As discussed in the chapter, the most direct method to identify the sites at which trans cription factors bind in vivo is to use chromatin immunoprecipitation, known as ChIP. The name accurately summarizes the technique, which is depicted in Figure 14.20, but some elaboration could be helpful. Transcription factors bind to specific DNA sequences, but these DNA sequences are found embedded in the chromosome, as part of the DNA–protein complex known as chromatin. The goal of ChIP is to identify the DNA sequence and its location in the genome; by using this location in the genome, it is possible to infer which gene is regulated by the transcription factor. Thus, ChIP has to be able to isolate one specific protein and its short binding site, amid a complex of many proteins and long stretches of DNA.
The key to ChIP is an antibody, the “immuno” part of the name. Antibodies (or immunoglobulins) are proteins made by the B cells of the mammalian immune system and are used to neutralize proteins from invaders such as bacteria and viruses. The B cells recognize a segment of the foreign protein, known as an antigen, and stimulate the production of antibodies that are highly specific to that antigen. The specificity of the antigen:antibody interaction has provided a versatile and powerful tool for detecting, localizing, and isolating proteins from a cell.
As an experimental tool, antibodies are made by injecting the protein of interest into a mammal, such as a rabbit, which makes antibodies against it; these antibodies can then be purified from the serum of the rabbit. The antibodies themselves have a constant and variable region; the variable region is the part of the antibody that interacts with the antigen, while the constant region, which is common to many different antibodies with different variable regions, is the part used to isolate or detect the antibody. For ChIP, the transcription factor of interest is used as an antigen.
Since production of antibodies specific to each of the hundreds of transcription factors in an organism is often a laborious step, another procedure is to clone the gene for the transcription factor into a plasmid with a common protein sequence, so that every transcription factor is produced as a fusion protein with the same common sequence. This common sequence is known as an epitope. Antibodies to the epitope can be used to isolate any transcription factor of interest; this is known as epitope tagging.
In whichever way an antibody to a transcription factor is produced, these form the basic components of a ChIP experiment. The process itself is summarized in Figure 14.20. Nuclei are isolated, and the protein interactions and protein–DNA interactions in chromatin are stabilized by chemical cross-links. The chromatin is fractionated, such as by sonication, to produce randomly sized fragments in solution. The antibody to the transcription factor is then added to the solution where it interacts with its antigen. The antibody:antigen complex is precipitated out of solution and purified. The cross-links that hold the complexes together are then reversed to release the individual components. Thus, immunoprecipitation is also widely used to study protein–protein interactions, although this is not the goal of ChIP. For ChIP, the associated DNA is isolated, and its sequence is determined.
Originally, the sequence of the DNA was determined using microarrays; since microarrays are nicknamed “chips,” this procedure is known as ChIP-chip. It is now more common to sequence the DNA directly, so the procedure is known as ChIP-seq.
The DNA purified from immunoprecipitation is 50–200 bases long. Part of this sequence is the binding site for the transcription factor, which is typically about 8–12 bases. The other part of the sequence is the region flanking the binding site. This can be compared to the genome sequence of the organism to determine the location at which the transcription factor is bound.
Thousands of ChIP assays have been performed, so a large volume of data is available. Nonetheless, they have a few limitations. Since eukaryotes have hundreds of transcription factors, each of which needs to be analyzed individually, many transcription factors have not yet been analyzed. More notably, the binding of transcription factors to their sites is highly dynamic, changing with different cell types and different conditions. ChIP provides a snapshot of where a transcription factor was bound in a cell at a particular time but does not yet capture the changes that occur in all the many cells under all of the conditions. Even with these recognized limitations, ChIP experiments dominate the literature on transcription factor binding sites and chromatin structure.