The chromatin immunoprecipitation (ChIP) assay is a powerful technique to determine the presence of a protein on a specific DNA region, depending on the cellular context. This method is commonly used to study chromatin extracted from cell culture (1). Briefly, it is based on three major steps: (i) cell fixation with formaldehyde to crosslink proteins and DNA; (ii) chromatin fragmentation by sonication; and (iii) selective immunoprecipitation of DNA-protein complexes with a specific antibody. The immunoprecipited DNA is analyzed by PCR. Recently, the ChIP assay has been used to study in vivo the mechanisms of transcriptional regulation by thyroid hormone receptors during Xenopus laevis development (2,3).

We adapted for the first time the ChIP assay to whole Danio rerio (zebrafish) embryos, in order to study the mechanisms controlling early brain patterning. The krox20 (krx20 in zebrafish) gene is involved in this process, and its expression pattern has been studied previously by in situ hybridization (4). krx20 expression is first activated at 100% epiboly [10 h postfertilization (hpf)], and the gene is expressed in two hindbrain segments until around 30 hpf. A good indicator of the activation of a gene is the hyperacetylation of its promoter (5). We investigated the level of histone acetylation on the krx20 promoter by ChIP, using an antibody directed against acetylated histone H4 (Upstate Biotechnology, Lake Placid, NY, USA) at different stages of zebrafish development, from 30% epiboly (4.7 hpf) to six-somites (12 hpf).

First, to remove chorions, the embryos were incubated for 2 min at room temperature in 1 mg/mL Pronase in Embryo medium E3 (6), then transfered in E3 where the chorion was removed with forceps. To eliminate the lipid-rich vitellus and isolate nuclei, dechorionated embryos were homogenized in 1 mL of a 2.2 M sucrose, 3 mM CaCl₂, 10 mM Tris-HCl, pH 7.5, 0.5% Triton® X-100, 1 mM phenylmethylsulphonyl fluoride (PMSF), 1 µg/mL aprotinin, 1 µg/mL pepstatin solution (7) and centrifuged at 100,000× g for 3 h at 4°C in a TL55 rotor (Beckman Coulter, Fullerton, CA, USA). Next, pelleted nuclei were resuspended in 1 mL of a 0.25 M sucrose, 10 mM Tris-HCl, pH 7.5, 3 mM CaCl₂, 1 mM PMSF, 1 µg/mL aprotinin, 1 µg/mL pepstatin solution (2), then centrifuged at 6000× g for 5 min at 4°C, and resuspended in 360 µL of this solution. Proteins were cross-linked to DNA by the addition of 10 µL of a 37% formaldehyde solution, with a 10-min incubation on ice, then a 20-min incubation at room temperature. The nuclei were pelleted at 6000× g for 5 min at 4°C, resuspended in 400 µL sodium dodecyl sulfate (SDS) lysis buffer with anti-proteases (2) (this buffer is provided in the ChIP assay kit; Upstate Biotechnology), and incubated for 10 min on ice before sonication. The average amount of chromatin recovered per embryo was 0.4 µg at the blastula stage, 1 µg at 100% epiboly, 1.3 µg at the four-somite stage, and 1.5 µg at 24 hpf.

The ChIP assay requires the segmentation of the chromatin into 200- to 800-bp immunoprecipitable fragments. We determined the optimal sonication conditions for whole embryos at different stages ((Figure 1)). The embryo lysate was subjected to 16 sets of 10-s pulses using a Vibra Cell sonicator 75041 (Bioblock Scientific, Illkirch, France) equipped with a 2-mm tip and set to 150 W. The lysate was left for 4 min on ice between each pulse. Two 10-µL samples were taken from the sonicated lysate after a different number of pulses. One of the two series of samples was centrifuged at 15,500× g for 10 min at 4°C to recover the supernatant. The two series were then loaded on a 1% agarose gel with ethidium bromide ((Figure 1)). We observed that shearing zebrafish chromatin into 200-to 800-bp fragments required 14 pulses independent of the developmental stage, from blastula to 24 hpf, for chromatin amounts between 50 and 150 µg. Neither RNase treatment nor the use of embryos from different stages modified the sonication profile (data not shown).

Figure 1.

We then carried out the ChIP assay to determine the acetylation level of the krx20 promoter during zebrafish development. Each immunoprecipitation and input (chromatin isolated before immunoprecipitation) required 10 µg chromatin. The IP protocol was done using the ChIP assay kit mentioned above, in the conditions recommended by the supplier. After the ChIP protocol, the recovered DNA was resuspended in 20 µL water for ChIP samples and 40 µL for the input. Eight micrograms of anti-acetylated histone H4 antiserum were added to 1 mL chromatin solution, from 30% epiboly (4.7 hpf), one-somite stage (11 hpf), four-somite stage (11.5 hpf), and six-somite stage (12 hpf) embryos. All PCRs were performed in 50 µL with 25 µL PCR Master Mix (AB gene, Courtaboeuf, France), 2 µL each primer (2 µM each; MWG-Biotech, Roissy, France), and 2 µL of the immunoprecipitated DNA sample. After a first denaturation of 5 min at 94°C, between 35 and 40 cycles were performed as follows: 1 min at 94°C, 1 min at annealing temperature, and 1 min at 72°C. A final elongation step of 5 min at 72°C followed. PCR analysis was performed using primers framing the -425 to -115 region of the krx20 promoter ((Table 1)). Forty cycles of PCR were performed, with an annealing temperature of 56°C. We observed that histone H4 on the krx20 promoter was not acetylated at 30% epiboly ((Figure 2)A). Acetylation was observed from the one-somite stage onward and persisted at the four- and six-somite stages ((Figure 2)A). These results are consistent with the krx20 expression pattern observed by in situ hybridization (4). To demonstrate that the procedure does not preferentially isolate nuclei from one tissue or cell type relative to others, we studied the acetylation of histone H4 in six-somite stage embryos on the promoters of myoD and keratin8, genes that are specifically expressed, respectively, in somitic mesoderm and in the enveloping layer of the embryo (8,9). For primers framing the -121 to +9 region of the myoD promoter ((Table 1)), 35 cycles were performed with an annealing temperature of 55 °C. For primers framing the -471 to -255 region of the keratin8 promoter ((Table 1)), 40 cycles were performed, with an annealing temperature of 56°C. Histone H4 was found to be acetylated on the promoters of these two genes ((Figure 2)A).