|Home | About | Journals | Submit | Contact Us | Français|
In order to fully understand the functions of a DNA-binding protein it is necessary to identify all of its binding sites in chromosomes, so that the role of each site in the overall biological function of the factor can be assessed. An approach called ChIP-on-Chip, which combines the chromatin immunoprecipitation technique with chromosomal DNA microarray analysis, has proven to be a powerful means for the chromosome-wide identification of protein binding sites. This approach can also be used to characterize chromosome-wide variations in patterns of post-translational protein modifications, for example histone modifications. This chapter presents methodologies for the ChIP-on-Chip analysis, using as an example the identification of chromosome-wide binding sites for the TATA-binding protein in mitotic cells.
One method for identifying new binding sites of factors for which a consensus recognition sequence has been determined is to perform database searches to identify matches to this sequence in the genome. The chromatin immunoprecipitation technique could then be used to test whether these identified sites are actually bound by the factor. However, this approach is not only inefficient, in that it would likely involve analyzing many sites that turn out not to be actual binding sites for the protein, but would also likely miss many actual binding sites whose sequences are similar to but don’t exactly match the test sequence(s) used to perform the database searching. In the ChIP-on-Chip approach, DNA fragments isolated by chromatin immunoprecipitation using an antibody against the protein of interest are used to probe a DNA Chip containing an array of genomic DNA sequences. Thus, this experimental methodology has the advantages of being able a) to identify binding sites of a protein in a way that is not biased with respect to the DNA sequence bound, and b) to simultaneously identify binding sites over very large regions of genomic DNA. This technique can not only provide important new insights into the genomic DNA binding sites of DNA-binding proteins, but also into the variations in patterns of post-translational modifications of proteins such as histones throughout chromosomes (1–8). In this chapter we describe the key experimental details involved in performing ChIP-on-Chip experiments. To illustrate the types of data that can be obtained using these methodologies we present figures showing the results of chromatin immunoprecipitation and ChIP-on-Chip identification of chromosome-wide binding sites of the general transcription factor TATA-binding protein (TBP) in mitotic cells (9).
In this first part of the ChIP-on-Chip experiment, cells are subjected to the chromatin immunoprecipitation assay using an antibody against the protein of interest in order to obtain protein-bound DNA fragments that will later be labeled and used to probe the Chip arrays containing genomic DNA sequences in order to identify new binding sites for the protein. In addition, by performing PCR analysis of the resulting DNA fragments obtained from the samples isolated by the antibodies against the protein vs. non-specific control antibodies using primers that amplify a known binding site for the protein as well as a site which is known not to bind the factor, one can also verify the specificity of the ChIP part of the procedure. This is important to prevent wasting substantial time and effort that would result from using sub-optimal DNA fragment sets to probe the DNA Chip arrays in the second part of the ChIP-on-Chip experiment. Figure 1 shows an example of such PCR analysis, in this case quantitative PCR analysis, to show the specific binding of TBP to the histone H4 promoter, a known binding site for this protein, in mitotic cells.
If the results of the analysis described at the end of part 3.1 above shows the specificity of the ChIP DNAs, these DNA fragments can then be used to probe the DNA Chip arrays containing the genomic sequences. In the first part of this second half of the ChIP-on-Chip, the DNA fragments isolated by immunoprecipitation using antibodies against the protein of interest, and the control Input DNA sample, is amplified by PCR and then labeled with biotin. These labeled DNAs are then incubated with the Chips containing the genomic DNA arrays and the DNA spots that hybridize with the probes identified using specialized instruments designed for these purpose (e.g. Affymetrix equipment). The binding sites for the factor of interest are then determined by analyzing the data using computer software. Figures 2, ,3,3, and and44 illustrate the types of data output that can result from ChIP-on-Chip studies, in this case from our studies that characterized the binding sites of the TBP protein within chromosomal DNA of mitotic cells(9).
We are very grateful to David Rodgers for allowing us to use computers in his lab to analyze the data from our ChIP-on-Chip experiments.
1Non-adherent cells are often chosen because a) it is easier to grow the large numbers of cells required for these types of experiments, for example using spinner flasks, compared to adherent cells, and b) it is likely that cross-linking is more efficient in such cells because more of their surface is exposed to the medium.
2The sonication conditions used, such as the energy and duration, are dependent on the sonicator employed and must be empirically tested to find the conditions that yield fragments in this size range as observed on polyacrylamide gels stained with ethidium bromide.
3In the Affymetrix protocol we followed, the initial round of linear amplification was carried out using random primer A (5′-GTTTCCCAGTCACGGTC(N)9-3′), and utilized 4 cycles of: 95°C, 4 minutes, 10°C, 5 minutes, ramp from 10°C to 37°C over 9 minutes, followed by 8 minutes at 37°C, using sequenase.
4In the second round of amplification, in the Affymetrix protocol the DNA from the first round of amplification is then amplified using primer B (5′-GTTTCCCAGTCACGGTC-3′) by Taq polymerase PCR using 32 cycles of: 95°C, 30s, 45°C, 30s, 55°C, 30s, 72°C, 1 minute.
5For our experiments we used the Affymetrix Genechip Human tiling array 2.0 c, which contains genomic DNA spanning human chromosomes 3, 21, 22, X, and Y, including intergenic regions, promoters, untranslated regions, exons, introns etc. However, DNA Chips are also available in which only known promoter regions are included, which may be desirable if you are studying a transcription factor and your goal is just to determine which promoters are bound by the factor of interest. The benefit of such a selective genomic DNA Chip is that one Chip can contain all the sequences one is interested in examining, whereas you would have to buy and hybridize many different Chips if you want to examine all of the genomic sequence. Of course, the disadvantage of using a selective genomic DNA sequence Chip is that it would miss any binding sites for your factor that are not in the promoter sequences selected for inclusion on the Chip (e.g. not-yet-identified promoters, control DNA sequences that are within introns, or far upstream of proximal promoter regions, etc.). To have confidence in the results of the probing, it is suggested that 3 of each Chip be probed in each experiment (i.e. probing done in triplicate).
6The steps of the procedure involving Chip hybridization, staining with fluorescent streptavidin-phycoerythrin, and scanning are typically done in a dedicated Microarray Facility that has the required specialized machines. For example, in our study the University of Kentucky Microarray Facility performed these steps using equipment including an Affymetrix GeneChip hybridization incubator 640, Fluidics Station 450, and GeneChip Scanner 3000 7G.
7In our study data were collected and analyzed using GeneChip Operating Software (GCOS) and TAS software (Affymetrix). The results (from three TBP IP DNA-probed chips and three genomic input DNA-probed chips) were quantile-normalized within treatment/control replicate groups. A Wilcox Rank Sum test was applied to the transformation log2 for data, testing the null hypothesis of the equality of the two population distribution functions against the alternative of a positive difference on location between the probability distribution of the treatment and that of the control. The Wilcoxon test was applied in a sliding window across the chromosomes. The chromosomal positions bound by TBP were identified based on a P value cut off of 0.005. Results were visualized using IGB software (Affymetrix) and the UCSC GenomeViewer, and compared to the RefSeq database (NCBI).