|Cancer||Cell death||Cell cycle||Cytoskeleton||Exo/endocytosis||Differentiation||Division||Organelles||Signalling||Stem cells||Trafficking|
Cell Biology International (2008) 32, 13021309 (Printed in Great Britain)
Interaction of p14ARF with Brca1 in cancer cell lines and primary breast cancer
Lizhi Heab, Catherine Fanab, Xiaoming Ningc, Xinchang Fengab, Yun Liud, Biao Chenab and Damu Tangab*
aDivision of Nephrology, Department of Medicine, McMaster University, Hamilton, ON, Canada
bFather Sean O'Sullivan Research Institute, St. Joseph's Hospital, Hamilton, ON, Canada
cLaboratory of Pathology, Heilongjiang Tumor Hospital, Harbin Medical University, Heilongjiang, PR China
dLaboratory of Veterinary Surgery, College of Veterinary Medicine, Northeast Agricultural University, Heilongjiang, PR China
We report an association between p14ARF and Brca1 in which both proteins co-immunoprecipitate (co-IP) in DU145 cells. The N-terminal 64 residues of p14ARF encoded by exon 1β are sufficient for this association. Inside the cell, ectopic p14ARF co-localizes with ectopic and endogenous Brca1 in A375 cells. Endogenous p14ARF co-localizes with endogenous Brca1 in DU145 cells but not in H1299 cells. Since p14ARF interacts with B23 in the nucleolus, Brca1 co-localizes with B23 in DU145 but not in H1299 cells. While ectopic ARF potently inhibited DU145 cell proliferation, it had no effect on the proliferation of H1299 cells, suggesting that the interaction between ARF and Brca1 contributes to ARF-mediated tumor suppression. Consistent with this notion, ectopic p14ARF modulates endogenous Brca1 expression in MCF7 breast cancer cells and p14ARF co-localizes with Brca1 in normal breast epithelial cells. This co-localization is enhanced in primary breast cancer. Taken together, the results show that p14ARF associates with Brca1, which may play a major role in tumor suppression.
Keywords: p14ARF, Brca1, Prostate cancer, Breast cancer.
*Corresponding author. T3310, St. Joseph's Hospital, 50 Charlton Ave East, Hamilton, ON, L8N 4A6 Canada. Tel.: +1 905 522 1155x35168; fax: +1 905 540 6549.
The INK4A/ARF locus encodestwo tumor suppressors, p16INK4A and ARF (p14ARF in humans and p19ARF in murine) by alternative splicing and use of different reading frames (Quelle et al., 1995; Scott et al., 1998). ARF functions in tumor surveillance through both p53-dependent and -independent pathways (Sherr, 2001). In primary fibroblasts, expression of oncogenic Ras activates p53 in an ARF-dependent manner (Serrano et al., 1997; Palmero et al., 1998). Expression of oncogenes such as c-Myc, E1A, E2F1, v-Abl and β-catenin in normal cells also leads to p53 activation through ARF (Zindy et al., 1998; de Stanchina et al., 1998; Dimri et al., 2000; Cong et al., 1999; Damalas et al., 2001). Furthermore, transgenic mice heterozygous for p19ARF engineered to express c-Myc in B-cells exhibit accelerated B-cell lymphomas; 80% of such tumors had lost the wild type p19ARF allele (Eischen et al., 1999), lending in vivo support to the notion that ARF plays an essential role in p53 activation by hyperproliferation signals.
Mechanistically, p14ARF binds directly to Mdm2 in the nucleoplasm, resulting in p53 activation (Llanos et al., 2001; Kashuba et al., 2003; Korgaonkar et al., 2005). Mdm2 interacts with the N-terminal region of p53, which results in direct inhibition of the transcriptional activity of p53 (Momand et al., 1992), suppression of the p300/CBP-mediated p53 acetylation that stabilizes p53 (Ito et al., 2001), ligation of ubiquitins on p53 (Honda et al., 1997), and export of p53 from the nucleus into cytoplasmic proteasomes for degradation (Roth et al., 1998). ARF interacts directly with Mdm2, thereby inhibiting the ubiquitin ligase activity of Mdm2 in vitro (Honda and Yasuda, 1999) and abolishing the inhibitory activity of Mdm2 on p53 acetylation (Ito et al., 2001).
p14ARF also functions independently of p53. It predominantly resides in the nucleolus, where it interacts with Topo I and B23, leading to enhancement of the relaxation activity of Topo I and the inhibition of B23-mediated rRNA processing by prompting B23 degradation (Itahana et al., 2003; Bertwistle et al., 2004). These results suggest a scenario where nucleolar ARF contributes to its observed p53-independent function (Sherr, 2001).
Breast cancer-associate gene 1 (Brca1) functions not only in DNA damage response but also in the G2/M checkpoint control (Narod and Foulkes, 2004). Brca1 regulates key G2-M transition factors, including Cdk1, Wee1 kinase, and Cdc25C phosphatase (Yarden et al., 2002), and initiates G2/M cell cycle arrest in an ERK1/2 kinase-dependent manner (Yan et al., 2005). Furthermore, Aurora-A kinase promotes mitosis by inhibiting Brca1 function (Sankaran et al., 2007).
The fact that ARF also negatively impacts cell cycle progression (Sherr, 2001) suggests a functional connection between Brca1 and ARF. In line with this concept, Brca1 has been shown to stabilize p53 only in p14ARF positive but not in p14ARF negative cells (Somasundaram et al., 1999). We report here that p14ARF associates with Brca1 in several human cancer cell lines (DU145 prostate cancer and A375 melanoma cells) and in primary human breast cancer.
2 Materials and methods
2.1 Materials, cell culture, and plasmids
DU145 (human prostate cancer), H1299 (human lung carcinoma), MCF7 (human breast cancer), A375 (human melanoma) and 293T (human embryonic) cells were purchased from ATCC. H1299 cells were cultured in RPMI1640, 10% FCS (fetal calf serum). MCF7, A375, and 293T cells were cultured in DMEM, 10% FCS. DU145 cells were cultured in MEM, 10% FCS. All the cells expressed endogenous Brca1. While DU145, H1299 (Fig. 2A, also see Scott et al., 1998), and 293T (our unpublished observation) cells contain endogenous p14ARF, A375 and MCF7 cells do not express p14ARF (Scott et al., 1998).
Human p14ARF cDNA was cloned from C33A cervical carcinoma cells by RT-PCR amplification and expressed as a C-terminal FLAG tagged protein using pcDNA3.1 and pBabe (a retroviral vector) vectors. p14ARF mutants containing the N-terminal 64 residues (ARF-N) and the C-terminal residues (65–132) (ARF-C) were generated by PCR-based mutagenesis (Tang et al., 2002).
2.2 Retroviral infection
Retroviral infection was performed following our previously published procedure (Li et al., 2004; Tang et al., 2001). Briefly, a gag-pol expressing vector and an envelope expressing vector (VSV-G) (Stratagene) were transiently co-transfected with a retroviral vector into 293T cells. After 48
2.3 Cell lysis and Western blot
Cell lysate was prepared and Western blot performed according to our published procedure (Narod and Foulkes, 2004). Primary antibodies and concentrations used were: anti-FLAG (M2 at 3
Cell lysates containing 200
Double immunofluorescent staining was carried out by fixing cells with prechilled (−20
For tissue dual-immunofluorescence (IF) staining, tissues were deparaffinized, rehydrated, and subjected to antigen-retrieval and endogenous peroxidase-quenching. Tissue sections were then blocked for 1
2.6 Collection of primary breast cancer tissues
Breast cancer specimens were collected at Heilongjiang Tumor Hospital, Harbin Medical University, Harbin, Heilongjiang, China following the Institute’s regulations. Cancers were examined and graded by the leading pathologist (Dr Xiaoming Ning) of the hospital. Patient information is presented in Table 1.
Clinical characteristics of breast cancer patients
3.1 Interaction between p14ARF and Brca1
Brca1 activates p53 in part through the induction of p14ARF (Somasundaram et al., 1999), suggesting a functional connection between p14ARF and Brca1. To investigate this connection, we examined whether both proteins are physically associated. When co-expressed in 293T cells, Brca1 was co-immunoprecipitated (co-IP) with p14ARF (Fig. 1A), demonstrating a physical association between p14ARF and Brca1. Attempts to co-IP p14ARF via Brca1 were unsuccessful (data not shown), which may be caused by possible interference of the anti-Brca1 antibody with the interaction between Brca1 and ARF. p14ARF consists of the N-terminal fragment (aa 1–64) encoded by exon 1β and the carboxy fragment encoded by exon 2 (aa 65–132) (Scott et al., 1998). While the N-terminal fragment mediates the association of p14ARF with Mdm2, E2F1, p120E4F, and B23 (Zhang et al., 1998; Mason et al., 2002; Rizos et al., 2003; Itahana et al., 2003), the carboxy fragment (aa 665–132) binds to Topo I (Ayrault et al., 2003). To map the Brca1 binding regions, we generated ARF-N (aa 1–64) and ARF-C (aa 65–132), and determined their association with Brca1 (Fig. 1A). ARF-N, but not ARF-C, was clearly co-IP with Brca1 (Fig. 1A), revealing that residues 1–64 are sufficient to mediate the interaction between p14ARF and Brca1. To consolidate the observed interaction, we demonstrated that IP of endogenous p14ARF co-precipitated endogenous Brca1 in DU145 (prostate cancer) cells (Fig. 1B). Taken together, the above shows that p14ARF binds to Brca1.
p14ARF interacts with Brca1 via the N-terminal fragment encoded by exon 1β (aa 1–64). (A) FLAG-tagged p14ARF, the N-terminal p14ARF fragment (aa 1–64, ARF-N), or C-terminal fragment (aa 65–132, ARF-C) was individually co-expressed with Brca1 in 293T cells as indicated for 48
To investigate the association in detail, we checked whether p14ARF co-localizes with Brca1 in the cell. When ectopically expressed in A375 human melanoma cells and 293T cells, p14ARF, ARF-N, but not ARF-C co-localized with endogenous Brca1 (data not shown). Furthermore, endogenous p14ARF also co-localized with endogenous Brca1 in DU145 (Fig. 2A). This co-localization was not observed in H1299 cells (Fig. 2A), although H1299 cells express Brca1 and high levels of p14ARF (data not shown).
Co-localization of p14ARF with Brca1. (A,B) DU145 and H1299 cells were double immunofluorescently stained for p14ARF (green) and Brca1 (red) and for Brca1 (red) and B23 (green). Nuclei were counter-stained with DAPI (blue). Images were taken using a fluorescent microscope (Carl Zeiss, Axiovert 200).
Since p14ARF interacts with B23 in the nucleolus (Bertwistle et al., 2004; Zhou and Elledge, 2000; Ayrault et al., 2003), we suspected that Brca1 may co-localize with B23 in DU145 cells. Indeed, double immunofluorescent staining showed a limited level of co-localization between Brca1 and B23 in DU145 (Fig. 2B). The co-localization or proximity of Brca1 with B23 may be caused via its interaction with p14ARF, as Brca1 does not co-localize with B23 in H1299 cells (Fig. 2B). Taken together, the findings support the concept that p14ARF interacts with Brca1.
3.2 Brca1 specifically associates with p14ARF in cells sensitive to ARF-mediated growth inhibition
The fact that ARF co-localizes with Brca1 in DU145 but not in H1299 cells prompted us to examine whether this interaction contributes to ARF-mediated growth inhibition in these cells. We infected DU145 and H1299 cells with either an empty vector (pBabe) or p14ARF retrovirus (Fig. 3A) and selected the infections with puromycin. While p14ARF potently reduced the number of colonies formed in DU145 cells, it had no effect on the number of colonies formed in H1299 cells (Fig. 3B,C), which demonstrates that the reduction in DU145 cell colonies imposed by p14ARF retrovirus did not result from a possible lower titer of p14ARF retrovirus than that of pBabe retrovirus. Since DU145 and H1299 lines are p53 null functional (Torosyan et al., 2006; Li et al., 2006), the differences in their sensitivity to p14ARF-mediated growth inhibition are thus p53-independent and may be attributable to the observed co-localization between p14ARF and Brca1 in DU145, but not in H1299 cells (Fig. 2A).
Ectopic p14ARF inhibits DU145 but not H1299 cell proliferation. DU145 and H1299 cells were infected with an empty vector (−) or p14ARF retrovirus (+). The expression of ectopic ARF was detected by Western blot (A). The infection was selected with puromycin and surviving cells were stained with 0.1% Crystal Violet (B) and the numbers of surviving colonies were graphed (C).
To confirm this concept, we determined whether reduction in Brca1 function compromised p14ARF-mediated growth inhibition, using a Brca1 mutant cell line, breast cancer HCC1937 (Kitagawa et al., 2004; Rodriguez et al., 2004). These cells, however, cannot sustain transient DNA transfection using either plasmid or retrovirus, which may be caused by its chromosome instability due to loss of Brca1 function (data not shown). We proceeded to knockdown endogenous Brca1 using specific Brca1 siRNA (Lou et al., 2005) in MCF7, DU145, and 293T cells. Knockdown of Brca1 in MCF7 cells substantially inhibited their proliferation, which makes the subsequent detection of ARF-mediated growth inhibition impossible (data not shown). For the rest of the cell lines, Brca1 siRNA did not attenuate ectopic p14ARF-mediated growth inhibition, when compared to control siRNA treated cells (data not shown). While these observations may not support the notion that interaction with Brca1 makes a major contribution to p14ARF-mediated inhibition of cell proliferation, these observations do not exclude this possibility. This is because the knockdown of Brca1 may induce chromosome instability, which will initiate the cellular DNA damage response, leading to sensitization of cells to p14ARF action (see Section
3.3 p14ARF modulates Brca1 expression in MCF7 breast cancer cells
Since Brca1 plays an important role in preventing breast cancer tumorigenesis, we examined the relationship between p14ARF and Brca1 in MCF7 cells. Unexpectedly, ectopic p14ARF induced a substantial reduction in Brca1 expression in MCF7 cells, when compared to empty vector (pBabe) infected cells (Fig. 4A). This was not caused by retrovirus insertion which may alter gene expression, as transfection of MCF7 cells with pcDNA3.1/p14ARF also reduced Brca1 expression (Fig. 4B, compare Brca1 staining in ectopic p14ARF positive versus negative cells). When 200 p14ARF positive cells were counted from several randomly selected fields, >80% of p14ARF positive MCF7 cells expressed reduced Brca1 (data not shown). This seems specific for MCF7 cells, as it does not occur in A375 cells (data not shown), H1299 and DU145 cells (Fig. 4C). The reduction of Brca1 in p14ARF expressing cells makes it difficult to detect a co-localization between ectopic p14ARF and endogenous Brca1. Nonetheless, the observed effect of p14ARF on Brca1 expression in MCF7 cells supports the notion that p14ARF interacts with Brca1 in breast cancer cells.
Ectopic p14ARF reduces Brca1 expression in MCF7 breast cancer cells. (A) MCF7 cells were infected with an empty retrovirus (pBabe) or p14ARF retrovirus. Infections were selected for 2
3.4 Co-localization of p14ARF with Brca1 in primary breast cancer
To determine the interaction between p14ARF and Brca1 in breast cancer, co-localization between p14ARF and Brca1 in primary human breast cancers was studied. Primary breast cancer and the paired non-cancer breast tissues were obtained from 20 patients who have underwent mastectomy to remove their breast cancers in the Heilongjiang Tumor Hospital, Harbin Medical University, Heilongjiang, PR China (Table 1). Samples were prepared to include tumor and its respective non-tumor breast tissue mounted on the same slide. Two-color immunofluorescence staining showed that p14ARF co-localizes with Brca1 in certain regions in both normal breast epithelial cells and carcinoma (Fig. 5A, arrows), while control IgG did not detect any signals (data not shown). To quantify this co-localization, we randomly counted 500 cells from each normal and the paired cancer tissue, and found that 24% of normal breast epithelial cells and 33% of breast cancer cells displayed co-localization of p14ARF and Brca1 (Fig. 5B). A two-tailed t-test showed that the increases in co-localization in breast cancer tissues over normal epithelial cells are statistically significant (P
Co-localization of p14ARF and Brca1 in primary breast cancer. (A) Normal breast epithelial cells and carcinoma cells, which were confirmed by H&E staining (data not shown), were immunofluorescently stained for ARF (red) and Brca1 (green). Nuclei were stained with DAPI (blue). The merged images of the ARF and Brca1 channels in the presence of DAPI staining are shown. Arrows indicate cells that show the co-localization between Brca1 and p14ARF. (B) Five hundred cells from each normal breast and carcinoma tissue were counted. The percentages of cells positive for the association between ARF and Brca1 are graphed.
From a variety of angles it is clear that p14ARF binds to Brca1 in cell lines, including A375 and DU145 prostate cancer cells. In these lines, ectopic p14ARF reduces their proliferation (Fig. 3 and data not shown). On the other hand, ectopic p14ARF does not co-localize with Brca1 in H1299 lung carcinoma cells (Fig. 2A) and is also incapable of inhibiting their proliferation (Fig. 3) (Li et al., 2004). Although this indicates a possible contribution of Brca1 to p14ARF-mediated inhibition of cell proliferation, knockdown of Brca1 using siRNA did not affect p14ARF’s ability to reduce DU145 cell proliferation (data not shown). Since knockdown of Brca1 causes chromosome instability which may activate the cellular DNA damage response, this may sensitize p14ARF-mediated inhibition of cell proliferation. As such, knockdown of Brca1 may not necessarily compromise p14ARF-mediated inhibition of cell proliferation. Thus, the impact of Brca1 on p14ARF-mediated inhibition of cell proliferation requires further investigation.
We unexpectedly found that ectopic p14ARF reduces endogenous Brca1 expression, detected by both Western blot and immunofluorescent staining of MCF7 cells (Fig. 4A,B), which might be due to feedback mechanisms. Interaction with p14ARF may enhance Brca1 function, which activates potential feedback pathways, leading to the observed reduction of the Brca1 protein. However, this certainly is not the only explanation. While ectopic p14ARF inhibits the proliferation of DU145 but not H1299 cells (Fig. 3), it does not reduce the endogenous Brca1 protein in these cells (Fig. 4C), making further investigation into the physiological significance of ARF-mediated Brca1 reduction necessary.
While the physiological relevance of the interaction between p14ARF and Brca1 in tumor suppression needs further investigation, the fact that p14ARF co-localizes with Brca1 in primary breast epithelial cells and breast carcinoma (Fig. 5) supports a role of this association in preventing breast cancer formation. We observed increases in the association between p14ARF and Brca1 in breast carcinoma in comparison with normal breast epithelial cells (Fig. 5B). Our patient population consists of one ductal carcinoma in situ (DCIS), which is an early lesion of a metastatic disease (van de Vijver, 2005), and 17 cases of ductal invasive carcinoma (Table 1). Increases in co-localization between p14ARF and Brca1 in breast carcinoma over normal breast epithelial cells in ductal invasive carcinoma indicate a role of this association in the surveillance of breast tumorigenesis at a later stage, consistent with the well-established involvement of p14ARF in tumor surveillance. However, these increases might be attributable to treatments such as radiotherapy and chemotherapy (Table 1).
Brca1 plays essential roles in DNA damage repair and cell cycle checkpoint control, which provides the mechanistic basis for Brca1 being a tumor suppressor. While germline mutations in Brca1 result in familial breast and ovarian cancer, somatic Brca1 mutations are not frequently detected in sporadic breast cancer or in other cancers (Narod and Foulkes, 2004). It is thus an intriguing possibility that disruption of its association with p14ARF may account for loss of Brca1 function in the majority of breast and other types of cancers.
This work was supported by a grant (Grant# 013009) from the National Cancer Institute of Canada to DT. We are very grateful to Dr Paul Harkin, The Queen’s University of Belfast, Northern Ireland, for providing the Brca1 plasmid.
Ayrault O, Karayan, L, Riou, JF, Larsen, CJ, Seite, P. Delineation of the domains required for physical and functional interaction of p14ARF with human topoisomerase I. Oncogene 2003:22:1945-54
Damalas A, Kahan, S, Shtutman, M, Ben-Ze'ev, A, Oren, M. Deregulated beta-catenin induces a p53- and ARF-dependent growth arrest and cooperates with Ras in transformation. EMBO J 2001:20:4912-22
Dimri GP, Itahana, K, Acosta, M, Campisi, J. Regulation of a senescence checkpoint response by the E2F1 transcription factor and p14(ARF) tumor suppressor. Mol Cell Biol 2000:20:273-85
Eischen CM, Weber, JD, Roussel, MF, Sherr, CJ, Cleveland, JL. Disruption of the ARF-Mdm2-p53 tumor suppressor pathway in Myc-induced lymphomagenesis. Genes Dev 1999:13:2658-69
Itahana K, Bhat, KP, Jin, A, Itahana, Y, Hawke, D, Kobayashi, R. Tumor suppressor ARF degrades B23, a nucleolar protein involved in ribosome biogenesis and cell proliferation. Mol Cell 2003:12:1151-64
Ito A, Lai, CH, Zhao, X, Saito, S, Hamilton, MH, Appella, E. p300/CBP-mediated p53 acetylation is commonly induced by p53-activating agents and inhibited by MDM2. EMBO J 2001:20:1331-40
Kashuba E, Mattsson, K, Klein, G, Szekely, L. p14ARF induces the relocation of HDM2 and p53 to extranucleolar sites that are targeted by PML bodies and proteasomes. Mol Cancer 2003:2:18
Korgaonkar C, Hagen, J, Tompkins, V, Frazier, AA, Allamargot, C, Quelle, FW. Nucleophosmin (B23) targets ARF to nucleoli and inhibits its function. Mol Cell Biol 2005:25:1258-71
Momand J, Zambetti, GP, Olson, DC, George, D, Levine, AJ. The mdm-2 oncogene product forms a complex with the p53 protein and inhibits p53-mediated transactivation. Cell 1992:69:1237-45
Quelle DE, Zindy, F, Ashmun, RA, Sherr, CJ. Alternative reading frames of the INK4a tumor suppressor gene encode two unrelated proteins capable of inducing cell cycle arrest. Cell 1995:3:993-1000
Rizos H, Diefenbach, E, Badhwar, P, Woodruff, S, Becker, TM, Rooney, RJ. Association of p14ARF with the p120E4F transcriptional repressor enhances cell cycle inhibition. J Biol Chem 2003:278:4981-9
Roth J, Dobbelstein, M, Freedman, DA, Shenk, T, Levine, AJ. Nucleo-cytoplasmic shuttling of the hdm2 oncoprotein regulates the levels of the p53 protein via a pathway used by the human immunodeficiency virus rev protein. EMBO J 1998:17:554-64
Sankaran S, Crone, DE, Palazzo, RE, Parvin, JD. Aurora-A kinase regulates breast cancer associated gene 1 inhibition of centrosome-dependent microtubule nucleation. Cancer Res 2007:67:11186-94
Scott FJ, Bates, S, James, MC, McConnell, BB, Starborg, M, Brookes, S. The alternative product from the human CDKN2A locus, p14(ARF), participates in a regulatory feedback loop with p53 and MDM2. EMBO J 1998:17:5001-14
Serrano M, Lin, AW, McCurrach, ME, Beach, D, Lowe, SW. Oncogenic ras provokes premature cell senescence associated with accumulation of p53 and p16INK4a. Cell 1997:88:593-602
Tang D, Wu, D, Hirao, A, Lahti, JM, Liu, L, Mazza, B. ERK activation mediates cell cycle arrest and apoptosis after DNA damage independently of p53. J Biol Chem 2002:277:12710-7
Torosyan Y, Dobi, A, Naga, S, Mezhevaya, K, Glasman, M, Norris, C. Distinct effects of annexin A7 and p53 on arachidonate lipoxygenation in prostate cancer cells involve 5-lipoxygenase transcription. Cancer Res 2006:66:9609-16
Zhang Y, Xiong, Y, Yarbrough, WG. ARF promotes MDM2 degradation and stabilizes p53: ARF-INK4a locus deletion impairs both the Rb and p53 tumor suppression pathways. Cell 1998:92:725-34
Zindy F, Eischen, CM, Randle, DH, Kamijo, T, Cleveland, JL, Sherr, CJ. Myc signaling via the ARF tumor suppressor regulates p53-dependent apoptosis and immortalization. Genes Dev 1998:12:2424-33
Received 9 March 2008/24 April 2008; accepted 15 July 2008doi:10.1016/j.cellbi.2008.07.018