Fig 2: UTX interacts with MiDAC under endogenous conditions.Western blot of UTX and DNTTIP1 IPs from WT and Mll3/4 DKO mESCs confirming interaction of UTX with MiDAC under endogenous conditions. A UTX antibody immunoprecipitates UTX and the MiDAC subunits DNTTIP1, ELMSAN1, and HDAC1. A DNTTIP1 IP identified DNTTIP1, ELMSAN1, and HDAC1 as members of MiDAC along with UTX. Nuclear extracts were used as input. An IgG antibody was used as an IP control. Panel 1 (DNTTIP1) and panel 4 (UTX) were assembled from two separate Western blots each with the assembly point marked by a black dashed line.

Fig 3: The MLL3/4 complexes associate with the mitotic deacetylase complex (MiDAC).Western blot of FLAG-DNTTIP IP from HEK293 cells confirming interaction of DNTTIP1 with the MiDAC components ELMSAN1, HDAC1, HDAC2, and members of the MLL3/4 complexes including UTX, the H3K4 methyltransferases MLL3 and MLL4, and the two core components RBBP5 and ASH2L. KAT2A was used as a negative control. HEK293 cells with a FLAG-tag expressing plasmid were used as an IP control. Nuclear extracts were used as input. Actin was used as a loading control for the inputs.

Fig 4: The interaction interface between UTX and ELMSAN1 forms the nexus between the MLL3/4 complexes and MiDAC.(A) CRISPR strategy to restore MLL3 expression and knockout MLL4 in HCT116 cells. A single base pair deletion in the MLL3 locus was repaired by using a single-stranded oligo donor that contained a blocking modification (G>A mutation at the highlighted position) to create MLL3+/+ MLL4−/− clone 2F11. The newly added G to restore the MLL3 sequence back to WT in HCT116 cells is highlighted in red. The MLL4 promoter was targeted with gRNAs SM124.KMT2D.g4 and SM125.KMT2D.g3 resulting in the deletion of a 9,522-bp sequence within the MLL4 promoter to create MLL3−/− MLL4−/− clone 2G5. The bases at the deletion junction are marked by green and purple, respectively. In addition, a G to A substitution in allele 1 and 2 (highlighted in red) was created 33 bp from the deletion junction. (B) Western blot for members of the MLL3/4 complexes including UTX, the H3K4 methyltransferases MLL3 and MLL4 and the core components RBBP5 and ASH2L from nuclear extracts of MLL3+/+ MLL4+/+ (2F11), MLL3−/− MLL4+/+ (control), and MLL3−/− MLL4−/− (2G5) HCT116 cells. Actin was used as a loading control. (C) Western blot of FLAG-UTX IP from MLL3+/+ MLL4+/+ (2F11), MLL3−/− MLL4+/+ (control), and MLL3−/− MLL4−/− (2G5) HCT116 cells after transfection with FLAG-UTX. UTX is able to associate with MiDAC (DNTTIP1, ELMSAN1, and HDAC1) in the absence of MLL3 and MLL4 suggesting that UTX mediates the interaction between the MLL3/4 complexes and MiDAC. HCT116 cells of the same genotypes transfected with a FLAG-tag expressing plasmid were used as IP controls. Nuclear extracts were used as input. Actin was used as a loading control for the inputs.

Fig 5: MiDAC is a global regulator of H4K20ac.(A, B) The heat maps show the log2 fold change in the relative abundance of (A) various H3 and H4 acetylation sites or (B) H4 post translational modifications (PTMs) between WT and two different Dnttip1 KO mESC clones as assessed by mass spectrometry. Single, double and triple asterisks indicate adjusted P < 0.05, adjusted P < 0.01 and adjusted P < 0.001, respectively (unpaired t test, P-values were corrected for multiple testing using the Benjamini–Hochberg method). Unmod = not acetylated on H4K5, H4K8, H4K12, H4K16, or H4K20. 1ac = one acetyl group detected on one lysine of H4K5, H4K8, H4K12, and H4K16. 2ac = two acetyl groups detected on two lysines of H4K5, H4K8, H4K12, and H4K16. 3ac = three acetyl groups detected on three lysines of H4K5, H4K8, H4K12, and H4K16. 4ac = acetyl groups detected on all lysines of H4K5, H4K8, H4K12, and H4K16. (C) Mass spectrometry quantification of the relative PTM abundance of H4K20ac expressed as a percentage (%) in WT and two different Dnttip1 KO mESC clones. Data are the means ± SD of three independent experiments. Single and triple asterisks indicate adjusted P < 0.05 and adjusted P < 0.001, respectively (unpaired t test, P-values were corrected for multiple testing using the Benjamini–Hochberg method). H4K20ac levels are strongly increased in the absence of MiDAC function.