Fig 2: FoxO3a modulates integrin α5 expression in BC. (a) WB analysis was conducted on 10 μg of total lysates from TamR and MCF-7 parental cells to detect indicated proteins. β-Actin was used as a loading control. A double set of cells was transfected in suspension with F3aAAA and pcDNA3.1 plasmids for 48 h. FoxO3a and integrin α5 expression was assessed at the (b) RNA and (c) protein level. mRNA content was normalized vs. the relative 18S rRNA content, and GAPDH was employed for loading controls. (d) 30 μg of total proteins extracted from TamR/TetOn-AAA-derived tumors, explanted from Dox-treated mice, and from TamR/TetOn-V control tumors [9] were used to evaluate FoxO3a and integrin α5 expression. β-Actin was used as a loading control. (e) TamR were seeded and, the day after, they were starved for 24 h, pre-treated with alpelisib 1 μM for 30′ and then treated with IGF 10 ng/mL for 15′. Cytosolic proteins were used to detect phosphorylated proteins. β-actin was employed as a loading control. (f) TamR were seeded and treated as in (e). Cytosolic and nuclear extracts were subjected to WB analysis for the detection of indicated proteins. Lamin B1 and GAPDH were used as loading controls. (g) TamR were seeded, and the day after, they were starved for 8 h. Alpelisib 1 μM was added for 18 h, followed by IGF-1 treatment for 15′. Total lysates were used for integrin alpha5 detection. β-actin was used as a loading control. (h) WB analysis on total proteins extracted from MCF-7 and TamR cells subjected to FoxO3a silencing (siF3a). FoxO3a and integrin α5 expressions were evaluated. β-actin was served as a loading control. Results represent the mean ± s.d. of at least three independent experiments (* p ≤ 0.05, *** p ≤ 0.001). (i) mRNA transcripts analysis through cBioPortal large-scale was used to investigate FoxO3a (FOXO3) and integrin α5 (ITGA5) correlation in ER+ (left panel) and ER− (right panel) breast tumors. Spearman and Pearson correlation coefficients with the respective p-value, regression line and R2 are reported.

Fig 3: Dissociation of FOXO3a and 14-3-3 sensitizes C33A cells to combination treatment with BKM120 and HCQA. Cells were seeded in 96-well plates overnight and then treated with 1 μM BKM120, 20 μM HCQ, and 5 μM BV02, alone or in combination, for 72 h. Cell viability was measured by the MTS-based assay. Columns, means of six replicate determinations; bars ± SD; *P < 0.01. B. Cells were transiently transfected with 14-3-3ζ siRNA and then treated with 1 μM BKM120 alone or in combination with 20 μM HCQ for 72 h. Columns, means of six replicate determinations; bars, ± SD; *P < 0.01. C. C33A cells were treated with 1 μM BKM120, 20 μM HCQ, and 5 μM BV02 alone or in combination for 72 h. Cell lysates were immunoblotted with the indicated antibodies. Actin was used as the loading control. D. Cells were transiently transfected with 14-3-3ζ siRNA and treated with 1 μM BKM120 alone or in combination with 20 μM HCQ for 72 h. Cell lysates were immunoblotted with the indicated antibodies. Actin was used as the loading control.

Fig 4: Cellular localization of FOXO3a upon BKM120 treatment plays an important role in autophagy induction in PIK3CA-mutant cancer cellsA. Cells were transfected with control siRNA or FOXO3a siRNA for 24 h and then treated with 1 μM BKM120 for an additional 24 h. The lysates were immunoblotted with designated antibodies. Actin was used as the loading control. LC3-II level was quantified by densitometry and normalized with Actin level. After normalization, LC3-II level was compared to that of non-treated control. B. Cells were cultured on coverslips to approximately 60% confluency and then treated with 1 μM BKM120 for an additional 24 h. Immunofluorescence staining was performed using an anti-FOXO3a antibody and the nuclei were stained with DAPI. Scale bar, 20 μm. C. Cells were treated with 1 μM BKM120 for 24 h, followed by lysis of whole-cell extracts and parallel subcellular fractionation. Lysates were immunoblotted with the indicated antibodies. α-tubulin and H3 were used as cytosolic and nuclear markers, respectively. Relative intensity was quantified by densitometry and normalized to the loading controls. The percentage of cytoplasmic and nuclear FOXO3a was determined as follows: cytoplasmic FOXO3a density/(cytoplasmic FOXO3a density + nuclear FOXO3a density) × 100 and nuclear FOXO3a density/(cytoplasmic FOXO3a density + nuclear FOXO3a density) × 100, respectively.

Fig 5: FoxO3a over-expression inhibits the growth of engineered TamR/TetOn derived tumor xenografts. (A) TamR/TetOn-V and TamR/TetOn-AAA cells were injected subcutaneously in female nude mice (see Materials and Methods). When the tumors reached average ~0.2 cm3, mice were treated with Doxycycline hyclate (Dox) for 35 days. Tumor growth was monitored by caliper, measuring the visible tumor sizes at indicated time points. Data represent the mean ± SD of measurements. * p < 0.05 vs. TamR/TetOn-Vector xenografts. (B) At the end of the experiment, tumors were explanted and representative tumor images are shown. The expressions of FoxO3a (C) and Ki-67 (E), as well as the apoptotic index (TUNEL assay) (F) were evaluated in tumor sections deriving from Dox-treated mice injected with engineered TamR/TetOn cells. NC: negative control. (D) An amount of 50 µg of lysates from explanted tumors were subjected to WB analysis to evaluate Dox-induced FoxO3a overexpression. β-actin was used as the loading control.