The putative oncogene CEP72 inhibits the mitotic function of BRCA1 and induces chromosomal instability

Abstract

BRCA1 is a tumor-suppressor gene associated with, but not restricted to, breast and ovarian cancer and implicated in various biological functions. During mitosis, BRCA1 and its positive regulator Chk2 are localized at centrosomes and are required for the regulation of microtubule plus end assembly, thereby ensuring faithful mitosis and numerical chromosome stability. However, the function of BRCA1 during mitosis has not been defined mechanistically. To gain insights into the mitotic role of BRCA1 in regulating microtubule assembly, we systematically identified proteins interacting with BRCA1 during mitosis and found the centrosomal protein Cep72 as a novel BRCA1-interacting protein. CEP72 is frequently upregulated in colorectal cancer tissues and overexpression of CEP72 mirrors the consequences of BRCA1 loss during mitosis. In detail, the overexpression of CEP72 causes an increase in microtubule plus end assembly, abnormal mitotic spindle formation and the induction of chromosomal instability. Moreover, we show that high levels of Cep72 counteract Chk2 as a positive regulator of BRCA1 to ensure proper mitotic microtubule assembly. Thus, CEP72 represents a putative oncogene in colorectal cancer that might negatively regulate the mitotic function of BRCA1 to ensure chromosomal stability.

Introduction

BRCA1 is a major tumor-suppressor gene, but it is still unclear how BRCA1 exerts its tumor-suppressor function.1, 2 This is because of the fact that BRCA1 represents a multi-functional protein that is involved in various cellular processes including the regulation of transcription, mRNA splicing, chromatin remodeling, DNA damage checkpoint control and homologous recombination repair of DNA double strand breaks.2, 3 In addition to its nuclear functions, BRCA1 has also been implicated in the regulation of interphase centrosomes.4 Interestingly, BRCA1 associates with centrosomes throughout the cell cycle and this localization is critical in the S and G2 phases when BRCA1 is required for the suppression of centrosome hyper-amplification, at least in breast-derived cells.5, 6 In contrast to the centrosomal function of BRCA1 in interphase, less is known about the role of centrosomal BRCA1 during mitosis. BRCA1 interacts with several key mitotic regulators such as TPX2, NuMA and RHAMM involved in mitotic spindle formation and its loss causes spindle abnormalities during mitosis.7 More recently, we demonstrated that the phosphorylation of BRCA1 by the checkpoint kinase and tumor suppressor Chk2 during an unperturbed mitosis is essential for proper chromosome segregation and for the maintenance of numerical chromosome euploidy.8 Furthermore, we discovered that the CHK2-BRCA1 axis is required to ensure proper microtubule plus end assembly within mitotic spindles, which is crucial for correct microtubule-kinetochore attachments and faithful chromosome segregation.9 Thus, the loss of BRCA1 or loss of its positive regulator CHK2 results in increased microtubule plus end assembly and provides the basis for a bona fide chromosomal instability (CIN) phenotype.8, 9 CIN is defined as the perpetual gain or loss of whole chromosomes during mitotic cell division and represents a major hallmark of human cancer that can support tumorigenesis and tumor progression.10 Therefore, the loss of BRCA1 or, alternatively, loss of CHK2 might induce CIN promoting the generation and progression of cancer. However, details on BRCA1 regulation at mitotic centrosomes are currently scarce.

Results and discussion

To gain insights into the regulation of BRCA1 during mitosis, we immunoprecipitated BRCA1 from mitotic cell lysates and identified interacting proteins by mass spectrometry analyses. In these experiments, we found known BRCA1 interactors such as BARD1, HSP90 and HMMR among others,11, 12, 13 and also identified the centrosomal protein Cep72 as a novel BRCA1 interacting protein (Figure 1a, Supplementary Table S1). This interaction was confirmed by immunoprecipitation experiments for BRCA1 and Cep72 and was found to be more pronounced in interphase than in mitotic cells (Figures 1b–d, Supplementary Figure S1a). So far, Cep72 is little studied, but was shown to interact with the centrosomal satellite proteins PCM1 and Cep290 being involved in regulating cilia formation in resting cells.14 In addition, Cep72 interacts with the centrosomal protein Kizuna and is required for the recruitment and stabilization of γ-tubulin ring complexes (γTuRC) and other centrosomal components during mitosis.15 Consequently, the knockdown of CEP72 causes severe spindle pole fragmentation leading to multipolar spindles and cell death.15 As Kizuna was shown to interact with various pericentriolar matrix proteins including pericentrin, γ-tubulin, ODF2 and CG-NAP/AKAP450,16 we investigated additional interactions of BRCA1 and Cep72 with pericentriolar matrix proteins. In fact, we found interactions of BRCA1 and Cep72 with pericentrin, γ-tubulin and ODF2 indicating that these proteins might be part of a common signaling complex at centrosomes (Figures 1b and c).

Figure 1
figure1

Identification of Cep72 as a novel BRCA1-interacting protein overexpressed in colorectal cancer. (a) Schematic summary of LS-MS/MS results for proteins interacting with BRCA1. Cells were synchronized in mitosis by treatment with 2 μM of the Eg5-inhibitor dimethylenastron (DME) and BRCA1 was immunoprecipitated from whole-cell lysates using anti-BRCA1 antibodies (D-9, Santa Cruz, Dallas, TX, USA). A tryptic in-gel-digestion of dissected gel slices was performed and peptides were extracted and analyzed using an Orbitrap-FT analyzer. The full list of identified proteins is given in Supplementary Table S1. The pictograms indicate examples of protein regions identified by CID fragmentation of peptides (green). The number of unique peptides with false discovery rates (FDRs)0.01 and the percentage of the total sequence coverage are given. For Cep72, all unique peptides identified, the number of spectral counts, the peptide modifications, the molecular masses of the protonized peptides (MH+), the SequestHT cross correlation values of the peptides (XCorr) and the Mascot IonScores are presented. (b, c) Co-immunoprecipitation of BRCA1 or Cep72 and detection of associated centrosomal proteins in mitotic cells. HCT116 cells (ATCC, Manassas, VA, USA; tested for mycoplasma contamination) were synchronized in mitosis by treatment with DME for 16 h, and BRCA1 (b) or Cep72 (c) was immunoprecipitated from whole-cell lysates using anti-BRCA1 antibodies (D-9, Santa Cruz) or anti-Cep72 antibodies (A301-297A, Bethyl Laboratories, Montgomery, TX, USA), respectively. Proteins were separated by sodium dodecyl sulfate–polyacrylamide gel electrophoresis and the indicated proteins were detected on western blots using the following antibodies: anti-BRCA1 (1:350, C-20, Santa Cruz), anti-BARD1 (1:1000, H-300, Santa Cruz), anti-Cep72 (1:100, A301-297A, Bethyl Laboratories), anti-pericentrin (1:1000, D-4, Santa Cruz), anti-γ-tubulin (1:2000, T3559, Sigma Aldrich, Taufkirchen, Germany), anti-ODF2 (1:300, a kind gift of S. Hoyer-Fender, Goettingen) and secondary antibodies conjugated to horseradish peroxidase (1:10000, Jackson ImmunoResearch, West Grove, PA, USA). (d) Detection of BRCA1-Cep72 interactions in interphase and mitosis. BRCA1 was immunoprecipitated from cell lysates derived form asynchronously growing cells (interphase) or from cells synchronized in mitosis by treatment with DME for 16 h and associated proteins were detected on western blots as in (b). (e) mRNA expression profiling of CEP72 in matched tissue samples from rectal carcinomas and normal mucosa. mRNA expression was evaluated in pretherapeutic biopsy material from human patients, which was labeled and hybridized to oligonucleotide-based Whole Human Genome Microarrays (4 × 44 K v2, Agilent Technologies, Santa Clara, CA, USA). Data were normalized and log2 differences of tumor—mucosa from the same patients were calculated. The box and whisker plot shows the range, mean and quartile of log2 expression (t-test, n=181). Data were deposited at GEO: (http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?token=fzwtjqkewgyiyre&acc=GSE40492). (f) Detection of Cep72 protein by immunohistochemistry analyses in tissue sections from normal mucosa and colorectal adenocarcinomas. For immunohistochemistry staining, the polyclonal Cep72-antibody (affinity isolated, rabbit, #HPA058235, Sigma Aldrich) was used, and staining intensity and the percentage of immunoreactive cells were determined as described.28 Examples and overall quantification of tumors with CEP72 overexpression compared with normal tissues are given (scale bar=100 μm, n=357).

As Cep72 represents a novel BRCA1-interacting protein, we reasoned whether CEP72 might be altered in human cancer. First, we evaluated CEP72 mRNA expression from public TCGA microarray datasets (www.oncomine.org) and found that CEP72 expression was not grossly altered in breast cancer where BRCA1 alterations are frequent (Supplementary Figure S1b). In contrast, CEP72 was found to be significantly, 3.2-fold, upregulated in colon cancer when compared with normal mucosa tissues (Supplementary Figure S1c). To corroborate this observation, we investigated the mRNA expression of CEP72 in 181 matched samples from pre-therapeutic rectal carcinomas and corresponding mucosa tissues by microarray analyses. In fact, in line with the data for colon cancer, we found a 3.2-fold increase in mRNA expression for CEP72 in tumor samples compared with normal tissues (Figure 1e, Supplementary Table S2). This alteration for CEP72 ranges among the top 0.1% of more than 19 000 genes evaluated in our microarray dataset. Moreover, we also analyzed protein levels of Cep72 in 357 colorectal adenocarcinoma tissues by immunohistochemistry analyses and found overexpression of CEP72 in 57% of the cases when compared with normal mucosa where CEP72 was only expressed in dividing cells at the base of the crypts (Figure 1f, Supplementary Table S3). However, in tumor tissues, high CEP72 expression, if present, was detected in virtually all cells and did not correlate with proliferation as evaluated by Ki67 staining (Supplementary Figure S1d, Supplementary Table S3). Interestingly, additional analyses of TCGA colon cancer datasets revealed a correlation between high CEP72 mRNA expression and an increased ‘weighted genome integrity index (wGII)’17 as well as an enhanced ‘Chromosomal Instability (CIN) index’18 (Supplementary Figures S1e and f) indicating that high CEP72 expression is associated with genomic instability in these tumors. It is intriguing that CEP72 overexpression is frequently found in colorectal cancer where BRCA1 mutations are usually rare. Vice versa, CEP72 expression seems to be rarely increased in breast cancer where BRCA1 inactivation is prevalent. Such an epistatic relationship is also found for mutations in CHK2, encoding for a positive regulator of BRCA1. In fact, mutated CHK2 confers an elevated risk for breast cancer preferentially in non-BRCA1 mutant carriers.19 Similarly, CHK2 expression is frequently lost in tumor entities such as colorectal and lung cancer8, 9, 20 that are rarely affected by BRCA1 mutations. Together, these observations support the hypothesis that CEP72 overexpression might represent an alternative genetic lesion leading to BRCA1 inactivation in colorectal cancer that contributes to the induction of genomic instability as typically seen in BRCA1-deficient tumors.21, 22 In such a scenario, overexpression of CEP72 might counteract the mitotic function of BRCA1. This would be in line with our observation that the BRCA1-Cep72 interaction is reduced in mitosis (Figure 1d), but enhanced when CEP72 is overexpressed (Supplementary Figure S2e).

Recently, we demonstrated that BRCA1 and its positive regulator Chk2 are required for the maintenance of chromosomal stability by ensuring proper microtubule plus end assembly rates during mitosis.8, 9 To investigate whether overexpression of CEP72, as a possible inhibitor of BRCA1, also affects microtubule plus end assembly rates, we tracked EB3-GFP fusion proteins in living colon cancer cells and determined the growth rates for 600 individual microtubules within mitotic spindles. Indeed, repression of BRCA1, loss of CHK2 as well as overexpression of CEP72 very similarly increased microtubule assembly rates during mitosis, but not in interphase cells (Figure 2a, Supplementary Figures S2a–c). Control experiments showed that reconstituting normal CEP72 levels restored normal microtubule assembly rates indicating specificity of the knockdown (Supplementary Figure S2d). Moreover, we previously showed that increased microtubule plus end assembly rates in mitosis can be efficiently reduced to normal levels by treatment of mitotic cells with low doses of Taxol, a drug stabilizing microtubule plus ends, or by partial repression of the plus end-directed microtubule polymerase CH-TOG.9, 23, 24 Both means were also efficient upon overexpression of CEP72 further demonstrating that overexpression of CEP72 specifically increases microtubule plus end assembly rates (Figure 2a). Interestingly, whereas the loss of CHK2 causes an increase in BRCA1-bound centrosomal Aurora-A kinase activity, which might contribute to a functional inactivation of BRCA1,9 this was not the case upon overexpression of CEP72 (Supplementary Figures S2e and f). This might indicate that the routes leading to a functional inhibition of mitotic BRCA1 in response to the loss of CHK2 and upon overexpression of CEP72 might be distinct.

Figure 2
figure2

Overexpression of CEP72 mimics the loss of BRCA1 function during mitosis. (a) Overexpression of CEP72 or partial loss of BRCA1 or CHK2 causes increased spindle microtubule assembly rates during mitosis. HCT116 cells were transfected with control short interfering RNAs (siRNAs) (5′-CUUACGCUGAGUACUUCGAUU-3′), BRCA1 siRNAs (5′-GGAACCUGUCUCCACAAAG-3′) or CHK2 siRNAs (5′-CCUUCAGGAUGGAUUUGCCAAUC-3′) using INTERFERin transfection reagent (Polyplus-Transfection, Illkirch-Graffenstaden, France) according to the manufacturer’s protocol. After 24 h, cells were transfected with pEGFP-EB3 (provided by Linda Wordeman, Seattle, WA, USA) and either empty pcDNA3 (Life Technologies, Carlsbad, CA, USA) or pcDNA3-CEP72 (CEP72 cDNA purchased from BioCat (Heidelberg, Germany) and cloned into pcDNA3) by electroporation (300 V, 500 μF, Bio-Rad (Hercules, CA, USA) electroporator). Cells were seeded onto glass bottom dishes (Ibidi, Martinsried, Germany) and cultivated in the presence of either dimethyl sulfoxide or 0.2 nM Taxol for 24 h. Two hours prior to live cell analysis, cells were treated with 2 μM DME and GFP-EB3 signals were tracked in monopolar spindles using a Deltavision ELITE microscope (GE Healthcare, Chalfont St. Giles, UK) equipped with a CoolSnap-HQ2 (Photomertics, Tucson, AZ, USA) and a PCO Edge sCMOS camera (PCO, Kehlheim, Germany) as described.9 Images of four sections with a Z-optical spacing of 0.4 μm were taken every 2 s, deconvolved and analyzed using SoftWorx 5.0/6.0 and softWoRx Explorer 1.3.0 software (Applied Precision, Issaquah, WA, USA). Average microtubule plus end assembly rates were calculated from 20 individual microtubules (mean±s.e.m., t-test, n=30 cells from three independent experiments). (b) Detection and quantification of abnormal metaphase spindles in HCT116 cells upon overexpression of CEP72 or after partial loss of BRCA1. HCT116 cells were transfected with control or BRCA1 siRNAs and empty pcDNA3 (control vector) or pcDNA3-CEP72 as in (a). Cells were synchronized in metaphase by release from a double thymidine block into medium containing 20 μg MG132 for 3 h. After PFA/methanol fixation, cells were stained with anti-α-tubulin (1:650, B-5-1-2, Santa Cruz), anti-Crest (1:800, Europa Bioproducts, Ely, UK), secondary antibodies conjugated to Alexa Fluor-488/−594 (1:1000, Life Technologies) and Hoechst 33342 (Sigma Aldrich) and analyzed using a Leica DM6000B microscope (Leica, Wetzlar, Germany) or a DeltaVision ELITE microscope. Images were taken with a Z-optical spacing of 0.2 μm, subsequently deconvolved and analyzed using the SoftWorx 5.0/6.0 and softWoRx Explorer 1.3.0 software. Left panel: Representative examples of curved and distorted spindles of CEP72-overexpressing cells (spindles, anti-α-tubulin: green; kinetochores, CREST: red; chromosomes, Hoechst 33342: blue; scale bar, 10 μm). Right panel: Quantification of abnormal metaphase spindles with or without treatment with 0.2 nM Taxol for 24 h prior to analysis (mean±s.d.; t-test, n=300 bipolar spindles of three independent experiments). (c) Determination of the average microtubule length in normal and curved/distorted metaphase spindles in cells transiently overexpressing CEP72. Cells were transfected, synchronized in metaphase and analyzed by immunofluorescence microscopy as described in (b). Centrosomes were visualized by anti-γ-tubulin antibodies (1:550, T3559, Sigma Aldrich) in immunofluorescence microscopy experiments. Left panel: Example and formula to calculate average microtubule length in bipolar spindles. Right panel: Box and whisker plot showing the range, mean and quartile of the measurements (t-test, n=61–64 cells from three independent experiments). (d) Determination of the pole-to-pole distance in normal and curved/distorted metaphase spindles in cells transiently overexpressing CEP72. Cells were transfected and synchronized in metaphase as in (b), and the pole-to-pole distance was determined by immunofluorescence analysis of cells stained for α- and γ-tubulin and Hoechst 33342. The box and whisker plot shows the range, mean and quartile of the measurements (t-test, n=61–64 cells from three independent experiments).

Because proper microtubule dynamics is pivotal for accurate mitotic spindle assembly, we analyzed bipolar spindle formation in metaphase. Intriguingly, the repression of BRCA1 or CEP72 overexpression induced mitotic spindles that appeared in a curved and distorted manner (Figure 2b). Those abnormally shaped spindles are a direct consequence of abnormal microtubule dynamics because they were efficiently suppressed upon restoration of normal microtubule assembly rates by treatment with low doses of Taxol (Figures 2a and b). In addition, as measurable parameters directly associated with curved spindles, we found increased microtubule length (Figure 2c) and increased pole-to-pole distance (Figure 2d) within the metaphase spindles.

The phenotypic similarity of high CEP72 expression and loss of BRCA1 in mitotic cells prompted us to investigate whether CEP72, like BRCA1, might act as a bona fide CIN gene. To test this, we generated single-cell clones derived from chromosomally stable HCT116 cells overexpressing CEP72 (Supplementary Figure S3a). These cell clones were cultured for a defined time span of 30 generations, and karyotype analyses using chromosome counting and CEP-FISH were performed to determine karyotype heterogeneity that evolved over time. In fact, cells stably overexpressing CEP72 and exhibiting an increase in microtubule assembly rates (Supplementary Figure S3b) generated a three- to four-times higher karyotype variability within 30 generations than control clones (Figure 3a, Supplementary Figures S3c and d, Supplementary Table S4), which indicates a bona fide CIN phenotype.

Figure 3
figure3

CEP72 overexpression or loss of BRCA1 causes CIN, which is mediated by an increase in mitotic microtubule assembly rates. (a) Determination of the chromosome number variability in single-cell clones derived from control or CEP72 overexpressing HCT116 cells. Stable cell lines were generated as described in Supplementary Figure S1a. Single-cell clones were cultured for 30 generations and subjected to karyotype analyses by chromosome counting as described.9 A representative example for a metaphase chromosome spread is given (scale bar, 10 μm) and the quantification is based on the analyses of 100 metaphase spreads for each cell clone as indicated. (b) Quantification of cells exhibiting lagging chromosomes during anaphase. Control cells and single-cell clones stably overexpressing CEP72 were synchronized in G1/S by a double thymidine block and released into medium for 9.5 h. Anaphase cells were analyzed by immunofluorescence microscopy detecting CREST-positive chromosomes lagging in the area of the equatorial plane. A representative example of a cell showing a lagging chromosome is given (DNA: blue; kinetochores, CREST: red; scale bar, 10 μm), and for the quantification, 200 anaphase cells were evaluated for each cell clone. (c) Restoration of normal microtubule plus end assembly rates by treatment with low doses of Taxol. Single-cell clones derived from the indicated stable cell lines were generated in the absence (dimethyl sulfoxide) or presence of 0.2 nM Taxol and were grown for 30 generations as indicated in Supplementary Figure S3e. Cells were transfected with pEGFP-EB3 and subjected to microtubule plus end assembly rate measurements by live cell microscopy. Scatter dot plots show average microtubule assembly rates based on measurement of 20 microtubules per cell (n=10 cells; t-test). (d) Determination of the proportion of cells exhibiting lagging chromosomes in response to restoration of normal microtubule assembly rates. The indicated single-cell clones were grown in the absence or presence of 0.2 nM Taxol and anaphase cells showing lagging chromosomes were determined as described in (b). The graph shows mean values±s.e.m. (n=3 independent experiments with 300 cells evaluated). (e) Determination of karyotype variability as a measure for CIN in single-cell clones overexpressing CEP72 or showing BRCA1 knockdown and grown in the absence or presence of 0.2 nM Taxol for 30 generations. The graph represents the proportion of cells showing karyotype deviations from the modal of 45 chromosomes (n=50 metaphase spreads). Detailed chromosome numbers for each metaphase spread are given in Supplementary Table S4.

In human cancer cells, chromosome missegregation is strongly associated with the appearance of lagging chromosomes during anaphase, which arise in response to unresolved erroneous (merotelic) microtubule-kinetochore attachments.25 Moreover, increased microtubule assembly rates upon the loss of CHK2-BRCA1 facilitate the generation of such faulty kinetochore attachments.8 Similarly, the overexpression of CEP72 induced anaphase cells with lagging chromosomes providing another independent measure for chromosome missegregation in these cells (Figure 3b). To demonstrate that the CIN phenotype upon CEP72 overexpression or after loss of BRCA1 is indeed mediated by an increase in microtubule plus end polymerization rates, we generated single-cell clones in the absence or presence of low doses of Taxol (Supplementary Figures S3e and f), which restored proper microtubule plus end assembly in those cell clones (Figure 3c). Importantly, this treatment efficiently suppressed the generation of lagging chromosomes (Figure 3d) and restored chromosomal stability in cells either overexpressing CEP72 or repressing BRCA1 (Figure 3e, Supplementary Figure S3g, Supplementary Table S4). These results establish CEP72 as a novel bona fide CIN gene and demonstrate that overexpression of CEP72 and loss of BRCA1 result in congruent mitotic defects, namely abnormally increased microtubule plus end dynamics that lead to abnormal spindle assembly and perpetual chromosome missegregation during mitosis. Interestingly, previous work has suggested that the loss of BRCA1 is associated with centrosome amplification during interphase leading to subsequent chromosome missegregation and CIN.5, 6 However, this was only observed in cell lines derived from breast tissues and was not detected in other cell lines.6 The reason for this tissue-specificity remains enigmatic, but we also found no centrosome amplification in colon cancer cells upon overexpression of CEP72 (Supplementary Figure S3h). In contrast to that, our results strongly suggest that the loss of BRCA1, loss of its positive regulator CHK2 or overexpression of its putative inhibitor CEP72 induce CIN by triggering abnormal microtubule dynamics in mitosis.

On the basis of the results showing that overexpression of CEP72 phenotypically mirrors the loss of BRCA1, we hypothesize that overexpressed CEP72 inhibits the mitotic function of BRCA1 to regulate microtubule plus end assembly and thus, counteracts Chk2 as a positive regulator of BRCA1 (illustrated in the model in Figure 4a). To investigate this hypothesis, we determined microtubule plus end assembly rates as a readout for the mitotic function of BRCA1 in response to CEP72 overexpression and concomitant overexpression or repression of CHK2. In fact, elevating the positive regulator Chk2 in the presence of CEP72 overexpression efficiently suppressed the increase in microtubule dynamics. In contrast, simultaneous repression of CHK2 did not further accelerate microtubule plus end assembly rates induced by CEP72 overexpression alone, suggesting that Cep72 and Chk2 are acting indeed in the same pathway to regulate microtubule plus end assembly (Figure 4b, Supplementary Figure S4a). Vice versa, repression of CEP72 restored proper microtubule assembly rates upon partial loss of CHK2, whereas overexpression of CEP72 did not, which further supports the expectation derived from our model (Figure 4c, Supplementary Figure S4b). Interestingly, the repression of CEP72 did not restore normal mitotic microtubule assembly rates in the complete absence of the positive regulator Chk2 (upon homozygous deletion of CHK2 in isogenic HCT116-CHK2−/− cells;26) indicating that Chk2 is essential for BRCA1 to mediate proper microtubule assembly during mitosis (Supplementary Figure S4c). Importantly, when BRCA1 is partially repressed, which is sufficient to cause an increase in microtubule plus end assembly, only the overexpression of CHK2 or the repression of CEP72 can restore proper microtubule assembly rates (Figure 4d, Supplementary Figure S4d). Thus, the positive regulator Chk2 and the negative regulator Cep72 seem to act in an opposing manner on BRCA1 to regulate microtubule plus end assembly during mitosis. To further investigate the functional importance of the balance between Chk2 and Cep72 for the mitotic role of BRCA1, we analyzed mitotic metaphase spindles and the generation of lagging chromosomes during anaphase as an indicator for chromosome missegregation in response to alterations in the Chk2-Cep72 balance. In full accordance with the results from the microtubule plus end assembly measurements, we found that abnormal metaphase spindles as well as lagging chromosomes and thus, chromosome missegregation, were suppressed in the presence of CEP72 overexpression only when CHK2 expression was concomitantly elevated. Vice versa, abnormal spindles and chromosome missegregation were rescued upon partial loss of the positive regulator CHK2 only when the dosage of CEP72 was simultaneously decreased (Figures 4e and f). These results support a model, in which high levels of Cep72 inhibit BRCA1 in mitosis and thus, represent a counterpart for its positive regulator Chk2 (Figure 4a). However, it is currently unclear how Cep72 acts on BRCA1 mechanistically. As an interacting protein, it might interfere with the ubiquitin ligase activity of the BRCA1/BARD1 complex, which appears to be indeed required during mitosis to ensure normal microtubule assembly rates (our unpublished results). This would be compatible with only little BRCA1-Cep72 interaction present during a normal mitosis (Figure 1d) leaving the normal mitotic function of BRCA1 intact. However, overexpression of CEP72 does not disrupt the BRCA1-BARD1 interaction per se (Supplementary Figure S2e). It also does not affect the binding of Aurora-A to BRCA1 (Supplementary Figure S2e), which might be relevant to BRCA1 ubiquitin ligase activity because Aurora-A can directly phosphorylate BRCA1 leading to an inhibition of its ubiquitin ligase activity.5 So far, physiologically relevant targets for the BRCA1/BARD1 ubiquitin ligase are still largely unknown,2, 27 but it will be crucial to identify targets for BRCA1/BARD1 that are restricted to mitosis to understand how BRCA1 restrains microtubule plus end assembly during mitosis.

Figure 4
figure4

Cep72 acts as an opponent for Chk2 to negatively regulate BRCA1 function during mitosis. (a) Model for the mitotic regulation of BRCA1 by Chk2 and Cep72 to ensure proper spindle microtubule assembly during mitosis. (b) Measurements of mitotic spindle microtubule plus end assembly rates in HCT116 cells after CEP72 overexpression and concomitant knockdown or overexpression of CHK2. HCT116 cells were transfected with control or CHK2 short interfering RNAs (siRNAs). Four hours later, cells were additionally transfected with 2 μg of pcDNA3, pcDNA3-CEP72 or pCEF-CHK2, and microtubule assembly rates were determined 48 h after transfection. Scatter dot plots show average plus end assembly rates based on measurement of 20 microtubules per cell (mean±s.e.m., t-test n=30 cells from three independent experiments). (c) Measurements of mitotic spindle microtubule assembly rates after CHK2 repression and concomitant knockdown or overexpression of CEP72. Cells were transfected with control, CHK2 or CEP72 (5′-TTGCAGATCGCTGGACTTCAA-3′) siRNA and either pcDNA3 or pcDNA3-CEP72 as in (b). Forty-eight hours later, microtubule plus end assembly rates were determined. Scatter dot plots show average plus end assembly rates based on measurement of 20 microtubules per cell (mean±s.e.m., t-test n=30 cells from three independent experiments). Note that measurements for CHK2 siRNA plus CEP72 overexpression were measured in the same set of experiments and are shared between panel (b) and (c). (d) Measurements of mitotic spindle microtubule plus end assembly rates in cells with normal or partially reduced expression of BRCA1 after concomitant CHK2 and CEP72 repression or overexpression. Cells were transfected with control, BRCA1, CHK2 or CEP72 siRNA and pcDNA3, pcDNA3-CEP72 or pCEF-CHK2 as indicated in (b). Microtubule plus end assembly rates were determined in cells synchronized in mitosis 48 h after transfection. Scatter dot plots show average plus end assembly rates based on measurement of 20 microtubules per cell (mean±s.e.m., t-test n=30 cells from three independent experiments). (e) Quantification of abnormal metaphase spindles in HCT116 cells after overexpression of CEP72 or repression of CHK2 and concomitant knockdown or overexpression of CHK2 or CEP72, respectively. Cells were transfected as described in (b) and (c), and the proportion of cells exhibiting curved and distorted metaphase spindles were determined. The graph shows mean values±s.d.; t-test, n=300 bipolar spindles from three independent experiments. (f) Quantification of the proportion of cells exhibiting lagging chromosomes after overexpression of CEP72 or repression of CHK2 and concomitant knockdown or overexpression of CHK2 or CEP72, respectively. Cells were transfected as in (b) and (c), and lagging chromosomes were detected in anaphase cells. The graph shows mean values±s.e.m. (n=3 independent experiments with 300 cells evaluated in total).

On the other hand, one might also speculate that BRCA1 can act as an inhibitor of Cep72. In fact, we found evidence that BRCA1 might be part of a centrosomal signaling complex containing Cep72 and various pericentriolar matrix proteins (Figures 1b and c).15, 16 It is possible that loss of BRCA1 in cancer cells might result in Cep72 hyperactivity leading to increased microtubule assembly and thus, would mimic an overexpression of CEP72 as seen in colorectal cancer. Defining the mechanistic roles of BRCA1 and Cep72 at mitotic centrosomes will require further detailed investigations.

References

  1. 1

    Miki Y, Swensen J, Shattuck-Eidens D, Futreal PA, Harshman K, Tavtigian S et al. A strong candidate for the breast and ovarian cancer susceptibility gene BRCA1. Science 1994; 266: 66–71.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  2. 2

    Silver PD, Livingston MD . Mechanisms of BRCA1 tumor suppression. Cancer Discov 2012; 2: 679–684.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  3. 3

    Savage IK, Harkin PD . BRCA1, a 'complex' protein involved in the maintenance of genomic stability. FEBS J 2014.

  4. 4

    Kais Z, Parvin DJ . Regulation of centrosomes by the BRCA1-dependent ubiquitin ligase. Cancer Biol Ther 2008; 7: 1540–1543.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  5. 5

    Sankaran S, Crone ED, Palazzo ER, Parvin JD . Aurora-A kinase regulates breast cancer associated gene 1 inhibition of centrosome-dependent microtubule nucleation. Cancer Res 2007; 67: 11186–11194.

    CAS  Article  PubMed  Google Scholar 

  6. 6

    Starita ML, Machida Y, Sankaran S, Elias JE, Griffin K, Schlegel BP et al. BRCA1-dependent ubiquitination of gamma-tubulin regulates centrosome number. Mol Cell Biol 2004; 24: 8457–8466.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  7. 7

    Joukov V, Groen AC, Prokhorova T, Gerson R, White E, Rodriguez A et al. The BRCA1/BARD1 heterodimer modulates ran-dependent mitotic spindle assembly. Cell 2006; 127: 539–552.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  8. 8

    Stolz A, Ertych N, Kienitz A, Vogel C, Schneider V, Fritz B et al. The CHK2-BRCA1 tumour suppressor pathway ensures chromosomal stability in human somatic cells. Nat Cell Biol 2010; 12: 492–499.

    CAS  Article  Google Scholar 

  9. 9

    Ertych N, Stolz A, Stenzinger A, Weichert W, Kaulfuß S, Burfeind P et al. Increased microtubule assembly rates influence chromosomal instability in colorectal cancer. Nat Cell Biol 2014; 16: 779–791.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  10. 10

    Lengauer C, Kinzler WK, Vogelstein B . Genetic instabilities in human cancers. Nature 1998; 396: 643–649.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  11. 11

    Wu CL, Wang WZ, Tsan TJ, Spillman MA, Phung A, Xu XL et al. Identification of a RING protein that can interact in vivo with the BRCA1 gene product. Nat Genet 1996; 14: 430–440.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  12. 12

    Pujana AM, Han DJ, Starita ML, Stevens KN, Tewari M, Ahn JS et al. Network modeling links breast cancer susceptibility and centrosome dysfunction. Nat Genet 2007; 39: 1338–1349.

    CAS  Article  PubMed  Google Scholar 

  13. 13

    Stecklein RS, Kumaraswamy E, Behbod F, Wang W, Chaguturu V, Harlan-Williams LM et al. BRCA1 and HSP90 cooperate in homologous and non-homologous DNA double-strand-break repair and G2/M checkpoint activation. Proc Natl Acad Sci USA 2012; 109: 13650–13655.

    CAS  Article  PubMed  Google Scholar 

  14. 14

    Stowe RT, Wilkinson JC, Iqbal A, Stearns T . The centriolar satellite proteins Cep72 and Cep290 interact and are required for recruitment of BBS proteins to the cilium. Mol Biol Cell 2012; 23: 3322–3335.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  15. 15

    Oshimori N, Li X, Ohsugi M, Yamamoto T . Cep72 regulates the localization of key centrosomal proteins and proper bipolar spindle formation. EMBO J 2009; 28: 2066–2076.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  16. 16

    Oshimori N, Ohsugi M, Yamamoto T . The Plk1 target Kizuna stabilizes mitotic centrosomes to ensure spindle bipolarity. Nat Cell Biol 2006; 8: 1095–1101.

    CAS  Article  PubMed  Google Scholar 

  17. 17

    Burrell AR, McClelland ES, Endesfelder D, Groth P, Weller MC, Shaikh N et al. Replication stress links structural and numerical cancer chromosomal instability. Nature 2013; 494: 492–496.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  18. 18

    Carter SL, Eklund AC, Kohane IS, Harris LN, Szallasi Z . A signature of chromosomal instability inferred from gene expression profiles predicts clinical outcome in multiple human cancers. Nat Genet 2006; 38: 1043–1048.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  19. 19

    Meijers-Heijboer H, Ouweland vd A, Klijn J, Wasielewski M, de Snoo A, Oldenburg R et al. Low-penetrance susceptibility to breast cancer due to CHEK2(*)1100delC in noncarriers of BRCA1 or BRCA2 mutations. Nat Genet 2002; 31: 55–59.

    CAS  Article  Google Scholar 

  20. 20

    Zhang P, Wang J, Gao W, Yuan BZ, Rogers J, Reed E . CHK2 kinase expression is down-regulated due to promoter methylation in non-small cell lung cancer. Mol Cancer 2004; 3: 14.

    Article  PubMed  PubMed Central  Google Scholar 

  21. 21

    Xu X, Weaver Z, Linke PS, Li C, Gotay J, Wang XW et al. Centrosome amplification and a defective G2-M cell cycle checkpoint induce genetic instability in BRCA1 exon 11 isoform-deficient cells. Mol Cell 1999; 3: 389–395.

    CAS  Article  Google Scholar 

  22. 22

    Weaver Z, Montagna C, Xu Xm, Howard T, Gadina M, Brodie SG et al. Mammary tumors in mice conditionally mutant for BRCA1 exhibit gross genomic instability and centrosome amplification yet display a recurring distribution of genomic imbalances that is similar to human breast cancer. Oncogene 2002; 21: 5097–5107.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  23. 23

    Derry BW, Wilson L, Jordan AM . Low potency of taxol at microtubule minus ends: implications for its antimitotic and therapeutic mechanism. Cancer Res 1998; 58: 1177–1184.

    CAS  PubMed  Google Scholar 

  24. 24

    Brouhard JG, Stear HJ, Noetzel LT, Al-Bassam J, Kinoshita K, Harrison SC et al. XMAP215 is a processive microtubule polymerase. Cell 2008; 132: 79–88.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  25. 25

    Gregan J, Polakova S, Zhang L, Tolić-Nørrelykke IM, Cimini D et al. Merotelic kinetochore attachment: causes and effects. Trends Cell Biol 2011; 21: 374–381.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  26. 26

    Jallepalli VP, Lengauer C, Vogelstein B, Bunz F . The Chk2 tumor suppressor is not required for p53 responses in human cancer cells. J Biol Chem 2003; 278: 20475–20479.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  27. 27

    Starita ML, Parvin DJ . Substrates of the BRCA1-dependent ubiquitin ligase. Cancer Biol Ther 2006; 5: 137–141.

    CAS  Article  PubMed  Google Scholar 

  28. 28

    Weichert W, Roske A, Gekeler V, Beckers T, Ebert MP, Pross M et al. Association of patterns of class I histone deacetylase expression with patient prognosis in gastric cancer: a retrospective analysis. Lancet Oncol 2008; 9: 139–148.

    CAS  Article  PubMed  Google Scholar 

Download references

Acknowledgements

We thank Linda Wordeman, Sigrid Hoyer-Fender, Bert Vogelstein, Ingrid Hoffmann and Olaf Stemmann for materials and Heike Krebber for microscopy support. We thank Dennis Vollweiter and Eric Schoger for general lab support. We thank the TCGA Research Network (http://cancergenome.nih.gov/) for the open access of gene expression data. This work was supported by the Deutsche Forschungsgemeinschaft (HB and GHB) and by a DFG funded Heisenberg professorship awarded to HB.

Author information

Affiliations

Authors

Corresponding author

Correspondence to H Bastians.

Ethics declarations

Competing interests

The authors declare no conflict of interest.

Additional information

Supplementary Information accompanies this paper on the Oncogene website

Supplementary information

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Lüddecke, S., Ertych, N., Stenzinger, A. et al. The putative oncogene CEP72 inhibits the mitotic function of BRCA1 and induces chromosomal instability. Oncogene 35, 2398–2406 (2016). https://doi.org/10.1038/onc.2015.290

Download citation

Further reading

Search

Quick links