BAG3 interacts with p53 in endometrial carcinoma

Dear Editor,

BAG3 (74 kDa) belongs to a family of proteins sharing a BAG domain, which interacts with the heat shock protein (HSP) 70 and regulates its activity. BAG3 can also bind to other proteins through its WW domain, proline-rich (PXXP) region and IPV (Ile-Pro-Val) motif. Due to its interactions with several partner proteins, BAG3 has been found to regulate diverse cellular processes, including autophagy, mechanotransduction, cytoskeleton organization, cell motility, cell cycle progression and apoptosis [1]. BAG3 is also constitutively expressed in some tumors, where it contributes to sustaining cell survival and proliferation [1, 2]. BAG3 has been found to be expressed in human endometrial carcinomas [3]. A putative trait of these tumors is a lack of p53 activity, due to mutations and/or allelic loss of its gene or to increased expression of its negative regulators, such as MDM2 [4].

To investigate the involvement of BAG3 in endometrial carcinoma development, we first assessed the expression of its protein in human primary benign endometrial hyperplasias (BEH), atypical endometrial hyperplasias (AEH) and AEHs associated with endometrial cancer (EC). In addition, we assessed the interaction of BAG3 with p53 in the human endometrial carcinoma cell line Ishikawa [5]. We used immunohistochemistry (IHC) to assess BAG3 expression in removed uterine polyps from 94 patients (age range: 26–84 years; mean age: 54.7 years) with diagnoses of: benign endometrial hyperplasia (BEH, 48 patients), atypical endometrial hyperplasia (AEH, 33 patients) and AEH associated with endometrial cancer (13 patients). We scored the samples based on the level of intensity and extent of staining, using a 3-point scale (0 = no visible staining; 1 = low positivity; 2 = high positivity). BAG3 expression could not be detected in BEH (mean IHC score: 0.00), while it was detectable in AEH, either associated or not associated with endometrial cancer. In AEH not associated with endometrial cancer, we detected a high BAG3 expression in 2 cases and a low expression in the remaining samples. Finally, in AEH associated with endometrial cancer, 100% of the samples were BAG3 positive, and the majority of them (9 out of 13 samples) showed high BAG3 levels (Fig. 1). These findings indicate that BAG3 protein appears in atypical hyperplasia and that its level increases during progression to cancer.

Fig. 1.

BAG3 expression in human endometrial hyperplasias. Patients who underwent hysteroscopy were enrolled in this study at the University Hospital Azienda Universitaria S. Giovanni e Ruggi d’Aragona of Salerno, Italy. Inclusion criteria were: blood loss post-menopause, BMI > 25, any personal medical history and/or a family history of breast cancer, and/or uterus and annexes cancers, any personal medical history of hypertension and/or diabetes, endometrial thickness greater than 5 mm measured by transvaginal ultrasound. All the patients were subjected to uterine biopsy during hysteroscopy. After histological confirmation and evaluation of compliance with the inclusion criteria, uterine polyps were collected, labeled with numeric codes for privacy, and further processed and analyzed. Formalin-fixed and paraffin-embedded tissues were subjected to IHC analysis for BAG3 expression. IHC was performed using a BenchMark XT immunostainer (Ventana MedicalSystems, Tucson, AZ, USA) and Ventana reagents. To detect BAG3 expression, samples were incubated with a murine anti-BAG3 monoclonal antibody (BIOUNIVERSA srl), 3 μg/ml. The signal was enhanced using a Ventana amplification kit. The slides were counterstained with hematoxylin, dehydrated and mounted. Representative images of AEH-EC (a), AEH (b) and BEH (c) (magnification: 20x) are shown. d Comparison between groups was performed using the Kruskal-Wallis test for one-way ANOVA and post- hoc Dunn’s test. P-values from 0.01 to 0.05, from 0.001 to 0.01, or < 0.001 were considered significant, very significant or highly significant, respectively

To investigate BAG3 interaction with p53, we immunoprecipitated BAG3 from Ishikawa cell lysate using an anti-BAG3 murine monoclonal antibody. In doing so, we revealed the presence of both BAG3 and p53 in the immunoprecipitate (Fig. 2a). Therefore, we conclude that BAG3 binds to p53. Next, we analyzed the influence of BAG3 on p53 localization. For this purpose, we measured p53 levels in the cytosol and in the nucleus of Ishikawa cells transfected with a bag3 siRNA or a control, non-targeted siRNA. We found that the nuclear levels of p53 were increased in bag3-silenced cells compared to control cells (Fig. 2b). We, therefore, conclude that the BAG3 protein retains p53 in the cytosol and that its down-modulation allows p53 translocation to the nucleus in these cells.

Fig. 2.

BAG3 interacts with p53 in a human endometrial adenocarcinoma cells. a BAG3 was immunoprecipitated from Ishikawa cell lysate. In brief, a murine anti-BAG3 mAb (BIOUNIVERSA srl) was coupled to Dynabeads™ using a Co-Immunoprecipitation Kit (14321D, Invitrogen) following the manufacturer’s instructions. About 500 μg of total lysate was immunoprecipitated 1 h at 4 °C and then analyzed by Western blotting. Anti-BAG3 HRP-conjugated polyclonal antibody (BIOUNIVERSA srl), anti-p53 antibody (DO-1)-HRP (sc-126 HRP, Santa Cruz Biotechnology), anti-Lamin A/C antibody (sab4200236, Sigma-Aldrich) and anti-Calregulin antibody (sc-11,398, Santa Cruz Biotechnology) were used at 1:1000 dilutions in TBST containing 5% nonfat dry milk overnight at 4 °C. Immunoreactivity was detected by sequential incubation with horseradish peroxidase-conjugated secondary antibodies (used at 1:5000 dilution) and ECL detection reagents (Pierce™ ECL Western Blotting Substrate, Thermo Scientific). b Ishikawa cells were transfected with a bag3 siRNA (5′-AUCGAAGAGUAUUUGACCAAA-3′) or a non-targeted (NT) siRNA (5′-CAGUCGCGUUUGCGACUGG-3′) at a final concentration of 200 nmol/L using TransIT-X2® (Mirus Bio, Medison, USA), for 72 h. Next, the cells were harvested, lysed and fractionated into cytosolic and nuclear fractions using a NE-PER Nuclear and Cytoplasmic Extraction Kit (Thermo Fisher) following the manufacturer’s instructions. Finally, both cytosolic and nuclear fractions were analyzed by Western blotting with the indicated antibodies.

We thus report a novel mechanism by which BAG3 can participate in tumorigenesis by interacting with the tumor suppressor p53. p53 is known to play an important role in regulating cell survival and proliferation. Upon DNA damage, p53 induces the expression of the cyclin-dependent kinase (CDK) inhibitor p21, resulting in cell cycle arrest in the G1 phase and allowing DNA repair before replication at S1. Upon extensive DNA damage, p53 induces apoptosis by activating signal molecules such as BAX, PUMA, Noxa and PERP. p53 can be sequestered by MDM2, that in response to DNA damage releases it, resulting in p53 activation [6]. Here we show that BAG3 can form a complex with p53 and, by doing so, prevent its translocation to the nucleus, thereby impeding p53-mediated induction of genes that induce cell cycle arrest and/or apoptosis. BAG3 has been reported to regulate cell survival and proliferation through more than one mechanism, depending on the cellular context. In melanoma and osteosarcoma cells, BAG3 protects the IKK subunit gamma from HSP70- mediated delivery to the proteasome. This results in increased NF-κB activity and survival. In thyroid carcinoma cells, a similar activity is exerted by BAG3 on BRAF, whose levels are sustained by interaction with BAG3. In tumor-associated neo-angiogenesis, endothelial cell proliferation is maintained by BAG3 binding to ERK kinase and its phosphatase DUSP6. When BAG3 is removed, DUSP6 binding to the kinase is reduced, and unaffected ERK phosphorylation results, in turn, in increased p21 and p15 levels and concomitant cell cycle arrest in the G1 phase. In glioblastoma cells, Bax is retained through BAG3 binding in the cytosol and fails to translocate to mitochondria, leading to suppression of apoptosis [1]. These results illustrate the pleiotropic activity of BAG3 in the regulation of cellular pathways [1]. Our current finding of BAG3-mediated regulation of p53 translocation to the nucleus adds a piece of information to the knowledge of mechanisms that govern the activity of this tumor suppressor, and may contribute to a better understanding of the biogenesis of endometrial cancer and the molecular steps underlying its neoplastic progression.

Authors’ contributions

Conceptualization: MDM, JT, AR, MG, LM; Methodology: ADA, RI, RM, MC, FS; Formal analysis and investigation: LG, PC; Writing - original draft preparation: MDM; Writing - review and editing: AR, MG, LM; Funding acquisition: JT, MG; Supervision: MG, LM.

Funding information

This work was supported in part by University of Salerno (FARB 2017) grants to MG and by Theoreo srl spin-off of the University of Salerno to JT.

Data availability

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Code availability

Not applicable.

Compliance with ethical standards

Conflict of interest

M.D.M., A.R. and L.M. are shareholders of BIOUNIVERSA s.r.l. that provided anti-BAG3 antibodies. The remaining authors declare no conflict of interest.

Ethical approval

The ethics committee of the University Hospital approved this study and informed consent was obtained from all patients.

Consent to participate

All the authors have seen and approved the manuscript being submitted.

Consent for publication

Not applicable.