Introduction

Macroautophagy (hereafter referred to as autophagy) is an evolutionarily conserved process implicated in cellular homeostasis and response to stress which can selectively remove protein aggregates and damaged or excess organelles [1]. The induction of autophagy by anticancer drugs always produces resistance effects on viability of tumor cells. Thus, developing pharmacological autophagy inhibitors is a very active area of research in cancer therapy. ATG4B (autophagy-related 4B) is a cysteine protease required for autophagy formation. C-terminal amino acid of pro-LC3 is cleaved by ATG4B to its LC3-I isoform (18 kDa), followed by ATG7-mediated conjugation of LC3-I with phosphatidylethanolamine (PE) to yield LC3-II (16 kDa), which integrates into the autophagosome precursor membranes under the control of the ATG5–ATG12 complex [2, 3]. Briefly, ATG4B acts as an activator in the process of autophagy. However, the expression and stability of ATG4B in human lung cancer remains largely unknown. In this study, the role of ATG4B in the progression of lung cancer was investigated. qRT-PCR indicated that the expression level of ATG4B was upregulated in human lung cancer tissues as well as in lung cancer cell lines. Then, tandem affinity purification/mass spectrometry (TAP/MS) were applied to the investigation of the interaction between ATG4B and other candidate proteins in A549 cell lines. Intriguingly, as the results showed, ATG4B protein expression was highly relevant to SLC27A4 in lung cancer cell lines and in lung tumor tissues.

SLC27A4 encoded fatty acid transport protein 4 (FATP4, membrane protein) that facilitates uptake of fatty acids indirectly by mediating their etherification [4–6]. SLC27A4 is expressed in multiple tissues and has therefore been proposed to be a major importer of dietary fatty acids. In this study, one interesting aspect was discovered that ATG4B protein expression was increased in parallel with SLC27A4 in lung cancer cell lines as well as in lung tumor tissues. However, relevant functional research of SLC27A4 in autophagy or cancer therapy has not been investigated before. To investigate the basis of the phenotype, co-immunoprecipitation (Co-IP) and GST-pull down assays were employed to investigate the potential interaction between SLC27A4 and ATG4B. Results showed that there is direct protein interaction between SLC27A4 and ATG4B in mammalian cells. Therefore, the function of SLC27A4/ATG4B complex in mammalian cells still needs more investigations.

As ATG4B acts as activator in autophagy system, the function of SLC27A4/ATG4B complex in lung cancer cells related to autophagy was investigated in this study. The expression level of ATG4B was decreased while the SLC27A4 expression was blockade by si-SLC27A4, whereas SLC27A4 expression level was almost unchanged when cells were treated with si-ATG4B. By proving of serials of experimentation, our study suggested that SLC27A4/ATG4B complex was helpful to maintain ATG4B activity and could activate autophagy rapidly during stress of chemotherapy in human lung cancer cells and blocked SLC27A4 expression by si-RNA also showed the enhanced therapeutic efficiency of several chemotherapy drugs. In conclusion, our study explored the potential values of SLC27A4/ATG4B complex related to autophagy pathway and uncovered a new mechanism of how cancer cells could trigger rapid autophagy under adverse condition. SLC27A4/ATG4B complex might act as a new therapeutic target of tumor chemotherapy and has a promising application in the future.

Materials and methods

Lung cancer tissues and cell lines

Human lung cancer cell lines A549, H460, H1299, H358, SK-MES-1, SPC-A1, H157, H1975, and human bronchial epithelial (HBE) cells were purchased from Cell Resource Center of Institute of Basic Medical Sciences, Chinese Academy of Medical Sciences and Peking Union Medical College (Beijing, China) and were cultured in Dulbecco’s modified Eagle medium (DMEM), respectively, supplemented with 10 % FBS, 100 mg/ml streptomycin, and 100 IU/ml penicillin at 37 °C with 5 % CO2. Cell transfection was performed when cells were grown to 80 % confluence, using the Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s instructions. Total 50 paired lung cancer patients’ tissues and compared non-tumor tissues were obtained from People’s Hospital of Bohou during 2013–2015. Histologic diagnosis of tumors was made and agreed upon by two senior pathologists at Department of Pathology of the Hospital based on World Health Organization (WHO) criteria. All the tissues were immediately stored in liquid nitrogen until used. This study was approved by the Institutional Review Board and Human Ethics Committee of Second People’s Hospital of Bozhou.

Regents and antibodies

Antibodies were obtained from the following sources: LC3I/II (Abcam, ab58610, 1:1000), SLC27A4 (Sigma, SAB4500189, 1:1000), p62 (CST, 8025S,1:1000), ATG4B (Abcam, ab154843, 1:1000), total caspase-3 (CST, D3E9, 1:2000), and β-actin (Abcam, ab6276, 1:2000).

Quantitative real-time PCR analysis and interference RNA

Total RNA from cells and tissues was isolated using Trizol Reagent (Invitrogen) and reverse transcribed using Revert Aid First-Strand cDNA Synthesis Kit (Thermo Scientific) according to the manufacturer’s protocol. Real time PCR was performed using Maxima SYBR Green/Fluoresce in qPCR Master Mix (Thermo Scientific) on the Real-Time PCR Detection System (iQ5, Bio-Rad). Interference RNAs (si-RNA) kits were purchased from Thermo Fisher Scientific company. The relative expression level of agents was normalized to that of internal control U6 by using 2−ΔΔCt cycle threshold method.

Cell proliferation assay and cell apoptosis assay

Cells were seeded in 96-well culture plates at 40 % confluence 1 day prior to transfection and were evaluated by MTT assays. Medium was replaced by 0.1 ml of DMSO (Sigma) and 96-well culture plates were shaken at room temperature for 1 min. The absorbance was measured at OD 490 nm. Cell apoptosis was identified by Hoechst assay. Cells were seeded in 6-well culture plates at 40 % confluence 1 day prior to transfection and washed with PBS then fixed by 4 % paraformaldehyde for 15 min after 24 h treatment. Subsequently, Hoechst 33342 diluted by PBS was added into each well for 10 min, followed by washing with PBS for 10 min twice. The blue stained nuclei were observed and statistics of apoptosis rate were calculated under fluorescence microscopy (Olympus Optical Co., Hamburg, Germany).

Detection of GFP–LC3 autophagic dots

pcDNA3.1–GFP–LC3 vectors were a gift from Professor Li Yu, Tsinghua University. The A549 and H460 cells which stably expressed high level of GFP-LC3 protein were established by pcDNA3.1–GFP–LC3 transfected and were selected by G418. The A549 and H460 stably expression GFP–LC3 was detected and confirmed by fluorescence microscope and used for the following experiments (referred to as A549–GFP–LC3 or H460–GFP–LC3). After A549–GFP–LC3 and H460–GFP–LC3 cells were transfected with different small interfering RNA or treated with RAD001 in 12 h, the percentage of GFP–LC3 punctate-positive cells was quantified and analysis by automated image acquisition using a threshold of ≥5 dots/cell. Data were shown as the mean ± SD and were representative of three independent experiments.

Immunoprecipitation, Western blot

The concentrations of protein were quantified by the DC protein assay kit (Bio-Rad, 500–0121) after cell protein was extracted in RIPA buffer (Milipore). Cellular lysates were resolved on SDS-PAGE and electrophoretically transferred to polyvinylidene difluoride membranes (PVDF, Milipore) according to the standard protocol. The immunoreactivities were visualized by enhanced chemiluminescence reagents (Millipore, WBKLS0500). β-actin was used as an internal control.

GST-pull down assay

GST-fusion proteins were purified from E. coli and immobilized for 30 min on glutathione–Sepharose (Amersham Pharmacia Biotech) at 4 °C. Beads were washed extensively with binding buffer (50 mM Tris–HCl pH 8.0, 250 mM NaCl) and were subsequently incubated for 90 min with HA-tagged or Flag-tagged protein. Bound proteins were extracted with loading buffer and analyzed by immunoblotting with the indicated antibodies.

Statistical analysis

All statistical analyses were performed using SPSS 13.0 software. The data were expressed as mean ± SD. Student’s t test was used to determine the significant differences of results. The level of significance was set at P < 0.05.

Results

The expression level of ATG4B is upregulated in lung cancer tissues and cell lines

qRT-PCR was used to detect the ATG4B expression in human lung cancer cell lines and 50 paired lung cancer patients’ tissues and the compared normal non-tumor tissues. As the results showed, the expression of ATG4B in human lung cancer cell lines was significantly upregulated when compared with HBE cells (Fig. 1a). The relative average expression level of ATG4B in lung cancer tissues was significantly lower compared with the adjacent non-neoplastic tissues (84 %, 42 out of 50 patients; Fig. 1b). Further investigate showed that the expression levels of ATG4B were not associated with age, gender, or clinical stage but interestingly associated with the response to chemotherapy, which suggested that the function role of ATG4B might be involved in cell resistance to chemotherapy (Fig. 1c). Since ATG4B acts as activator in autophagy and the ability to activate autophagy within cancer cells plays a protective role in cell survival, ATG4B expression might be valuable in the screening, diagnosing, and prognosis predicting in human lung tumors.

Fig. 1
figure 1

ATG4B expression is upregulated in NSCLC. a The expression levels of ATG4B in HBE cells and NSCLC cell lines were measured by qRT-PCR. The values represent the mean ± standard deviation of triplicates (*P < 0.05). b 50 pairs of NSCLC tissues and matched adjacent normal non-tumor tissues from patients were assessed by qRT-PCR analysis (*P < 0.05). The values were standardized to the U6 expression level. c Relationship between clinicopathological features and ATG4B expression level in NSCLC patients. qRT-PCR real-time quantitative polymerase chain reaction, AN adjacent normal non-tumor tissue, NSCLC non-small cell lung cancer

Full size image

Identification of protein interactions between ATG4B and other candidate proteins

According to the above results, the expression level of ATG4B was upregulated in lung cancer tissues as well as lung cancer cell lines. To investigate the reason for obtaining ATG4B stability in lung tumor cells, tandem affinity purification/mass spectrometry (TAP/MS) were applied to the investigation of the interaction between ATG4B and potential proteins [7]. By proving experimentation, five candidate proteins were identified as LRPPRC, SLC27A4, PFKP-6, GABARAP, and ATP5A1, which was shown in the table (Fig. 2a). Furthermore, subsequent experiments indicated that only SLC27A4 encoded fatty acid transport protein 4 (FATP4, membrane protein) that facilitates uptake of fatty acids, exerted characteristic that the protein expression of SLC27A4 was increased in parallel with ATG4B protein in numerous lung cancer cell lines (Fig. 2b).

Fig. 2
figure 2

SLC27A4 protein expression increased in parallel with ATG4B in lung tumor tissues and in tumor cell lines. a Tandem affinity purification/mass spectrometry (TAP/MS) were applied to the investigation of the interaction between ATG4B and other proteins in human cancer cell A549. Candidate proteins were listed in the chart. b SLC27A4 protein expression was increased in parallel with ATG4B in lung cancer cell lines and the relative gray value (/β-actin) was calculated in the right of the diagram. c SLC27A4 protein expression was increased in parallel with ATG4B in lung tumor tissues (7/8, randomly selected). d The expression of SLC27A4 in lung tumor tissues was upregulated and the fold change of ATG4B and SLC27A4 mRNA was measured by qRT-PCR,*P < 0.05

Full size image

SLC27A4 encoded fatty acid transport protein 4 (FATP4) that facilitates uptake of fatty acids indirectly by mediating their etherification. Interestingly, this expression phenomenon was also discovered in the lung tumor tissues (eight tissue samples out of 50 lung cancer patients were randomly selected) and the SLC27A4 and ATG4B protein expression was detected respectively in the tumor tissue samples as well as in the matched adjacent normal non-tumor tissues. As the results showed, SLC27A4 expression level was increased in majority of the lung tumor tissues (7/8), while the tendency of ATG4B expression level was same as the SLC27A4 expression (Fig. 2c). Furthermore, expression of SLC27A4 in lung tumor tissues was also consistent with the findings of prior results in this study (Fig. 2d). However, reports of SLC27A4 related to the processes of autophagy or the treatment of cancers were very limited. The underlying mechanism of how SLC27A4 could affect ATG4B expression and further involved in the occurrence of autophagy in human cancer cells still needs further researching and discussing.

SLC27A4 interacts ATG4B directly in lung cancer cells

Previous studies have confirmed that the protein expression of SLC27A4 in numerous lung cancer cell lines was increased in parallel with ATG4B protein expression. In order to better explore the relevant functions between ATG4B and SLC27A4 in human lung cancer cells, co-immunoprecipitation and GST-pull down assays [8] were employed to verify the protein interactions of ATG4B and SLC27A4 in human lung cancer cells. As the results showed, SLC27A4 and ATG4B proteins could interact with each other, which provided a new approach for studying the function role of SLC27A4 gene in human cancers. Furthermore, starvation treatment activated autophagy and promoted the separation of ATG4B from SLC27A4 in A549 cells, with the inducing of autophagosome translocation of LC3 and increasing expression level of Beclin-1 [9] as the markers of autophagy activation (Fig. 3a). The results have interesting implications that SLC27A4/ATG4B complex might be conducive to the occurrence of autophagy in human cancer cells. In order to verify our hypothesis, cell immunofluorescence method was used to identify the protein interaction between SLC27A4 and ATG4B on the cellular level. As expected, SLC27A4 and ATG4B proteins could interact with each other in both A549 and H460 cells without further treatment. However, ATG4B was stripped off from the protein conjugate rapidly while autophagy was stimulated by everolimus (RAD001) in a relatively short period of time (Fig. 3b). Western blot analysis of autophagy protein markers confirmed that autophagy was activated and autophagosomes were accumulated since the cells were treated with RAD001 for 1 h, which was also coincided well with the immunofluorescence experiments (Fig. 3c). Meanwhile, A549 cells stably expression GFP–LC3 detected by fluorescence microscope (referred as A549–GFP–LC3) also characterized of GFP–LC3 punctate-positive dots under the treatment of RAD001 (Fig. 3d).

Fig. 3
figure 3

SLC27A4 interacts ATG4B directly in lung cancer cells. a SLC27A4–ATG4B interaction in vivo in A549 cells. IB immunoblotting, IP immunoprecipitation. SLC27A4–ATG4B interaction under the medium with serum (Nutrient-rich) or serum deprivation (Starv.) in 2 h. WCL whole-cell lysate. GST-pull down was employed to detect the interaction of SLC27A4–ATG4B. b Detection of SLC27A4 and ATG4B expression in the A549 and H460 cells with the presence or absence of autophagy activator (RAD001, 20 nM) in 2 h by immunofluorescence. c Western blot analysis of autophagy markers LC3I/II, P62, SLC27A4, and ATG4B expression from lysates of A549 cells treated RAD001 (20 nM) in various hours. β-actin was used as the internal control. d A549–GFP–LC3 cells which stable expressing GFP–LC3 were used to detect autophagy after cells were treated with RAD001 in 4 h (RAD001, 20 nM)

Full size image

SLC27A4/ATG4B directly linking sensing of RAD001-induced autophagy in human lung cancer cells

Prior results showed a link between SLC27A4 and autophagy process in human cancer cells for the first time, suggested the protein-binding of SLC27A4/ATG4B was beneficial for cancer cells to resist chemotherapy drug toxicity by stimulating autophagy. To our best knowledge, ATG4B acts as a cysteine protease that cleaves the C-terminal amino acid of ATG8 family proteins to reveal a C-terminal glycine which is necessary for ATG8 proteins conjugation to phosphatidylethanolamine (PE) and insertion to autophagosome precursor membranes. Thus, the concentration of the ATG4B protein in cell cytoplasm will be helpful to the occurrence of autophagy. In order to further clarify SLC27A4 function in human cancer cells, especial related to autophagy, small interference RNA (si-RNA) was employed. As shown in Fig. 4a, RNAi knockdown of SLC27A4 (si-SLC27A4) decreased the expression level of ATG4B, while knockdown of ATG4B by si-ATG4B had limited influence to the expression of SLC27A4. Results indicated that high expression SLC27A4 was crucial to maintain the stability of ATG4B protein in cytoplasm (Fig. 4a). Furthermore, transfection of si-SLC27A4 could block autophagy response which was activated by RAD001 treatment in lung cancer cells (Fig. 4b). Consistent with these above findings, A549–GFP–LC3 cells showed clear evidence of a decline in punctate-positive numbers when cells were transfection of si-SLC27A4 or si-ATG4B after RAD001 treatment (Fig. 4c). Collectively, SLC27A4/ATG4B complex could help to maintain ATG4B activity in tumor cells and activate autophagy rapidly during stress of chemotherapy or conditions of reduced nutrient availability (Fig. 4d).

Fig. 4
figure 4

Protein-binding of SLC27A4/ATG4B directly linking sensing of autophagy in human lung cancer cells. a si-SLC27A4-RNA and si-ATG4B-RNA were used to explore the potential mechanism of SLC27A4/ATG4B complex related to autophagy in A549 cells, respectively. b After A549 cells treated with si-SLC27A4 or si-ATG4B, respectively, in 24 h, Western blot analysis of autophagy markers LC3I/II, P62, SLC27A4, and ATG4B expression from lysates of A549 cells was detection with the absence of RAD001 (20 nM) in 4 h. β-actin was used as the internal control. c A549–GFP–LC3 cells which stable expressing GFP–LC3 protein were used to detect autophagy flux after cells were treated with RAD001 (20 nM) with the transfected of si-SLC27A4 or si-ATG4B in various hours. d Graphical illustration of SLC27A4/ATG4B complex mechanism related to autophagy pathway in human cancer cells

Full size image

Reduced SLC27A4 expression enhanced chemosensitivity of RAD001, doxorubicin, and cisplatin through autophagy inhibition

In order to verify our hypothesis of SLC27A4 related to autophagy, three common chemotherapy drugs, including everolimus (RAD), doxorubicin (Dox), and cisplatin (DDP) were used in the subsequent experiments. A549–GFP–LC3 cells stably expressed high level of GFP–LC3 protein were exposed to those drugs and the percentage of GFP–LC3 puncta-positive cells was significantly increased in the drug treatment groups, while the presence of si-SLC27A4 could reduced the percentage of GFP–LC3 puncta-positive cells effectively (Fig. 5a). To further determine the influence of autophagy inhibition by reduced SLC27A4 on the chemoresistance in lung cell lines, MTT assay was used to detect cell viability in A549 cells. As shown in Fig. 5b, reduced SLC27A4 by si-RNAs sensitized A549 cells to rapamycin, doxorubicin, and cisplatin, which suggested that the presence of si-SLC27A4 could improve therapeutic response of those agents. Cells treated with both si-SLC27A4 and chemotherapy agents showed enhanced cell apoptosis compared with groups treated with chemotherapy agents alone through Hoechst assay (Fig. 5d). Importantly, RAD-, Dox-, DDP-treatment groups all activated the cleavage of caspase3 (marker of apoptosis), while si-SLC27A4 enhanced the apoptosis ability of those agents (Fig. 5e). Cell apoptosis rate of the above groups showed in Fig. 5f. These results suggested that si-SLC27A4 reversed the resistance of chemotherapeutics through autophagy inhibition in lung cancer cells.

Fig. 5
figure 5

Reduced SLC27A4 expression enhanced chemosensitivity of RAD001, doxorubicin and cisplatin through autophagy inhibition. a GFP–LC3 puncta-positive cells were significantly increased in the drug treatment groups and si-SLC27A4 reduced the percentage of GFP–LC3 puncta-positive cells effectively in 24 h. b The percentage of GFP–LC3-positive cells with GFP–LC3 dots of the above experiments in 24 h. c Western blot analysis of autophagy markers LC3I/LC3II of the above experiments in 24 h. d Cell apoptosis was identified via Hoechst assay after cells was treated with RAD001 20 nM, Dox 0.1 g/ml, or DDP 10 μM in each groups with the presence or absence of si-SLC27A4 in 24 h treatment. e Western blot analysis of apoptosis marker cleaved-caspase3 of the above experiments in 24 h. f Cell apoptosis rate of the above groups. RAD RAD001, Dox doxorubicin, DDP cisplatin

Full size image

Discussion

The chemo-resistant chemotherapeutic agents are still a major obstacle to effective therapy, but the underlying mechanisms of chemo-resistance still remain largely unknown. Recently, substantial research have proved that autophagy always functions as a protective mechanism by degrading and reusing abnormal proteins and organelles as energy resources to promote cancer cell survival in various cancer treatments [10–12]. Upregulated autophagy has been identified in a wide variety of cancer cells after therapeutic stress. Therefore, regulating the activities of autophagy may be a new target of cancer therapy. However, the potential underlying mechanisms of the ability to activate autophagy within those cancer cells could prove invaluable in the treatment of various types of tumors [13, 14].

Autophagy relate genes (ATGs) had been proved to be regulated by upstream of mTOR signaling, including PDK1, Akt, and TSC1/2 [15]. Among of the proteins, ATG4B, a cysteine proteinase that activates LC3B, acts as an activator in autophagy process. Recently, developing pharmacological autophagy inhibitors is an active area of cancer research, including ATG4B inhibitor. Tran et al. found that overexpression of dominant negative ATG4B can either amplify the effects of cytotoxic therapies or contribute to treatment resistance in prostate cancer therapy [16]. Dunn et al. also proved that ATG4B antagonist inhibited autophagy and had a negative impact on osteosarcoma tumors [17]. However, detailed studies on the function of ATG4B during anticancer therapy are still very lacking. In our study, ATG4B expression and function roles were investigated in lung cancer for the first time. As shown in the results, expression level of ATG4B was significantly upregulated in human lung cell lines as well as in lung cancer tissues. Further investigate showed that the expression levels of ATG4B were not associated with age, gender, or clinical stage but associated with the response to chemotherapy, which suggested that the function role of ATG4B might be involved in cell resistance to chemotherapy. To reveal the mechanism governing the stability of ATG4B in lung cancer cells, tandem affinity purification/mass spectrometry (TAP/MS) were applied to the investigation of the interaction between ATG4B and other proteins. Intriguingly, only SLC27A4, which encoded fatty acid transport protein 4 (FATP4), showed the special characteristic that the protein expression of SLC27A4 was increased in parallel with ATG4B in lung tumor tissues, while the tendency of ATG4B expression level was same as the SLC27A4 expression in those tissues. This is a phenomenon worthy of attention and a very interesting research topic.

SLC27A4 has been hypothesized to be bifunctional, exhibiting both fatty acid transport and acyl-CoA synthetase activities that work in concert to mediate fatty acid influx across biological membranes [18, 19]. However, there are no reports of SLC27A4 related to cancer therapy before. In this study, we found that SLC27A4 and ATG4B proteins could interact with each other in human lung cancer cells by co-immunoprecipitation and GST-pull down assays. Through cell immunofluorescence analysis in human lung cancer cell lines A549 and H460, ATG4B was found to be stripped off from the SLC27A4/ATG4B protein conjugate rapidly while autophagy was stimulated by RAD001 in a relatively short period of time (1 h). The releasing of ATG4B from the protein conjugate to the cytoplasm was beneficial to the occurrence of autophagy in cancer cells, which increasing resistance to chemotherapeutics at the beginning of the treatment and eventually developed into the tolerance to chemotherapy. Further studies found that RNAi knockdown of SLC27A4 decreased the expression level of ATG4B, while RNAi knockdown of ATG4B had limited influence to the expression of SLC27A4. This result suggested that SLC27A4 played a predominant function in the SLC27A4/ATG4B complex to maintain the ATG4B protein stability in human lung cancer cells. Furthermore, blocked SLC27A4 expression by si-RNA also showed the enhanced therapeutic efficiency of everolimus, doxorubicin, and cisplatin in human lung cancer cells through autophagy inhibition.

Blocking cancer cell autophagy is emerging as a novel approach to enhance the efficiency of chemotherapy in cancer treatment [20–24]. In this study, higher-level expression level of SLC27A4 in cancer cells, forming amounts of SLC27A4/ATG4B proteins complex, showed the capacity to respond faster under the stress of cytotoxic chemotherapy by releasing ATG4B to cytoplasm and then stimulating autophagy. By this way, autophagy degrades and reuses abnormal proteins and organelles as energy resources to promote cancer cell survival in chemotherapeutic treatment. The results have interesting implications that SLC27A4/ATG4B complex might be conducive to the occurrence of autophagy in human cancer cells, which is meaningful investigations toward the aim of developing autophagy-targeting drugs and have significant values in clinical application.