Fig 1: Nuclear localisation and HDAC activity of the endogenous MiDAC complex.a Cytoplasmic and nuclear proteins were prepared from U2OS cells before western blotting using antibodies against MIDEAS, DNTTIP1 and HDAC1. Antibodies against GAPDH and LaminB were used, respectively, as cytoplasmic and nuclear housekeeping controls. b Fluorescent-based HDAC activity assay of IgG control, MIDEAS and DNTTIP1 immunoprecipitates from U2OS nuclear lysates (mean ± s.e.m, n = 8 independent experiments, ***P = 0.001, **P = 0.0012, one-way ANOVA with Holm–Sidak post hoc test). c Fluorescent- based HDAC activity assay of MIDEAS and DNTTIP1 immunoprecipitates from nuclear lysates that had been pre-incubated with 100 μM InsP6 or 5 μM SAHA for 30 min. IgG controls were performed in parallel and used to subtract background activity before normalisation to untreated complex HDAC activity (mean ± s.e.m, n = 4 independent experiments, ****P < 0.0001, one-way ANOVA with Holm–Sidak post hoc test). d Confocal images of PFA fixed U2OS cells dual stained with either HDAC1 (grey), MIDEAS (green) or DNTTIP1 (yellow) antibodies and anti-α-tubulin (magenta). DNA (blue) was visualised with Hoechst 33258 (scale bars: interphase, 10 μm; mitotic cells, 5 μm). e Western blot for DNTTIP1, MIDEAS-FLAG and Tubulin using lysates from MIDEAS-FLAG CRISPR mouse ES cells that had been incubated for 14 h with 10 µM CDKi RO-3306 (Ci), followed by a further 2-h incubation with nocodazole (30 ng/ml). Cells were harvested at various time points after nocodazole release. f Western blot for MIDEAS-FLAG and DNTTIP1 from cell-cycle synchronised cells following co-immunoprecipitation (Co-IP) using a anti-FLAG antibody. Cells synchronised as in panel e. g Western blot for MIDEAS, DNTTIP1, HDAC2 and HDAC1 from co-IP’s using Rb anti-MIDEAS and Rb anti-DNTTIP1 antibodies. U2OS cells were blocked for 16 h in the G1/S with aphidicolin (1.6 µg/ml) or M with nocodazole (3.3 µg/ml). A mock IP was also carried out with rabbit IgG as a control. Antibodies were cross-linked to beads prior to IP. h Fluorescent-based HDAC activity assay of MIDEAS and DNTTIP1 immunoprecipitates from U2OS cells blocked in the G1/S or M as in panel g. Results are normalised to cell count after detachment from plates (mean ± s.e.m, n = 3 independent experiments, *P = 0.022, one-way ANOVA with Holm–Sidak post hoc test). Source data are provided as a Source data file.
Fig 2: The structure of the dimeric MiDAC complex.a Post-processed cryo-EM map of the dimeric complex from Relion3 with protein chains shown as cartoons. Map contour level is 0.024—Chimera. b Closeup of the dimerisation domain of DNTTIP1 and a closeup of the dimerisation domain of DNTTIP1 interacting with MIDEAS. Map contour level is 0.024—Chimera. c Cartoon representation of the dimeric MiDAC complex. d Comparison of the MiDAC complex with the NuRD complex. The active sites are shown with magenta arrows. The dotted magenta arrow indicates that the active site is at the back. e Structural alignment of MIDEAS with MTA1. The alpha-helices are shown as cylinders, and the beta-strand as an arrow. The ELM2 domain is shown in red, and the SANT domain in blue. Identical residues are shaded grey, and residues at the tetramer interface are indicated by magenta spots.
Fig 3: MiDAC protein co-dependency and role in chromosome alignment.a Western blot for MIDEAS (top) and DNTTIP1 (bottom) following treatment of U2OS cells with siRNA for 72 h targeting MIDEAS or DNTTIP1. Blots shown are representative. LaminB was detected in parallel from the same lysate sample, and was used as a reference for normalisation (mean ± s.e.m., n = 3 independent experiments, **P = 0.0029, ***P < 0.001 (P = 0.0003 for control versus DNT_si1, 0.0004 for control versus DNT_si2), ****P < 0.0001, one-way ANOVA with Dunnett’s post hoc test). b Confocal images representative of normal and misaligned chromosomes. U2OS cells were treated with siRNA as in (a) before fixation with PFA and staining for α-tubulin (green), CENPA (magenta). DNA (blue) was visualised with Hoechst 33258 (scale bars: 5 μm). c Quantification of metaphase cells with misaligned chromosomes in U2OS cells after siRNA treatment shown as percent misaligned (mean ± s.e.m, n = 3 MID_si2, DNT_si1, DNT_si2, n = 4 MID_si1, (50 metaphase cells counted over 3/4 independent experiments), ***P < 0.001 (P = 0.0008 for control versus MID_si1, 0.0003 for control versus MID_si2), ****P < 0.0001, one-way ANOVA with Holm–Sidak post hoc test). d Quantification of metaphase cells with misaligned chromosomes in HeLa cells after MIDEAS siRNA treatment shown as percent misaligned (mean ± s.e.m., n = 3 (50 metaphase cells counted over 3 independent experiments). e Western blot using nuclear proteins isolated from stable, DOX inducible, siRNA-resistant FLAG-DNTTIP1 U2OS cell line showing DOX induces FLAG-DNTTIP1. f Quantification of metaphase cells with misaligned chromosomes in U2OS cells after siRNA treatment and induction of siRNA-resistant FLAG-DNTTIP1 with DOX (mean ± s.e.m., n = 3 (50 metaphase cells counted over three individual experiments), *P = 0.0416, **P = 0.0041, one-way ANOVA with Holm–Sidak post hoc test). Source data are provided as a Source data file.
Fig 4: Cryo-electron microscopy of the MiDAC complex.a Schematic of the domain structures of MIDEAS, HDAC1 and DNTTIP1: components of the MiDAC complex. b SDS-PAGE of the gel-filtration purification of the MiDAC complex on a Superdex-S200 column. c Section of an electron micrograph of the MiDAC complex. Scale bar: 20 nm. d 2D class averages of the dimer complex from Relion3. e 2D class averages of the tetramer complex from Relion3.
Fig 5: Analysis of mice embryos and MEF’s lacking MIDEAS or DNTTIP1.a Images of wild-type, heterozygous and homozygous MIDEAS-del1 and DNTTIP1-del1 embryos isolated at e16.5 (scale: 5 mm). b Images of sections from e16.5 wild-type, MIDEAS−/− and DNTTIP1−/− embryos demonstrating absence of erythrocytes in the heart, enlarged pericardium and deformed ventricle morphology in the knockouts compared with wild-type (green arrows) (scale: 500 µm) (representative images from n = 2 biologically independent animals). c Quantification of metaphase cells with misaligned chromosomes in mouse embryonic fibroblasts (MEFs) from e13.5 MIDEAS-del1 (MID-del1) or DNTTIP1-del1 (DNT-del1) wild-type and homozygous embryos (mean ± s.e.m., n = 3 MID-del1, n = 2 DNT-del1 (40 metaphase cells were counted over 3/2 individual experiments), **P = 0.0048, two-tailed unpaired Student’s t test). d Venn diagram depicting the number of overlapping genes identified as differentially expressed in MIDEAS and DNTTIP1 knockout MEFs. Differential expression was based on a P-value of <0.1 as calculated using DESEQ2 in R. e The number of down- and upregulated genes identified as differentially expressed in MIDEAS and DNTTIP1 knockout MEFs and in the overlapping gene set list from panel d. Differential expression calculated from n = 3 (MIDEAS knockout MEFs) and n = 2 (DNTTIP1 knockout MEFs) independent MEF lines from e13.5 embryos, P-value of <0.1 as calculated using DESEQ2 in R. f Plot of the 468 overlapping perturbed gene set showing fold change for homozygous knockout MIDEAS (x-axis) and DNTTIP1 (y-axis) genes. (R2 calculated using Pearson correlation). g DAVID gene ontology (GO) biological processes analysis using the overlapping, upregulated gene list. List shows changes in biological process with more than 12 genes upregulated from the overlapping gene list and a Benjamini post hoc test with P < 0.05. h DAVID GO tissue association of the overlapping, upregulated gene list. Source data are provided as a Source data file.
Supplier Page from Abcam for Anti-DNTTIP1/TDIF1 antibody