Defects in nuclear pore assembly lead to activation of an Aurora B-mediated abscission checkpoint

Abstract

Correct assembly of nuclear pore complexes (NPCs), which directly and indirectly control nuclear environment and architecture, is vital to genomic regulation. We previously found that nucleoporin 153 (Nup 153) is required for timely progression through late mitosis. In this study, we report that disruption of Nup 153 function by either small interfering RNA-mediated depletion or expression of a dominant-interfering Nup 153 fragment results in dramatic mistargeting of the pore basket components Tpr and Nup 50 in midbody-stage cells. We find a concomitant appearance of aberrantly localized active Aurora B and an Aurora B-dependent delay in abscission. Depletion of Nup 50 is also sufficient to increase the number of midbody-stage cells and, likewise, triggers distinctive mislocalization of Aurora B. Together, our results suggest that defects in nuclear pore assembly, and specifically the basket structure, at this time of the cell cycle activate an Aurora B-mediated abscission checkpoint, thereby ensuring that daughter cells are generated only when fully formed NPCs are present.

Figure 1.

Nup153 depletion leads to distinctive defects in nuclear pore basket assembly in midbody-stage cells. (A and B) HeLa cells transfected with either control or Nup153-specific siRNA oligos were analyzed by immunofluorescence using antibodies directed against the NPC basket components Tpr (A) and Nup50 (B). α-Tubulin was used to identify midbody-stage cells (brackets). (C) Quantification of the prevalence of the mislocalized Nup phenotypes in the cell population (Ø indicates a mean of 0%; n = 3). (D) Montages of time-lapse imaging illustrating the recruitment of POM121-3GFP to chromatin (H2B-mCherry) immediately after anaphase onset (t = 0:00) in siControl or siNup153-1–treated cells. (E) Quantification of POM121-3GFP recruitment to chromatin shows similar kinetics between control and Nup153-depleted cells (n = 3). (F–H) siRNA-treated cells stained for the central NPC component Nup133 (F), the NPC basket–associated protein Mad1 (G), and Lamin A/C (H). Error bars indicate mean ± SD. Bars, 10 µm.

Figure 2.

Nup153 depletion leads to an Aurora B–dependent delay in abscission. (A) Montages illustrating the prolonged timing of midbody persistence in Nup153-depleted cells expressing GFP-tubulin and H2B-mCherry, where the first frame in which a fully formed midbody was observed (t = 0:00; closed arrows) until the frame in which the midbody was no longer present (open arrows) was tracked. Bars, 10 µm. (B) Quantification of midbody persistence. Black bars indicate the median time of midbody persistence in cells from two to three independent experiments (siControl = 1.5 h, 238 cells analyzed; siNup153-1 = 3.25 h, 220 cells analyzed; siNup153-2 = 4 h, 87 cells analyzed). (C) Control or Nup153-depleted HeLa cells were briefly treated with the Aurora B inhibitor ZM447439 or DMSO and analyzed after 60 min for the presence of midbody cells (arrowheads) by detection of α-tubulin. Bar, 20 µm. (D) Quantification of midbody-stage cells and cells with more than one nucleus present after treatment with ZM447439 at the indicated time points (n = 3). Error bars indicate mean ± SD.

Figure 3.

Aurora B is mislocalized and aberrantly active in midbody-stage cells depleted of Nup153. (A and B) siRNA-treated cells were stained for Aurora B (A) and pAurora B (B). Note that Aurora B (and pAurora B) mislocalizes in cytoplasmic foci upon Nup153 depletion. Insets show normal Aurora B localization in interphase cells. (C and D) Localization of Aurora B (C) and the spindle checkpoint protein Mad1 (D) during prometaphase (left) and metaphase (right). CREST antiserum was used to indicate centromeres. (E) Detection of pAurora B and CRIK illustrates that mislocalized pAurora B foci are neither ectopic nor residual midbodies (arrowheads). Insets show a cell with what is likely a residual midbody. (F) Cells stained for pAurora B and INCENP. Brackets indicate midbody-stage cells. Bars: (A, B, E, and F) 10 µm; (C and D) 5 µm.

Figure 4.

Rescue of both delayed abscission and aberrant Aurora B localization in midbody-stage cells requires the same domains of Nup153. (A) Schematic drawing of the domain structure of Nup153, including the N terminus (N), zinc finger region (Z), and FG-rich C terminus (C), and the different constructs used to determine the domains of Nup153 required to rescue function after Nup153 depletion. (B) Rescue of midbody accumulation after depletion of endogenous Nup153 is seen with expression of full-length and N + C constructs. Brackets indicate a nonsignificant difference between control and Nup153 siRNA treatment (n = 3). (C) Significant reduction in Aurora B foci is seen after Nup153 depletion when cells express full-length or N + C constructs compared with GFP (*, P < 0.005; n = 3). Control siRNA-treated cells never displayed mislocalized Aurora B (Ø indicates a mean of 0%). Error bars indicate mean ± SD.

Figure 5.

Induced expression of Nup153-C fragment dominantly interferes with NPC basket assembly and results in mislocalized active Aurora B in midbody-stage cells. (A) HeLa-T-REx cell lines without (−DOX) and with (+DOX) induced expression of GFP or a Nup153 C-terminal fragment were analyzed for the presence of midbody-stage cells (arrowheads). (B) Immunostaining of GFP or Nup153-C–GFP-expressing cells shows that localization of both Nup153 and Nup133 is unaffected, whereas Tpr and Nup50 mislocalize to cytoplasmic foci in midbody-stage cells expressing Nup153-C–GFP. Detection of α-tubulin (not depicted) was used to identify midbody-stage cells. (C) Quantification of the incidence of midbody-stage cells and the mislocalized Nup phenotypes within the cell population (Ø indicates a mean of 0%; n = 3). (D and E) Staining for both Aurora B (D) and pAurora B (E) indicates that expression of Nup153-C–GFP leads to mislocalized and aberrantly active Aurora B in midbody-stage cells. (F–H) HeLa cells treated with Nup50-specific siRNA oligos display normal localization of both Nup153 (F) and Tpr (G) but display increased numbers of midbody-stage cells, many of which contain mislocalized pAurora B (H). (I) Quantification of the incidence of pAurora B foci in a midbody-stage cell after either siRNA treatment or fragment expression as indicated (n = 3). (J) Model for cross talk between nuclear pore assembly and the abscission checkpoint. Under normal conditions, resolution of cytokinesis ensues after proper nuclear pore assembly and Aurora B dephosphorylation (dashed red circle). pAurora B (solid red circles) is stabilized throughout the cytoplasm, including at the midbody, when assembly of the nuclear basket is disrupted, whether by a discrete change, such as the depletion of Nup50 (purple ovals), or a more potent disruption, as results from either interfering with Nup153 function (but not localization) by dominant interference (C terminus) or depleting Nup153. Activated Aurora B delays abscission, but the specific contributions of cytoplasmic versus midbody pools of Aurora B remain to be elucidated. (A, B, and D–H) Brackets indicate midbody-stage cells. Error bars indicate mean ± SD. Bars, 10 µm.