Fig 2: High-throughput RNAi screen identifies regulators of CENP-A mislocalization. (A) Bar chart showing the ratio between mean YFP signal intensities and mean number of nuclei in HeLaYFP–CENP-A-high cells transfected with the indicated siRNAs for 72 h in 384-well plates. (B) Median Z-score of lead candidate genes obtained from the primary screen of an siRNA chromatin library for increased nuclear YFP intensity. (C) Secondary analysis of lead candidate siRNAs selected from the primary RNAi screen of genes that regulate epigenetic processes. The nuclear YFP intensity data are presented relative to that of cells transfected with a scrambled siRNA (siNeg.1). The results from siHIRA.2 are shown as a positive control.

Fig 3: CHAF1B-depleted cells show an enrichment of CENP-A in chromatin and increased levels of the histone H3.3 chaperone DAXX. (A) Western blots of cellular fractionated lysates prepared from HeLaYFP–CENP-A-low cells transfected with the indicted siRNAs for 72 h and analyzed using antibodies as indicated. (B) Quantification of CHAF1B, TAP-tagged CENP-A and endogenous (‘End.’) CENP-A in the soluble (left) and chromatin (right) fractions of HeLaCENP-A–TAP cells transfected with the indicated siRNAs. The protein levels were normalized against those of a-tubulin and H2B for the soluble and chromatin fractions, respectively, and expressed as fold increase or decrease relative to those of control cells. (C,D) Gels (C) and bar charts (D) from semi-quantitative RT-PCR and RT-qPCR, respectively showing the mRNA levels of CHAF1B and CENPA in HeLaCENP-A–TAP cells transfected with the indicated siRNAs. GAPDH was used as loading control. In D, the levels were normalized against those of GAPDH and expressed as a log2 fold change. DAXX mRNA levels were also analyzed as part of these experiments (Fig. S5A,B), and the data for CHAF1B and GAPDH are shown again in Fig. S5A,B for comparison with the DAXX results. (E,F) Western blot analysis (E) and quantification (F) of lysates prepared from HeLaCENP-A–TAP cells transfected with the indicated siRNAs for 72 h and analyzed using antibodies as indicated. GAPDH was used as a loading control and used to normalize the levels of other proteins to express as a fold increase or decreased relative to those of control cells. For B,D,F, error bars represent s.e.m. from three independent experiments and P-values were calculated using unpaired, two-tailed t-test.

Fig 4: Genome-wide analysis of CUT&RUN sequencing shows the mislocalization of CENP-A upon depletion of CHAF1B in HeLaCENP-A–TAP cells. (A) CUT&RUN sequencing of TAP-tagged CENP-A showing the distribution of CENP-A across chromosome 19 (CHM13v2) in asynchronous HeLaCENP-A–TAP cells transfected with the indicated siRNAs for 72 h. The analysis was performed using two different antibodies against CENP-A. An ideogram of chromosome 19 is shown above the CUT&RUN sequencing tracks and the red box indicates the endogenous centromere. (B) Peaks of CENP-A identified by CUT&RUN sequencing in CHAF1B-depleted cells are shown relative to those of siNeg-transfected control cells. The pie chart shows the distribution of CENP-A peaks across various genomic regions. TSS, transcription start site; UTR, untranslated region. (C) Representative distributions of CENP-A at the gene level are shown for two genes: PSEN1 (Chr14) and CABIN1 (Chr22). (D) Overlap of CENP-A CUT&RUN profile from HeLaCENP-A–TAP cells transfected with the indicated siRNAs plotted against previously identified ATAC-seq peaks (GSM2830382; Cho et al., 2018) and DNase I-hypersensitive sites (GSM763533; Mercer et al., 2013).

Fig 5: Depletion of HIRA, a positive control for the imaging-based screen, shows the mislocalization of CENP-A and CIN phenotypes in HeLaYFP–CENP-A-low cells. (A) Western blots of lysates prepared from HeLaYFP–CENP-A-low cells transfected with the indicated siRNAs for 72 h and analyzed using the antibodies as indicated. (B) Representative images of metaphase chromosome spreads showing the localization of CENP-A at centromeric and non-centromeric regions in HeLaYFP–CENP-A-low cells transfected with the indicated siRNAs. Metaphase chromosome spreads were prepared 72 h post transfection, and cells were immunostained using an antibody against CENP-A and stained with DAPI. Scale bars: 5 µm. (C) Quantification of CENP-A signal intensities at centromeric (left) and non-centromeric (right) regions in metaphase chromosome spreads of HeLaYFP–CENP-A-low cells transfected with the indicated siRNAs. Each circle represents a spot on a centromeric or non-centromeric region. n denotes the number of chromosomes analyzed. Red lines indicated mean±s.e.m. for YFP signal intensities across areas measured in the number of spots indicated as n from three independent experiments. A.U., arbitrary units. (D,E) Immunostained images (D) and bar chart (E) showing the proportion of cells with defective chromosome segregation in HeLaYFP–CENP-A-low cells transfected with the indicated siRNAs. The yellow arrow shows lagging chromosomes. Error bars represent s.e.m. across three independent experiments. n denotes the number of cells analyzed from three experiments. P-values shown in C, E were calculated using unpaired, two-tailed t-test.