Fig. 4

Composition of HIV DNA in blood samples from HIV-1 infected patients. Percentages of a uDNA among total HIV DNA, b 2-LTR circles among total HIV DNA and c 2-LTR circles among the totality of uDNA forms in Off-ART patients (n = 12), viremic (n = 14) and aviremic (n = 29) On-ART patients, and untreated infected controller patients (HIC, n = 20). Each bar within a group represents the datum from a single sample. The percentages refer to data expressed as HIV DNA copies per µg of DNA. The percentages of uDNA or 2-LTR circles in total HIV DNA and of 2-LTR circles in uDNA were reduced to 0% when uDNA or 2-LTR circle copy numbers were < QL of the assays. When both uDNA and 2-LTR circle copy numbers were < QL, the fractions are reported as non-determinable (ND). d The percentages from the above a-b panels are presented as mean ± SEM. Subtracting uDNA and 2-LTR circles from total HIV DNA, the remaining fraction is constituted by integrated HIV DNA (black dotted line). e The percentages from the above c panel are presented as mean ± SEM. Subtracting the 2-LTR circle DNA fraction from uDNA, the remaining fraction is constituted by the sum of 1-LTR circles, 2-LTR circles with anomalous junction sequence, rearranged circles and linear HIV DNA (red dotted line). The Kruskal–Wallis test and Dunn’s multiple comparison test were used for comparisons among groups of patients (black asterisks), the Wilcoxon signed rank test was used for matched samples (red asterisks), only p < 0.05 are reported; *p < 0.05; **p < 0.01; ***p < 0.001