MYBBP1A suppresses breast cancer tumorigenesis by enhancing the p53 dependent anoikis

MCF-10A, human mammary epithelial cells, were maintained in Dulbecco’s modified Eagle’s medium-F12 (DMEM-F12) (Invitrogen, Carlsbad, CA) supplemented with 0.5 μg/ml hydrocortisone (Sigma-Aldrich, St Louis, MO), 10 μg/ml insulin (Sigma-Aldrich, St Louis, MO), and 20 ng/ml recombinant human EGF (Peprotech, Rocky Hill, NJ). MCF-7 human breast cancer cells were maintained in DMEM (Sigma-Aldrich, St Louis, MO). ZR-75-1 human breast cancer cells were maintained in RPMI 1640 (Nacalai Tesque, Kyoto, Japan). All media were supplemented with 10% fetal bovine serum (FBS) and 1% penicillin–streptomycin solution (Nacalai Tesque, Kyoto, Japan). Transfection was performed using Lipofectamine LTX (Invitrogen, Carlsbad, CA).

cDNAs encoding full-length p53 and MYBBP1A were amplified using PCR and subcloned into pcDNA3 plasmids (Invitrogen, Carlsbad, CA) and pQCXIP (Clontech, Mountain View, CA) containing sequences encoding FLAG sequences. Anti-β-Actin (Sigma-Aldrich, St Louis, MO) and anti-human-p53 (Santa Cruz, Santa Cruz, CA) monoclonal antibodies and rabbit anti-p53-K382Ac(Cell Signaling Technology, Danvers, MA) polyclonal antibody were used according to the manufacturers’ instructions. Rabbit anti-human MYBBP1A antibody was raised against a synthetic peptide corresponding to amino acids 1265–1328 of human MYBBP1A.

The Oncomine database and gene microarray analysis tool, a repository for published complementary DNA microarray data [18, 19], were explored (July 2012) for MYBBP1A mRNA expression in non-neoplastic and breast cancer tissues. Statistical analysis of the differences in MYBBP1A expression between these tissues used Oncomine algorithms, which provided multiple comparisons among different studies [20‐22]. Data sets obtained from TCGA Breast, Finak Breast, and Richardson Breast 2 included various stage, and all cancer samples were invasive.

Human breast cancer tissue microarrays were purchased from SuperBioChips Laboratories (Seoul, Korea) included various stages, and all cancer samples were invasive. Formalin-fixed tissues were dewaxed in xylene and rehydrated in alcohol. For antigen retrieval, the sections were subsequently heated in EDTA buffer (1 mM, pH 8.0) in a microwave oven for 5 min. Endogenous peroxidase activity was suppressed using a solution of 3% hydrogen peroxide in methanol for 6 min. The samples were stained with avidin–biotin–peroxidase complexes using a Histofine SAB-PO Immunohistochemical Staining Kit (Nichirei, Tokyo, Japan) according to the manufacturer’s instructions. Rabbit anti-MYBBP1A antibody was used at a dilution of 1:50.

Methods for stable RNA interference followed those described by Kajiro et al.[23]. To generate a shRNA retroviral supernatant, GP2-293 cells (Clontech, Mountain View, CA) were co-transfected with a p10A1 vector encoding an envelope protein and pRETRO-SUPER (Oligo Engine, Seattle, WA) vector containing either the MYBBP1A or luciferase (control) target sequence. MCF-7 cells were incubated with the retroviral supernatant in the presence of 8 μg/ml polybrene. Infected cells were selected using 1 μg/ml of puromycin. The target sequences were 5^′- TTCAGGGCATATTTCATCTCGGACC-3^′ for MYBB1A #1, 5^′-TTCAGGGCATATTTCATCTCGGACC-3^′ for MYBBP1A #2, and 5^′-GAAGCTGCGCGGTGGTGTTGT-3^′ for luciferase. For siRNA transfection, cells at 30%–50% confluency were transfected using Lipofectamine RNAi MAX (Invitrogen, Carlsbad, CA) according to the manufacturer’s protocol. All siRNAs were purchased from Invitrogen. The siRNA duplexes were MYBBP1A, 5^′-UCUUUCAGUCAGGUCGGCUGGUGAA-3^′ and p53, 5^′-CCAGUGGUAAUCUACUGGGACGGAA-3^′. Stealth RNAi negative control was used as a negative control.

All animal experiments were performed in accordance with institutional guidelines. The tumor xenograft models have been described previously [23]. Each mouse was subcutaneously injected with 100 μl of cell suspension (5 × 10⁶) in both flanks. At the time points indicated in the figures, the tumors were excised, weighed, and fixed or stored in liquid nitrogen.

For soft agar assays, 22,000 cells were mixed with 1 ml of 0.35% top agar and plated onto a 35-mm six-well plate containing bottom plugs (0.6% agar, 10% FBS, 1× DMEM). After the top agar had solidified (about 2 h at 37°C), 200 μl of DMEM was added into each well to prevent dehydration. This covering medium was changed every 2 or 3 days during culture. After allowing growth for 2 weeks at 37°C, colonies with a diameter of >100 μm were counted.

MCF-10A cells were plated at a density of 3 × 10⁵ cells per well in six-well plates in an ultra-low attachment culture dish (Corning, Tewksbury, MA) and maintained in DMEM-F12 (Invitrogen, Carlsbad, CA) supplemented with 10% FBS, 0.5 μg/ml hydrocortisone (Sigma-Aldrich, St Louis, MO), 10 μg/ml insulin (Sigma-Aldrich, St Louis, MO), and 20 ng/ml recombinant human EGF (Peprotech, Rocky Hill, NJ). MCF-7 cells were plated at a density of 3 × 10⁵ cells per well in DMEM with 5% FBS. ZR-75-1 cells were plated at a density of 3 × 10⁵ cells per well in RPMI 1640 with 10% FBS. All media were supplemented with 1% penicillin–streptomycin solution (Nacalai Tesque, Kyoto, Japan). Before analysis, the samples were treated with trypsin to disperse the cells. Cell viability after detachment was determined by trypan blue dye exclusion. Viable cells were determined by MTT assay using an MTT cell counting kit (Nacalai Tesque, Kyoto, Japan). Apoptotic cell proportion was determined by fluorescence-activated cell sorting (FACS) analysis using PI and fluorescein isothiocyanate (FITC)-conjugated Annexin V (MBL, Tokyo, Japan) staining.

Real-time RT-PCR was performed as described previously [24]. Cells were homogenized in 1 ml Isogen (Nippon Gene, Tokyo, Japan), and the total RNA was extracted according to the instruction manual. cDNA was synthesized from total RNA using ReverTra Ace reverse transcriptase (Toyobo, Osaka, Japan) and oligo dT primers. Real-time PCR was used to amplify fragments representing the indicated mRNA expressions. The primer sequences used were as follows:

GAPDH fw primer: 5^′- GTATGACTCCACTCACGGCAAA -3^′

GAPDH rv primer: 5^′-GGTCTCGCTCCTGGAAGATG-3^′

p21 fw primer: 5^′-GGAGACTCTCAGGGTCGAAA-3^′

p21 rv primer: 5^′-TTAGGGCTTCCTCTTGGAGA-3^′

Bax fw primer: 5^′-AGCAAACTGGTGCTCAAGG -3^′

Bax rw primer: 5^′-CTTGGATCCAGCCCAACAG -3^′

PUMA fw primer: 5^′-GGGCCCAGACTGTGAATCCT-3^′

PUMA rv primer: 5^′-ACGTGCTCTCTCTAAACCTATGCA-3^′

Cells were lysed in TNE buffer [10 mM Tris–HCl (pH 7.8), 1% Nonidet P-40 (NP-40), 0.15 M NaCl, 1 mM ethylenediaminetetraacetic acid (EDTA), 1 M phenylmethylsulfonyl fluoride (PMSF), 1 g/ml aprotinin]. The extracted proteins were immunoprecipitated with antibody-coated protein G Sepharose beads (GE Healthcare Japan, Tokyo, Japan). Bound proteins were separated by SDS–PAGE, transferred to polyvinylidene difluoride membranes (Millipore, Temecula, CA), and detected with appropriate primary antibodies and horseradish peroxidase-conjugated secondary antibodies. Specific proteins were visualized using an enhanced chemiluminescence (ECL) western blot detection system (GE Healthcare Japan, Tokyo, Japan).

ChIP assay was performed according to the published procedures [24]. The primers for real-time PCR were as follows: forward, TAATCCCAGCGCTTTGGAAG; reverse, TTGCTAGATCCAGGTCTCTGCA for the upstream region of the Bax gene.

Cells were fixed in 3.7% formaldehyde in PBS for 10 min. After rinsing twice with PBS, the cells were permeabilized in 0.1% Triton X-100 in PBS and later blocked with TBS-T buffer containing 0.5% bovine serum albumin and 10% goat serum for 1 h at room temperature. Subsequently, the cells were incubated with anti-MYBBP1A and anti-p53 antibodies for 1 h, stained with Alexa Fluor-conjugated secondary antibodies (Invitrogen, Carlsbad, CA) for 1 h, and mounted with Vectashield (Vector Laboratories, Burlingame, CA). Immunofluorescent images were obtained by Biozero immunofluorescence microscopy (Keyence, Osaka, Japan).

We previously reported that MYBBBP1A was involved in activating p53 function. Therefore, we assumed that MYBBP1A would have a cancer prevention function via p53. To examine the relationship between MYBBP1A expression and breast cancer progression, we examined the MYBBP1A expression profiles in breast carcinomas compared to those of normal tissue using the Oncomine database, which provides publicly available datasets of gene expression in cancer. Of the 12 datasets, 11 contained gene chip profiles classified as normal or breast carcinoma tissues, which indicated that MYBBP1A mRNA levels were significantly lower in breast carcinomas than in normal tissues. Three representative results from independent datasets characterized by large population sizes are shown in Figure 1 (MYBBP1A expression levels in normal vs. breast carcinoma; P = 2.95E-25, 2.10E-6, and 7.17E-4). Next, we used IHC to assess MYBBP1A expression in non-neoplastic and breast cancer tissues using a human breast cancer tissue microarray. As shown in Figure 1B, MYBBP1A expression was significantly decreased in the breast cancer tumors. To confirm these results, we compared MYBBP1A protein levels in MCF-10A, MCF-7, and ZR-75-1 cells. The MYBBP1A protein levels were much lower in MCF-7 and ZR-75-1 cells, the cancer cell lines derived from human breast cancer cells, than in MCF-10A cells, a normal breast tissue cell line (Figure 1C). These results suggested that MYBBP1A had an inhibitory effect on carcinogenesis in these cancer patients.

To investigate a possible relationship between MYBBP1A and breast cancer cell growth, we generated MCF-7 cells that had stably knocked down or overexpressed MYBBP1A (Figure 2A and 2B). MCF-7 is a breast cancer cell line that expresses wild type p53. As shown in Figure 2C, the colony numbers in soft agar were markedly increased when using MYBBP1A knockdown cells (shMYBBP1A #1 and shMYBBP1A sh#2). Conversely, MYBBP1A overexpression decreased the number of these colonies (Figure 2C: OE MYBBP1A).

Next, we performed xenograft experiments to test the effect of MYBBP1A expression on tumorigenicity in vivo. MYBBP1A knockdown cells formed tumors that were significantly larger than those of control cells. In contrast, MYBBP1A overexpressing cells formed smaller tumors than control cells (Figure 2D, 2E, and 2F). These results indicated that MYBBP1A could suppress breast cancer tumor growth.

To examine how MYBBP1A could suppress tumor formation we focused on anoikis, because p53 plays a critical role in anoikis [7‐12]. Thus, we examined whether MYBBP1A was involved in anoikis in the context of the p53 pathway. MCF-10A mammary epithelial cells were transfected with sip53 or siMYBBP1A and cultured in non-adherent plates for 24 h. The number of viable cells under detached conditions increased with p53 or MYBBP1A knockdown (Figure 3A). MCF-7 and ZR-75-1 breast cancer cells were also cultured under detached conditions. The number of viable cells increased with p53 knockdown, while they decreased when p53 was overexpressed. Similarly, MYBBP1A knockdown increased and MYBBP1A overexpression decreased the numbers of viable cells under detached conditions (Figure 3B and 3C). These results indicated that MYBBP1A was involved in anoikis in breast tissues.

To confirm that MYBBP1A was involved in anoikis in context of p53 activation, we tested the combination of p53 knockdown and MYBBP1A overexpression in anoikis assay using MCF-7 cells (Figure 4A). Based on previous results (Figure 3B), MYBBP1A overexpression decreased the number of viable cells (compare lanes 1 and 2 in Figure 4A). However, MYBBP1A overexpression in p53 knocked-down cells did not show any significant effects (compare lanes 3 and 4 in Figure 4A). Similar results were obtained in ZR-75-1 cells (Figure 4B).

To further examine the role of MYBBP1A in anoikis, we examined cellular apoptosis as determined by Annexin V–FITC/PI staining, followed by flow cytometric analysis. In accordance with the results shown in Figures 3 and 4, apoptosis decreased after p53 and MYBBP1A knockdown and increased when these genes were overexpressed (Figure 5A). Moreover, MYBBP1A overexpression in p53 knocked-down cells did not result in any significant effects (Figure 5B). These results indicate that MYBBP1A regulates p53-dependent anoikis.

Next, the effects of MYBBP1A knockdown on the induction of p53-target genes were examined. The mRNA levels of Bax, PUMA, and p21 were increased under detached conditions, whereas the increases in these mRNA levels were suppressed when MYBBP1A was knocked down using siRNA in MCF-7 cells (Figure 6A). Similar results were obtained with ZR-75-1 cells (Figure 6B). These results suggest that MYBBP1A regulates p53-dependent anoikis by enhancing p53 activation.

The acetylation levels of p53 are increased in response to stress and correlate well with p53 activation and stabilization [26‐28]. Accumulating evidence supports the conclusion that acetylation stabilizes p53 and is indispensable for p53 activation [29, 30]. Therefore, to study the molecular mechanism by which MYBBP1A induced anoikis in a p53-dependent manner, we examined whether MYBBP1A was involved in the accumulation of p53 protein and the acetylation of p53 K382 under detached conditions. Immunoblotting revealed that detached conditions induced the accumulation and acetylation of p53 (Figure 7A lane 3). However, p53 accumulation and acetylation were not observed when MYBBP1A was knocked down using siRNA (Figure 7A lane 4). This suggested that MYBBP1A was required for p53 activation in anoikis.

Consistent with these results, p53 recruitment to the Bax promoter was significantly enhanced under detached conditions, while p53 recruitment was abrogated by MYBBP1A knockdown (Figure 7B).

A previous report showed that MYBBP1A activates p53 by facilitating its direct interaction with p53 in response to stress when MYBBP1A translocated from the nucleolus to the nucleoplasm [13]. Therefore, we examined the localization of MYBBP1A and the interaction between MYBBP1A and p53 under detached conditions. Immunostaining revealed that MYBBP1A translocated from the nucleolus to the nucleoplasm under detached conditions (Figure 7C). Moreover, co-immunoprecipitation showed that endogenous MYBBP1A was bound to p53 in MCF-7 cells under detached conditions (Figure 7D). These results indicated that, under detached conditions, MYBBP1A translocates from the nucleolus to the nucleoplasm, and then binds to p53. Thus, MPBBP1A enhances p53 target gene transcription.