Epigenetic centromere specification directs aurora B accumulation but is insufficient to efficiently correct mitotic errors

Abstract

The nearly ubiquitous presence of repetitive centromere DNA sequences across eukaryotic species is in paradoxical contrast to their apparent functional dispensability. Centromeric chromatin is spatially delineated into the kinetochore-forming array of centromere protein A (CENP-A)-containing nucleosomes and the inner centromeric heterochromatin that lacks CENP-A but recruits the aurora B kinase that is necessary for correcting erroneous attachments to the mitotic spindle. We found that the self-perpetuating network of CENPs at the foundation of the kinetochore is intact at a human neocentromere lacking repetitive alpha-satellite DNA. However, aurora B is inappropriately silenced as a consequence of the altered geometry of the neocentromere, thereby compromising the error correction mechanism. This suggests a model wherein the neocentromere represents a primordial inheritance locus that requires subsequent generation of a robust inner centromere compartment to enhance fidelity of chromosome transmission.

Figure 1.

Epigenetic determinants of centromere identity are intact at the neocentromere of the PD-NC4 chromosome variant. (A) ACA stains both the epigenetically silenced centromere and the neocentromere. Insets show higher magnification views of the boxed area. (B) Immunoblot of PD-NC4 lysates with ACA reveals two bands with electrophoretic mobility consistent with ∼80 kD CENP-B (Earnshaw and Rothfield, 1985) and ∼17 kD CENP-A (Earnshaw and Rothfield, 1985). (C) CENP-A and CENP-B clearly differentiate the active versus inactive ACA staining centromeres on the PD-NC4 chromosome. (D) The CATD is sufficient for targeting H3^CATD to the neocentromere. (E) CENP-A^NAC components, each stably expressed as LAP fusions, track with CENP-A to the neocentromere. Outer kinetochore components Mad1 (Fig. S3 A) and Ndc80 (Fig. S3 B) were recruited to the neocentromere. Arrowheads indicate the neocentromere, and asterisks indicate the silenced centromere. Bars, 2 µm.

Figure 2.

Altered aurora B localization on a neodicentric chromosome. (A) In mitotic spread preparations, aurora B is present on normal centromeres at the inner centromere location, whereas on the PD-NC4 chromosome, the kinase has vacated the silenced centromere and now tracks to an ∼1-µm region of chromosome 4 adjacent to the neocentromere that runs along the length of each metaphase chromatid arm. Image capturing to robustly detect aurora B on the neodicentric chromosome on metaphase spread preparations overexposes aurora B staining at normal centromeres. (B and C) In intact, monastrol-arrested cells with monopolar spindles, the total pool sizes of neocentromeric aurora B (B) and INCENP (C) are within the range observed at normal centromeres in the same cells (aurora B: normal chromosome range, 5,623–11,249 arbitrary fluorescence units; PD-NC4, 7,019; INCENP: normal chromosome range, 2,871–7,173; PD-NC4, 3,650). In monastrol-treated cells, many of the chromosomes are rotated and have small interkinetochore distances, so for consistency, the enlargements (B and C) are examples where the sister centromeres overlap in x and y planes. Arrowheads indicate the neocentromere, and asterisks indicate the silenced centromere. Insets show higher magnification views of the boxed areas. Bars, 2 µm.

Figure 3.

Aurora B–dependent error correction is crippled on the neodicentric chromosome. (A) Experimental scheme to detect the rate of failure of the PD-NC4 to biorient on the metaphase plate as cells recover from monastrol treatment. (B) The PD-NC4 chromosome has an elevated rate of alignment failure after spindle recovery after initial monastrol treatment. The measured values from three independently performed experiments are plotted showing SEM. (C) An example of the PD-NC4 chromosome failing to align when the normal chromosomes have corrected syntelic attachments after monastrol withdrawal and are nearly aligned on the metaphase plate. Insets show higher magnification views of the boxed areas. Arrowheads indicate the neocentromere, and asterisks indicate the silenced centromere. Bar, 2 µm.

Figure 4.

Measuring the mispositioning of aurora B at a location adjacent to the neocentromere of the PD-NC4 chromosome. (A–D) Aurora B location on misoriented normal chromosomes (A and B) or misoriented PD-NC4 chromosomes (C and D) was determined by indirect immunofluorescence in intact cells that were fixed after monastrol treatment and recovery as in Fig. 3. The indicated distance measurements (mean ± SD) for misoriented normal chromosomes (B) and misoriented PD-NC4 chromosomes (D) reveal mispositioning in the PD-NC4 chromosome of aurora B away from its kinetochore targets. The arrowhead indicates the neocentromere, and the asterisk indicates the silenced centromere. Bar, 2 µm.

Figure 5.

Mispositioned aurora B renders the neocentromere insensitive to monastrol-induced attachment errors. (A) In monopolar cells after monastrol treatment, aurora B activity at the PD-NC4 kinetochore, measured by phospho–CENP-A detection, is inappropriately silenced. Insets show higher magnification views of the boxed areas. Bar, 2 µm. (B) The phospho–CENP-A intensity values in cells treated as in A are plotted as SEM. (C and D) When measured simultaneously (C), the longer kinetochore–aurora B distances and reduced phospho–CENP-A intensity on the PD-NC4 chromosome segregate away from the shorter kinetochore–aurora B distances and higher phospho–CENP-A intensity of normal chromosomes (D). The circles indicate the area of measurement of phospho–CENP-A intensities at each centromere. Arrowheads indicate the neocentromere, and asterisks indicate the silenced centromere.