Figure 1.
The functional chromatin landscape of CLL as defined by HDAC1 recruitment. (A) Comparison of HDAC1 and BRD4 levels in whole cell lysates of CLL (n = 8; CLL 1-8 = CLL251228, CLL250401, CLL251776, CLL250063, CLL250112, CLL250383, CLL250384, CLL250746) and independent healthy CD19+ selected B cells (n = 4). Cell lysates were sequentially probed for HDAC1, BRD4, and GAPDH on the same immunoblot. (B) Quantitation of HDAC1 and BRD4 in CLL (n = 16, CLL251228(U, 11q-/13q-), CLL250401(U, 11q-/13q-/17p-), CLL251776(Un, 11q-/13q-/17p-), CLL250063(Un, 13q-), CLL250112(Un, NK), CLL250383(Un, 11q-), CLL250384(U, 11q-), CLL250746(Un, NK), CLL251124(U, 13q-), CLL250776, CLL251275(U, 11q-/13q-), CLL250028 (M, 11q-/13q-), CLL250109(M, 13q-), CLL250146(M, 13q-), CLL250943(U, NK), CLL251456(M, NK)) and healthy CD19+ B-cell samples (n = 8). Expression of HDAC1 and BRD4 in healthy B cells and CLL cells were normalized to GAPDH (used as a loading control) and the statistical significance derived using Student two-tailed t tests; ∗P < .5, ∗∗∗P < .001, Graph pad software. (C) ChIP-seq densities ranked in decreasing order for the genome-wide combined occupancy signals of HDAC1 with BRD4, H3K9Ac, H3K27Ac, open ATAC signatures, and RNA Pol2 engagement (cluster I), HDAC1 corecruited with H3K9Ac, ATAC, and RNA Pol2 (cluster II), and HDAC alone (cluster III) centered ±5 kb of the TSS window in 3 CLL samples (CLL251288, CLL251766, and CLL251275). (D) Venn diagram showing the intersection of HDAC1 with BRD4, H3K9Ac, H3K27Ac, and RNA Pol2 recruitment common to 3 primary CLL samples; cluster I (HDAC-BRD4-H3K9Ac-H3K27Ac-ATAC-RNA Pol2) comprises 4799 (in red) peaks with robust HDAC-BRD4-H3K9Ac-H3K27Ac-ATAC-RNA Pol2 tag densities close to the TSS of 5122 total genes, of which 4192 were PCG and 123 were microRNA genes, cluster II (HDAC1-H3K9Ac-Pol2) comprises 5909 peaks (in blue) peaks with robust HDAC1-H3K9Ac-ATAC- Pol2 tag densities close to the TSS of total 5676 genes, of which 4934 were PCG and 164 microRNA genes, and cluster III (HDAC-low/no H3K9Ac-no Pol2) comprises 4626 peaks (in purple) with discrete HDAC1 tag densities with low or no H3K9Ac and minimal Pol2 tag densities close to the TSS of 1077 total genes, of which 927 were PCG and 49 were microRNA genes. (E) Transcriptional output of genes within the HDAC1-containing chromatin. Cluster I accounted for 25% of the CLL transcriptome with 2724 high expressors, 1625 medium expressors, and 773 low expressors. Cluster II accounted for 29% of the CLL transcriptome with 2165 high expressors, 2421 medium expressors, and 1090 low expressors. Cluster III accounted for 6% of the CLL transcriptome with 154 high expressors, 373 medium expressors, and 550 low expressors. In addition, there were 646 high expressors, 1270 medium expressors, and 3277 low expressors not marked by any of the regulatory chromatin modifiers in our study. Analysis of variance was used to compare values between high, medium, and low expressers in clusters I to III and in the group with none of the marks evaluated in our study; ∗∗∗∗P < .0001. 11q-, del11q, 13q-, del13q; 17p-, del17p; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; M, IgVH mutated; NK, normal karyotype; U, IgVH unmutated; Un, unknown.

The functional chromatin landscape of CLL as defined by HDAC1 recruitment. (A) Comparison of HDAC1 and BRD4 levels in whole cell lysates of CLL (n = 8; CLL 1-8 = CLL251228, CLL250401, CLL251776, CLL250063, CLL250112, CLL250383, CLL250384, CLL250746) and independent healthy CD19+ selected B cells (n = 4). Cell lysates were sequentially probed for HDAC1, BRD4, and GAPDH on the same immunoblot. (B) Quantitation of HDAC1 and BRD4 in CLL (n = 16, CLL251228(U, 11q-/13q-), CLL250401(U, 11q-/13q-/17p-), CLL251776(Un, 11q-/13q-/17p-), CLL250063(Un, 13q-), CLL250112(Un, NK), CLL250383(Un, 11q-), CLL250384(U, 11q-), CLL250746(Un, NK), CLL251124(U, 13q-), CLL250776, CLL251275(U, 11q-/13q-), CLL250028 (M, 11q-/13q-), CLL250109(M, 13q-), CLL250146(M, 13q-), CLL250943(U, NK), CLL251456(M, NK)) and healthy CD19+ B-cell samples (n = 8). Expression of HDAC1 and BRD4 in healthy B cells and CLL cells were normalized to GAPDH (used as a loading control) and the statistical significance derived using Student two-tailed t tests; ∗P < .5, ∗∗∗P < .001, Graph pad software. (C) ChIP-seq densities ranked in decreasing order for the genome-wide combined occupancy signals of HDAC1 with BRD4, H3K9Ac, H3K27Ac, open ATAC signatures, and RNA Pol2 engagement (cluster I), HDAC1 corecruited with H3K9Ac, ATAC, and RNA Pol2 (cluster II), and HDAC alone (cluster III) centered ±5 kb of the TSS window in 3 CLL samples (CLL251288, CLL251766, and CLL251275). (D) Venn diagram showing the intersection of HDAC1 with BRD4, H3K9Ac, H3K27Ac, and RNA Pol2 recruitment common to 3 primary CLL samples; cluster I (HDAC-BRD4-H3K9Ac-H3K27Ac-ATAC-RNA Pol2) comprises 4799 (in red) peaks with robust HDAC-BRD4-H3K9Ac-H3K27Ac-ATAC-RNA Pol2 tag densities close to the TSS of 5122 total genes, of which 4192 were PCG and 123 were microRNA genes, cluster II (HDAC1-H3K9Ac-Pol2) comprises 5909 peaks (in blue) peaks with robust HDAC1-H3K9Ac-ATAC- Pol2 tag densities close to the TSS of total 5676 genes, of which 4934 were PCG and 164 microRNA genes, and cluster III (HDAC-low/no H3K9Ac-no Pol2) comprises 4626 peaks (in purple) with discrete HDAC1 tag densities with low or no H3K9Ac and minimal Pol2 tag densities close to the TSS of 1077 total genes, of which 927 were PCG and 49 were microRNA genes. (E) Transcriptional output of genes within the HDAC1-containing chromatin. Cluster I accounted for 25% of the CLL transcriptome with 2724 high expressors, 1625 medium expressors, and 773 low expressors. Cluster II accounted for 29% of the CLL transcriptome with 2165 high expressors, 2421 medium expressors, and 1090 low expressors. Cluster III accounted for 6% of the CLL transcriptome with 154 high expressors, 373 medium expressors, and 550 low expressors. In addition, there were 646 high expressors, 1270 medium expressors, and 3277 low expressors not marked by any of the regulatory chromatin modifiers in our study. Analysis of variance was used to compare values between high, medium, and low expressers in clusters I to III and in the group with none of the marks evaluated in our study; ∗∗∗∗P < .0001. 11q-, del11q, 13q-, del13q; 17p-, del17p; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; M, IgVH mutated; NK, normal karyotype; U, IgVH unmutated; Un, unknown.

Close Modal
This Feature Is Available To Subscribers Only

or Create an Account

Close Modal
Close Modal