DNA Immunoprecipitation (DIP) for the Determination of DNA-Binding Specificity

Andrea J. Gossett   and   Jason D. Lieb ¹

Department of Biology, University of North Carolina at Chapel Hill, Chapel Hill, NC 27599, USA

¹ Corresponding author ( jlieb@bio.unc.edu )

INTRODUCTION

Knowledge of the DNA-binding specificity of a transcription   factor aids in understanding the function of that factor in   the regulation of gene transcription. One popular method of ^{identifying the genomic DNA sites bound by a given protein in   vivo is the Chromatin ImmunoPrecipitation with microarray analysis   (ChIP-chip) technique. However, this method reveals a binding   pattern influenced by in vivo phenomena that may mask the actual   DNA-binding specificity of the factor, such as chromatin effects   and competitive or cooperative protein-protein interactions.   ChIP-chip also requires adequate expression of the protein in   the cell type chosen to create the extract used for immunoprecipitation.   DNA Immunoprecipitation (DIP) is an alternative technique that   allows one to take advantage of the favorable properties of   both in vivo and traditional in vitro techniques (e.g., electromobility   shift assay [EMSA] and binding site selection [SELEX]). DIP   utilizes naked genomic DNA as a binding substrate for one or   more purified recombinant proteins. Because genomic DNA is used   as a template in DIP reactions, the results are directly comparable   to ChIP-chip or ChIP-seq data. DIP can be carried out in the   presence of cofactors such as heterodimer partners, competitors,   or small molecule binding inhibitors. After DNA is isolated   by DIP, it is most efficiently detected using a highly parallel   genomic technique such as a DNA microarray (DIP-chip) or high-throughput   sequencing (DIP-seq). In this protocol, we describe a DIP of   a yeast ( Saccharomyces cerevisiae ) protein with yeast genomic   DNA.}

RELATED INFORMATION

For more information about the DIP-chip technique, see   Liu et al. (2005 ,   2006) .

MATERIALS

Reagents

Amylose resin (3 mg/mL maltose-binding protein [MBP]-binding   capacity; New England Biolabs)

  DIP binding/wash buffer

  DIP elution buffer

DNA (genomic, purified and sheared to an average size of ~600   bp)

Genomic yeast DNA from the strain S288C was purified by CsCl   gradient and sheared by sonication.

The protein and genomic DNA used for DIP need not be from the   same species, and genomic DNA from any species with a sequenced   genome can be used.

Protein of interest (MBP-tagged, purified)

Reagents for detection (i.e., a DNA microarray corresponding   to the genomic DNA used, or reagents for high-throughput sequencing;   see Step 11)

Equipment

DNA purification columns (e.g., Zymo-Spin I; Zymo Research)

Equipment for detection (i.e., DNA microarray or high-throughput   sequencing capacity; see Step 11)

Ice

Microcentrifuge (capable of 2000 g   for microcentrifuge tubes)

Microcentrifuge tubes (1.5 mL)

Rotator (Labquake e.g., Barnstead International 400110 or comparable   end-over-end rotator) or micropipettor (see Step 4)

METHOD

This protocol was modified from the method by   Liu et al. (2005) .

 

1. To prepare a 100 µL reaction, combine the purified   protein of interest (MBP-tagged) to a final concentration of   40 nM and the sheared genomic DNA to a final concentration of   0.3 pM in 100 µL DIP binding/wash buffer.  
For yeast   DIPs, 0.3 pM genomic DNA is equivalent to 200 ng in   a 100-µL   reaction. The suggested protein concentration   of 40 nM is based   on a typical DNA-binding protein K _d , and may   be varied according   to the protein used.  
 

2. Incubate for 30 min at room temperature   (~22°C).  
 

3. Wash 10 µL of amylose resin with 100   µL DIP binding/wash   buffer in a 1.5-mL microcentrifuge   tube. Centrifuge at 2000 g   for 5 min in a microcentrifuge at   room temperature. Store on ^{ice until Step 4.}  

4. Add the protein/DNA   reaction mixture (100 µL) to the   washed amylose resin.   Rotate end-over-end on a Labquake rotator   for 15 min at room   temperature.  
It is essential that the resin remains in suspension   during   this step. Nutation is insufficient to maintain suspension.   If end-over-end rotation is unavailable, frequent (every 30-60   sec) pipetting may be substituted.  
 

5. Centrifuge the resin   at 2000 g   for 5 min.  
 

6. Wash the resin with 50 µL DIP   binding/wash buffer and   centrifuge at 2000 g   for 5 min.  
 

7.   Repeat Step 6 three additional times for a total of four   washes.   Remove the supernatant after the final wash.  
 

8. Elute the   protein by adding 50 µL DIP elution buffer   to the resin.  
 

9. Centrifuge the resin at 2000 g   for 5 min and collect the   supernatant.  
 

10. Isolate the DNA from the supernatant with   a Zymo-Spin I   column (or equivalent DNA purification method)   and elute into   the desired volume.  
A starting amount of 200   ng of yeast DNA (Step 1) typically   results in a total yield   of 16-30 ng of isolated DNA. The elution   volume may be adjusted   to achieve the desired final concentration.  
 

11. Amplify and   detect the isolated DNA.  
DNA isolated by DIP has been successfully   amplified by random   priming, ligation-mediated PCR, and Whole   Genome Amplification   (WGA) (Genomeplex WGA2; Sigma-Aldrich).   Detection options include   DNA microarray (DIP-chip) or high-throughput   sequencing (DIP-seq)   methods. Follow the sample preparation   recommendations for the   preferred detection technique; detailed   methods for the different   detection platforms vary by manufacturer   and are typically well-described.  

^{TROUBLESHOOTING}

^{Problem:   Protein recovery fails.}

^{[Step 8]}

^{Solution:   Confirm that the resin remains suspended in Step 4.   If the resin is not suspended, the protein will not bind efficiently.}

^{Problem:   Little to no enrichment of DNA is achieved.}

^{[Step 11]}

^{Solution:   Consider the following:}

 

• ^{Determine the optimal number of washes for the protein of interest.   The wash conditions were established using MBP-Leu3DBD, but   may vary based on the K _d   of the protein used. Cross-linking   the protein and DNA complex with formaldehyde or a similar reversible   cross-linker may enhance isolation of DNA when proteins with   a low-affinity for DNA are used, but this has not been rigorously   tested.}  
  
  
• ^{Confirm that the protein is capable of binding DNA,   because   some recombinant proteins fail to fold properly. If   the consensus   sequence is already known, traditional methods   (such as EMSA)   can be used to check the protein’s ability   to bind DNA.   As an alternative to using recombinant protein,   native protein   may be purified from the organism of interest.   Treatment of   the purified protein with DNase prior to DIP-chip   will likely   be required to remove residual native DNA.}  
  
  

^DISCUSSION

^{Traditional techniques for studying the in vitro binding specificity   of purified proteins, such as EMSA (Garner and Revzin 1981 ;   Ellington and Szostak 1990 ) and SELEX (Brenowitz et al. 1986 ;   Tuerk and Gold 1990 ), bypass the difficulties associated with   the in vivo ChIP-chip method (see Introduction). However, EMSA   typically requires some prior knowledge of binding specificity   and cannot assay a wide diversity of DNA sequences simultaneously, while SELEX is subject to over-selection of bound sequences,   which may cause significant but lower-affinity interactions   to be missed. More recently, several high-throughput techniques   have been developed; one example is protein binding microarray   (PBM), in which purified protein is directly incubated with   a DNA microarray. Detection of DNA fragments bound in the reaction   is achieved by fluorescently labeling the protein. Although   the arrays used can be comprised of PCR products or oligonucleotides   representing the entire genome, recent studies have focused   on using "universal" oligonucleotides designed to represent   every possible   k -mer (Mukherjee et al. 2004 ;   Berger and Bulyk 2006 ;   Bulyk 2007 ).}

^{Both PBMs and DIP-chip can be used to determine in vitro consensus   sequences for proteins by assaying millions of artificial or   natural DNA sequences in parallel. DIP-chip has been shown to   perform as well as traditional assays (SELEX and EMSA) and the   ChIP-chip method in identifying consensus sequences (Liu et al. 2005 ).   During DIP, the protein and DNA are in solution during the binding   reaction, allowing the protein to sample the DNA in three dimensions   and eliminating any physical hindrance caused by having the   DNA tethered to a surface. DIP also allows controlled manipulation   of the binding conditions, enabling quantitative conclusions   to be made about the effect of reaction parameters on DNA-binding   specificity. DIP-chip results are directly comparable to ChIP-chip   results, because the same genomic DNA template is used for binding.   By adding protein cofactors to DIP reactions and then comparing   the loci bound in vitro and in vivo, it may be possible to quantify   the relative contributions of various factors, such as chromatin   structure (Liu et al. 2006 ) and cofactor-assisted targeting.}

^REFERENCES

^[Medline]

^[Medline]

^[Medline]

^[Medline]

^{[Abstract/ Free  Full Text]}

^{[Abstract/ Free  Full Text]}

^{[Abstract/ Free  Full Text]}

^[Medline]

^{[Abstract/ Free  Full Text]}