16, 2017
actively transcribed genes tend to colocalize, possibly with
their transcriptional regulators (23), revealing chromatin structural dynamics involving distal DNA interactions throughout
the chromatin. These interactions may mediate long-range
regulatory mechanisms affecting gene expression and help
reveal structural three-dimensional interactions beyond the
linear effects investigated in studies of TF binding and nucleosome positioning.
All these studies highlight the complexity of genome regulation and illustrate the importance of a wide range of DNAprotein interactions, which mediate transcription. Yet, despite
several large-scale efforts, only a limited number of known
DNA-binding proteins and common histone modifications
have been revealed and studied. It is likely that we have only
uncovered a small fraction of the vast array of DNA-binding
proteins that are associated with chromatin and regulate the
genome, and additional efforts and approaches are needed to
uncover the full complement of DNA-binding proteins. De
novo identification of novel DNA-binding proteins or chromatin-associated proteins remains a challenge, however, and
novel tools addressing this need are only beginning to emerge.
These new methods focus on uncovering previously unknown
proteins that directly interact with the genome at specific loci.
The purpose of this review is to describe and discuss these new
approaches and to provide an overview of the potential applications of these technologies for a more detailed analysis of
gene expression regulation and genome biology.
Review
443
METHODS FOR DNA-PROTEIN INTERACTIONS
method was developed in yeast targeting histones H2A and its
variant Htz1p. Compared with traditional IP, mChIP showed
higher sensitivity, detecting 98 Hta2 associated proteins compared with 42 detected by traditional IP, 13 of which were
found in both methods. Furthermore, the identifications obtained by mChIP showed higher proportions of nuclear annotations, several of them with relevant chromatin associations.
Among them, core histones, RNA polymerase subunits, and
replication factors were found. As mentioned above, the complexes enriched during mChIP depend directly on the shearing
method. Consistent with this observation, it was seen, for
example, that the histone chaperone Nap1p, which strongly
associates to Htz1, was present in all samples regardless of
how the DNA got sheared, while the topoisomerase Top2p
required larger chromatin regions, thus not appearing in
DNase-processed samples. This method is not limited to histone analysis. Other nonhistone proteins, like Lge1, Mcm5, and
Yta7, were also shown to be suitable for mChIP analysis in this
study. This technology was applied in a high-throughput manner in an effort to globally define the chromatin-associated
interactome in yeast (18). In that effort, 102 known chromatinrelated proteins were used as bait, generating a curated list of
724 associated proteins with 2,966 high-confidence protein
associations. The mChIP technology appears to be a robust and
flexible method that has already been proven to be amenable to
high-throughput global analysis. The dependence on TAPtagged modifications, however, might limit the systems to
which it can be applied, but it is an excellent tool to study a
wide range of general protein-chromatin interactions in Saccharomyces cerevisiae.
Chromatin-Interacting Protein Mass Spectrometry
Another similar method outlined in Fig. 1B, chromatininteracting protein mass spectrometry (ChIP-MS) (36), was
recently adapted from earlier work by Guerrero et al. (16),
termed QTAX for quantitative analysis of tandem affinity
purified in vivo cross-linked protein complexes. ChIP-MS
uses an engineered affinity tag (HTB-tag) on the COOH
terminus of a target DNA-binding protein consisting of a
hexahistidine sequence, a cleavage site for the tobacco etch
virus protease, and a bacterially derived 75-amino acid peptide
that induces biotinylation in vivo (10). As with mChIP, the
HTB tag modification does not result in overexpression of the
protein but significantly improves the affinity purificationbased enrichment of the targeted chromatin fragments. The
ability of the protein to become biotinylated by endogenous
biotin ligase allows the use of streptavidin-coated beads for an
easy immobilization of the target region, thus eliminating the
IP step and the requirement for a suitable antibody. Mimicking
the proteomics of isolated chromatin segments (PICh) method
(Table 1), the ChIP-MS method uses strong cross-linking
conditions (3% formaldehyde for 30 min) to firmly stabilize the
DNA-protein interactions and shears the chromatin mechanically, through mild sonication. To increase enrichment yield
and reduce background caused by endogenously biotinylated
cytoplasmic proteins, the cross-linking step is performed after
the nuclei extraction. The strong affinity chemistry behind this
method enables fully denaturing conditions during the capture
process. These conditions allow for stringent washing for
removal of background proteins. An on-bead trypsin digestion
is performed prior to LC-MS/MS analysis. ChIP-MS was
employed to study Drosophila dosage compensation. There,
the male-specific lethal complexes 2 and 3 (MSL2, MSL3),
which are known to associate with active genes (1), were
genetically engineered to contain an HTB-tag on the COOH
terminus. Cells lacking the HTB-tag were used as negative
controls, and proteins identified in these samples were designated as false positives. The approach was successful in capturing the target regions and revealed several of the canonical
MSL associations (MSL1, MSL2, MSL3, and MOE, among
many others), as well as histone modifications (H4 Lys 16
acetylation, H3 Lys 36 methylation, among others). One of the
proteins found to be enriched at high levels in the MSL
complex was CG4747. The ChIP-MS protocol also works if
TAP-tagged targets are used instead of HTB-tags, but this
results in decreased sensitivity, identifying significantly fewer
proteins. Western blotting of pull-downs, as well as reciprocal
enrichments using uncovered proteins, validated their findings
Physiol Genomics • doi:10.1152/physiolgenomics.00041.2014 • www.physiolgenomics.org
Downloaded from http://physiolgenomics.physiology.org/ by 10.220.33.6 on June 16, 2017
Fig. 1. Illustration of chromatin enrichment using chromatin immunoprecipitation (ChIP)-like
methods. A: modified chromatin immunopurification (mChIP) uses histones with tandem affinity purification (TAP) tags to enrich local
chromatin fragments. DNA shearing can be accomplished mechanically through sonication or
enzymatically through micrococcal nuclease or
DNase I digestion. B: chromatin-interacting
protein in mass spectrometry (ChIP-MS). HTBtags capable of undergoing endogenous biotinylation are attached to a target DNA-binding
protein and used to capture a chromatin fragment by using streptavidin-coated magnetic
beads. C: chromatin proteomics (ChroP). Nontreated chromatin can be sheared through micrococcal nuclease digestion, or cross-linked
chromatin can be mechanically sheared through
sonication. Chromatin fragments are enriched
with antibodies against trimethylated histone
variants.
Review
444
METHODS FOR DNA-PROTEIN INTERACTIONS
Table 1. Methods comparison
Cell Source
mChIP
ChIP-MS
N-ChroP
X-ChroP
PICh
ChAP-MS
TAL-ChAP-MS
yeast
fly
human
human
human
yeast
yeast
Cell Input
X-link, %
10
10
109
108
108
109
1011
1011
3
0.75
3
1.25
1.25
Fragmentation
Target
MN/son/DNase
mild sonication
MN
sonication
mild sonication
mild sonication
mild sonication
histones
DNA BP
3me Histones
3me Histones
telomeres
single locus
single locus
SILAC
yes
yes
WT-amenable
Ref. No.
yes
yes
yes
19
36
26
26
11
7
8
X-link, cross-linked with formaldehyde. WT-amenable, method does not require any genetic modification or plasmid insertions. MN, micrococcal nuclease
digestion. Son, sonication. DNA BP, DNA binding protein. mChIP, modified chromatin immunopurification; ChIP-MS, chromatin-interacting protein mass
spectrometry; N-ChroP, native chromatin proteomics; X-ChroP, cross-linked chromatin proteomics; PICh, proteomics of isolated chromatin segments;
ChAP-MS, chromatin affinity purification with mass spectrometry; TAL, transcription activator-like.
demonstration of this, X-ChroP was used to identify other
proteins interacting at those domains. An obvious advantage of
this technology is that, unlike many technologies of its kind,
ChroP does not require any sort of genetic engineering. Thus,
this technology is more accessible for potential use in other
systems and organisms. It appears to be a robust technology
well suited to analyze histone modifications and their relationship with other chromatin-interacting proteins, in both euchromatin and heterochromatin conformations.
Chromatin Proteomics
LOCUS-SPECIFIC METHODS
ChroP, which stands for chromatin proteomics, is another
ChIP-like technology designed for identification of DNAprotein interactions (26). ChroP expands previously established IP methods by using peptide baits for trimethylated
lysines on histone 3 to study, with the help of stable isotope
labeling by/with amino acids in cell culture (SILAC), associated interactomes (32, 33). This method can be performed with
native or cross-linked chromatin (N-ChroP and X-ChroP),
offering complementary results (Fig. 1C). N-ChroP utilizes
micrococcal nuclease digestion to fragment the chromatin,
producing mononucleosome-sized fragments, while X-ChroP
shears the cross-linked (0.75% formaldehyde) chromatin
through sonication, producing chromatin fragments of ⬃500
bp in length (roughly 3 nucleosomes). Samples are immunoprecipitated with the desired antibody (H3K9me3, H3K4me3
in this case) and further separated by SDS-PAGE prior to
LC-MS/MS analysis. Chemical alkylation of proteins with
D6-acetic anhydride is used to further analyze histone methylation patterns. Of all the methods discussed in the present
review, ChroP requires the least amount of starting material
(Table 1). N-ChroP was used to analyze the associated profiles
of H3K9me3 (commonly found in densely packed silent heterochromatin), as well as H3K4me3 (commonly found in
active euchromatin) in HeLa cells. Histone posttranslational
modifications (PTMs) were identified and quantified showing a
depletion of H3k4me1 and H3K4me3, as would be expected in
transcriptionally inactive chromatin (14). By using a bottom-up
approach during the MS analysis (analysis of tryptic peptides
by MS), the authors bring attention to the limitation that such
a strategy entails, since only short-range histone modifications
are detected and a more challenging top-down approach (analysis of nonfragmented intact proteins by MS) would paint a
clearer picture about cis/trans modifications. While N-ChroP
seems better suited to study histone PTMs, X-ChroP can be
applied to different kinds of DNA-binding proteins. As a
PICh
There are two methodologies that focus on DNA sequences
to extract a desired locus for protein analysis (Fig. 2). PICh is
the only one so far that uses hybridization to target and isolate
a specific chromatin locus (11). This technology was used to
target and capture telomeric regions of chromatin in HeLa
cells. Telomeres provide a significant technical advantage,
considering that they consist of six base pairs repeats that
comprise a few thousand base pairs. For hybridization purposes, these repeats act as high copy number targets, which
increase the efficiency of the process. Furthermore, each cell
contains ⬃100 telomeres, which greatly decreases the amount
of input material needed. PICh uses strong cross-linking parameters (3% formaldehyde, 30 min) that allow the rest of the
process to occur under extremely stringent conditions. The 25
mer oligonucleotides used as capture probes contain locked
nucleic acids, which increase the melting temperature by enhanced base stacking (34). A scrambled oligonucleotide with
the same original composition is used as the negative control.
Desthiobiotin, the moiety used on the capture oligonucleotides
for isolation of the target on streptavidin-coated beads, binds
with slightly less affinity than biotin, which allows biotin to act
as an elution agent. An intricate sequence of hybridization and
washes is used to extract the targeted regions, which are then
separated in an acrylamide gel prior to mass spectrometry
analysis. PICh was applied to study alternative lengthening of
telomeres, which is a mechanism common in some cancers,
where telomere maintenance can occur in the absence of
telomerase. Thus different cell populations known to have
distinct telomere maintenance pathways were selected for analysis (HeLa S3 and WI38-VA13). Close to 200 proteins per
condition were detected, half of which were present in both
samples. About 85% of the known telomere interactors were
identified in a single experiment showcasing the robustness of
Physiol Genomics • doi:10.1152/physiolgenomics.00041.2014 • www.physiolgenomics.org
Downloaded from http://physiolgenomics.physiology.org/ by 10.220.33.6 on June 16, 2017
(36). The strong biotin-streptavidin affinity allows ChIP-MS to
work with relatively low numbers of cells (109). It can be
assumed that many other targets would be amenable to ChIP-MS
so long as genetically modifying the genome is a possibility for a
particular application. Endogenous biotinylation enzymes need
also to be present in the system, as is the case with yeast (16, 29).
A more global approach for this technology has not yet been
pursued and would require introducing the HTB-tag in all the
DNA-binding proteins to be studied.
Review
METHODS FOR DNA-PROTEIN INTERACTIONS
445
the approach. A subset of the novel telomere-interacting proteins was positively validated through immunostains, including
Fanc-J, RIP140, and NXP-2.
PICh was also recently applied to Drosophila telomereassociated sequence (TAS) repeats (3). Enrichment and protein
identification of TAS repeats regions signifies an improvement
in the technology since, even though human telomeres and
Drosophila TAS repeats represent equivalent percentage of the
entire genome, the longer TAS repeats (450 bp compared with
6 bp in human telomeres) significantly reduce the number of
places to which each probe can bind. The successful identification and validation of novel TAS-binding proteins show that
this technology is adaptable to less abundant target regions. It
has yet to be seen if this method can be further adapted to
single copy regions. A single copy locus would suggest that
100⫻ larger samples would be needed. Would the same
capture approach work in regions that do not comprise repeats?
It is clear that to study telomeres, PICh is a robust, capable
technology, the first one of its kind, not requiring previous
protein knowledge or antibody affinity purification to isolate
and analyze a specific chromatin region.
Chromatin Affinity Purification with Mass Spectrometry
Chromatin affinity purification with mass spectrometry
(ChAP-MS) is a technology that relies on an engineered LexA
DNA binding site, introduced by homologous recombination in
proximity to a given region of interest within the genome. A
plasmid constitutively expressing a LexA-Protein A fusion
protein is introduced in the cells, and, as shown in Fig. 1B,
once the LexA-PrA is bound to its target in the genome, it can
be used as a handle to extract that region (7). Samples are
cross-linked in vivo, and fragmentation is achieved by mild
sonication to generate chromatin fragments of ⬃1,000 bp in
length. The tight bond between LexA-PrA and its engineered
target binding sequence allows for efficient affinity-based enrichment. Additionally, similar to the ChroP method mentioned above, this technology relies on a variation of SILAC
called iDIRT (isotopic differentiation of interactions as random
or target) to reduce background noise due to nonspecific
protein binding (28). Cells containing the constructs are grown
in light media, while control cells lacking the LexA binding
site are placed in heavy media, with the intention to discern at
the time of the MS analysis which proteins are truly enriched
and which ones come from non-specific background. Captured
chromatin fragments are resolved in an SDS-PAGE gel prior to
LC-MS/MS analysis. Unlike all the previous methods discussed in this review, the ChAP-MS approach allows interrogation of one single-copy locus at a time. ChAP-MS was used
to study the 5=-region of the GAL1 gene in S. cerevisiae. The
LexA binding sites were placed on the 3=-end of the upstream
activator sequence of the GAL genes, just upstream of GAL1.
Cells were grown in the presence of either glucose or galactose, to transcriptionally repress or activate, respectively, the
GAL1 locus. This analysis produced a list of ⬃250 and 350
mass spectrometry-identified proteins found under transcriptionally silent (glucose) or active (galactose) conditions, respectively. Among those, H3K36me3 was found to be enriched
in cells grown in the presence of glucose, while Gal3, Spt16,
Rpb2, and a number of acetylated histones were found when
grown in the presence of galactose. A small subset of the
resulting proteins was validated with ChIP.
A modified version of this technology termed TAL-ChAP-MS
has recently been developed (Fig. 2C). This approach substitutes the LexA binding strategy for a modified transcription
activator-like (TAL) effector. These proteins come from Xanthomonas and serve as transcription activators (24). In this case
the TAL fusion protein is capable of recognizing a specific and
unique 18-nucleotide sequence in the UASGAL, close to where
the LexA binding site was located in the original method. This
allows affinity-based enrichment of the target chromatin region. A transformation to introduce a plasmid capable of
constitutively expressing the TAL protein is still required, but
it does eliminate the need to genetically modify the genome
(8). Furthermore, the SILAC requirement is eliminated, which
naturally decreases cost and complexity, making the technology more accessible. Unfortunately, chromatin enrichment was
not possible under glucose conditions that result in a transcriptionally silent chromatin around the Gal10 and Gal1 genes,
where the target for the TAL is located. It remains to be seen
if this result is region-specific, or if TAL-ChAP-MS is only
amenable to transcriptionally active loci. Nevertheless, in transcriptionally active cells grown under galactose, TALChAP-MS was able to also identify Rpb2, Spt16, and Gal3,
which would be expected to be found during active transcrip-
Physiol Genomics • doi:10.1152/physiolgenomics.00041.2014 • www.physiolgenomics.org
Downloaded from http://physiolgenomics.physiology.org/ by 10.220.33.6 on June 16, 2017
Fig. 2. Illustration of chromatin enrichment using locusbased methods. A: Proteomics of isolated chromatin segments (PICh). Cross-linked chromatin is sheared mechanically through sonication and captured using probes containing locked nucleic acids (LNA). Probes hybridize to
telomere regions. B: chromatin affinity purification with
mass spectrometry (ChAP-MS). A LexA binding site is
genetically engineered immediately upstream of the GAL1
gene. A plasmid constitutively expressing LexA-PrA is introduced and used for affinity-based purification. Cross-linked
chromatin is mechanically sheared through sonication. C:
transcription activator-like-ChAP-MS (TAL-ChAP-MS). A
plasmid constitutively expressing TAL-PrA, capable of binding a unique 18 nt sequence upstream of the GAL1 gene, is
introduced and used for affinity-based purification.
Review
446
METHODS FOR DNA-PROTEIN INTERACTIONS
tion of GAL1 (22). The presence of several histone PTMs were
also detected.
Both versions of the ChAP-MS approach represent the first
and only effort that has targeted a single copy genomic region
for enrichment and protein analysis. As would be expected, the
cell number requirements are high (Table 1), and there is a
need to modify the genome and/or introduce a plasmid to make
ChAP-MS possible. For transcriptionally active sites, the TALChAP-MS version of the method seems to be a strong tool to
study the associated proteome at a particular locus. ChAP-MS
can be an important tool to study genomic regions where little
information about interacting proteins is available.
In the present review, we summarized a number of novel
technologies for studying DNA-protein interactions. The methods described here focus on the analysis of proteins bound to
chromatin, rather than use the bound DNA as a readout. A
number of methods benefit from the use of mass spectrometry
for discovery of novel DNA-interacting proteins, which is not
possible with ChIP-chip and ChIP-Seq, where the focus is on
the analysis of the DNA sequences involved in DNA-protein
interactions. Mass spectrometry for chromatin-associated proteins is now becoming a field in itself (27). The importance of
this field has been highlighted before (25) and will undoubtedly
contribute to basic understanding of transcription regulation.
The roads these technologies are carving will surely contribute
to the discovery of extensive lists of DNA-interacting proteins
that had not been considered before.
The methodologies discussed here fall into two general
categories. In the first category, a known DNA-binding protein
is used to extract the genomic loci with which it interacts
(mChIP, ChroP, Chip-MS), while in the second approach, a
specific genomic locus is the target, regardless of which proteins are bound to it (PICh, ChAP-MS). Some of the difficulties
and challenges among these methods are shared. An important
one is the amount of sample needed for each experiment,
ranging between 108 and 1011 cells per experiment (Table 1).
These input amounts limit the organisms to be studied and
more importantly the applications for which it can be used.
Clinical applications seem still far-fetched at this point, since a
biopsy wouldn=t yield sufficient material for analysis at current
sensitivities. Another shared challenge is the ability to analyze
both euchromatin and heterochromatin. Chromatin has different properties in active and repressed conformation that should
be taken into consideration for isolation purposes (30). Many
methods do not well when the chromatin is in a packed (inactive)
conformation, which was evident in the TAL-ChAP-MS
method. There are also intrinsic benefits and limitations that
are not shared and are worth noting. An obvious limitation of
ChIP-like approaches is their dependence on previous knowledge of a given DNA-binding protein and on the availability of
suitable antibodies for the protein selected as bait. This becomes an even greater issue when working with nonhuman
systems such as rodent model organisms commonly used in
physiological and genetic research. Both mChIP and ChIP-MS
do not require specialized antibodies but come with the requirement of genetically introducing tags for the purification
step. These approaches are also severely limited if the focus is
to study a single locus in the genome. That is not the intention
GRANTS
The authors acknowledge the support of the Wisconsin Center of Excellence in Genomics Science (National Human Genome Research Institute Grant
P50 HG-004952).
DISCLOSURES
No conflicts of interest, financial or otherwise, are declared by the author(s).
Physiol Genomics • doi:10.1152/physiolgenomics.00041.2014 • www.physiolgenomics.org
Downloaded from http://physiolgenomics.physiology.org/ by 10.220.33.6 on June 16, 2017
CONCLUSIONS
of the ChIP-like technologies, however. These methods are
suited to study more generalized chromatin behaviors like
histone PTMs and their relationship to the associated proteome
across the genome. ChIP-like technologies also allow us to
look at a single TF and to inquire not only to which loci it
binds, but how it interacts with other DNA-binding proteins or
even protein complexes that are not in direct contact with the
DNA, thus building the protein networks that interact with
DNA and are essential in regulating gene expression.
Many of the difficulties observed in ChIP-like methods are
overcome by DNA-/locus-specific methods. Most importantly,
by using specific DNA sequences as bait, one can capture a
snapshot of a particular locus without the need for either prior
knowledge of at least some of the interacting proteins or
availability of suitable antibodies. This kind of tool becomes
extremely useful when novel or noncanonical interactions are
the focus of the study. It is obvious, however, that bringing
these DNA-based technologies up to high-throughput and/or
genome-wide scales will be intrinsically more challenging. But
just as ChIP-like methods are not meant for single locus
studies, DNA-based methods are not meant (at least not yet)
for genome-wide analyses. Their power lies in discovery of
novel DNA binding proteins at a specific locus under specific
conditions, which in itself is a very ambitious goal. Also, none
of the methodologies described here are easily adaptable for
the combined genome-wide analysis of both proteins and
DNA. As a result, ChIP approaches are often used to validate
the protein-focused findings from these novel approaches, and
it will be exciting to explore the synergies between ChIP and
these novel methods in analyzing genome regulation.
The potential for the novel technologies presently discussed
is great for elucidating many questions that so far have remained a challenge. Why do some drug treatments affect the
expression of a particular gene in some patients but not in
others? How do promoter variants affect protein binding and
therefore gene expression? Which proteins cause well-established TF pathways to act differently on their targets in different disease variants? There are a number of important questions related to gene expression regulation currently eluding
the scientific community that these novel technologies promise
to address. It is likely that these methodologies, and others that
will further expand the repertoire of technologies, will have an
important impact on the analysis of genome function and
biology for years to come. Further analyses and comparisons
will be required (and likely additional techniques will emerge)
to use several of the described approaches on the same target(s)
before it will become clear which approach holds the greatest
potential, but even the initial reports so far clearly suggest that
different questions (and different targets) are likely better
suited for certain approaches.
Review
METHODS FOR DNA-PROTEIN INTERACTIONS
AUTHOR CONTRIBUTIONS
Author contributions: H.G.-A. drafted manuscript; H.G.-A., M.R.S.,
L.M.S., and M.O. approved final version of manuscript; M.R.S., L.M.S., and
M.O. edited and revised manuscript.
REFERENCES
21. Park PJ. ChIP-seq: advantages and challenges of a maturing technology.
Nat Rev Genet 10: 669 –680, 2009.
22. Platt A, Reece RJ. The yeast galactose genetic switch is mediated by the
formation of a Gal4p-Gal80p-Gal3p complex. EMBO J 17: 4086 –4091,
1998.
23. Schoenfelder S, Sexton T, Chakalova L, Cope NF, Horton A, Andrews
S, Kurukuti S, Mitchell JA, Umlauf D, Dimitrova DS, Eskiw CH, Luo
Y, Wei CL, Ruan Y, Bieker JJ, Fraser P. Preferential associations
between co-regulated genes reveal a transcriptional interactome in erythroid cells. Nat Genet 42: 53–61, 2010.
24. Scholze H, Boch J. TAL effector-DNA specificity. Virulence 1: 428 –432,
2010.
25. Smith LM, Shortreed MR, Olivier M. To understand the whole, you
must know the parts: unraveling the roles of protein-DNA interactions in
genome regulation. Analyst 136: 3060 –3065, 2011.
26. Soldi M, Bonaldi T. The proteomic investigation of chromatin functional
domains reveals novel synergisms among distinct heterochromatin components. Mol Cell Proteomics 12: 764 –780, 2013.
27. Soldi M, Cuomo A, Bremang M, Bonaldi T. Mass spectrometry-based
proteomics for the analysis of chromatin structure and dynamics. Int J Mol
Sci 14: 5402–5431, 2013.
28. Tackett AJ, DeGrasse JA, Sekedat MD, Oeffinger M, Rout MP, Chait
BT. I-DIRT, a general method for distinguishing between specific and
nonspecific protein interactions. J Proteome Res 4: 1752–1756, 2005.
29. Tagwerker C, Flick K, Cui M, Guerrero C, Dou Y, Auer B, Baldi P,
Huang L, Kaiser P. A tandem affinity tag for two-step purification under
fully denaturing conditions: application in ubiquitin profiling and protein
complex identification combined with in vivo cross-linking. Mol Cell
Proteomics 5: 737–748, 2006.
30. Torrente MP, Zee BM, Young NL, Baliban RC, LeRoy G, Floudas
CA, Hake SB, Garcia BA. Proteomic interrogation of human chromatin.
PLoS One 6: e24747, 2011.
31. van Berkum NL, Lieberman-Aiden E, Williams L, Imakaev M,
Gnirke A, Mirny LA, Dekker J, Lander ES. Hi-C: a method to study the
three-dimensional architecture of genomes. J Vis Exp: e1869, 2010.
32. Vermeulen M, Eberl HC, Matarese F, Marks H, Denissov S, Butter F,
Lee KK, Olsen JV, Hyman AA, Stunnenberg HG, Mann M. Quantitative interaction proteomics and genome-wide profiling of epigenetic
histone marks and their readers. Cell 142: 967–980, 2010.
33. Vermeulen M, Mulder KW, Denissov S, Pijnappel WW, van Schaik
FM, Varier RA, Baltissen MP, Stunnenberg HG, Mann M, Timmers
HT. Selective anchoring of TFIID to nucleosomes by trimethylation of
histone H3 lysine 4. Cell 131: 58 –69, 2007.
34. Vester B, Wengel J. LNA (locked nucleic acid): high-affinity targeting of
complementary RNA and DNA. Biochemistry 43: 13233–13241, 2004.
35. Walhout AJM. Unraveling transcription regulatory networks by proteinDNA and protein-protein interaction mapping. Genome Res 16: 1445–
1454, 2006.
36. Wang CI, Alekseyenko AA, LeRoy G, Elia AE, Gorchakov AA,
Britton LM, Elledge SJ, Kharchenko PV, Garcia BA, Kuroda MI.
Chromatin proteins captured by ChIP-mass spectrometry are linked to
dosage compensation in Drosophila. Nat Struct Mol Biol 20: 202–209,
2013.
37. Warren CL, Kratochvil NC, Hauschild KE, Foister S, Brezinski ML,
Dervan PB, Phillips GN Jr, Ansari AZ. Defining the sequence-recognition profile of DNA-binding molecules. Proc Natl Acad Sci USA 103:
867–872, 2006.
38. Wozniak GG, Strahl BD. Hitting the ‘mark’: interpreting lysine methylation in the context of active transcription. Biochim Biophys Acta:
S1874 –9399(14)00053– 4, 2014.
39. Wu CH, Chen S, Shortreed MR, Kreitinger GM, Yuan Y, Frey BL,
Zhang Y, Mirza S, Cirillo LA, Olivier M, Smith LM. Sequence-specific
capture of protein-DNA complexes for mass spectrometric protein identification. PLoS One 6: e26217, 2011.
40. Zhang J, Poh HM, Peh SQ, Sia YY, Li G, Mulawadi FH, Goh Y,
Fullwood MJ, Sung WK, Ruan X, Ruan Y. ChIA-PET analysis of
transcriptional chromatin interactions. Methods 58: 289 –299, 2012.
Physiol Genomics • doi:10.1152/physiolgenomics.00041.2014 • www.physiolgenomics.org
Downloaded from http://physiolgenomics.physiology.org/ by 10.220.33.6 on June 16, 2017
1. Alekseyenko AA, Larschan E, Lai WR, Park PJ, Kuroda MI. Highresolution ChIP-chip analysis reveals that the Drosophila MSL complex
selectively identifies active genes on the male X chromosome. Genes Dev
20: 848 –857, 2006.
2. Allfrey VG, Faulkner R, Mirsky AE. Acetylation and methylation of
histones and their possible role in the regulation of RNA synthesis. Proc
Natl Acad Sci USA 51: 786 –794, 1964.
3. Antão JM, Mason JM, Déjardin J, Kingston RE. Protein landscape at
Drosophila melanogaster telomere-associated sequence repeats. Mol Cell
Biol 32: 2170 –2182, 2012.
4. Beisel C, Paro R. Silencing chromatin: comparing modes and mechanisms. Nat Rev Genet 12: 123–135, 2011.
5. Boeva V, Surdez D, Guillon N, Tirode F, Fejes AP, Delattre O, Barillot
E. De novo motif identification improves the accuracy of predicting
transcription factor binding sites in ChIP-Seq data analysis. Nucleic Acids
Res 38: e126, 2010.
6. Bulyk ML. Computational prediction of transcription-factor binding site
locations. Genome Biol 5: 201, 2003.
7. Byrum SD, Raman A, Taverna SD, Tackett AJ. ChAP-MS: a method
for identification of proteins and histone posttranslational modifications at
a single genomic locus. Cell Rep 2: 198 –205, 2012.
8. Byrum SD, Taverna SD, Tackett AJ. Purification of a specific native
genomic locus for proteomic analysis. Nucleic Acids Res 41: e195, 2013.
9. The Consortium ENCODE. Identification and analysis of functional
elements in 1% of the human genome by the ENCODE pilot project.
Nature 447: 799 –816, 2007.
10. Cronan JE. Biotination of proteins in vivo. A post-translational modification to label, purify, and study proteins. J Biol Chem 265: 10327–10333,
1990.
11. Déjardin J, Kingston RE. Purification of proteins associated with specific
genomic loci. Cell 136: 175–186, 2009.
12. Dillon N. Heterochromatin structure and function. Biol Cell 96: 631–637,
2004.
13. Elnitski L, Jin VX, Farnham PJ, Jones SJM. Locating mammalian
transcription factor binding sites: a survey of computational and experimental techniques. Genome Res 16: 1455–1464, 2006.
14. Garcia BA, Pesavento JJ, Mizzen CA, Kelleher NL. Pervasive combinatorial modification of histone H3 in human cells. Nat Meth 4: 487–489,
2007.
15. Goh Y, Fullwood MJ, Poh HM, Peh SQ, Ong CT, Zhang J, Ruan X,
Ruan Y. Chromatin Interaction Analysis with Paired-End Tag Sequencing
(ChIA-PET) for mapping chromatin interactions and understanding transcription regulation. J Vis Exp: 37702012.
16. Guerrero C, Tagwerker C, Kaiser P, Huang L. An integrated mass
spectrometry-based proteomic approach: quantitative analysis of tandem
affinity-purified in vivo cross-linked protein complexes (qtax) to decipher
the 26 s proteasome-interacting network. Mol Cell Proteomics 5: 366 –
378, 2006.
17. Kadonaga JT. Regulation of RNA polymerase II transcription by sequence-specific DNA binding factors. Cell 116: 247–257, 2004.
18. Lambert JP, Fillingham J, Siahbazi M, Greenblatt J, Baetz K, Figeys
D. Defining the budding yeast chromatin-associated interactome. Mol Syst
Biol 6: 448, 2010.
19. Lambert JP, Mitchell L, Rudner A, Baetz K, Figeys D. A novel
proteomics approach for the discovery of chromatin-associated protein
networks. Mol Cell Proteomics 8: 870 –882, 2009.
20. Mittler G, Butter F, Mann M. A SILAC-based DNA protein interaction
screen that identifies candidate binding proteins to functional DNA elements. Genome Res 19: 284 –293, 2009.
447