Shown as the mean ?standard purchase GW610742 deviation (error bars) of at least
Shown as the mean ?standard deviation (error bars) of at least three independent sets of experiments. The number of selected clones is also shown at the top of the histogram.templates even at low histones concentrations (Figure 2C). Similar results were obtained using different enzyme constructs, preparations, purification procedure and various viral DNA substrates (see data reported for HIV-1 IN in Additional file 4: Figure S4). To better ascertain the relationship between integration, nucleosomes positions and chromatin structure, we mapped more precisely the integration sites obtained from retroviral integrases on the different chromatin structures of the acceptor plasmid.Effect of nucleosomes density on in vitro HIV-1, PFV, ASV and MLV integration selectivityThe integration site targeting in the two regions of the plasmid differing in their nucleosomes density and stability was first studied. Fifty integrants previously obtained with naked and chromatinized p5S acceptors using HIV-1, PFV, ASV and MLV IN were cloned, sequenced and the positions of the integration sites werecompared. As shown in Figure 3, most of the HIV-1 integration sites mapped on the chromatinized templates were found in the region of low nucleosome occupancy outside the 5S-G5E4 nucleosome positioning region 1 (86 of total analyzed sites), thus confirming previously reported results [36]. Similar profile was observed with ASV integrase with a more drastic redistribution of the integration sites in the lower nucleosome density region of the chromatinized vector. However, in contrast to HIV-1 and ASV integration, the 5S-G5E4 positioning region 1 of the plasmid containing dense nucleosomes was not found refractory to PFV and MLV integration. Indeed these INs were found to accommodate both regions of the plasmid with a significant preference for the nucleosome dense region 1 (64 of sites analyzed were found within the 5S-G5E4 fragment for PFV and 74 for MLV). To correlate more precisely the nucleosomes positions with integration sites we analyzed more in details theirBenleulmi et al. Retrovirology (2015) 12:Page 6 ofFigure 3 Localization of the HIV-1, PFV, MLV and ASV integration sites in both regions of the naked or chromatinized acceptor DNA. Fifty PFV, HIV-1, ASV and MLV integrants carrying the correct target DNA duplications (respectively 4, 5, 6 and PubMed ID: 4 bp) were selected from the concerted integration shown in Figure 2 using p5S vector, either under the naked structure or chromatinized with a DNA/histones ratio of 1/1.3. The positions of the different integration events were identified and shown in region 1 or region 2. The values are plotted as the number of ampicillin-, kanamycin- and tetracycline-resistant selected clones as a percentage of integration reaction control performed with naked vectors and as the mean ?standard deviation (error bars) of at least three independent sets of experiments. The number of selected clones is also shown at the top of the histogram. A Student test was performed on serial values and the significant p values are reported in the figure.localization in the 5S-G5E4 region 1 containing nucleosomes assembled at known positions predicted PubMed ID: from the algorithm described in [53] (and derived from [54]) and experimentally validated (see Additional file 2: Figure S2). The analysis was performed on 50 integration sites previously selected within the 5S-G5E4 region for each enzyme. PFV and MLV sites were found to be preferentially enriched.