Fig 2: Metastatic cell growth and lung colonization do not require KMT2 complex components.(A) Schematic of subunit composition of several KMT2 complexes. (B) Western blot analyses of the indicated proteins in LM2 cells transduced with inducible shRNA targeting KMT2A, RBBP5, DPY30, HCFC1, CXXC1, and WDR82. Cells were collected after 3 days of doxycycline (DOX) (1 µg/mL) treatment. (C) RT-quantitative PCR (qPCR) analysis of BOD1L1 in LM2 cells transduced with two independent hairpins targeting BOD1L1. Cells were collected after 3 days of DOX treatment (n=4, unpaired two-side Student’s t test). (D–E) Colony formation assay of LM2 shCtrl or shKMT2A cells (shKMT2A-1 and shKMT2A-2) after 9 days of treatment with control or DOX. Representative images (D) and quantification (E) are shown (n=3, unpaired two-side Student’s t test). (F) Normalized bioluminescence unit (BLU) signals of lung metastasis at day 66 of mice injected intravenously with LM2 cells from (B) and fed with DOX chow. The data represent mean ± SEM. Significance determined using unpaired two-tailed Mann–Whitney test. ns, not significant; p=0.2857 (shCtrl versus shKMT2A-1) and p=0.5556 (shCtrl versus shKMT2A-2). (G) RT-qPCR analysis of the noted RP genes in LM2 cells transduced with the indicated inducible shRNAs. Cells were collected after 3 days of DOX treatment. Significance determined by comparing each treatment to shCtrl. (H–I) Colony formation assay of LM2 cells from (G) after 9 days of either control or DOX treatment. Representative images (H) and quantification (I) are shown (n=3, unpaired two-side Student’s t test). ***p=0.0005 (shHCFC1 #1 -DOX versus shHCFC1 #1 +DOX) and **p=0.0039 (shHCFC1 #2 -DOX versus shHCFC1 #2 +DOX). For gel source data, see Figure 5—source data 1. Figure 5—source data 1.Original western blots for Figure 5B.

Fig 3: KSHV resolves the PROX1 bivalent chromatin by recruiting MLL2 and CFP1 and removing PRC2 to the PROX1 Promoter.Mock- and KSHV-infected PDLSCs were subjected to ChIP assay with antibodies against different histone methyltransferase components, including MLL1, MLL2, CFP1, UTX, RBBP5, WDR5, and EZH2 (a component of PRC2). Error bars represent SD. Statistics analysis was performed using the one-way ANOVA test. P-values < 0.05 were considered significant (*P<0.05, **P<0.01 and ***P<0.001) and N.S represents no significance.

Fig 4: MiDAC stability does not depend on the MLL3/4 complexes.Western blot for various subunits of the MLL3/4 complexes and MiDAC from nuclear extracts of WT, Mll3/4 DCD, and Mll3/4 DKO mESCs. As a result of MLL3/4 DKO, UTX protein levels are strongly decreased, whereas protein levels of RBBP5 and MiDAC subunits (DNTTIP1, ELMSAN1, HDAC1, and HDAC2) are unaffected in the absence of MLL3/4. Actin was used as loading control.

Fig 5: The MLL3/4 complexes associate with the mitotic deacetylase complex (MiDAC).(A) FLAG-UTX immunoprecipitation (IP) followed by mass spectrometry (MS) identifies all known subunits of the MLL3/4 complexes along with ELMSAN1, and DNTTIP1. SC, spectral counts; TP, peptide counts; abundance = SC x 50 (kD)/protein size (kD). (B) Western blot (WB) of FLAG-UTX IP from HEK293 cells confirming interaction of UTX with ELMSAN1, DNTTIP1, HDAC1, and HDAC2. UTX interacts with the H3K4 methyltransferases MLL3 and MLL4, RBBP5 a core component of the MLL3/4 complexes, ELMSAN1, DNTTIP1, HDAC1, and HDAC2, but does not interact with KAT2A a histone acetyltransferase. 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. (C) WB of FLAG-ELMSAN1 IP from HEK293 cells confirming interaction of ELMSAN1 with the MiDAC components DNTTIP1, 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. (D) Glycerol gradient sedimentation after FLAG-UTX IP reveals co-fractionation of the MiDAC subunits ELMSAN1, DNTTIP1, HDAC1, and HDAC2 with components of the MLL3/4 complexes (red box). Each antibody panel was assembled from three separate WBs with the assembly point marked by a black dashed line. (E) WB of FLAG IP from ELMSAN1 TRERF1 double knockout (ELMSAN1 TRERF1 DKO) HEK293 cells expressing FLAG-DNTTIP1, FLAG-ELMSAN1, or FLAG-TRERF1. Among the MiDAC subunits, only ELMSAN1 interacts with UTX. Reexpression of FLAG-ELMSAN1 or FLAG-TRERF1 in ELMSAN1 TRERF1 DKO cells results in restoration of DNTTIP1 levels. * marks a nonspecific band. (F) Diagram showing that ELMSAN1 and UTX form the nexus between MiDAC and the MLL3/4 complexes.