Notes

1. Use at least 100 mouse islets for each immunoprecipitation. Take into account that a complete experiment includes at least two immunoprecipitations (i.e., the antibody of interest and a control immunoprecipitation performed with nonimmune serum or an unrelated antibody). It is important to proceed to fixation shortly after isolation, as islets tend to aggregate with time, constituting large structures that become difficult to sonicate. Using small amounts of DNase (e.g., 0.1 mg/mL) during the pancreas digestion can also be necessary to prevent the formation of large islet clusters.

2. Fixation is one of the critical steps of the process. Thus, temperature and time need to be optimized for each new antibody tested (higher temperatures accelerate the process of fixation). A time-course experiment at room temperature is an advisable first step when testing a new antibody. Excess crosslinking can affect sonication quality of the samples and results in reduced antigen availability, thus interfering with the immunoprecipitation step. Ten minutes at room temperature represents a typical fixation condition.

3. Adjust resuspension volume according to islet number. Use about 150-200 iL for a minimum of300-400 islets; otherwise, sonication is inefficient when smaller volumes are used. The volume can be increased to 300 iL if more tissue is used. Note that because the sample is later diluted 10-fold prior to immunoprecipitation, approx 50 iL of this solution corresponds to one immunoprecipitation reaction.

4. Effective sonication is essential for a successful ChIP assay. Sonication effectiveness depends on the quality of the sample, the extent of the fixation, the volume of the sample, and the sonication technique. The sonifier probe and the tube need to be in a firmly fixed position. Preliminary experiments may be needed to determine the optimal position of the probe tip relative to the liquid surface and the bottom of

Fig. 3. Example of multiplex PCR analysis of a ChIP experiment using mouse isolated islets. Approximately 400 islets were processed as described in the text and divided in aliquots, so that 100 islets were immunoprecipitated with anti-tetra-acetylated his-tone H4 antibody (UBI) (AcH4). The selected DNA was analyzed by multiplex PCR and the results were compared to (1) the amplification pattern of the precipitation performed with a nonimmune serum (NI), corresponding to nonspecifically selected DNA, (2) the amplification of a series of dilutions of the input DNA, which reveals the expected amplification pattern in the absence of enrichment, and (3) the amplification pattern of the DNA selected by the antibody in another tissue—in this case, mouse hepatocytes. We used primers for a promoter known to be active in both islets and liver (E-Cadherin, positive control), a promoter active in liver but not in islets (hnf-4a P1 promoter, negative control), and a promoter active in islets but not in the liver (hnf-4a P2 promoter, our problem gene) (12). The results show that P2 but not P1 chromatin is hyperacety-lated in islets, whereas the opposite pattern is observed in hepatocytes.

Fig. 3. Example of multiplex PCR analysis of a ChIP experiment using mouse isolated islets. Approximately 400 islets were processed as described in the text and divided in aliquots, so that 100 islets were immunoprecipitated with anti-tetra-acetylated his-tone H4 antibody (UBI) (AcH4). The selected DNA was analyzed by multiplex PCR and the results were compared to (1) the amplification pattern of the precipitation performed with a nonimmune serum (NI), corresponding to nonspecifically selected DNA, (2) the amplification of a series of dilutions of the input DNA, which reveals the expected amplification pattern in the absence of enrichment, and (3) the amplification pattern of the DNA selected by the antibody in another tissue—in this case, mouse hepatocytes. We used primers for a promoter known to be active in both islets and liver (E-Cadherin, positive control), a promoter active in liver but not in islets (hnf-4a P1 promoter, negative control), and a promoter active in islets but not in the liver (hnf-4a P2 promoter, our problem gene) (12). The results show that P2 but not P1 chromatin is hyperacety-lated in islets, whereas the opposite pattern is observed in hepatocytes.

the tube to allow the most efficient sonication results without excessive foaming. We have found that for small volumes such as that required for islet cell experiments, sonication in buffers that do not contain SDS results in an excessive amount of high molecular-weight DNA. Overheating during the sonication process needs to be avoided as it can potentially result in crosslink reversal. Glass beads (1/5 volume) are used by some investigators to increase physical shearing of the DNA.

5. High-salt buffer is commonly used, but low-salt buffer can be used to favor immunoprecipitation in cases when it is necessary (e.g., when working with limiting samples or when low specific enrichment is obtained with a particular antibody).

6. Use preimmune serum, nonimmune IgGs, or unrelated serum, preferably against a non-nuclear protein. This negative control is intended to detect the DNA fragments that are nonspecifically precipitated by interacting with protein A or by forming aggregates. If saving material is an issue, this control may be avoided and the specificity of the precipitation may be checked by testing the presence or absence of specific DNA sequences in the precipitate. However, this is only feasible if appropriate positive and negative gene controls are known.

7. Alternatively, take an aliquot prior to immunoprecipitation, but this saves material if limiting. Note that this DNA cannot be used for checking the extent of sonication, as it contains salmon sperm DNA.

8. The volume of resuspension depends on the initial cell number. Resuspend in as small a volume as possible (e.g., 20-30 iL) and test PCR on undiluted and diluted samples. Generally, input DNA needs to be diluted a minimum of 50 times.

9. The purpose of coamplifying more than one amplimer is to ensure that the ChIP experiment has worked properly in all samples. Spurious carryover of DNA fragments (rather than enrichment as a result of specific antibody recognition) in some samples can occur, as well as sample-to-sample differences in the recovery of specifically immunoprecipitated chromatin. Intra-assay variation in PCR efficiency resulting from carryover of inhibitors can also occur. Using small amounts of tissue as starting material can increase the occurrence of such problems, leading to completely incorrect conclusions. In our view, the best way to overcome these problems is to ensure that the immunoprecipitated DNA exhibits a pattern of target amplification relative to negative and positive control segments that is consistently distinct from preimmune antibody precipitation and input DNA samples. This can be achieved with two sets of duplex DNA pairs or with a single triplex combination, as discussed in the protocol. Nonetheless, coamplification requires optimization of mutually compatible primer pairs. An alternative option is to run multiple parallel reactions with different primer pairs. Real-time PCR can be used to increase the ability to accurately quantitate small changes in target DNA.

The use of appropriate PCR controls as discussed here does not ensure that the DNA segments that have been enriched in the test antibody sample correspond solely to chromatin segments containing the intended epitope. Additional experiments are required to verify the specificity of the antibody under the immunopre-cipitation conditions used. Ideally, this can be accomplished by comparison of control cellular sources that are known to either contain or lack the intended epi-tope. Some examples are cell lines that are transfected with a DNA-binding protein of interest versus a control vector, different tissues that are known to express or not such a protein, or a genetic null mutant model.

0 0

Post a comment