BEX2 regulates autophagy by inhibiting PIK3CA-p85 interaction in non-small-cell lung cancer cells Authors: Yu Wang, Ning Mu, Xinke Wang, Jianhao Liu, Yiqing Qu, Xiangguo Liu, Ling Su Published: 2025 ABSTRACT BackgroundAutophagy is an evolutionarily conserved cellular process that maintains homeostasis. It enables tumor cells to survive and grow, which may lead to drug resistance. The PI3K/AKT/mTOR pathway is reported to play a key role in autophagy induction. BEX2 promotes cell growth and survival in cancer cells. However, its role in autophagy remains unclear.MethodsHuman non-small cell lung cancer (NSCLC) cell lines (A549, H1792 and H1299) and 293FT cells were used to investigate the function of BEX2 in autophagy. The experimental research was conducted via western blot, co-immunoprecipitation and confocal microscopy.ResultsOur data show that rapamycin induces BEX2 protein levels. Western blot and confocal microscopy analysis demonstrate that BEX2 regulates autophagy. Furthermore, BEX2 impairs PI3K/AKT/mTOR signaling. Specifically, BEX2 interacts with PIK3CA (PI3K catalytic subunit). BEX2 impairs the interaction of PIK3CA and p85 (PI3K regulatory subunit), therefore inhibiting PI3K activity. In summary, BEX2 regulates autophagy through the PI3K/AKT/mTOR signaling pathway by modulating the activity of PI3K, specifically dependent on the presence of PIK3CA.ConclusionsWe discovered that BEX2 promotes autophagic flux via PI3K/AKT/mTOR signaling. BEX2 interacts with PIK3CA and impairs PIK3CA and p85 interaction, which hinders activation of PI3K/AKT/mTOR signaling and promotes autophagy induction.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12964-025-02385-8. Introduction Autophagy is an evolutionarily conserved cellular process by which dysfunctional proteins or organelles are directed to autophagosomes for degradation via lysosomes to maintain homeostasis. Moreover, autophagy also serves as a dynamic recycling system that produces new materials and energy for cellular reformation and homeostasis [1, 2]. Originally, autophagy was regarded as having a suppressive effect on cancer development. However, increasing evidence has shown that once cancer is established, increased autophagic flux often enables tumor cells to survive and grow better, eventually leading to drug resistance. In response to chemotherapy, autophagy tends to be competent, not deficient [3]. BEX2 gene belongs to the brain expressed X-linked gene family (brain expressed X-linked gene 2). The BEX2 subfamily has four isoforms, and the longest isoform encodes a 160-amino acid protein [4]. Additionally, BEX2 participates in diverse cellular processes. BEX2 is regulated by NF-κB, ERBB2 signaling [5, 6]. However, BEX2 also stimulates NF-κB, c-jun/JNK pathway [6]. Previous studies have indicated that BEX2 may play a pro-oncogenic role. BEX2 inhibits apoptosis, promotes cell cycle and facilitates metastasis [7–10]. However, the role of BEX2 in autophagy remains poorly understood. PI3K consists of a catalytic and a regulatory subunit. Mammals express four class I catalytic isoforms (p110α, β, γ, and δ encoded by PIK3CA, PIK3CB, PIK3CG, and PIK3CD) that catalyze the phosphorylation of PtdIns-4,5-P2 to generate PtdIns-3,4,5-P3. PIK3CA associates with one of five different regulatory subunits (p85α, p55α, p50α, encoded by PIK3R1; p85β, PIK3R2; p55γ, PIK3R3) [11]. PIK3CA, which encodes the p110α, is frequently mutated in cancer [12–14]. The PI3K/AKT/mTOR pathway plays a key role in autophagy induction [15]. RHEB is a small GTPase that responds to various signals to activate mTOR [16–18]. mTOR negatively regulates autophagy through phosphorylation and inhibition of ULK1 [19]. In this study, we showed that BEX2 enhanced rapamycin-induced autophagy. Moreover, BEX2 promotes autophagy via the PI3K/AKT/mTOR pathway. Finally, we confirmed that BEX2 impairs the interaction of PIK3CA and p85 and therefore promotes autophagy. Our data suggest that BEX2 may be a therapeutic candidate for NSCLC treatment. Materials and methods Cell lines and cell culture A549, H1792, and H1299 cell lines were cultured in RPMI-1640 2.0 (1640, 2.0 g NaHCO3/L, Sigma-Aldrich) supplemented with 10% (v/v) fetal bovine serum (FBS, TOCYTO). HEK293FT cell lines were cultured in Dulbecco’s modified Eagle’s medium (DMEM, 3.7 g NaHCO3/L, Sigma-Aldrich) supplemented with 10% FBS. All cell lines were maintained at 37℃ in a humidified atmosphere consisting of 5% CO2 and tested to ensure no mycoplasma contamination. Reagents and antibodies The chemicals used in our experiments were Rapamycin (S1039) and Bafilomycin A1 (S1413) were purchased from Selleckchem. The primary antibodies used in the western blot and immunoprecipitation assays were as follows: anti-ACTB (A1978), and anti-MYC (C3956) were obtained from Sigma-Aldrich; anti-LC3B (2775), anti-PIK3CA (4249), anti-ATG5 (12994), anti-p-AKT (9271), anti-AKT (9272), anti-p-RPS6KB1 (9234), anti-RPS6KB1 (9202), anti-p-EIF4EBP1 (9451), anti-EIF4EBP1 (9452) were obtained from Cell Signaling Technology; anti-p85 (60225-1-Ig), anti-GST (66001-2-Ig) were obtained from Proteintech; anti-BEX2 (SC-398486), anti-PIK3CA (SC-293172) were obtained from Santa Cruz Biotechnology. Plasmids and siRNA transfection The cells were transfected with jetPRIME transfection reagent (PolyPlus-transfection) or LipoMax reagent (Sudgen Biotechnology) in serum-free Opti-MEM (Gibco) according to the instruction manual. All siRNAs were synthesized by GenePharma (Shanghai, China). BEX2#1 and BEX2#2 siRNA target sequences: 5’-GCCCTACCTTTGAATGTTA-3’ and 5’-GGAGCAAGTTGCTAATAAA-3’, respectively. ATG5 siRNA target sequences: 5’-CCTTTGGCCTAAGAAGAAA-3’. Human BEX2 and PIK3CA coding regions were amplified from cDNA and subcloned into the PEBG-GST or pcDNA3.1 vector. mCherry-GFP-LC3B plasmid was purchased from Sangon Biotech. Western blot and co-immunoprecipitation The cells were harvested and lysed in lysis buffer on ice for 30 min and then purified via centrifugation at 13,200 rpm for 15 min at 4 °C. Equivalent protein quantities were subjected to 10%−14% SDS-PAGE, transferred to PVDF membranes, and probed with primary antibodies, followed by the appropriate peroxidase-conjugated secondary antibodies (anti-mouse, AP124P; anti-rabbit, AP307P). A chemiluminescence kit (WBKLS0500) was used to visualize the immunoreactive bands. Images were taken using Amersham Imager 680 (GE). For co-immunoprecipitation, cells were transiently transfected with the indicated plasmids for a specified time. Cells were collected and lysed in IP lysis buffer for 30 min and then purified via centrifugation at 13,200 rpm for 15 min at 4 °C. FLAG M2 beads (A2220, Sigma-Aldrich), glutathione sepharose beads (16100, Invitrogen) were incubated with the clarified lysates at 4 °C for 6 h, or the clarified lysates were immunoprecipitated with a specific antibody and protein A-sepharose (101142, Invitrogen) or protein G-sepharose (101243, Invitrogen) for 4 h at 4 °C. The immune complexes were washed 3 times with IP lysis buffer, and the bead-conjugated proteins were denatured in 2×SDS loading buffer for 10 min at 100 °C. Samples were then separated by SDS-PAGE and detected by immunoblotting. Immunofluorescence microscopy Immunofluorescence microscopy assays were performed as previously described [20]. Briefly, GFP or GFP-mCherry expressing cells were fixed and permeabilized in PHEMO buffer (0.025 M HEPES, 0.068 M PIPES, 0.003 M MgCl2·6H2O, 0.015 M EGTA·Na2, 10% DMSO, pH = 6.8). Additional reagents were added before use, with final concentrations of 0.05% glutaraldehyde, 0.5% Triton X-100, and 3.7% formaldehyde, and then incubated at room temperature for 10 min. After being washed with PBS three times, cells were blocked in 3% BSA, and nuclei were stained with DAPI (Sigma-Aldrich) for 5 min. Finally, the cells were washed with PBS three times. After mounting, the cells were visualized using a confocal microscope (ZEISS, LSM 880, 63×oil). Analysis was performed using ImageJ software. Statistics GraphPad Prism software was used for the statistical analysis. All data are presented as the mean ± SD. Differences between groups were identified using a t-test and one-way ANOVA. P < 0.05 was considered statistically significant. Statistical significance was also taken as *P < 0.05, **P < 0.01, ***P < 0.001 and ****P < 0.0001. Results BEX2 controls rapamycin-induced autophagy in NSCLC cells Our previous work reported that crotonylated BEX2 plays an important role in inhibiting chemotherapeutic agent-induced apoptosis via enhancing mitophagy in human lung cancer cells [21]. To further characterize the potential role of BEX2 in the regulation of autophagy, we treated the NSCLC cell line H1299 with rapamycin, a popular autophagy inducer, at different concentrations and for different times. We observed that rapamycin induced the expression of BEX2 (Figure S1A and B), suggesting that BEX2 might play an important role in autophagy. Therefore, we monitored autophagy by confocal microscopy in A549 cells after transfection with mCherry-GFP-LC3B, which switches from yellow to red upon autolysosome formation. Overexpression of BEX2 significantly increased the number of red puncta in rapamycin-treated cells, suggesting activation of autophagy (Fig. 1A and B). Consistently, BEX2 overexpression further induced the accumulation of LC3B-II under rapamycin treatment in NSCLC cell lines A549 and H1792 (Fig. 1C and D). Accordingly, BEX2 depletion significantly reduced rapamycin-induced autophagy, as determined by autolysosome formation (Fig. 2A and B) and LC3B-II levels in NSCLC cell line H1299 (Fig. 2C and D). Collectively, these data suggested that BEX2 controls rapamycin-induced autophagy in NSCLC cells. Fig. 1BEX2 overexpression enhances rapamycin-induced autophagy in NSCLC cells. A The cells were transfected with mCherry-GFP-LC3B and NC or BEX2 plasmids for 48 h, then the cells were treated with rapamycin (10 µM) for 4–6 h. Autophagic flux was monitored by confocal microscopy. B The average number of autophagosomes per cell (yellow puncta in merged image) and autolysosomes per cell (red puncta in merged image) in A was counted. Scale bars: 10 μm. C A549 and H1792 cells were transiently transfected with NC and BEX2, and the cells were treated with rapamycin (10 µM) for 4 h. Cell lysates were analyzed by western blotting with the indicated antibodies. D The relative protein levels in C were further evaluated by densitometry analysis using ImageJ software and quantified for the ratio of LC3B-II: ACTB. Data in B and D are shown as the mean ± SD (n = 3), and statistical significance was assessed by t-test. **P < 0.01, ***P < 0.001 BEX2 overexpression enhances rapamycin-induced autophagy in NSCLC cells. A The cells were transfected with mCherry-GFP-LC3B and NC or BEX2 plasmids for 48 h, then the cells were treated with rapamycin (10 µM) for 4–6 h. Autophagic flux was monitored by confocal microscopy. B The average number of autophagosomes per cell (yellow puncta in merged image) and autolysosomes per cell (red puncta in merged image) in A was counted. Scale bars: 10 μm. C A549 and H1792 cells were transiently transfected with NC and BEX2, and the cells were treated with rapamycin (10 µM) for 4 h. Cell lysates were analyzed by western blotting with the indicated antibodies. D The relative protein levels in C were further evaluated by densitometry analysis using ImageJ software and quantified for the ratio of LC3B-II: ACTB. Data in B and D are shown as the mean ± SD (n = 3), and statistical significance was assessed by t-test. **P < 0.01, ***P < 0.001 Fig. 2BEX2 knockdown inhibits rapamycin-induced autophagy in NSCLC cells. A The cells were transfected with mCherry-GFP-LC3B and siCTRL or siBEX2 siRNAs for 48 h, then the cells were treated with rapamycin (10 µM) for 4–6 h. Autophagic flux was monitored by confocal microscopy. B The average number of autophagosomes per cell (yellow puncta in merged image) and autolysosomes per cell (red puncta in merged image) in A was counted. Scale bars: 10 μm. C H1299 cells were transfected with siRNA and then treated with rapamycin (10 µM) for 4 h. Cell lysates were analyzed by western blotting with the indicated antibodies. D The relative protein levels in C were further evaluated by densitometry analysis using ImageJ software and quantified for the ratio of LC3B-II: ACTB (D). All data are shown as the mean ± SD (n = 3), and statistical significance was assessed by t-test. **P < 0.01, ***P < 0.001 BEX2 knockdown inhibits rapamycin-induced autophagy in NSCLC cells. A The cells were transfected with mCherry-GFP-LC3B and siCTRL or siBEX2 siRNAs for 48 h, then the cells were treated with rapamycin (10 µM) for 4–6 h. Autophagic flux was monitored by confocal microscopy. B The average number of autophagosomes per cell (yellow puncta in merged image) and autolysosomes per cell (red puncta in merged image) in A was counted. Scale bars: 10 μm. C H1299 cells were transfected with siRNA and then treated with rapamycin (10 µM) for 4 h. Cell lysates were analyzed by western blotting with the indicated antibodies. D The relative protein levels in C were further evaluated by densitometry analysis using ImageJ software and quantified for the ratio of LC3B-II: ACTB (D). All data are shown as the mean ± SD (n = 3), and statistical significance was assessed by t-test. **P < 0.01, ***P < 0.001 BEX2 facilitates the complete autophagic flux The ATG16L1-ATG5-ATG12 complex plays an essential role in Autophagy [19]. Genetic inhibition of ATG5 blocked BEX2-enhanced accumulation of LC3B-II in NSCLC cell lines A549 and H1792 (Fig. 3A, B, C and D), which suggested that BEX2-enhanced autophagy requires integrated autophagic flux. We further used the lysosomal inhibitor bafilomycin A1 (BafA1) to inhibit lysosomal protein degradation. BEX2 overexpression further elevated LC3B-II levels upon bafilomycin A1 treatment in NSCLC cell line H1792 (Fig. 3E and F), indicating that BEX2 facilitates autophagy rather than inhibiting lysosomal protein degradation. In general, these data indicated that BEX2 facilitates the initiation of rapamycin-induced autophagy. Fig. 3BEX2 facilitates the complete autophagic flux. A A549 cells were transfected with BEX2 and ATG5 siRNA and then treated with rapamycin (10 µM) for 4 h. Cell lysates were analyzed by western blotting with the indicated antibodies. B The relative protein levels in A were further evaluated by densitometry analysis using ImageJ software and quantified for the ratio of LC3B-II: ACTB. C H1792 cells were transfected with BEX2 and ATG5 siRNA and then treated with rapamycin (10 µM) for 4 h. Cell lysates were analyzed by western blotting with the indicated antibodies. D The relative protein levels in C were further evaluated by densitometry analysis using ImageJ software and quantified for the ratio of LC3B-II: ACTB (D). E H1792 cells were transfected with NC or BEX2 and then incubated with bafilomycin A1 (BafA1, 20 nM) for 4 h. Cell lysates were analyzed by western blotting with the indicated antibodies. F The relative protein levels in E were further evaluated by densitometry analysis using ImageJ software and quantified for the ratio of LC3B-II: ACTB. Data in B and D are presented as the mean ± SD (n = 3 independent experiments), and statistical significance was assessed by one-way ANOVA. ***P < 0.001, **P < 0.01. Data in F are presented as the mean ± SD (n = 3 independent experiments), and statistical significance was assessed by t-test. ***P < 0.001 BEX2 facilitates the complete autophagic flux. A A549 cells were transfected with BEX2 and ATG5 siRNA and then treated with rapamycin (10 µM) for 4 h. Cell lysates were analyzed by western blotting with the indicated antibodies. B The relative protein levels in A were further evaluated by densitometry analysis using ImageJ software and quantified for the ratio of LC3B-II: ACTB. C H1792 cells were transfected with BEX2 and ATG5 siRNA and then treated with rapamycin (10 µM) for 4 h. Cell lysates were analyzed by western blotting with the indicated antibodies. D The relative protein levels in C were further evaluated by densitometry analysis using ImageJ software and quantified for the ratio of LC3B-II: ACTB (D). E H1792 cells were transfected with NC or BEX2 and then incubated with bafilomycin A1 (BafA1, 20 nM) for 4 h. Cell lysates were analyzed by western blotting with the indicated antibodies. F The relative protein levels in E were further evaluated by densitometry analysis using ImageJ software and quantified for the ratio of LC3B-II: ACTB. Data in B and D are presented as the mean ± SD (n = 3 independent experiments), and statistical significance was assessed by one-way ANOVA. ***P < 0.001, **P < 0.01. Data in F are presented as the mean ± SD (n = 3 independent experiments), and statistical significance was assessed by t-test. ***P < 0.001 BEX2 hampers the interaction between PIK3CA and p85 To determine the underlying mechanisms of BEX2-regulated autophagy, we searched for potential proteins that interact with BEX2 in the BioGRID database (https://thebiogrid.org/) (Fig. 4A). PIK3CA (p110α, catalytic subunit of class I PI3K) is a candidate protein. We first evaluated the interaction between BEX2 and PIK3CA in 293FT cells. Immunoprecipitation (IP) assays with either PIK3CA or BEX2 tags verified their interaction (Figure S2A and B). Our data also verified the interaction of endogenous BEX2 and PIK3CA in NSCLC cell line H1299 (Fig. 4B). Moreover, colocalization of BEX2 and PI3KCA was confirmed by immunohistochemistry (Fig. 4C). Fig. 4BEX2 hampers the interaction between PIK3CA and p85. A The interaction network of BEX2 was analyzed in the BioGRID database (https://thebiogrid.org/). B Co-IP assays were carried out with the PIK3CA antibody or IgG in H1299 cells, followed by western blotting using anti-BEX2 and anti-PIK3CA antibodies. C Immunofluorescence staining of endogenous PIK3CA and BEX2 was performed in H1299 cells. D H1792 cells were transiently expressing BEX2; co-IP assays were carried out with the PIK3CA antibody, and the co-eluted proteins were detected by western blot assays with p85, PIK3CA, and BEX2 antibodies. E The content of PIP3 in A549 cells treated with BEX2 overexpression was quantified by an ELISA kit. F The content of PIP3 in H1299 cells treated with BEX2 depletion was quantified by an ELISA kit. G Immunoblot of AKT, p-AKT(S473) and BEX2 in H1299 cells treated with BEX2 depletion in the presence of LY294002 (10 µM for 24 h) BEX2 hampers the interaction between PIK3CA and p85. A The interaction network of BEX2 was analyzed in the BioGRID database (https://thebiogrid.org/). B Co-IP assays were carried out with the PIK3CA antibody or IgG in H1299 cells, followed by western blotting using anti-BEX2 and anti-PIK3CA antibodies. C Immunofluorescence staining of endogenous PIK3CA and BEX2 was performed in H1299 cells. D H1792 cells were transiently expressing BEX2; co-IP assays were carried out with the PIK3CA antibody, and the co-eluted proteins were detected by western blot assays with p85, PIK3CA, and BEX2 antibodies. E The content of PIP3 in A549 cells treated with BEX2 overexpression was quantified by an ELISA kit. F The content of PIP3 in H1299 cells treated with BEX2 depletion was quantified by an ELISA kit. G Immunoblot of AKT, p-AKT(S473) and BEX2 in H1299 cells treated with BEX2 depletion in the presence of LY294002 (10 µM for 24 h) Because the interaction of p110 with p85, the standard regulatory subunit of class I PI3K [22], is critical for the stability and activation of PI3K, we next evaluated the effect of BEX2 on PIK3CA-p85 interaction. Co-immunoprecipitation (co-IP) assays showed that the overexpression of BEX2 impaired the interaction of PIK3CA and p85 (Fig. 4D and Figure S2C). We next estimated the effect of BEX2 on PI3K activity. phosphatidylinositol-3,4,5- triphosphate (PIP3) is directly regulated by PI3K. Elisa data showed that BEX2 knockdown upregulated PIP3 abundance, whereas BEX2 overexpression downregulated PIP3 level (Fig. 4E and F). Furthermore, LY294002, a PI3K inhibitor, reversed the enhancement of Akt phosphorylation in NSCLC cells by BEX2 depletion (Fig. 4G). These data demonstrated that BEX2 potentially inhibits PI3K activity through competitively binding to PIK3CA. BEX2 promotes autophagy via PI3K/AKT/mTOR signaling PI3K/AKT/mTOR signaling pathway negatively regulates autophagy initiation [23]. Therefore, we monitored downstream signaling molecules of PI3K and found that phosphorylation of AKT, RPS6KB1 and EIF4EBP1 was downregulated upon BEX2 overexpression in NSCLC cell lines A549 and H1792 (Fig. 5A, B, C and D). Fig. 5BEX2 promotes autophagy via PI3K/AKT/mTOR signaling. A A549 cells transfected with NC and BEX2 and cell lysates were analyzed by western blotting with the indicated antibodies. B H1792 cells transfected with NC and BEX2, and cell lysates were analyzed by western blotting with the indicated antibodies. C The relative protein levels in A were further evaluated by densitometry analysis using ImageJ software and quantified for the ratio of LC3B-II: ACTB. D The relative protein levels in B were further evaluated by densitometry analysis using ImageJ software and quantified for the ratio of LC3B-II: ACTB. E H1792 cells transfected with BEX2 and MYC-RHEBQ64L, and cell lysates were analyzed by western blotting with the indicated antibodies. F H1792-GFP-LC3B cells were expressed BEX2 and MYC-RHEBQ64L, and then autophagic puncta were monitored by confocal microscopy. Scale bars: 10 μm. G The average number of autophagic puncta per cell in F was counted. For C, D and G, data are presented as the mean ± SD (n = 3 independent experiments), and statistical significance was assessed by t-test. ***P < 0.001, **P < 0.01, ns, no significance BEX2 promotes autophagy via PI3K/AKT/mTOR signaling. A A549 cells transfected with NC and BEX2 and cell lysates were analyzed by western blotting with the indicated antibodies. B H1792 cells transfected with NC and BEX2, and cell lysates were analyzed by western blotting with the indicated antibodies. C The relative protein levels in A were further evaluated by densitometry analysis using ImageJ software and quantified for the ratio of LC3B-II: ACTB. D The relative protein levels in B were further evaluated by densitometry analysis using ImageJ software and quantified for the ratio of LC3B-II: ACTB. E H1792 cells transfected with BEX2 and MYC-RHEBQ64L, and cell lysates were analyzed by western blotting with the indicated antibodies. F H1792-GFP-LC3B cells were expressed BEX2 and MYC-RHEBQ64L, and then autophagic puncta were monitored by confocal microscopy. Scale bars: 10 μm. G The average number of autophagic puncta per cell in F was counted. For C, D and G, data are presented as the mean ± SD (n = 3 independent experiments), and statistical significance was assessed by t-test. ***P < 0.001, **P < 0.01, ns, no significance To further demonstrate whether mTOR signaling is necessary for BEX2-regulated autophagy. A constitutively active form of RHEBQ64L, which enhanced the phosphorylation of RPS6KB1 and EIF4EBP1 through activation of mTOR kinase activity [24], was transfected into the NSCLC cell line H1792. RHEBQ64L diminished BEX2-induced decline of RPS6KB1 and EIF4EBP1 phosphorylation level (Fig. 5E). Simultaneously, overexpression of RHEBQ64L eliminated the number of GFP-LC3B puncta induced by BEX2 overexpression (Fig. 5F and G). Collectively, these data suggested that mTOR signaling plays the dominant role in BEX2-regulated autophagy. BEX2 augments autophagy depending on PIK3CA To examine whether PIK3CA exerts indispensable effects in BEX2-regulated autophagy, we simultaneously interfere with or overexpress BEX2 and PIK3CA. We found that BEX2 knockdown significantly promoted AKT phosphorylation and reduced LC3B-II protein level in the control group, but not in the PIK3CA knockdown group (Fig. 6A). Similarly, BEX2 interference failed to reduce GFP-LC3B puncta in the PIK3CA knockdown group (Fig. 6B and C). Consistently, ectopic expression of BEX2 inhibited AKT phosphorylation, increased LC3B-II protein level and promoted GFP-LC3B puncta in the control group, but not in the PIK3CA knockdown group (Fig. 6D, E and F). Together, these results suggested that PIK3CA is indispensable in BEX2-regulated autophagy.Fig. 6BEX2 enhances rapamycin-induced autophagy through PIK3CA. A H1299 cells line were translated with BEX2 and PIK3CA siRNA, and cell lysates were analyzed by western blotting with the indicated antibodies. B H1299-GFP-LC3B cells were translated with BEX2 and PIK3CA siRNA, and autophagic puncta were monitored by confocal microscopy. C The average number of autophagic puncta per cell in B was counted. Scale bars: 10 μm. D H1792 cells were transfected with BEX2 and PIK3CA, cell lysates were analyzed by western blotting with the indicated antibodies. E H1792-GFP-LC3B cells were transfected with BEX2 and PIK3CA, and autophagic puncta were monitored by confocal microscopy. F The average number of autophagic puncta per cell in E was counted. Scale bars: 10 μm. For C and F, data are presented as the mean ± SD (n = 3 independent experiments), and statistical significance was assessed by t-test. ***P < 0.001, ns, no significance BEX2 enhances rapamycin-induced autophagy through PIK3CA. A H1299 cells line were translated with BEX2 and PIK3CA siRNA, and cell lysates were analyzed by western blotting with the indicated antibodies. B H1299-GFP-LC3B cells were translated with BEX2 and PIK3CA siRNA, and autophagic puncta were monitored by confocal microscopy. C The average number of autophagic puncta per cell in B was counted. Scale bars: 10 μm. D H1792 cells were transfected with BEX2 and PIK3CA, cell lysates were analyzed by western blotting with the indicated antibodies. E H1792-GFP-LC3B cells were transfected with BEX2 and PIK3CA, and autophagic puncta were monitored by confocal microscopy. F The average number of autophagic puncta per cell in E was counted. Scale bars: 10 μm. For C and F, data are presented as the mean ± SD (n = 3 independent experiments), and statistical significance was assessed by t-test. ***P < 0.001, ns, no significance Discussion Autophagy is a highly conserved intracellular degradation mechanism that plays a crucial role in cell survival and stress reactions, as well as in cancer development and metastasis. We observed that the expression of BEX2 was increased when autophagy was induced. Previous work revealed that BEX2 is regulated by NF-κB, ERBB2 signaling [5, 6]. We found that BEX2 mRNA increased at 2 h and then decreased to nearly a basal level at 4 h upon rapamycin treatment (data not shown). This implied that BEX2 is under rigorous regulation upon rapamycin treatment. However, the mechanism by which rapamycin regulates BEX2 expression still needs to be explored. BEX2 overexpression promoted, while BEX2 depletion inhibited rapamycin-induced autophagy. The autophagy process involves several steps, including sequestration, the fusion of autophagosomes with lysosomes, and the degradation of the substrates [25]. During the process, many key ATG molecules and important complexes are involved, such as the ATG12-ATG5-ATG16L1 complex, ULK1 complex, and PIK3C3-BECN1 complex [24]. ATG5 was knocked down to determine at which step autophagy was affected by BEX2. The results showed that ATG5 deficiency attenuated BEX2-induced autophagy, demonstrating that BEX2 may act upstream of the autophagic pathway. It is well known that the PI3K/AKT/mTOR pathway plays an important role in the initiation of autophagy [26]. Importantly, we found that BEX2 interacted with PIK3CA. Next, we found that overexpression of BEX2 did not affect the expression of PIK3CA and p85 (data not shown) but competitively bound PIK3CA with p85 to induce autophagy. Indeed, BEX2 hindered the activation of PI3K. However, the mechanism by which BEX2 hampered the interaction between PIK3CA and p85 is obscure. Our data also showed that the phosphorylation levels of AKT, EIF4EBP1 and RPS6KB1 were decreased in BEX2-overexpressing cells. Moreover, the constitutively active form of RHEBQ64L or PIK3CA overexpression deprived PI3K/AKT/mTOR pathway inhibition and autophagy induction caused by BEX2 overexpression, suggesting that BEX2 regulates autophagy through the PI3K/AKT/mTOR pathway (Fig. 7).Fig. 7A graphical summary of how BEX2 regulates autophagy in NSCLC cells BEX2 interacted with PIK3CA, which impairs the interaction of PIK3CA and p85. Therefore, BEX2 inhibits the PI3K/AKT/mTOR signaling pathway and activates autophagy A graphical summary of how BEX2 regulates autophagy in NSCLC cells BEX2 interacted with PIK3CA, which impairs the interaction of PIK3CA and p85. Therefore, BEX2 inhibits the PI3K/AKT/mTOR signaling pathway and activates autophagy Although the molecular mechanism of BEX2-regulated autophagy is well demonstrated, whether BEX2 affects the tumor growth of NSCLC in vivo is obscure. Nevertheless, our previous work had proved that BEX2 promoted tumor growth in vivo. The previous work had certified that BEX2 promotes mitophagy, a kind of selective autophagy, by facilitating the interaction between the autophagy receptor NDP52 and LC3B. This work further demonstrated that BEX2 directly regulates PI3K activity to influence autophagy. This paper is an extension and complement of the previous work. Taken together, these data provide a more comprehensive understanding of how BEX2 affects autophagy. PIK3CA mutation has been found in high frequencies in many types of cancer, including NSCLC [27]. Although several PIK3CA-specific inhibitors have been approved for clinical use or are currently under clinical development for cancer treatment, numerous challenges and problems remain [14]. The clinical relevance of BEX2 and NSCLC progression has not been investigated in this paper. However, our previous work had demonstrated that BEX2 is overexpressed in lung adenocarcinoma and is associated with poor prognosis in lymph node metastasis-free patients and clinical stage (I + II) patients by tissue microarrays [21]. Our data suggest that BEX2 may be a therapeutic candidate for treating PIK3CA mutation-positive NSCLC. Conclusions We discovered that BEX2 promotes autophagic flux via PI3K/AKT/mTOR signaling. BEX2 interacts with PIK3CA and impairs PIK3CA and p85 interaction, which hinders the activation of PI3K/AKT/mTOR signaling and promotes autophagy induction. Supplementary Information Additional file 1: Figure S1.Rapamycin induces BEX2 expression in NSCLC cells. A H1299 cells were treated with 10μM rapamycin for different indicated times. B H1299 cells were treated with different indicated concentrations of rapamycin for 2h. Additional file 1: Figure S1.Rapamycin induces BEX2 expression in NSCLC cells. A H1299 cells were treated with 10μM rapamycin for different indicated times. B H1299 cells were treated with different indicated concentrations of rapamycin for 2h. Additional file 2: Figure S2. BEX2 hampers the interaction between PIK3CA and p85. A HEK293FT cells were cotransfected with PIK3CA-FLAG and GST-BEX2 or the negative control plasmid; co-IP assays were carried out with glutathione sepharose beads, followed by western blotting using the indicated antibodies. B HEK293FT cells were cotransfected with BEX2 and PIK3CA-FLAG or the negative control plasmid; co-IP assays were carried out with anti-FLAG M2 affinity gel, followed by western blotting using the indicated antibodies. C Cell lysates were prepared from H1792 cells transiently co-expressing PIK3CA-FLAG and BEX2. Co-IP assays were carried out with anti-FLAG M2 affinity gel, and the eluted proteins were detected by western blot assays with p85, PIK3CA, and BEX2 antibodies. Additional file 2: Figure S2. BEX2 hampers the interaction between PIK3CA and p85. A HEK293FT cells were cotransfected with PIK3CA-FLAG and GST-BEX2 or the negative control plasmid; co-IP assays were carried out with glutathione sepharose beads, followed by western blotting using the indicated antibodies. B HEK293FT cells were cotransfected with BEX2 and PIK3CA-FLAG or the negative control plasmid; co-IP assays were carried out with anti-FLAG M2 affinity gel, followed by western blotting using the indicated antibodies. C Cell lysates were prepared from H1792 cells transiently co-expressing PIK3CA-FLAG and BEX2. Co-IP assays were carried out with anti-FLAG M2 affinity gel, and the eluted proteins were detected by western blot assays with p85, PIK3CA, and BEX2 antibodies. Additional file 3. Additional file 3. Additional file 4. Additional file 4.