Loss of CHFR in human mammary epithelial cells causes genomic instability by disrupting the mitotic spindle assembly checkpoint

Abstract

CHFR is an E3 ubiquitin ligase and an early mitotic checkpoint protein implicated in many cancers and in the maintenance of genomic stability. To analyze the role of CHFR in genomic stability, by siRNA, we decreased its expression in genomically stable MCF10A cells. Lowered CHFR expression quickly led to increased aneuploidy due to many mitotic defects. First, we confirmed that CHFR interacts with the mitotic kinase Aurora A to regulate its expression. Furthermore, we found that decreased CHFR led to disorganized multipolar mitotic spindles. This was supported by the finding that CHFR interacts with alpha-tubulin and can regulate its ubiquitination in response to nocodazole and the amount of acetylated alpha-tubulin, a component of the mitotic spindle. Finally, we found a novel CHFR interacting protein, the spindle checkpoint protein MAD2. Decreased CHFR expression resulted in the mislocalization of both MAD2 and BUBR1 during mitosis and impaired MAD2/CDC20 complex formation. Further evidence of a compromised spindle checkpoint was the presence of misaligned metaphase chromosomes, lagging anaphase chromosomes, and defective cytokinesis in CHFR knockdown cells. Importantly, our results suggest a novel role for CHFR regulating chromosome segregation where decreased expression, as seen in cancer cells, contributes to genomic instability by impairing the spindle assembly checkpoint.

Figure 1

Decreased CHFR expression causes increased aneuploidy. (A) SKY analysis of parental MCF10A cells shows the characteristic karyotype of this genomically stable hyperdiploid cell line. (B and C) SKY analysis of MCF10A cells stably expressing shRNA against CHFR are either minimally aneuploid (B) or nearly tetraploid (C) and show novel chromosome translocations. (D) Western blot analysis shows >80% decrease in CHFR expression in MCF10A cells transiently transfected with siRNA against CHFR (“CHFR siRNA”) compared to untransfected (“mock”) and nontargeting siRNA (“siControl”) transfected cells after 72 hours. (E) Metaphase spreads show that MCF10A: CHFR-siRNA cells (right panel) are more aneuploid. Scale bar, 25 µm. (F) Frequency of increased aneuploidy in transiently transfected MCF10A cells.

Figure 2

CHFR interacts with Aurora A and regulates its protein expression. (A) MCF10A:CHFR-siRNA cells overexpress Aurora A, as shown by Western blot analysis, compared to control cells. (B) Flag:Aurora A interacts with endogenous CHFR by coimmunoprecipitation. Lysates from MCF10A cells transiently transfected with Flag-tagged Aurora A were subjected to IP with an anti-Flag (M2) antibody then probed for CHFR or Flag by Western blot analysis using rabbit antibodies. “Input” on the left indicates 10% of the lysates used for the IP reaction. (C) Immunofluorescence for Aurora A (green) indicates that MCF10A:CHFR-siRNA cells (bottom row) have greater than two Aurora A foci when compared to two foci in negative control cells during metaphase. Cells were stained with DAPI (blue) for DNA and for α-tubulin (red) to see the spindle. Note the compacted, disorganized mitotic spindle (red) in CHFR-siRNA cells (subpanel l compared to subpanels d and h). (D) Quantification of the data in (C), showing that nearly 16% of MCF10A cells transfected with CHFR siRNA had greater than two Aurora A foci whereas less than 1.5% of control cells had greater than two foci.

Figure 3

CHFR ubiquitinates α-tubulin and regulates α-tubulin protein expression. (A) A GST pull down using a GST:CHFR fusion protein shows that CHFR can interact with α-tubulin from MCF10A whole-cell lysates as shown by Western blot analysis. The “input” is 10% of the MCF10A whole-cell lysates used for the GST pull down. MCF10A cells were either untreated (.Noc) or treated with nocodazole (+Noc) before lysate collection. (B) CHFR interacts with α-tubulin by coimmunoprecipitation. A Flag:CHFR construct was transfected into HEK293 cells and the lysates were used for IP with either anti-Flag or anti-α-tubulin mouse antibodies then Western blotted (WB) with either anti-Flag or anti-α-tubulin rabbit antibodies. Cells were either untreated or treated with nocodazole. The “input” indicates 5% of the lysates used for the IP reaction. (C) CHFR ubiquitinates α-tubulin in nocodazole-treated cells. MCF10A cells were cultured in MG132 and either untreated or simultaneously treated with nocodazole. Western blot analysis of immunoprecipitated α-tubulin for ubiquitin shows that the amount of ubiquitinated α-tubulin is dramatically decreased in MCF10A:CHFR-siRNA cells treated with nocodazole. The “input” indicates 10% of the lysates used for the IP reaction. (D) Western blot analysis reveals that MCF10A:CHFR-siRNA cells have a modest increase in unmodified and acetylated α-tubulin protein levels compared to control cells. (E) A graphic representation of the data presented in (D) from triplicate experiments. There was a reproducible 1.6-fold increase in unmodified α-tubulin and a twofold increase in acetylated α-tubulin in MCF10A:CHFR-siRNA cells.

Figure 4

CHFR interacts with MAD2 and decreased CHFR expression causes the mislocalization of spindle checkpoint proteins MAD2 and BUBR1 and alters MAD2/CDC20 complex formation. (A) Immunofluorescence indicates that endogenous CHFR (red) and MAD2 (green) colocalize (yellow) in the cytoplasm of interphase MCF10A cells and strongly colocalize in mitotic cells. DNA was stained blue with DAPI. Scale bar, 50 µm. Subpanels a to e are magnified images from the merged image indicating cells in interphase (a), prophase (b), metaphase (c), anaphase (d), and telophase/cytokinesis (e). (B) Endogenous MAD2 interacts with CHFR. HEK293 cells were transfected with a Flag-tagged CHFR construct, with or without nocodazole treatment. Immunoprecipitation with an anti-Flag antibody was performed to isolate CHFR and subsequently analyzed by Western blot analysis (WB) for the Flag:CHFR fusion protein and endogenous MAD2. “Input” indicates 5% of the lysates used for the IP reaction. (C and D) Immunofluorescence to visualize BUBR1 and MAD2 reveals that cells with CHFR siRNA (right panels) have diffuse BUBRI and MAD2 staining patterns (red, D and E, respectively) indicating mislocalization. Control cells have the characteristic punctate staining patterns for BUBR1 and MAD2. DNA was stained blue with DAPI. Scale bar, 5 µm. (E) Decreased CHFR expression by siRNA impairs MAD2/CDC20 complex formation. Whole-cell lysates from control and CHFR siRNA-transfected MCF10A cells treated with nocodazole were subjected to IP with an anti-MAD2 antibody. Samples were subsequently analyzed by Western blot (WB) for both CDC20 and MAD2. “Input” lysates, representing 15% of the samples used for IP, were probed by Western blot analysis for CDC20 and MAD2, for CHFR to show efficient knockdown of expression, and for β-actin as a loading control.

Figure 5

Decreased CHFR expression impairs the mitotic spindle checkpoint. (A) Chromosomes did not properly migrate to the metaphase plate in MCF10A:CHFR-siRNA cells (arrow, right panel). Immunofluorescence (IF) detected phosphorylated Histone H3-Ser28 (green) to identify metaphase chromosomes. (B) A graph of the data shown in (A); 24% of cells with CHFR siRNA have chromosomes improperly located during metaphase. (C) MCF10A:CHFR-siRNA cells have lagging chromosomes and chromosome bridges during anaphase (arrow, right panel). DNA was stained blue with DAPI. Scale bar, 5 µm. (D) Immunofluorescence to detect cytoskeletal α-tubulin (red) during interphase shows that MCF10A:CHFR-siRNA cells become binucleated (subpanels c and d; arrow) compared to the negative control cells (subpanels a and b). Scale bar, 50 µm. (E) Quantification of the data shown in (D), in which 6% of MCF10A cells transfected with CHFR siRNA are binucleated compared to less than 2% of control cells. One asterisk (*) indicates that P < .05, whereas two asterisks (**) indicates that P < .001.

Figure 6

A proposed model of how CHFR regulates genomic instability. Decreased or lost CHFR expression causes Aurora A, α-tubulin, and acetylated α-tubulin overexpression and MAD2 mislocalization. The increase in acetylated α-tubulin occurs by an unknown mechanism, possibly through HURP or SIRT2 and may stress the mitotic spindle. Aurora A overexpression causes centrosome amplification. Both Aurora A overexpression and MAD2 mislocalization result in an impaired spindle checkpoint, contributing to aneuploidy and/or failed cytokinesis. Both processes lead to mitotic defects causing genomic instability and possibly tumorigenesis.