Patent Application
20100041096
This invention relates to a method of polyubiquitinating a nucleophsmin including the reaction of nucleophosmin with BRCA1-BARD1.
BRCA1, which is a tumor suppressor gene for breast and ovarian cancers, is one of the most important genes in the field of breast cancer research1-4. Recently, the present inventors have found that BRCA1 and BARD1 form a RING heterodimeric ubiquitin ligase and that the activity of this ligase is completely inactivated by a missense mutation in BRCA1 which causes familial breast cancer1,9. Further, the inventors have reported that polyubiquitin chains catalyzed by BRCA1-BARD1 are not the conventional Lys48-linked ubiquitin chains which function as a signal for proteolysis by 26S proteasome, but are Lys6-linked ubiquitin chains that have not yet been reported in the past and that these polyubiquitin chains are deubiquitinated in vitro by 26S proteasome2,3. However, it has not been elucidated how the activity of the ubiquitin ligase is involved in the tumor suppressing function of BRCA1. The major reasons are that the substrate for the ligase activity has not been identified and that the upstream signals that control the activity have not been clearly understood.
It has been reported that nucleophosmin (NPM) is present in the centrosome during the G1 phase and that centrosome replication begins when NPM has been phosphorylated by CDK2-cyclin E during the G1/S transition phase and separated from the centrosome7,8. Since NPM returns to the centrosome at the time of mitosis and the centrosome of the daughter cells after mitosis has NPM, it has been indicated that NPM may be a licensing factor for centrosome replication. However, it has not been elucidated how NPM returns to the centrosome during the mitotic phase. On the other hand, it is known that deletion of BRCA1 causes excessive replication of the centrosome, which makes the genome unstable. This mechanism has not been clearly understood yet.
As mentioned above, it is not clear how the BRCA1-BARD1 ligase regulates various cellular processes such as DNA repair, cell-cycle progression, and centrosome replication4-6. The purpose of the present invention is to carry out polyubiquitination of nucleophosmin including the reaction of nucleophosmin with BRCA1-BARD1. The present inventors thoroughly investigated in an attempt to solve the aforementioned subject of concern and found that BRCA1-BARD1 bind to and catalyze the polyubiquitination of nuclear phosphoprotein, nucleophosmin/B23 (NPM), which causes NPM stabilization.
In one aspect of the invention, a method of polyubiquitinating a nucleophosmin comprised by reacting the nucleophosmin with BRCA1-BARD1 is provided. According to this method, the polyubiquitination may be carried out in vitro or in vivo.
In one aspect of the invention, a method of stabilizing nucleophosmin comprising the polyubiquitination of nucleophosmin is provided.
In another aspect of the invention, a method of inhibiting polyubiquitination of nucleophosmin comprising phosphorylating a BARD1 using CDK2-cyclin E and/or CDK2-cyclin A, and a method of dissociating BRCA1-BARD1, and a method of inactivating the activity of ubiquitin ligase are provided. In the present methods, phosphorylation sites of the BARD1 are at least 3 sites selected from the group consisting of S148, S251, S288 and T299, and the said sites can be S148, S288 and T299 and can be S148, S251, S288 and T299.
The other aspect of the invention is a method of transporting BRCA1 from nucleus to cytoplasm expression comprising of both BRCA1 and CDK2-cyclin E and/or CDK1-cyclin A.
The present invention relates to a method of polyubiquitinating a nucleophosmin comprising of reacting nucleophosmin with BRCA1-BARD1. NPM coexists with BRCA1 at the spindle pole at the M phase where it is ubiquitinated. It has been known that CDK2-cyclin E and CDK2-cyclin A phosphorylate NPM during the G1/S transition phase and S phase and NPM is dissociated from the centrasome7, 8. To our interest, CDK2 also phosphorylates BARD1. As a result, dissociation of BRCA1-BARD1 occurs, thus, NPM ubiquitination by BRCA1-BARD1 is suppressed. Due to deletion of BRCA1, excessive replications of the centrosome and chromosome distribution abnormalities occur. This new CDK2-BRCA1-NPM pathway was suggested to be the mechanism of playing the role of BRCA1 in the replication of the centrosome. This invention will be described in detail below.
Nucleophosmin in the present invention (hereinafter referred to as “NPM”) is a substrate of BRCA1-BARD1 ubiquitin ligase. It has a variety of biological activities in all cells such as ribosome biosynthesis, apoptosis inhibition, and histone chaperone functions, etc. For example, it is included nucleophosmin/B23/NO38 (NPM). The inventors have identified nucleophosmin/B231NO 38 (NPM) as the substrate for BRCA1-BARD1 ubiquitin ligase by two screening methods using a mass spectrometer (LC/MS/MS). NPM coexists with BRCA1 at the spindle pole during the mitotic phase (M phase) and is ubiquitinated. It has been known that CDK2-cyclin E and CDK2-cyclin A phosphorylate NPM and dissociate NPM from the centrosome, which is result in centrosome is divided into the daughter centrosomes7,8 during the G1/S transition phase and S phase. During the S phase, NPM cannot bind with the centrosome. Meanwhile, the daughter centrosomes grow to the centrosome and subsequently to the two spindle poles. As a control mechanism of centrosome replication, a CDK2-BRCA1-NPM pathway exists and NPM ubiquitination by BRCA1-BARD1 plays an important role in the re-localization of NPM to the spindle poles during the mitotic phase.
BRCA1 is a tumor suppressor gene for breast and ovarian cancers and is one of the most important genes in the field of breast cancer research. BRCA1 and BARD1 form a RING heterodimer to acquire high ubiquitin ligase activity1-4. Targeted disruption of the BRCA1 gene in mice resulted in centrosome hyperamplification and genomic instability13, 14 During mitosis, BRCA1 is localized to the centrosome through its binding to gamma-tubulin15, 16.
BARD1 was identified as a RING finger protein bound to BRCA1 (BRCA1 associated Ring Domain 1).
BRCA1-BARD1 form a RING heterodimeric ubiquitin ligase and the activity of this ligase is completely inactivated due to missense mutation in BRCA1, which causes familial breast cancer1,9
BRCA1-BARD1 ligase regulates a variety of cellular processes such as DNA repair, cell-cycle progression and centrosome replication. According to the immunocytostaining experiments, the behaviors of NPM, BRCA1 and BARD1 at various phases of cellular cycle were found to be as follows. Namely, NPM was present in the nucleosome during the interphase, whereas BRCA1 and BARD1 were present in the nucleus other than the nucleosome. However, they coexisted around the nucleus near the spindle body and in the centrosome (spindle pole) during the mitotic phase. According to the tests using Hela cells synchronized by a thymidine-nocodazole block at the time of mitotic phase of the cell cycle, NPM was found to be polyubiquitinated in a short time during the transition phase from the mitotic phase to the G1 phase. In order to observe subcellular localization and its coexistence with NPM during the cell cycle, a rabbit polyclonal antibody specific to the C-terminal of BARD1 can be utilized by preparing it. Synchronization of the cellular cycle is monitored by flowcytometry and in vivo NPM ubiquitination was assessed by IP-Western analysis.
Polyubiquitin chains catalyzed by BRCA1-BARD1 are not the conventional Lys48-linked ubiquitin chains which function as a signal for proteolysis by 26S proteasome, but Lys6-linked ubiquitin chains that have not yet been reported in the past and that these polyubiquitin chains are de-ubiquitinated in vitro by 26S proteasome2, 3. NPM is ubiquitinated and stabilized by BRCA1-BARD1 in vivo, but this ubiquitination does not function as a signal for the 26S proteasome-dependent proteolysis.
It is noteworthy17 that NPM redistributes to the spindle poles during the mitotic phase. NPM ubiquitination by BRCA1-BARD1 is important for redistribution so that its absence may cause excessive replication of the centrosome. Furthermore, many functions attributed to NPM such as upregulation after DNA damage18, a role during the cell cycle and apoptosis19,20, and implications in chromatin remodeling21,22 overlap with the functions of BRCA1. Therefore, NPM ubiquitination by BRCA1-BARD1 can be the mechanism carried by BRCA1 as one of the known functions.
Ubiquitination is a process wherein enzymes such as ubiquitin activation enzyme (E1), ubiquiting binding enzyme (E2), and ubiquitin ligase (E3), are coordinated to bind ubiquitin sequentially onto the protein as a substrate. The physiological implication of ubiquitination is to lead the substrate to the 26S proteasome-dependent proteolysis, which is due to ubiquitin-ubiquitin links via Lys-48 on ubiquitin11, 12. However, the present inventors and other researchers discovered recently that BRCA1-BARD1 catalyzes the formation of a non-conventional Lys-6 linked type polyubiquitin chain. It is deubiquitinated by purified 26S proteasome in vitro instead of being targeted for proteolysis2, 3.
“Self ubiquitination” used in this specification means that due to the presence of ubiquitin ligase activity of BRCA1-BARD1, BRCA1 itself becomes a substrate protein for ubiquitination so that BRCA1 itself binds ubiquitin rather than using other ubiquitin ligase.
NPM ubiquitination does not function as a signal for the 26S proteasome-dependent proteolysis. Whether or not BRCA1-BARD1-mediated NPM ubiquitination targets NPM for proteolysis can be determined by measuring its stability in vivo. In the case of expression with NPM alone and in the case of coexpression in the presence of BRCA1-BARD1, if the stability of NPM is higher in the case of coexpression, NPM ubiquitination by BRCA1-BARD1 is not a signal for the proteasome proteolysis. Whether or not BRCA1-BARD1-mediated NPM ubiquitination targets NPM for proteolysis can also be determined by the pulse-chase analysis. Also, it can be determined by treatment with proteasome inhibitors MG132 or LLnL. In this case, if the amount of ubiquitinated Flag-NPM does not increase, BRCA1-BARD1-mediated NPM ubiquitination is not a signal for proteasome proteolysis. These finding suggest that BRCA1-BARD1-mediated NPM ubiquitination affects the function of NPM through a non-proteolytic mechanism
It is also clear that NPM is ubiquitinated in vitro by BRCA1-BARD1. Whether BRCA1-BARD1 ubiquitinates NPM in vitro can be tested in an in vitro system using a recombinant protein alone. For example, tagged NPM, BRCA1 and BARD1 are incubated with E1 and E2 ubiquitins, and tagged NPM is detected by immunoblotting using an appropriate antibody. This demonstrates BRCA1-BARD1-dependent NPM ubiquitination in a purified system.
CDK is a cyclin-dependent kinase. CDK2-cyclin E implies a complex of CDK2 with cyclin E and CDK2-cyclin A implies a complex of CDK2 with cyclin A. Both are coordinated in the cell cycle regulatory system.
A complex of CDK2-cyclin E induces a cell to pass G1/S commitment (starting point), and a complex of CDK2-cyclin A is known to be necessary for the activation of the subsequent DNA replication apparatus. Cyclin B described later in this specification forms a CDK1 (cdc2) complex and induces the start of the miotic phase.
Both CDK2-cyclin E and CDK2-cyclin A completely abolish the ubiquitination of NPM by BRCA1-BARD1 in vivo. There are two possible mechanisms by which CDK2 could inhibit BRCA1-BARD1-mediated ubiquitination:
1) CDK2 phosphorylates NPM to disguise the BRCA1-BARD1 recognition elements or
2) CDK2 directly phosphorylates and inhibits the BRCA1-BARD1 ligase.
To distinguish between these two models, we first tested whether the BRCA1-BARD1 ligase is capable of ubiquitinating NPM, a mutant that abolishes the CDK2 phosphorylation sites. The aforementioned mutant is ubiquitinated by BRCA1-BARD1 ligase and the ubiquitination is inhibited by CDK2. BARD1 is a substrate of CDK2-cyclin E1 so that there is a possibility that inhibition of the BRCA1-BARD1 ubiquitin ligase activity directly affects phosphorylation of BARD1. However, BRCA1 autoubiquitination via non-phosphorylation mutant of BARD1 is also inhibited by CDK2-cyclin E1 suggesting that not all inhibition of NPM ubiquitination is attributed to phosphorylation of BARD1.
Phosphorylation of BARD1 by CDK when both BARD1 and CDK-cyclin are co-expressed can be analyzed by measuring the changes in the molecular weight of BARD1. The changes in floating positions disappear by treating the BARD1 immunoprecipitates with alkali phosphatase. This suggests that BARD1 is a direct substrate of CDK2 kinase. In this case, when using NH2 terminal segments of BARD1 and COOH terminal segments of BARD1, it is possible to determine in which areas phosphorylation of BARD1 occurs. Also, in order to investigate whether CDK directly phosphorylates BARD1, we need to investigate whether BARD1 can become a substrate of CDK in vitro.
It was indicated that BARD1 is phosphorylated at the NH2 terminal side by CDK1 and CDK2 in an independent mode from the heterodimer formation of BRCA1-BARD1. Four phosphorylation sites were identified to be S148/S251/S288/T299 at the NH2 terminal side of BARD1. The mutants at these four sites remained associated with BRCA1 when co-expressed with either CDK2-cyclin E1 or CDK1-cyclin B1, indicating little change in molecular weight. These four sites were determined to be phosphorylation sites. The aforementioned mutant of BARD1 remained to have a sensitivity for the inhibition of ubiquitin ligase activity by CDK2, and when CDK1 phosphorylated BARD1 at the same sites as CDK2, BRCA1-BARD1 ubiquitin ligase activity was not inhibited so that phosphorylation itself appeared to be not so important in the downregulation of ubiquitin ligase. The coexpression of BARD1 and CDK1-cyclin B induced phosphorylation of BARD1 in vivo, but no effect on binding with BARD1 and BRCA1, which suggested that the function of BRCA1-BARD1 is controlled respectively by the cell cycle kinase at different phases during the cell cycle.
Due to phosphorylation of BARD1 by the aforementioned CDK2-cyclin E and CDK2-cyclin A, BRCA1 and BARD1 were dissociated and as a result, ubiquitin ligase activity was inactivated. The effect of CDK2-cyclin E and CDK2-cyclin A on BRCA1-BARD1 heterodimer specific ligase activity can be tested by measuring BRCA1 autoubiquitination. According to this analysis, CDK1-cyclin B did not inhibit autoubiquitination of BRCA1, whereas both CDK2-cyclin E and CDK2-cyclin A completely inhibited autoubiquitination. The point to be noted is that both CDK2-cyclin E and CDK2-cyclin A decompose BRCA1-BARD1.
The expression of cyclin E at the protein level showed remarkable variations in the sporadic breast cancer and the high degree of expression levels significantly correlated to malignant prognosis23.
According to the present invention, regarding how CDK2-cyclin E1 and A1 have an effect on the expression of protein of BRCA1 and BARD1, the steady state levels of BRCA1 and BARD1 in response to ubiquitin ligase activity decreased dramatically with CDK1-cyclin E1/A1 in a dose-dependent manner instead of with CDK1-cyclin B1. In addition, when testing the [35S]-methionine-labeled cells by the pulse chase method, if a BRCA1-BARD1 complex was co-expressed by CDK2-cyclin E1, the half life in the cells decreased significantly, which suggested that a decrease in expression of protein was due to proteolysis.
According to the present invention, the presence of high levels of cyclin E indicated a decrease in the BRCA1 ubiquitin ligase activity. Therefore, BRCA1 ubiquitin ligase activity was found to be involved in sporadic breast cancer in addition to familial breast cancer.
4. Transport of BRCA1 from Nucleus to Cytoplasm
Previously, BRCA1 and BARD1 were reported previously to be degraded due to transportation from nucleus to cytoplasm. When CDK2-cyclin E1 and BRCA1 were co-expressed and the protein levels of BRCA1 in the nucleus and cytoplasm were quantitatively analyzed, it was detected that the protein level of BRCA1 in cytoplasm was higher than the protein level in nucleus. Thus, in the presence of CDK2-cyclin E1, BRCA1 was assumed to be transported from nucleus to cytoplasm.
The present invention is explained specifically with reference to the embodiments. However, the present invention will not be limited by these embodiments for the scope of technology.
Through the present embodiments, the following specific methods were employed.
Mouse monoclonal antibodies to HA (12CA5, Boehringer, Mannheim), Myc (9E10, BabCo), Flag (M2, Sigma), polyubiquitin (Affiniti), α- and β-tubulin (DMIA+BMIB, Neomarkers) and NPM (Sigma-Aldrich or Zymed) as well as rabbit polyclonal antibodies to BRCA1 (Santa Cruz c-20) were purchased commercially. The rabbit polyclonal antibody made against the C-terminal of BARD1 was generated to the synthesized peptide CVMSFELLPLDS (SEQ ID No. 3) and affinity purified before use. Its specificity for immunofluorescence in cells was confirmed by using the competing peptide.
cDNA for full length human NPM (B23.1) (SEQ ID No. 1) was amplified by PCR from a HeLa cell cDNA library using Pfx polymerase (Stratagene). The cDNA for full length human NPM (B23.1) was subcloned into the mammalian expression pcDNA3 vector in frame with the N-terminal Flag tag or into the E. coli expression pET vector in frame with the N-terminal 6×His-Flag tag. Mammalian expression plasmids for BRCA1, BARD1 and HA-ubiquitin were prepared by the methods described in Hashizume, R. et al. The RING heterodimer BRCA1-BARD1 is ubiquitin ligase inactivated by a breast cancer-derived mutation. J. Biol. Chem. 276, 14537-14540 (2001), Nishikawa, H. et al. Mass spectrometric and mutational analyses reveal Lys-6-linked polyubiquitin chains catalyzed by BRCA1-BARD1 ubiquitin ligase. J. Biol. Chem. in press [online resource]1,2
The point mutations including that for a stop codon to make truncated mutants were generated by site-directed mutagenesis (Stratagene). All plasmids used were verified by DNA sequencing. Mammalian expression plasmids for cyclins and CDKs were prepared by the methods described26, 27 in Ohta, T., Xiong, Y., Phosphorylation- and Skp1-independent in vitro ubiquitination of E2F1 by multiple ROC-cullin ligases. Cancer res. 61, 1347-1353 (2001) and Maeda, I. Ohta, T. Koizumi, H. Fukuda, M. In vitro ubiquitination of cyclin D1 by ROC1-CUL1 and ROC1-CUL3. FEBS Lett. 494, 181-185 (2001).
Rabbit E1 (Affiniti Research Products), bovine ubiquitin (Sigma) and Flag-ubiquitin (Sigma) were purchased commercially. His-UbcH5c, and truncated N-terminal fragment of His-BRCA1 and His-BARD1 were prepared by Hashizume et al. previously described1. Full length His-Flag-NPM was obtained by two-step purification using nickel agarose beads followed by anti-Flag cross-linked agarose beads. The purified proteins were stained with Coomassie Brilliant Blue, and their concentrations were determined through a comparison to protein standards using a densitometer (LAS 3000, Fuji film).
Cells were cultured in Dulbecco's Modified Eagle's Medium supplemented with 10% fetal calf serum (293T, HeLa and COS7) or 10% newborn calf serum (3T3 Swiss) and 1% antibiotic-antimycotic agent (Life Technologies, Inc). Cells were expressed using the standard calcium phosphate precipitation method as previously described24 in; Ohta, T., Michel, J. J., Schottelius, A. J., & Xiong, Y. ROC1, a homolog of APC11, represents a family of cullin partners with an associated ubiquitin ligase activity. Mol. Cell. 3, 535-541 (1999). To analyze the effect of proteasome inhibitors, cells were treated with MG132 (20 μM), LLnL (20 μM) or the same volume of DMSO solvent (1 μl/ml culture) for 10 hours before harvesting. Cell cycle synchronization by thymidine-nocodazole was performed by the method previously described25 in; Ohta, T., Michel, J. J., & Xiong, Y. Association with cullin partners protects ROC proteins from proteasome-dependent degradation. Oncogene 18, 6758-6766 (1999).
The cells were then stained with propidium iodide and subjected to flow cytometry using FACSCalibur (Becton Dickinson).
Immunoprecipitation and immunoblotting methods, including the detection of in vivo ubiquitinated substrates, were performed by the method proposed by Hashizume et al. and Nishikawa et al.1,2 except that Flag peptide was used to extract Flag-NPM from anti-Flag antibody-conjugated agarose beads (M2, Sigma) according to the manufacturer's instructions.
The in vitro ubiquitination assay was performed by the method proposed by Hashizume et al. and Nishikawa et al.1,2 as mentioned above except that His-Flag-NPM (0.5 μg) was added to the reaction.
The in vitro kinase assay was performed as reported26 in; Ohta, T., & Xiong, Y. Phosphorylation- and Skp1-independent in vitro ubiquitination of E2F1 by multiple ROC-cullin ligases. Cancer Res. 61, 1347-1353 (2001). For dephosphorylation of immunoprecipitated HA-BARD1, HA-BARD1 immobilized beads were incubated with 2 U of calf intestinal alkaline phosphatase (Takara) at 37° C. for 30 minutes.
Using anti-CDK2 or anti-HA immunocomplex as a kinase as precipitated from the 293T cells expressing CDK2 cyclin E1 or HA-CDK1 cyclin B1, a kinase assay was performed for the substrate protein (2 μg). For preparation of nucleus fractions and cytoplasm fractions, the cells were dissolved at 4° C. for 5 minutes in a buffer A [10 mM/L MgCl2, 100 mM/L NaCl, 0.5% Triton X-100, 2.5 mM/L MgCl2, 10 mM/L NaF, and a protenase inhibitor] and passed through three times with a 26-gauge syringe.
Cytoplasm supernatant fractions were obtained after centrifugal separation at 4,000×g for 2 minutes. In obtain nucleus fractions from the remaining pellets, ultrasonic waves were applied to the buffer B for extraction [50 mM/L Tris-HCl (pH 7.5), 150 mM/L NaCl, 0.5% NP-40, 50 mM/L NaF, 1 mM 1/L DTT, 1 mM/L NaVO3, and a protenase inhibitor].
After the 293 cells were labeled with [35S]-methionine for 1 hour for pulse labeling, pulse-chase analysis was monitored.
The ubiquitination reaction was carried out using Flag-ubiquitin, E1, His-UbcH5c and anti-Myc immunocomplex immobilized on protein A agarose beads precipitated from 293T cells expressing Myc-BRCA1 (1-772) and BARD1 by the same procedures as in the method by Nishikawa et al2. The supernatant from ten reactions (30 μl each) was collected and incubated with a 30 μl volume of anti-Flag antibody-cross linked beads (Sigma) to precipitate proteins conjugated with Flag-ubiquitin. The proteins conjugated with Flag-ubiquitin were eluted off the beads in 30 μl of 25 mM ammonium bicarbonate containing 0.1 mg/ml of Flag peptide and digested with 7.4 μg/ml of trypsin for 20 hours at 30° C.
Alternatively, proteins interacting with the complex of HA-BARD1(1-408) and either wild-type or mutant (I26A) Flag-BRCA1(1-222) were immunoprecipitated and extracted as described above, separated by SDS-PAGE, and stained with Sypro Ruby (Molecular Probe) according to the manufacturer's instructions. The target bands were excised from the gel and digested with trypsin using the In Gel Digest Kit (Millipore) according to the manufacturer's instructions.
The peptide fragments were analyzed by LC/MS/MS according to the method by Nishikawa et al. as described above2. The acquired collision-induced dissociation spectra were analyzed by Mascot software.
Proliferating cells were fixed with 3% formalin for 15 minutes and permeated through 0.2% Triton X-100 for 5 minutes. Cells were washed with phosphate-buffered saline (PBS), blocked with 0.5% BSA in PBS, and stained with indicated antibodies: anti-NPM antibody, anti-α/β tubulin antibody, anti-BARD1 antibody and anti-BRCA1 antibody.
For detecting both BRCA1 and NPM colocalized at spindle poles, cells were fixed with cold methanol and permeated through 0.1% Triton X-100 buffer as reported15 in Hsu, L. C, & White R L. BRCA1 is associated with the centrosome during mitosis. Proc. Natl. Acad. Sci. U.S.A. 95, 12983-12988 (1998). Primary antibodies were diluted in the blocking buffer at the following concentrations: anti-NPM (0.5 μg/ml); anti-α/β-tubulin (1 μg/ml); anti-BARD1 (3 μg/ml); and anti-BRCA1 (2 μg/ml). FITC or Rhodamine-conjugated secondary antibodies (Jackson Immunoresearch) were used at 1:50 dilution. The nucleus was counterstained with 0.5 μM TO-PRO-3 (Molecular probe), cells were then mounted with fluorescent mounting medium (BioLad) and examined with a confocal laser scan microscope (LSM 510, Carl Zeiss).
First, after hypothesizing that ubiquitin ligase immunocomplexes may contain substrates, we performed the following test. A protein conjugated with Flag-ubiquitin was prepared from the Myc-BRCA1 (1-722)-BARD1 immunocomplexes mentioned above. A reaction supernatant was analyzed by immunoblotting using an anti-Flag antibody (Left panel in
Second, presuming that substrate proteins would associate with a ubiquitin ligase transiently and that this association could be stabilized by the loss of ligase activity, we compared the proteins associated with wild-type BRCA1-BARD1 ligase and catalytically inactive BRCA1(I26A)-BARD1 ligase. Anti-Flag immunocomplex precipitated from the 293T cells expressing one of HA-BARD1 (1-408), wild type Flag-BRCA1 (1-222), or I26A mutant was separated by SDS-PAGE and stained with Sypro Ruby. The protein shifting to near 38-40 kDa (arrow) was digested to be provided to LC/MS/MS. The I26A mutation in BRCA1 retains the BARD1 association site, but it disrupts the association with UbcH5c resulting in inactivation of the catalytic activity of the BRCA1-BARD1 heterodimer enzyme9. A protein of approximately 38-40 kDa was present only in the mutant ligase complex at a higher level and was identified by LC/MS/MS as NPM (
The association between BRCA1-BARD1 and NPM in vivo was confirmed using cells expressed transiently by immunoprecipitation (IP)-Western analysis (
Myc-BRCA1(1-772), HA-BARD1 and Flag-NPM plasmid were combined and expressed simultaneously in the 293T cells. Total cellular solution (2 panels above
The 293T cell solution underwent immunoprecipitation with an antibody to BARD1, and was analyzed by Western blotting. As a result, BARD1 (
To determine whether NPM is a substrate of the BRCA1-BARD1 ubiquitin ligase, Flag-NPM was co-expressed in 293T cells with HA-tagged ubiquitin, Myc-BRCA1 (1-772) and BARD1 (
Another RING type E3 ligase, MDM2, is known to effectively accelerate ubiquitination of a cancer suppressing gene p53 that is a known substrate10.
In a parallel in vivo ligase assay, in vivo ubiquitinated Myc-p53 and Flag-NPM were detected as mentioned above. (
To eliminate the possibility that NPM ubiquitination was caused by misfolding due to ectopic overexpression, we examined whether endogenous NPM is also ubiquitinated by BRCA1-BARD1 (
We further tested whether BRCA1-BARD1 ubiquitinates NPM in a in vitro system using a recombinant protein alone (
As mentioned above, the present inventors and others recently discovered that BRCA1-BARD1 catalyzes untraditional Lys-6-linked polyubiquitin chain formations which are deubiquitinated by purified 26S proteasome in vitro instead of being targeted for proteolysis2, 3. Therefore, we tested whether BRCA1-BARD1-mediated NPM ubiquitination targets NPM for proteolysis by measuring its stability in vivo (
Subsequently, 293T cells expressing Flag-NPM, pcDNA3 and Myc-BRCA1/BARD1 were incubated with cyclohexamide (10 μM), and the results were monitored at 4, 8, and 12 hours. The amount of intracellular expression of Flag-NPM was analyzed by immunoblotting using anti-Flag antibody. Pulse-chase analysis also supported that BRCA1-BARD1 stabilizes NPM (
Furthermore, from the cells treated with one of the proteasome inhibitors MG132 (20 μM) or LLnL (20 μM) or DMSO solvent for 10 hours, in vivo BRCA1-BARD1-mediated ubiquitinated Flag-NPM was detected as described in Embodiment 2. However, the amount of ubiquitinated Flag-NPM did not increase. (
A rabbit polyclonal antibody specific to the C-terminal of BARD1 was prepared and used to examine its intracellular localization (subcellular) and its coexistence with NPM during the cell cycle (
Proliferating Swiss 3T3 cells (a) or COS7 cells (b, c) were fixed respectively with 3% formalin or cold methanol. These cells were stained with anti-NPM antibody, anti-α/β tubulin antibody, anti-BARD1 antibody, and anti-BRCA1 and subsequently with FITC (green) or Rhodamin (red) conjugated secondary antibodies. The nucleus was stained with TO-PRO-3 (blue).
The results revealed that in the interphase cells, BARD1 is localized within nucleoli other than nucleolus, while NPM is localized in the nucleolus, but both did not coexist (
Furthermore to determine whether mitotic association of BRCA1-BARD1 and NPM is correlated with cell cycle-dependent NPM ubiquitination, we synchronized HeLa cells at the G2/M boundary by a thymidine-nocodazole treatment, subsequent reentry to the cell cycle after 0.5, 1, 2, and 4 hours, followed by a release into a G1 phase (
We next attempted to determine the changes in the molecular weight of BARD1 due to the co-expression of CDK/cyclin in order to investigate whether BARD1 is phosphorylated by CDK.
Namely, NH2 terminal (1-320) fragments or COOH terminal (411-777) fragments of Myc-BRCA1-772 and HA-BARD1 were co-expressed with CDK-cyclin or pcDNA3 vector in 293T cells. When HA-BARD1411-777 (COOH terminal fragments 411-777 of BARD1) was co-expressed with CDK/cyclin, no changes were found. However, when HA-BARD11-320 (NH2 terminal fragments 1-320 of BARD1) was co-expressed with CDK/cyclin, a moderate movement of proteins were observed in the gel (
As a result, both His-BARD14-180 and HA-BARD11-320 were phosphorylated in vitro with CDK2-cyclin E1 and CDK1-cyclinc B1 (
As in the results of in vivo movement shift of HA-BARD11-320, in vitro phosphorylated HA-BARD11-320 indicated several bands and demonstrated one or more phosphorylation sites in the HA-BARD1-320 fragments. This means that the ability of in vitro phosphorylation of purified BARD1 is not particularly needed for BRCA1 to have interactions with the said kinase. The fact that the co-expression of BRCA1 does not have any effect on the amount of phosphorylation of BARD1 supports this concept. The inventors performed mapping of four phosphrylation sites in BARD1 by mutation analysis and identified mutants of BARD1S148A/S251 A/S288A/T299A.
These showed almost no changes in molecular weight when using CDK2-cyclin E1 (
The inventors attempted to understand the molecular mechanism that may be the cause for the NPM ubiquitination by BRCA1-BARD1 only during the mitosis phase. It has been reported that NPM is phosphorylated by CDK2-cyclin E at the G1/S transition phase resulting in dissociation of NPM from centrosomes to allow daughter centrioles to divide7, 8. NPM is prevented from associating with the centrosomes during S phase while daughter centrioles mature into centrosomes and, subsequently, into two spindle poles. Since NPM is regulated by phosphorylation of CDK2-cyclin during centrosome replication, we investigated the possibility that CDK2 may affect NPM ubiquitination by the BRCA1-BARD1 ligase. Myc-BRCA1 (1-772), BARD1, Flag-NPM and HA-tagged ubiquitin were expressed in 293 T cells. The in vivo ubiquitinated products by BRCA1-BARD1 were detected as in Embodiment 2. CDK-cyclin was also re-expressed. To our surprise, both CDK2-cyclin E and CDK2-cyclin A completely inhibited NPM ubiquitination in vivo (
In order to distinguish these two models, the inventors tested whether BRCA1-BARD1 ligase can catalyze ubiquitination of NPM (T199A) which is a mutant having no CDK2 phosphorylation sites8 (
Therefore, it was suggested that an inhibition of NPM ubiquitination by CDK2 is not attributed to phosphorylation of NPM.
Next, the inventors tested the effect of CDK2-cyclin E and CDK2-cyclin A on the intrinsic ligase activity of the BRCA1-BARD1 heterodimeric complex by measuring BRCA1 autoubiquitination. The procedures were the same as those described in Embodiment 6. An anti-Myc antibody was used in the immunoprecipitation for detecting Myc-BRCA1, which was autoubiquitinated with the co-expression of CDK-cyclin. In contrast to the fact that CDK1-cyclin B did not inhibit the autoubiquitination of BRCA1, both CDK2-cyclin E and CDK2-cyclin A completely inhibited autoubiquitination of BRCA1 (
Since BARD1 is the substrate of CDK2-cyclin, inhibition of the BRCA1-BARD1 ubiquitin ligase activity may have a direct influence on phosphorylation of BARD1. However, BARD1 is a non-phosphorylation mutant so that autoubiquitination of BRCA1 via BARD1 S148A, S288A/T299 can be inhibited by CDK2-cyclin E1 (
The aforementioned non-phosphorylation mutant of BARD1 remains to have a sensitivity to the inhibition of ubiquitin ligase activity by CDK2. Although CDK1 phosphorylates BARD1 at the same site as that for CDK2, BRCA1-BARD1 ligase activity is not inhibited so that it appears that phosphorylation itself is not so important in the downregulation of the BRCA1-BARD1 ligase.
The inventors investigated how CDK2 influences the levels of expression of proteins of BRCA1 and BARD1. Namely, Myc-BRCA11-772 and HA-BARD1 were expressed in 293T cells and followed by immunoblotting using an anti-HA antibody then subsequently an anti-Myc antibody as a re-probe.
In response to the ubiquitin ligase activity, the steady state of both levels in BRCA1 and BARD1 severely decreased by CDK2-cyclin E1/A1 (
The 293T cells expressing Myc-BRCA11-772, HA-BARD1 and parent pc DNA3 vector (upper column) or CDK2 cyclin E1 (lower column) were supplemented with [35S] for performing a pulse chase test. Subsequently, the cellular solution was treated with an anti-Myc antibody for immunoprecipitation and separated by SDS•PAGE. The results were analyzed by autoradiography. The colocalization of CDK2-cyclin E1 noticeably reduced the half life of the BRCA1-BARD1 complex in the cells (up until 2 hours or less between 4 and 6 o'clock:
BRCA1 and BARD1 were reported to be decomposed during the transportation from the nucleus to cytoplasm. The inventors investigated how CDK2-cyclin E1 has an effect on localization with BRCA1 when both CDK2-cyclin E1 and BRCA1 were co-expressed. Flag-BRCA1, HA-BARD1, and parent pcDNA3 vector or CDK2cyclin E1 were transiently expressed on 293T cells. In order to prevent the degradation of cytoplasmic BRCA1, carbobenzoxyl-leucinyl-leucinyl-leucinal (MG132), which is a proteasome inhibitor, was added to the cells 6 hours prior to the time for recovering cells. For quantitative analysis of the BRCA1 protein levels in the nucleus and cytoplasm, cells were fractionated into a nucleus (N) pool and a cytoplasm (C) pool and immunoblotting was performed. When CDK2-cyclin E1 was not present, the cytoplasmic BRCA1 protein level was much lower than the protein level found in the nucleus (
The following experiment was carried out in order to investigate how BARD1 is expressed in the cell cycle. HeLa cells were synchronized by performing a double thymidine inhibition (A) to the G1/S transition phase. Also, the cells were synchronized to the mitotic phase by a performing thymidine-nocodazole inhibition (B). After releasing the inhibition, the courses of the cell cycle were monitored by FACS analysis. The cells were also analyzed by immunoblotting using antibodies at an appropriate time after releasing inhibition. The results demonstrated that the level of expression of BARD1 was highest during the mitotic phase, showing a mild movement band (
They shifted to dominant rapid bands (
The following references are incorporated in the entire specification as references.
SEQ ID No. 3: Synthetic peptide
| Number | Date | Country | |
|---|---|---|---|
| 60541287 | Feb 2004 | US |
| Number | Date | Country | |
|---|---|---|---|
| Parent | 10588124 | Nov 2006 | US |
| Child | 12453723 | US |
