Introduction

The PI3K/AKT/mTOR (PAM) signalling pathway has been extensively explored in oncology [1, 2]. PI3K and AKT are the kinases of the pathway that participate in cell growth and differentiation. Mammalian target of rapamycin (mTOR), a downstream protein of the PI3K signalling pathway, has emerged as a pivotal node through which cells regulate their growth and nutrient-sensing pathways to control various cellular functions, including survival, proliferation, autophagy and metabolism [3]. mTOR is a catalytic subunit of two multiprotein complexes, namely, mTOR complex 1 (mTORC1) and mTORC2 [4]. mTORC1 can be activated through the phosphoinositide 3-kinase (PI3K)/AKT signalling pathway by growth factors, chemokines and nutrients [5].

Deregulation of the PAM pathway can promote non-small-cell lung cancer (NSCLC) tumorigenesis and contribute to the poor prognosis of patients [6]. There are three classes of PI3Ks that are different in structure and substrate specificity. The phosphorylation of Class IA PI3Ks is initiated by growth factors binding to receptor tyrosine kinases (RTKs), which promotes the release of p110α and the subsequent phosphorylation of PIP2 to PIP3 [7]. PIP3 first facilitates AKT phosphorylation and then activates mTOR. Clinical studies also indicate that abnormal activation of p110α promotes NSCLC progression [8].

Accordingly, blocking the PAM signalling pathway has been recommended as a therapeutic approach against NSCLC [9]. It has been reported that there is a close and inverse relationship between mTORC1 activation and macroautophagy (hereafter called “autophagy”) induction [10]. As the first identified mTOR inhibitor, in 2002 rapamycin was reported to have antitumour effects and has since been documented in a wide range of malignancies [11, 12]. Dactolisib (NVP-BEZ235) is a PI3K/mTOR dual inhibitor that can suppress tumour growth in numerous preclinical models and can thus prevent the phosphorylation of the ribosomal protein S6 kinase B2 (RPS6KB2) and other mTORC1 downstream effectors, eventually leading to a reduction in autophagy [13]. Autophagy is a “self-eating process”, the major process of cellular digestion that removes damaged macromolecules and organelles to maintain homeostasis [14]. It is a double-edged sword in the progression of NSCLC [15] because it may not only promote tumour survival by providing nutrients to tumour cells when limited but it may also inhibit tumour growth by leading to programmed cell death or apoptosis [16, 17].

In our previous study, a series of compounds based on the structure of VS-5584 were synthesized as PI3K inhibitors [18]. W941 was selected for further study because of its robust inhibitory activity against PI3Kα. In this study, we noticed that W941 could exert potent anti-NSCLC efficacy as a single agent both in vitro and in vivo. We also found that W941 could induce autophagy in an NSCLC cell line. However, we found in this study that autophagy actually protected cells from W941-mediated cytotoxicity. We pre-treated NSCLC cells with autophagy inhibitors and demonstrated more potent anti-proliferative activity with the co-treatment strategy than with W941 treatment alone. Therefore, the combination of a PI3K inhibitor with an autophagic inhibitor might provide a novel option for NSCLC therapy.

Materials and methods

Antibodies and reagents

W941 was synthesized as a new PI3K inhibitor by our laboratory (Department of Medicinal Chemistry, Xi’an Jiaotong University, Xi’an, China). A 10 mM stock solution was prepared by dissolving the W941 powder in dimethyl sulfoxide (DMSO), and the solution was stored at −20 °C. For the xenograft experiment, W941 was dissolved in DMSO, diluted with PEG400 and ddH2O (DMSO/PEG400/H2O = 1:7:2, volume ratio). The primary antibodies used in this study include anti-AKT (Cell Signalling Technology (CST), #9272), anti-phospho-S473-AKT (CST, #4060), anti-mTOR (CST, #2983), anti-phospho-mTOR (CST, #5536), anti-LC3A/B (CST, #12741) and anti-SQSTM1 (CST, #88588), anti-GAPDH (Sigma Aldrich Corporation (SAC), G9545), anti-Ki67 (Abcam, ab15580), and anti-cleaved caspase-3 (CST, #9664). Anti-rabbit IgG (HRP-linked antibody, CST#7074) was used as a secondary antibody. The chemicals used in this study include chloroquine (CQ) (SAC, C6628), spautin-1 (APExBIO, B5873), and bafilomycin A1 (APExBIO, A8627).

Cell culture

NSCLC cell lines, including A549, Hcc827, H1975 and H1299, were first obtained from the American Type Tissue Culture Collection (ATCC). All cell lines were cultured in a humidified atmosphere at 37 °C in DMEM (HyClone) medium supplemented with 10% foetal bovine serum (FBS), penicillin and streptomycin. The medium was replaced every 2 days, and the cells were sub-cultured at a confluency of 80%.

Cellular proliferation assay

Cellular proliferation was assessed using the MTT (Solarbio, M8180, Beijing, China) colorimetric assay. Briefly, cells were seeded at a density of 3000 cells/well in a 96-well plate and cultured overnight. The medium was replaced with 100 μL of W941 at designated concentrations for 48 h. MTT solution (5 mg/mL in PBS) was added (20 μL/well) to each well, and the plates were incubated for 4 h in the dark at 37 °C. The purple formazan crystals were subsequently dissolved in 150 μL of DMSO. Absorbance at 490 nm was read on a microplate reader, the results of which were expressed as a percentage (%) of the control. Assays were performed for a total of at least 3 independent experiments.

Colony formation assay

The long-term effects of W941 on tumour cell growth were assessed by using the colony formation assay as previously described [19]. Briefly, the cell suspension was inoculated in each well of 6-well plates at a density of 1000 cells/well. After overnight incubation, the cells were treated with W941 at different concentrations (0.2, 1 or 5 μM) and were allowed to grow for 2 weeks to visualize the colonies. After removing the medium by aspiration, the plates were rinsed with PBS and stained with 0.1% crystal violet solution (Solarbio, G1063) for 10 min, then imaged and counted. The results are expressed as a percentage (%) of the control.

In vivo xenograft model of NSCLC cell line A549

All animal experiments were approved by the Ethics Committee of Xi’an Jiaotong University Health Science Center, Xi′an, China (No. 2018–114). Four-week-old Balb/c female nude mice were purchased from Beijing Vitong Lihua Experimental Animal Technology Company (Beijing, China) and housed in the Animal Center of Xi’an Jiaotong University, Xi’an, China, in a 12:12-h light/dark cycle with food and water readily available. A549 cells (106 cells/mouse) were transplanted into the right flank of the Balb/c mice. The mice were randomly allocated to W941-treated or vehicle-treated groups (intragastrically once daily) when the average tumour volume reached 120 mm3. Tumour size was measured with a calliper and calculated as 0.5 × length×width2. The mice were sacrificed on the last day of the treatment, and the tumour masses were isolated and weighed.

Cellular immunofluorescence

The cells were grown as monolayers on 13-mm coverslips (Sarstedt, Germany) and treated with W941 (or pre-treated with autophagy inhibitors) for 24 h. After the treatment, the cells were rinsed in PBS and fixed for 20 min in 4% paraformaldehyde in PBS. After being washed in PBS 3 times, the cells were blocked in PBS with 0.1% Triton X-100 containing 3% donkey serum (Jackson ImmunoResearch) and subsequently incubated overnight at 4 °C with anti-LC3A/B (1:1000), anti-Ki67 (1:1000) or anti-cleaved caspase-3 (1:500). After being washed 3 times in PBS, the cells were incubated with Alexa Fluor 488-labelled secondary antibody (1:1000) and Hoechst 33342 (1 μg/ml) for 1 h at room temperature. After washing in PBS, the cells were mounted onto slides using Molecular Probes ProLong Gold Antifade Mountant (Thermo Fisher Scientific, P36930). The images were obtained using a ZESS LSM700 confocal microscope system (Zeiss, Germany).

Western blot analysis

Cell lysis was generated by adding RIPA buffer, protease and phosphatase inhibitor cocktails to the cultured cells, which were centrifuged at 12,000 g at 4 °C for 15 min for protein extraction. The protein concentrations were determined by BCA assay following the instructions. Equal amounts of proteins (15 μg) were loaded, separated by SDS-PAGE gel at 100 V and transferred to a polyvinylidene fluoride membrane. The membrane was blocked in 5% slim milk and then incubated with primary and secondary antibodies in sequence. The proteins were visualized using enhanced chemiluminescence (Bio-Rad, California, USA).

Statistical analysis

All data are presented as the mean ± SEM. The variation between the two treatment groups was determined by two-tailed unpaired Student’s t test using GraphPad Prism, V 6.0. A value of *P < 0.05 was considered statistically significant.

Results

Antiproliferative effects of W941 on NSCLC

The structure of W941 is depicted in Fig. 1a. NSCLC cell lines, including A549, Hcc827, H1975 and H1299, were treated with W941 for 48 h, and their IC50 values were measured using the MTT assay. As shown in Table 1, W941 exhibited a potent inhibitory effect on A549 cells with an IC50 value of 0.12 μM. It also significantly inhibited the proliferation of Hcc827 cells with an IC50 of 0.23 μM. The results demonstrated that W941 decreased cell viability in a time- and concentration-dependent manner; this was determined by measuring cell growth either after incubating A549 and Hcc827 cells with a gradient of W941 concentrations for 24 h or at various time points with a fixed W941 concentration of 10 μM (Fig. 1b). Next, we used the colony formation assay to assess the long-term effects of W941 on NSCLC cell growth. This assay revealed that W941 remarkably suppressed the formation of colonies at 1 μM in A549 and 0.2 μM in Hcc827 cells and that the formation of cell colonies was almost completely inhibited at a concentration of 5 μM (Fig. 1c).

Fig. 1
figure 1

Inhibition properties of W941 in NSCLC. a The chemical structure of W941. b A549 and Hcc827 cells were incubated with a gradient of W941 concentrations for 24 h or treated with W941 at 10 μM for different durations of time. W941 concentration- and time-dependently inhibited the proliferation of A549 and Hcc827 cells. c Single cells were plated and treated with W941 for 3 weeks, and then the colonies formed were stained and counted. The bar graph on the right represents the percentage of colony area in the whole well. d Balb/c nude mice were injected with A549 cells (106/mouse) in the right flank and were grouped and treated with the indicated concentrations of W941 when the average tumour volume reached 150 mm3. The tumour volume, tumour weight, mouse body weight and tumour image are presented. Statistical significance was determined using Student’s t test, n = 6; *P < 0.05; **P < 0.01 compared with the control group

Full size image
Table 1 IC50 values of W941 in NSCLC (μM)
Full size table

To investigate the efficacy of W941 in an A549 xenograft model, tumour-bearing mice were treated with 5, 15 or 50 mg/kg of W941 for 3 weeks. W941 treatment at a dose of 50 mg/kg led to significant tumour growth inhibition compared with the solvent group. In contrast to VS-5584 treatment (15 mg/kg), which suppressed tumour growth while reducing body weight, W941 was well tolerated at all doses tested without any weight loss. This indicates that the toxicity of W941 was minimal (Fig. 1d).

W941 blocked the PI3K downstream signalling pathway in NSCLC

In our previous study, we investigated the enzymatic inhibitory ability of W941 against PI3Kα and found that W941 inhibited PI3Kα extracellularly with an IC50 value of 21.7 nM [18]. To verify this result intracellularly, we detected the phosphorylated and total proteins of the PAM downstream pathway in A549 and Hcc827 cells after treatment with a gradient of W941 concentrations for 24 h. Measurement of AKT, ERK, p70S6K and 4E-BP1 via Western blotting showed that W941 dephosphorylated all four proteins in a concentration-dependent manner, whereas no observable change was found in the total protein levels (Fig. 2). The results indicate that the anti-NSCLC activity of W941 was related to the suppression of the PAM pathway.

Fig. 2
figure 2

Blockage of PI3K signalling by W941 treatment on NSCLC. a A549 cells were treated with a gradient of W941 concentrations for 24 h, and the phosphorylation levels of AKT, mTOR, p70S6K and 4E-BP1 were measured by Western blot analysis. b Similar experimental conditions were used to treat Hcc827 cells. Statistical significance was determined using Student’s t test, n = 3; **P < 0.01 compared with the 0 μM W941 treatment

Full size image

Treatment with W941 induced autophagy in NSCLC cells

As inhibition of mTOR is closely related to autophagy induction, we investigated whether W941 could induce autophagy in NSCLC cells. A549 and Hcc827 cells were treated with W941 at concentrations of 0.2, 1 and 5 μM for 24 h, and LC3 expression was determined by Western blotting. W941 treatment markedly increased LC3-II levels in the two cell lines in a dose-dependent manner (Fig. 3a and d). Consistently, an immunofluorescence assay also demonstrated that LC3-II expression was significantly enhanced in cells treated with W941 at 5 μM for 24 h compared with that of the controls (Fig. 3c and f). To determine whether W941 can induce or inhibit autophagy, both of which can cause a build-up of LC3-II and LC3-positive puncta, we investigated the impact of W941 on autophagic flux with or without autophagosome-lysosome fusion inhibitors, viz. chloroquine (CQ) [20] and bafilomycin A1 (Baf A1). As shown in Fig. 3b&E, the LC3-II/I ratio increased in cells with pretreatment with CQ or Baf A1 compared to cells with pretreatment with W941 alone. In addition, a concentration-dependent decrease in SQSTM1 was also observed in W941-treated cells. These results indicate that W941 acts as an autophagy inducer rather than an inhibitor in NSCLC cells.

Fig. 3
figure 3

W941 triggers autophagy and induces autophagic flux in NSCLC cells. a & d A549 (a) or Hcc827 (d) cells were treated with designated concentrations of W941 for 24 h, and the expression levels of LC3 and SQSTM1 were measured using a Western blotting assay. Statistical significance was determined using Student’s t test, n = 3; **P < 0.01 compared with 0 μM W941 treatment. b & e A549 (b) or Hcc827 (e) cells were treated with or without 1 μM W941, 10 μM CQ, 0.01 μM Baf A1 or the combinations for 24 h, and the expression of LC3 was detected by Western blotting. Statistical significance was determined using Student’s t test, n = 3; #P < 0.01 compared with the single W941 treatment. c & f A549 (C) or Hcc827 (f) cells were treated with or without 5 μM W941 for 24 h, and the expression of LC3 was visualized by cellular immunofluorescence

Full size image

W941-induced autophagy protected NSCLC cells from W941-induced cytotoxicity

To determine the relationship between autophagy and W941-induced cytotoxicity in NSCLC cells, autophagy inhibitors such as spautin-1, Baf A1 and CQ were used. As illustrated in Fig. 4a and b, cells co-treated with these inhibitors had lower viability than those treated with W941 alone. We also measured the effect of W941 on the cellular expression of Ki67 and cleaved caspase-3 using an immunofluorescence assay. The fluorescence intensity of Ki67 was lower in the W941 treatments than in the co-treatments. In addition, we did not find any expression of cleaved caspase-3 in treatments with only W941 or CQ. However, puncta of cleaved caspase-3 fluorescence were observed in CQ and W941 co-treated cells (Fig. 4c). We also utilized Annexin V-FITC/PI double staining to quantify the apoptosis induced by pretreatment with CQ. As shown in Fig. 4d, pretreatment with CQ led to almost three times more apoptotic cells than did treatment with W941 alone. In summary, W941-induced autophagy acted as a protective cellular reaction, indicating that pretreatment with autophagy inhibitors could enhance the ability of W941 to induce apoptosis.

Fig. 4
figure 4

W941-induced autophagy protects NSCLC cells from W941-induced cytotoxicity. A549 (a) and Hcc827 (b) cells were treated with W941 for 24 h in the presence or absence of autophagy inhibitors spautin-1, Baf A1 and CQ. Cell viability was analysed via MTT assay. Statistical significance was determined using Student’s t test, n = 3; *P < 0.05; **P < 0.01 compared with the control. c The cellular immunofluorescence assay was conducted to visualize the expression of Ki67 and cleaved caspase-3 in cells. d The A549 and Hcc827 cells were treated with or without 5 μM W941, 50 μM (A549) or 25 μM (Hcc827) of CQ, the combination of W941 and CQ for 36 h, and apoptosis was quantified by flow cytometry. Statistical significance was determined using Student’s t test, n = 3; #P < 0.01 compared with the control

Full size image

Discussion

The PAM pathway is involved in regulating many kinds of basic cellular functions, such as proliferation, differentiation, adhesion and invasion, and its alteration may lead to serious consequences, such as the development of non-small-cell lung cancer [21]. Although many PAM inhibitors have been designed for cancer therapy, including PI3K, AKT and mTOR inhibitors, they have limitations that curtail their clinical application [6]. For example, PI3K inhibitor displays modest clinical activity on solid tumours. AKT inhibitors are not effective enough to block the AKT-independent signalling pathway [22]. mTOR inhibitors usually do not have satisfactory efficacy since they act only on mTORC1 and not on mTORC2 [23]. Thus, exploration of PAM pathway inhibitors is still an ongoing process that calls for further study.

In this study, we investigated the antitumour effects of W941 in NSCLC cell lines both in vitro and in vivo and explored its mechanism. W941 turned out to suppress NSCLC cell proliferation and colony formation in vitro, and the xenograft experiment also demonstrated that W941 has single-agent antitumour efficacy in A549 non-small-cell lung cancer cells in vivo. Western blotting assays validated that W941 markedly blocked PAM signalling transduction by inhibiting the phosphorylation of AKT and mTOR in a dose-dependent manner. However, earlier reports revealed a close relationship between the mTOR downregulation and autophagy induction [10]. Thus, PAM blockers may promote autophagy at the early stage as a survival mechanism to counteract the pressure to undergo apoptosis.

Autophagy is an evolutionarily conserved mechanism that is essential for maintaining cellular homeostasis by degrading dysfunctional cellular proteins and recycling cellular components. Nutrient deprivation, hypoxia or chemicals may promote autophagy [24, 25]. Autophagy has been reported to either inhibit or promote the proliferation of cancer cells in different model systems, suggesting that its role in cancer is context-dependent [26]. Therefore, it is necessary to investigate the role of autophagy in cancer development because it may shed light on the discovery of some novel means by which cancer growth can be regulated [27]. Autophagy is considered a mechanism of cell survival in drug resistance and chemotherapy [28, 29]. Despite the current success of PAM blockers, PAM-induced autophagy may be a survival mechanism that confers resistance to apoptosis [30, 31]. Thus, we assume that inhibition of autophagy can improve the apoptotic potency of PAM inhibitors.

In this study, W941-induced autophagy was further confirmed by the conversion of LC3-I into LC3-II. The autophagic flux was validated by the decrease of P62, which proved that the LC3-II/I ratio was upregulated when cells were pre-treated with CQ and bafilomycin A1. Autophagy plays a paradoxical role in cancer cell survival and death. To better understand autophagy’s action in this case, we treated the cells with autophagy inhibitors and measured their cell viability. The results of this experiment indicate that inhibition of autophagy notably hampered the proliferation of A549 and Hcc827 cells, and inhibition of autophagy by 50 μM CQ increased W941-induced cell death (Fig. 5).

Fig. 5
figure 5

Mechanism of W941’s anti-NSCLC effects. W941 inhibits cell viability and causes cell cycle arrest through the PAM pathway. However, blocking the PAM pathway induces autophagy, which plays a protective role in cell survival. The combination of W941 and autophagy inhibitors leads to cell death

Full size image

It is essential to clarify what role autophagy plays in certain cases. The present study has revealed that W941-induced autophagy plays a protective role, suggesting that inhibition of autophagy by chemical inhibitors may enhance the anti-proliferative and apoptosis-inducing potential of W941 in NSCLC. The application of PI3K inhibitor alone in clinical trials displayed modest anticancer effects on solid tumours. Therefore, a rational combination of PI3K antagonists with other agents may be beneficial for treating patients [32]. Based upon the results of this study, we suggest that a combination of a PI3K inhibitor with an autophagy inhibitor might be a novel option for NSCLC treatment and provide some reference points for further autophagy-related studies.