|
|
|
Figure 1. Y RNA-Independent Initiation Activity in Xenopus Egg Extracts(A) Y RNA-dependent initiation of DNA replication in a human cell-free system. Human late G1 phase template nuclei were incubated in either buffer (top) or mock-treated or Y RNA-depleted cytosolic extracts.(B) Y RNA-independent initiation of DNA replication in Xenopus egg extracts. Human late G1 phase template nuclei were incubated in either buffer (top) or mock-treated or Y RNA-depleted Xenopus egg extracts. Representative immunofluorescence micrographs are shown with merged channels for total DNA (propidium iodide, red) or sites of DNA replication (digoxigenin-deoxyuridine triphosphate [dUTP] incorporation, green).(C) Experimental design for the detection and isolation of the initiation activity from Xenopus egg extracts by cross-species complementation assays.(D) Cross-species complementation assays. Left: template nuclei were incubated as in (A) in either buffer or mock-treated cytosolic or Y RNA-depleted human cytosolic extracts. Right: increasing amounts of Y RNA-depleted Xenopus egg extracts were added to replication initiation reactions in the presence of either buffer (light green) or Y RNA-depleted human cytosolic extract (dark green). Percentages of replicating template nuclei for each reaction are shown as mean values ± SD of n independent experiments.Scale bars, 10 μm.
|
|
Figure 2. Isolation and Identification of the Y RNA-Independent Initiation Factor from Xenopus Egg Extracts(A) Schematic representation of the fractionation scheme.(B) Activity profile of the sucrose minigradient fractions. Template nuclei were incubated in Y RNA-depleted human cytosolic extract supplemented with the indicated gradient fractions. Percentages of replicating nuclei are shown as mean values ± SD of n = 4 independent fractionation experiments. Brackets indicate results of t tests (unpaired, two-tailed with unequal variance) of experimental samples (fractions 6 and 7) against the control (∗∗∗p ≤ 0.001). Tests of all other fractions against the control were not significant. Sedimentation of calibrator protein complexes (cytochrome c, 12.5 kDa; BSA, 68 kDa; aldolase, 158 kDa; catalase, 240 kDa) are indicated at the top.(C) Proteins in active fractions as identified by mass spectrometry. Proteins identified in both of two independent experimental replicates that were enriched in the active gradient fractions 6/7 compared with inactive fraction 5 are ranked according to their abundance (determined as emPAI scores).(D) NuRD subunit sedimentation profiles. Fractions of the sucrose minigradient were analyzed by western blotting with antibodies specific for the indicated NuRD subunits.See also Figure S1 and Table S1.
|
|
Figure 3. The Xenopus NuRD Complex Has DNA Replication Initiation ActivityNuRD was immunoprecipitated with MTA2-specific antibodies from the 20%–45% ammonium sulfate fraction of the activated Xenopus laevis egg extract.(A) Co-immunoprecipitation of GATAD2/p66 and HDAC1 subunits with MTA2, as confirmed by western blotting. Input (In), supernatant (Su), and washed pellet (P) of the control and MTA2 coIPs are shown, 10% of the total was loaded per lane for each.(B) Mass spectrometry analysis of co-immunoprecipitated material. Xenopus proteins (n = 447) uniquely identified only in the immunoprecipitate and not in the control IP were ranked according to their quantitative emPAI values after normalization to emPAI (MTA2) = 1. The top 14 most abundant proteins are plotted together with additional NuRD subunits. NuRD subunits are shown in blue. The mean normalized emPAI value of all proteins is indicated by a dashed line. Separate and additional isoforms were identified in the Uniprot database for GATAD2/p66 (∗LOC398154; ∗∗LOC100158394, isoform X2) and MTA2 (∗∗∗MGC83056), respectively.(C) Immunodepletion. xNuRD was co-immunoprecipitated with MTA2-specific antibodies as indicated above. Control immunoprecipitations were performed with either empty beads (mock) or with antibodies specific for Drosophila Vasa protein. The indicated protein amounts of immunodepleted supernatants were added to Y RNA-depleted DNA replication initiation reactions. Mean values ± SD of percentages of replicating template nuclei are plotted from n = 3 independent experiments. Brackets indicate results of t tests (unpaired, two-tailed with unequal variance) of experimental and control samples against the untreated input samples (ns, not significant; ∗∗∗p ≤ 0.001).See also Table S2.
|
|
Figure 4. Human NuRD Complexes Are Structurally and Functionally Distinct from xNuRD(A–F) Proliferating human HeLa cells were fractionated into nuclear (A–C) and cytosolic extracts (D–F), and each extract was partially sub-fractionated by precipitation with 20%–45% ammonium sulfate and ultracentrifugation through preparative sucrose gradients.(A and D) Western blot analyses of human NuRD subunits. Fractions of the preparative sucrose gradients of nuclear (A) and cytosolic extracts (D) were analyzed by western blotting with antibodies specific for the indicated human NuRD subunits.(B and E) Mass spectrometry analysis of the MTA2 co-immunoprecipitations. Fractions 6–8 of the preparative sucrose gradients of nuclear (B) and cytosolic extracts (E) were pooled, and human NuRD was immunoprecipitated with anti-MTA2 antibodies. Human proteins identified in the immunoprecipitates (n = 126 and n = 183 for nuclear and cytosolic extracts, respectively) were ranked according to their quantitative emPAI values after normalization to emPAI (MTA2) = 1, and all human NuRD subunits detected are plotted with their rank. The mean normalized emPAI values of all proteins are indicated by dashed lines.(C and F) Activity profiles of the sucrose fractions. Template nuclei were incubated in Y RNA-depleted human cytosolic extract supplemented with the indicated gradient fractions of the nuclear (C) and cytosolic extracts (F). Mean values ± SD of percentages of replicating template nuclei are plotted of n independent experiments. Brackets indicate results of t tests (unpaired, two-tailed with unequal variance) of the positive control (xNuRD) and experimental samples against the Y RNA-depleted background (∗∗∗p ≤ 0.001).See also Figures S2 and S3 and Table S3.
|
|
Figure 5. Inhibition of xNuRD Leads to an Inhibition of Y RNA-Independent DNA ReplicationShown is inhibition of DNA replication using HDAC inhibitors and NuRD-specific antibodies.(A) Template nuclei were incubated in Y RNA-depleted human cytosolic extract supplemented with partially purified xNuRD and the indicated concentrations of the HDAC inhibitors trichostatin A (TSA), SAHA, and MS-275.(B) Template nuclei were incubated with human cytosolic extract containing human NuRD and Y RNAs in the presence of the indicated inhibitors.(C) Template nuclei were incubated in Y RNA-depleted human cytosolic extract supplemented with partially purified xNuRD and the indicated NuRD-specific antibodies.(D) Template nuclei were incubated with human cytosolic extract containing human NuRD and Y RNAs in the presence of the indicated NuRD-specific antibodies. Mean values ± SD of the percentages of replicating template nuclei are plotted from n independent experiments. Results of t tests (unpaired, two-tailed with unequal variance) of treated experimental samples against the no addition controls are indicated (∗∗p ≤ 0.01; ∗∗∗p ≤ 0.001).
|
|
Figure 6. The Replication Initiation Activity and Complex Formation of xNuRD Declines after MBT(A) Expression of xNuRD during development of Xenopus laevis. Extracts of staged embryos were analyzed by western blotting. Pre-MBT and post-MBT stages are indicated. Dephosphorylation of hyperphosphorylated nucleoplasmin (P-NPL) is a control for MBT (stage 8.5).(B) DNA replication initiation activity of xNuRD declines after MBT. Template nuclei were incubated in Y RNA-depleted human cytosolic extract supplemented with partially purified xNuRD from activated eggs, stage 6 pre-MBT embryos, and stage 18 post-MBT embryos, as indicated. Mean values ± SD of the percentages of replicating template nuclei are plotted from n independent experiments.(C) Mass spectrometry analyses of xNuRD complexes. MTA2 was immunoprecipitated from pre-MBT (stage 6) and post-MBT (stage 18) embryo extracts. Xenopus proteins identified in the immunoprecipitates were ranked according to their quantitative emPAI values in the pre-MBT IP after normalization to emPAI (MTA2) = 1. All xNuRD-specific subunits detected are plotted (left, dark blue bars for pre-MBT; right, light blue bars for post-MBT). The mean normalized emPAI values for all detected proteins are indicated. Separate isoforms were identified in the Uniprot database for GATAD2/p66 (∗LOC398154; ∗∗LOC100158394, isoform X2).(D) Relative enrichment of xNuRD subunits before MBT in immunoprecipitated complexes. Normalized emPAI values were taken from MTA2 co-immunoprecipitations of stage 6 pre-MBT and stage 18 post-MBT extracts, and their ratios are plotted for the indicated xNuRD subunits and the overall mean.(E and F) xNuRD complex formation in (E) pre-MBT and (F) post-MBT embryo extracts. xNuRD was partially sub-fractionated by precipitation with 20%–45% ammonium sulfate and ultracentrifugation through preparative sucrose gradients. MTA2, HDAC, and RBBP7 subunits were analyzed by western blot analysis of the preparative sucrose gradient fractions. A blue box shows the reference for the sedimentation of active xNuRD from egg extracts. Positions of sedimentation markers are indicated.See also Figure S4 and Table S4.
|
|
Figure 7. xNuRD Is Required for Development and DNA Replication before the MBT In Vivo(A and B) Phenotypes of developing Xenopus embryos after microinjection of NuRD-specific antibodies and Y3 RNA-specific antisense morpholino oligonucleotides (MOs). The indicated antibodies (panel A, 65ng serum/embryo; panel B, 5ng IgG/embryo) and MOs (control coMO or Y3 MO, 40 ng/embryo) were injected into the animal pole at the 1-cell stage, and representative embryos were photographed at the indicated times post fertilization (p.f.). See Movies S1 and S2 for the full time course of these experiments.(C and D) DNA replication before the MBT depends on xNuRD. Total embryonic DNA of experimental repeats shown in (A) and (B) was quantified relative to rRNA (C) and (D), respectively. Individual datasets were normalized to the mean of the control distributions and plotted as box and whisker plots superimposed with individual data points. Antibodies available to other xNuRD subunits were ineffective in this assay, regardless of whether Y RNA was co-depleted (Figure S6). Results of t tests (unpaired, two-tailed with unequal variance) of treated experimental samples against the controls are indicated (∗p = 0.07; ∗∗∗p ≤ 0.001).See also Figures S5 and S6 and Movies S1 and S2.
|
|
Figure S1. Polypeptide composition of the Y RNA-independent initiation factor from Xenopus
egg extracts. Related to Figure 2.
Protein composition analysis of final sucrose minigradient fractions by SDS-PAGE and silver staining.
Fraction numbers (1-10, collected from top to bottom of the gradient), pellet (P) and molecular weight
markers (M) are indicated. Sedimentation positions of calibrator protein complexes (cytochrome C,
12.5kDa; bovine serum albumin, 68kDa; aldolase, 158kDa; catalase, 240kDa) are indicated at the
bottom.
|
|
Figure S4. Complex formation of xNuRD in Xenopus egg extracts. Related to Figure 6.
xNuRD was partially purified by precipitation of egg extract with 20-45% ammonium sulphate and
ultracentrifugation through preparative sucrose gradients. MTA2, HDACm, HDAC1 and RBBP7/p46
subunits were analysed by Western blot analysis of the preparative sucrose gradient fractions. Fraction
numbers (1-10, collected from top to bottom of the gradient), the pellet fraction (P), and positions of
sedimentation markers are indicated.
|
|
Figure S5. xNuRD is required for early development before the MBT and for embryo viability.
Related to Figure 7.
Phenotypes of developing Xenopus laevis embryos after microinjection of NuRD-specific antibodies
and Y3 RNA-specific antisense morpholino oligonucleotides (MOs). The indicated antibodies
(5ng/embryo) and MOs (40 ng/embryo of coMO or xY3MO; Collart et al., 2011) were injected in the
animal pole at the 1 cell stage, and representative embryos were photographed at 20 h post fertilisation
(p.f.), when the control embryos had reached the neurula stage. Three experimental replicates of the
injections are presented from top to bottom
|
|
Figure S6. Embryonic DNA synthesis after microinjection with ineffective antibodies. Related to
Figure 7.
Antibodies directed against control GFP and the indicated subunits of xNuRD (5ng/embryo) were
(co)injected with the indicated MOs (40 ng/embryo of coMO or xY3MO; Collart et al., 2011) into the
animal pole at the 1 cell stage. Total embryonic DNA of the experimental repeats was quantified
relative to rRNA. Individual data sets were normalised to the mean of the control anti-GFP
distributions, and plotted as box and whisker plots superimposed with individual data points. Results of
T-tests (unpaired, two-tailed with unequal variance) of treated experimental samples against the
controls are indicated (ns, not significant).
|