The eukaryotic genome in the nucleus is compacted into a highly complex structure known as chromatin, which is an important component in the control of gene transcription. One of the first events required for gene expression is the remodeling of chromatin into an accessible state. Chromatin remodeling associated with gene activation can be measured by evaluating the accessibility of intact nuclei to micrococcal nuclease (1) or restriction endonuclease enzymes. Restriction endonuclease assays involve the treatment of nuclei with a restriction enzyme, followed by detection of chromatin cleavage by Southern blot hybridization (2), direct realtime PCR (3) or ligation-mediated PCR (LM-PCR) (4). The main limitation of the first two detection approaches is the requirement of higher numbers of cells; while LM-PCR overcomes that problem, it is poorly quantifiable. Furthermore, none of these techniques have originally been conceived to measure the differential accessibility of individual members of a multigene family.

Changes in chromatin accessibility in one gene member from a multigene V_H immunoglobulin family had been previously performed by using LM-PCR and Southern blot analysis (5). In that work, data were quantitated by comparing band intensities. Here we tested whether chromatin accessibility of individual members of a highly homologous gene family could be efficiently measured by combining LM-PCR with realtime PCR. We chose the immunoglobulin V_HS107 family because its three functional members have very different rearrangement frequencies, with V1 rearranging five times more than V11, and V13 rearranging forty times less frequently than V1 (6). Furthermore, we previously demonstrated a positive correlation between the extent of the association with acetylated histones and relative rearrangement frequency within the V_HS107 family (7). Based on those observations and the fact that gene rearrangement in the immune system is preceded by increased chromatin accessibility (8), we considered the V_HS107 family an ideal model to evaluate the efficacy of our assay.

We tested our approach in freshly isolated mouse pro-B cells, where the V_HS107 family actively rearranges. We took advantage of the existence of a shared NlaIII restriction site downstream of the recombination signal sequence (RSS) in all three V_HS107 family members (Figure 1A).

Figure 1. Combined use of ligation-mediated PCR (LM-PCR) with real-time PCR for quantifying chromatin accessibility of individual gene family members. (Click to enlarge)

Intact nuclei were prepared according to Hempel and Ferrier (9) with some minor modifications (see the supplementary material available online at www.BioTechniques.com). Approximately 10⁶ nuclei were treated with 0, 2, or 10 U NlaIII at 37°C for 10 min. After overnight incubation in 2× DNA extraction buffer, DNA was recovered by phenol/chloroform/isoamylalcohol extraction and overnight isopropanol precipitation. DNA was then dissolved in 1× Tris EDTA (TE), and the concentration was measured by PicoGreen® (Invitrogen, Carlsbad, CA, USA). Firststrand synthesis was performed on 200 ng using the forward family-specific primer V107-5′, which evenly picks up all three family members, followed by overnight ligation of the unidirectional BW linker (3) at 17°C (Table 1). Linkerligated DNA samples were recovered by overnight isopropanol precipitation, followed by resuspension in1×TE.

Table 1. Oligonucleotides Used in Chromatin Accessibility Assays (Click to enlarge)

Half of the sample volume was used in a first round of PCR using the V107-5′ and BW3 primers (18 cycles). DNA quantification was performed using PicoGreen, and all samples were diluted to 25 pg/L in 1× TE. Fifty picograms of each sample were used to perform real-time PCR with SYBR® Green. In this step, genespecific forward primers were used—each of them paired with the BW3 primer, which is complementary not only to the unidirectional BW linker, but also to the last two base pairs of the first-strand synthesized product (Figure 1B). The amount of product generated is proportional to the amount of digestion occurring across the regions amplified for the primers. Thus, if the region under study has an increased accessibility and hence increased digestion, more PCR product will be generated, which in real-time PCR is reflected by a lower number of amplification cycles.

We included in our final calculations the differential amplification efficiency of each pair of primers. This is a critical step, because omitting this calculation could lead to a misinterpretation of the real-time PCR data. Therefore, after PCR using a gene-specific forward primer and the BW3 reverse primer, the V1, V11, and V13 products were purified from agarose gels. The purified PCR products for V1, V11, and V13 were quantified for DNA concentration using PicoGreen. Samples were diluted to the same number of amplicon copies. Because there are differences in sizes between the V1, V11, and V13 PCR products, we considered it more appropriate to dilute samples based on the number of amplicon copies rather than on their DNA concentration. Amplicon copy calculations take into consideration both the fragment size and the mass of one amplicon molecule.

As a practical test for PCR efficiency, we constructed a standard curve by using 10-fold dilutions of each PCR product (10,11). The quantification was made by SYBR Green real-time PCR. For normalization purposes, the whole product signal was measured by using a family-specific forward primer (V107co-5′) and the BW3 primer. After determination of the cycle threshold (C_T) values for each primer combination (C_Tcom), the C_T value from the whole product signal (C_Tp) was subtracted from C_Tcom (C_Tcom-C_Tp). The difference was named C_T control. So, when differences in primer efficiency where taken into consideration, chromatin accessibility of individual gene family members were calculated as 2-(C_{T sample}-C_{T control}), where C_{T sample} is the C_T value for a specific gene, and C_{T control} is the efficiency of the respective primer combination. Data are finally presented as relative to the chromatin accessibility of the V1 gene. We found that at very limiting conditions of enzyme concentration and incubation time, V1 had the highest accessibility over V11 and V13, which agrees with its highest rearrangement frequency in that family (Figure 2).

Figure 2. Sensitive and quantitative measurement of chromatin accessibility of individual members of a highly homologous gene family. (Click to enlarge)

In the present study, we have improved the quantification of chromatin accessibility of individual gene family members by combining LM-PCR with real-time PCR. Our assay allows the use of a small amount of sample and is both sensitive and quantitative. Furthermore, by considering the primer efficiency differences in the final calculation, the quantification is more accurate as opposed to simply measuring and comparing individual C_T values.