The interaction of Atg4B and Bcl-2 plays an important role in Cd-induced crosstalk between apoptosis and autophagy through disassociation of Bcl-2-Beclin1 in A549 cells

Author’s Accepted Manuscript The interaction of Atg4B and Bcl-2 plays an important role in Cd-induced crosstalk between apoptosis and autophagy through disassociation of Bcl-2-Beclin1 in A549 cells Zhiguo Li, Qiujuan Li, Wei Lv, Liping Jiang, Chengyan Geng, Xiaofeng Yao, Xiaoxia Shi, Yong Liu, Jun Cao

PII: DOI: Reference:

www.elsevier.com

S0891-5849(18)31567-3 https://doi.org/10.1016/j.freeradbiomed.2018.11.020 FRB14042

To appear in: Free Radical Biology and Medicine Received date: 9 September 2018 Revised date: 23 October 2018 Accepted date: 16 November 2018 Cite this article as: Zhiguo Li, Qiujuan Li, Wei Lv, Liping Jiang, Chengyan Geng, Xiaofeng Yao, Xiaoxia Shi, Yong Liu and Jun Cao, The interaction of Atg4B and Bcl-2 plays an important role in Cd-induced crosstalk between apoptosis and autophagy through disassociation of Bcl-2-Beclin1 in A549 cells, Free Radical Biology and Medicine, https://doi.org/10.1016/j.freeradbiomed.2018.11.020 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

The interaction of Atg4B and Bcl-2 plays an important role in Cd-induced crosstalk between apoptosis and autophagy through disassociation of Bcl-2-Beclin1 in A549 cells Zhiguo Lia,1, Qiujuan Lia,1, Wei Lva, Liping Jianga, Chengyan Genga, Xiaofeng Yaoa, Xiaoxia Shia, Yong Liu*, b, Jun Cao*,a a

Department of Occupational and Environmental Health, Dalian Medical University, No. 9 W. Lvshun South Road, Dalian 116044, China

b

School of Life Science and Medicine, Dalian University of Technology, Panjin 124221, China

[email protected] [email protected] *

Co-Corresponding author: School of Life Science and Medicine, Dalian University

of Technology, Panjin 124211, China. *

Corresponding author at: Occupational and Environmental Health Department,

Dalian Medical University, Dalian 116044, China.

The authors declare that there are no conflicts of interest in the present work.

1

Both authors contributed equally to this work.

ABSTRACT Cadmium (Cd) is a highly ubiquitous detrimental metal in the environment. It is a well-known inducer of tumorigenesis, but the mechanism is not clear. In our previous study, we found that ROS-dependent Atg4B upregulation mediated Cd-induced autophagy and autophagy played an important role in Cd-induced proliferation and invasion in A549 cells. In this study, we found that Cd induced both apoptosis and autophagy in A549 cells, and apoptosis preceded autophagy. Z-VAD-FMK repressed Cd-induced LC3 and Beclin1, indicating that apoptosis was essential for Cd-induced autophagy. 3MA destroyed the recovery of mitochondrial membrane potential and increased Cd-induced CL-CASP9 and CL-CASP3 expression, suggesting that Cd-induced autophagy prevented A549 cells from apoptosis. Further study showed that Atg4B upregulation was mediated by mitochondrial dysfunction and conversely affected mitochondrial function by decreasing Bcl-2 protein expression and its localization in mitochondria, and played an important role in Cd-induced apoptosis. Moreover, Bcl-2 was involved in Cd-induced autophagy. Co-IP assay showed that Atg4B could directly bind to Bcl-2, and consequently promote disassociation of Bcl-2-Beclin1 and released autophagic protein Beclin1 to activate autophagic pathway. Taken together, our results demonstrated that the interaction of Atg4B and Bcl-2 might play an important role in Cd-induced crosstalk between apoptosis and autophagy through disassociation of Bcl-2-Beclin1. Cd-induced autophagy is apoptosis-dependent and prevents apoptotic cell death to ensure the growth and proliferation of A549 cells.

1

Graphical Abstract:

2

Abbreviations: Z-Val-Ala-DL-Asp

Cd,

cadmium;

3MA,

3-methyl-adenine;

(methoxy)-fluoromethylketone;

CCCP,

Z-VAD-FMK,

carbonyl

cyanide

m-chlorophenyl hydrazone; siRNA, small interfering RNA; AO, acridine orange; AVOs,

acidic

vesicular

organelles;

AVs,

autophagy

vesicles;

Co-IP,

co-immunoprecipitation; CL-CASP9, cleaved Caspase9; CL-CASP3, cleaved Caspase3; Atg, autophagy-related; MMP, mitochondrial membrane potential; CQ, chloroquine; DAPI, 4,6-diamidino-2-phenylindole; ER, endoplasmic reticulum

Keywords: apoptosis; Atg4B; autophagy; Bcl-2; cadmium

3

1. Introduction Cadmium (Cd) is a well-known environmental pollutant [1] and is listed as a class I carcinogen [2]. Occupational contamination by battery manufacturing and mining industry is the primary sources of cadmium exposure to humans, whereas tobacco smoke and foods are the main routes of non-occupational exposure [3]. Because of its low rate of excretion, Cd gradually accumulated into multiple tissues and caused injury to organs, even had a causality correlation with cancer development [4, 5]. The epidemiological evidence identified low-level environmental exposure to Cd as a risk factor for lung cancer [6]. A large number of studies reported that Cd induced apoptosis in many cell types from different tissues, and apoptosis was an important mechanism for Cd-induced toxicity [7-9]. Apoptosis usually induces cell death by two primary pathways, death receptor apoptotic pathway and mitochondrial apoptotic pathway [10, 11]. The mitochondria-mediated pathway is often activated by external stimulus including mitochondrial membrane potential (MMP) alteration, release of cytochrome c and other apoptogenic proteins [12]. Different mechanisms underlying Cd-induced apoptosis were illustrated. In HK-2 cells, Cd was reported to induce apoptosis via p53 accumulation by down-regulation of UBE2D2 and UBE2D4 gene expression [13]. Cd-induced apoptosis was also determined by the ratio of Bax/Bcl-2 and the activation of poly (ADP-ribose) polymerase (PARP) in primary rat osteoblast [14]. Meanwhile, Cd-induced autophagy was demonstrated in many studies. Cd induced autophagy through ROS-dependent activation of the LKB1-AMPK signaling 4

in skin epidermal cells [15]. In Saccharomyces cerevisiae, phosphatidylethanolamine is necessary for autophagy under Cd stress [16]. Autophagy is a cellular degradative process in which cytoplasmic proteins and organelles are engulfed within autophagosomes and transported to lysosomes for degradation. So, autophagy could promote cell survival by clearance of damaged organelles and cytosolic components [17]. Apoptosis and autophagy are not mutually independent pathways. A growing number of studies showed that there is a cooperative relationship between apoptosis and autophagy. Several well-known inducers of autophagy have been shown to activate apoptosis, such as Atg12, which can also induce apoptosis through binding to antiapoptotic Bcl-2 family members [18]. Similarly, activation of apoptosis can also mediate autophagy [19, 20]. Data from several reports suggested that the proteins of the Bcl-2 family regulated autophagy in addition to their function in controlling the pathways of apoptosis [21-23]. Stresses were reported to influence autophagy by alteration of the interaction of Bcl-2-Beclin1 [24, 25]. Therefore Bcl-2 represents one of the important points of crosstalk of the apoptotic and autophagic machinery. In our previous study, we found that Cd induced ROS formation and upregulation of Atg4B which caused autophagy in A549 cells [26]. Atg4B is one of the autophagy-related (Atg) genes, which was first discovered in yeast [27]. In the human genome, there are four independent genes encoding Atg4 (4A, 4B, 4C and 4D) [28]. Atg4B is necessary for activation of LC3 and the delipidation of LC3-II from autophagosomes for its recycling to induce autophagy [28, 29]. Although numerous 5

studies have demonstrated the correlation between Atg4B and autophagy, Pei-Feng Liu etc. discovered that Atg4B negatively regulated autophagy in human colorectal cancer cells [29]. Besides inducing autophagy, Atg4B was also reported to serve as an oncogene in many cancers, such as lung cancer [30] and osteosarcoma [31]. In this study, we first demonstrated that Atg4B was involved in Cd-induced crosstalk between apoptosis and autophagy through binding directly to Bcl-2 to result in disassociation of Bcl-2-Beclin1 in A549 cells. Cd-induced autophagy was apoptosis-dependent and the autophagy conversely prevented Cd-induced apoptosis to ensure the growth and proliferation of A549 cells. 2. Materials and Methods 2.1. Cell culture and reagents Human lung cancer (A549) cells were obtained from Cell Center for Peking Union Medical College (Peking, China) and were maintained in RPMI 1640 culture medium containing 8% fetal bovine serum (FBS) (BI, Israel) at 37°C, 5% CO2 in humidified environment. Then A549 cells were cultured in Cd (Cadmium chloride, CdCl2) (10108-64-2, Sigma-Aldrich, USA; purity of 99%) at concentration of 2μM for different times. The autophagy inhibitor 3-methyl-adenine (3MA) (M9281, USA) and chloroquine (CQ) (C6628, USA) were from Sigma-Aldrich. Caspase inhibitor Z-Val-Ala-DL-Asp (methoxy)-fluoromethylketone (Z-VAD-FMK) was from Selleck (187389-52-2, USA). 2.2. Small interfering RNA transfection The Atg4B small interfering RNA was 5’-UGCUGCUGCUGCUUGUGUATT-3’ 6

and 5’UACACAAGCAGCAGCAGCATT-3’. Bcl-2 small interfering RNA was 5’-GGGAGAUAGUGAUGAAGUATT-3’

and

5’-UACUUCAUCACUAUCUCCCTT-3’. The oligo nucleotides encoding scramble siRNA

were

5’-UUCUCCGAACGUGUCACGU-3’

and

5’-ACGUGA

CACGUUCGGAGAA-3’. All siRNAs were transfected into cells by using siRNA-mateTM (G04002, GenePharma, China) according to manufacturer’s protocol. 2.3. Plasmids transfection The

plasmids

pcDNA3.1-Atg4B,

pcDNA3.1-vector,

pEX-3-Bcl-2

and

pEX-3-vector were purchased from GenePharma Biotechnologies. The plasmids were transfected into A549 cells with Lipofectamine 3000 (1788337, Invitrogen, USA) according to specifications. 2.4. TUNEL assay At the end of the designated treatments, A549 cells were fixed with 4% paraformaldehyde and then incubated with 0.1% Triton X-100. After that, cells were labeled with fluorescence TUNEL (KGA7073, Keygen Biotech, China) reagent mixture in the dark. Finally, A549 cells were counterstained with DAPI to reveal cell nuclei. TUNEL and DAPI cells were calculated. Data were presented as the apoptosis index, which was calculated as the percentage of apoptotic nuclei/total nuclei number. 2.5. Hoechst Staining A549 cells were fixed with 4% paraformaldehyde for 30 min at room temperature, then incubated with 10μg/ml Hoechst 33342 (14533, Sigma, USA) for 30 min according to manufacturer’s protocol. Nuclear morphology was observed 7

under fluorescence microscopy. 2.6. Acridine orange (AO) staining AO (A3568, Invitrogen, USA) was used to measure the number of acidic vesicular organelles (AVOs) in cells. AO emits bright red fluorescence in acidic vesicles but green fluorescence in the cytoplasm and nucleus. A549 cells were treated with corresponding reagents. After the treatment, A549 cells were incubated with AO (1μg/ml). Then cells were observed by fluorescence microscopy. 2.7. Electronic microscopy After the treatment, A549 cells were fixed for 2 hrs at 4°C in 2.5% glutaraldehyde. The cells were washed in 2% osmium tetroxide for 2 hrs and then dehydrated, embedded and cut at 70 nm thickness using an ultramicrotome according to standard procedures. The samples were observed by transmission electron microscope. 2.8. MMP Assay JC-1 (C2006, Beyotime, China) was used to prove the effect of Cd on MMP in A549 cells. JC-1 is a cationic dye which selectively gets incorporated in mitochondria and can reversibly change colour from red to green as the membrane potential decreased. So, the ratio of red and green fluorescence is often used to measure the level of mitochondrial depolarization. After the treatment, A549 cells were incubated with JC-1 dye for 20 min and then washed twice with dyeing buffer. Cells were treated with carbonyl cyanide m-chlorophenyl hydrazone (CCCP) (0.1μM), a mitochondrial function inhibitor as a positive control. The staining was observed 8

using the fluorescence microscopy. Data were analyzed by Image-Pro-Plus 7.0 software. 2.9. Immunofluorescence staining After the treatment, A549 cells on coverslips were stained with trackers of mitochondria and endoplasmic reticulum (ER) (Mito-Tracker Green, C1048, Beyotime, China; ER-Tracker Red, KGMP016-1, Keygen Biotech, China) according to the protocols. After washed with RPMI 1640 medium lacking phenol red (90022, Solarbio, China), cells were fixed in 4% formalin solution for 2 min and blocked for 60 min in immunostaining blocking solution (B600060, Proteintech, USA). Then slides were incubated with Bcl-2 rabbit polyclonal antibody (AF6139, Affinity, USA) or Atg4B antibody (15131-1-AP, Proteintech, USA) over night at 4°C in a moist chamber. Excessive antibodies were washed and incubated by an anti-rabbit Alexa Fluor 594 conjugated secondary antibody (SA00006-4, Proteintech, USA) or an anti-rabbit Alexa Fluor 488 conjugated secondary antibody (SA00006-2, Proteintech, USA). After counterstained with 4, 6-diamidino-2-phenylindole (DAPI) (28718-90-3, ROCHE, Switzerland) for 10 min, slides were mounted with cover slip and observed by a fluorescence microscope (Olympus BX63, Japan, 40×10). The image analyzed using Image-Pro Plus 7.0. 2.10. Western blot analysis Total cell extracts were lysed in lysis buffer (KGP250/KGP2100, Keygen Biotech, China). Cytosolic and mitochondrial fractions were prepared using the Cell Mitochondria Isolation Kit (C3601, Beyotime, China). Protein concentrations 9

quantified using the Bicinchoninic Acid Protein Assay Kit (KGP902, Keygen Biotech, China). An equal amount of protein lysate was separated in 10%-15% SDS-polyacrylamide

gel

electrophoresis

(SDS-PAGE)

and

transferred

onto

polyvinylidene fluoride (PVDF) membranes (IPVH00010, Merck Millipore, USA) via a wet electrophoretic transfer method. Membranes were blocked with skimmed milk and probed with antibodies of interest. Protein bands were detected using BeyoECL Plus Kit (P0018M, Keygen Biotech, China), imaged on ChemiDoc XRS+ System (Bio-Rad, USA). Densitometry analysis was performed using ImageJ software to calculate the relative expression change after normalizing with GAPDH, β-actin or VDAC1. Primary antibodies used in the study were as follows: LC3 (4M4802V, Sigma, USA), Beclin1 (AM1818A, Abgent, USA), P62 (18420-1-APA, Proteintech, USA), Caspase-9 (10380-1-AP, Proteintech, USA), Caspase-3 (#9662, Cell Signaling, USA), Cytochrome C (10993-1-AP, Proteintech, USA), VDAC1 (66345-1-Ig, Proteintech, USA), β-actin (TA-09, ZSGB-BIO, China) and GAPDH (10494-1-AP, Proteintech, USA). Secondary antibodies, anti-Rabbit HRP (ZB-2301) and anti-Mouse HRP (ZB-2305) were from ZSGB-BIO. 2.10. Co-immunoprecipitation (Co-IP) experiment After the treatment, A549 cells were lysed with IP lysis buffer (10mM Tris pH 7.4, 25mM NaCl2, 5mM EDTA, 0.1% NP40, 1% protease inhibitor). Cell lysates were precleared with Protein A/G PLUS-Agarose (sc-2003, Santa, USA), and then mixed with Bcl-2 (60178-1-Ig, Proteintech, USA) or control IgG (B900620, Proteintech, USA) for 1 hrs at 4°C. The immunoprecipitates were captured on Protein A/G 10

PLUS-Agarose and analyzed by Western blot with antibodies against Bcl-2, Atg4B (15131-1-AP, Proteintech, USA) or Beclin1 (11306-1-AP, Proteintech, USA), respectively. 2.11. Statistical analysis The statistical analysis was carried out with SPSS 17.0, using the statistical one-way analysis of variance (ANOVA) test and p values to analyze the differences of statistical. All data were obtained independently from three times experiments. Values of P<0.05 and P<0.01 were taken as statistically significant. 3. Results 3.1. Cd induced time-dependent apoptosis and autophagy in A549 cells First, Cd-induced apoptosis and autophagy in A549 cells were observed. As shown in Fig. 1A, B, after treatment with Cd for 0, 12, 24, 36 and 48 hrs, apoptosis-related proteins, cleaved Caspase9 (CL-CASP9) and cleaved Caspase3 (CL-CASP3) were increased and Bcl-2 was down-regulated significantly during 12 to 24 hrs (P<0.05 and P<0.01). Cd treatment at 12 and 24 hrs also caused the release of cytochrome c from the mitochondria to the cytosol as measured by Western blot analysis of cytosolic protein fractions (P<0.01). The expression of CL-CASP9, CL-CASP3 and release of cytochrome c proteins in A549 cells were gradually decreased during 24 to 48 hrs and Bcl-2 level was increased significantly compared to that of 24 hrs (P<0.05 and P<0.01). Meanwhile, significant increases in LC3-II protein level and a decrease in P62 protein level were observed after Cd treatment for 24 hrs (P<0.01). The effects of Cd on apoptosis and autophagy were further examined 11

by direct observation of the nuclei fragmentation and AVOs formation using Hoechst and AO assay respectively. Cd-treated A549 cells showed nuclear shrinkage and fragmentation when treated for 12-24 hrs (P<0.01). However, the number of apoptotic cells decreased after 24 hrs (P<0.01) (Fig. 1C, E). As shown in Fig. 1D, F, the number of AVOs (positive red puncta) in Cd-treated cells significantly increased after 24 hrs in A549 cells (P<0.01). These results suggested that Cd-induced both apoptosis and autophagy in A549 cells, and apoptosis preceded autophagy.

12

Fig.1. Cd induced time-dependent apoptosis and autophagy in A549 cells. (A) Western blots were performed on the total protein or cytoplasmic protein of A549 cells treated with Cd at different time. (B) GAPDH was used as control. Relative expression of these proteins was expressed as a percentage of GAPDH. Each bar represents mean ± SD from three independent experiments (*P<0.05 vs. Control, **P<0.01 vs. Control; &P<0.05 vs. Cd treatment for 24 hrs 13

group, &&P<0.01 vs. Cd treatment for 24 hrs group). (C) Apoptotic morphological changes were observed by fluorescent microscopy using Hoechst 33342 staining. (D) Formation of AVOs in A549 cells was measured by AO staining. (E) Quantitation of formation of nuclei fragmentation and condensation. Each bar represents mean± SD from three independent experiments. Scale bar: 20 μm (**P<0.01 vs. Control; &&P<0.01 vs. Cd treatment for 24 hrs group). (F) Quantitation of formation of AVOs in Cd-treated A549 cells was done. Each bar represents mean± SD from three independent experiments. Scale bar: 20 μm (**P<0.01 vs. Control; &&P<0.01 vs. Cd treatment for 24 hrs group).

3.2. Crosstalk between Cd-induced apoptosis and autophagy in A549 cells To verify the relationship between Cd-induced apoptosis and autophagy, apoptosis inhibitor, Z-VAD-FMK and the autophagy inhibitor, 3MA were employed. As shown in Fig. 2A, B, Cd-induced LC3-II and Beclin1 protein expression was reduced, and Cd-reduced P62 was upregulated significantly in Z-VAD-FMK pretreatment group compared with that of only Cd-treated group (P<0.05 and P<0.01). Also, the percentage of AVOs induced by Cd was significantly decreased after pretreatment with Z-VAD-FMK compared with only Cd-treated cells (P<0.01) (Fig. 2C, D). These data suggested that apoptosis induced by Cd treatment played an important role in triggering Cd-induced autophagy.

14

Fig.2. Effect of Cd-induced apoptosis on autophagy in A549 cells. A549 cells were pretreated with or without Z-VAD-FMK and treated with Cd for 36 hrs. (A) To confirm whether apoptosis was involved in Cd-induced autophagy, the protein level of P62, LC3 and Beclin1 was analyzed by Western blot. (B) GAPDH was used as control. Relative expression of these proteins was expressed as a percentage of GAPDH. Each bar represents mean± SD from three independent experiments. (*P<0.05 vs. Control, **P<0.01 vs. Control; #P<0.05 vs. Cd alone, ##P<0.01 vs. Cd alone). (C) The effect of Z-VAD-FMK on Cd-induced AVOs formation in A549 cells was measured by AO staining. (D) Quantitation of formation of AVOs was done. Each bar represents 15

mean± SD from three independent experiments. Scale bar: 20 μm (**P<0.01 vs. Control; ##P<0.01 vs. Cd alone).

As shown in Fig. 3A, B, Hoechst staining illustrated that there was a significant increase in the number of apoptotic cells in 3MA pretreated groups compared to only Cd-treated cells (P<0.01). Pretreatment with 3MA also raised Cd-induced CL-CASP9 and CL-CASP3 significantly after incubation with Cd for 48 hrs compared to the only Cd-treated A549 cells (P<0.01). The release of cytoplasmic cytochrome c induced by Cd was further elevated after 3MA pretreatment compared to the only Cd-treated A549 cells (P<0.01) (Fig. 3C, D). These results demonstrated that autophagy induced by Cd inhibited Cd-induced apoptosis and might have a protective effect on Cd-induced apoptosis. To confirm the autophagic flux induced by Cd, chloroquine (CQ) which could inhibit lysosome-mediated proteolysis was used [32]. The results showed that Cd-induced upregulation of LC3 II was potentiated by CQ, and the downregulation of P62 was reversed by CQ. These results indicated that Cd induced autophagosome formation (Fig.3E, F).

16

Fig.3. Effect of Cd-induced autophagy on apoptosis in A549 cells. A549 cells were pretreated with or without 3MA and then treated with Cd for 48 hrs. (A) Apoptotic morphological changes were observed in A549 cells by Hoechst 33342 staining. (B) Quantitation of nuclei fragmentation and condensation was done. Each bar represents mean ± SD from three independent experiments. Scale bar: 20 μm (**P<0.01 vs. Control; ##P<0.01 vs. Cd;

&&

P<0.01 vs. Cd treatment for 24 hrs

group). (C) Western blots were performed on the total protein or cytoplasmic protein. (D) GAPDH 17

was used as control. Relative expression of these proteins was expressed as a percentage of GAPDH. Each bar represents mean ± SD from three independent experiments (*P<0.05 vs. Control, **P<0.01 vs. Control; #P<0.05 vs. Cd alone, ##P<0.01 vs. Cd alone). (E) Autophagic flux was determined in A549 cells by treatment with Cd in the presence or absence of CQ (10μM) for 48 hrs. Western blots were performed on the total protein. (F) The relative levels of LC3II and P62 were normalized to β-actin. Each bar represents mean ± SD from three independent experiments (*P<0.05 vs. Control; #P<0.05 vs. Cd alone, ##P<0.01 vs. Cd alone).

To further confirm the protective effect of autophagy on Cd-induced apoptosis, Cd-induced depolarization of MMP was detected. As illustrated in Fig. 4A, B, Cd treatment reduced MMP significantly around 12 to 24 hrs and then significantly rescued during 36-48 hrs compared to 12 to 24 hrs (P<0.01), but still lower than the control group (P<0.01). This result indicated that exposure to Cd caused a depolarization of MMP. After 3MA pretreatment, the rescue of MMP during 36-48 hrs was reduced significantly compared with only Cd-treated cells (P<0.01). This result suggested that Cd-induced autophagy might impose a protective effect on Cd-induced apoptosis via the ability of rescuing the damage to MMP induced by Cd.

18

Fig.4. Effect of autophagy on Cd-induced MMP depolarization in A549 cells. (A) A549 cells were pretreated with or without 3MA and then treated with Cd (2μM) for 0, 12, 24, 36 and 48 hrs. The MMP of A549 cells was analyzed using JC-1. (B) Quantitation of the ratio of red and green fluorescence intensity in A549 cells was done. Each bar represents mean± SD from three independent experiments. Scale bar: 20 μm (**P<0.01 vs. Control; ##P<0.01 vs. Cd; &&P<0.01 vs. 19

Cd treatment for 24 hrs group).

All the above results demonstrated that there was a crosstalk between Cd-induced apoptosis and autophagy in A549 cells. The stress induced by apoptosis might initiate autophagy to protect A549 cells from apoptotic death. 3.3. Atg4B played an important role in Cd-induced autophagy in A549 cells Previously, we were able to demonstrate that Atg4-mediated autophagy plays an important role in Cd-induced cell growth, migration and invasion in A549 cells [26]. In this study, as illustrated in Fig. 5A, B, blocking Atg4B expression by transfection with Atg4B siRNA significantly reduced Cd-induced LC3-II and Beclin1 protein expression and increased Cd-reduced P62 protein expression (P<0.05 and P<0.01). Furthermore, as shown in Fig. 5C, D, Atg4B siRNA decreased the number of the autophagy vesicles (AVs) per viable cells significantly compared to only Cd-treated cells (P<0.01). In order to figure out the effect of Atg4B on autophagy, plasmid Atg4B and empty vector were transfected in A549 cells. It was observed that the expression of LC3-II and Beclin1 increased in the plasmid Atg4B transfected cells compared with the empty vector cells (P<0.05 and P<0.01), whereas the level of P62 protein had a significant decrease in plasmid Atg4B transfected A549 cells as shown in Fig. 5E, F (P<0.01). Meanwhile, plasmid Atg4B transfection increased the number of AVs per viable cells significantly compared to empty vector cells (P<0.01) (Fig. 5G, H). These results suggested that Atg4B played an important role in Cd-induced autophagy in A549 cells.

20

Fig.5. The involvement of Atg4B in Cd-induced autophagy in A549 cells. (A) A549 cells were pretreated with Atg4B siRNA or Control siRNA, and then exposed to Cd for 48 hrs. Western blots were performed on the total protein of untreated and treated cells. (B) GAPDH was used as control. Relative expression of these proteins was expressed as a percentage of GAPDH. Each bar 21

represents mean± SD from three independent experiments (*P<0.05 vs. Si Control, **P<0.01 vs. Si Control; #P<0.05 vs. Si Control+Cd,

##

P<0.01 vs. Si Control+Cd). (C) Transmission electron

microscopy was used for observing AVs in A549 cells. Scale bar: 500 nm. (D) Quantitation of AVs in A549 cells was done. Each bar represents mean± SD from three independent experiments (**P<0.01 vs. Si Control;

##

P<0.01 vs. Si Control+Cd). (E) A549 cells were tranfected with

plasmid Atg4B or empty vector. The protein level of autophagy related proteins was analyzed by Western blot. (F) GAPDH was used as control. Relative expression of these proteins was expressed as a percentage of GAPDH. Each bar represents mean± SD from three independent experiments (*P<0.05 vs. Empty vector;

**

P<0.01 vs. Empty vector). (G) Transmission electron

microscopy was used for observing AVs. Scale bar: 500 nm. (H) Quantitation of AVs in A549 cells was done. Each bar represents mean± SD from three independent experiments (**P<0.01 vs. Empty vector).

3.4. Atg4B also played an important role in Cd-induced apoptosis in A549 cells In this study, as illustrated in Fig. 1A, the expression of Atg4B protein was obviously increased at 12 hrs and increased further at 24 hrs compared with the control group (P<0.01). Cd-induced apoptosis was also around this treatment time. So, experiments were initiated to examine the role of Atg4B in Cd-induced apoptosis in A549 cells. As illustrated in Fig. 6A, B, there was a significant decrease of CL-CASP9 and CL-CASP3 protein level in Cd-treated Atg4B-siRNA cells compared with the only Cd-treated cells (P<0.05 and P<0.01). Bcl-2 was increased and the release of cytochrome c from mitochondria decreased in Cd-treated Atg4B-siRNA A549 cells compared with the only Cd-treated cells (P<0.05). In addition, plasmid 22

Atg4B and empty vector were transfected in A549 cells. It was observed that both the expression of CL-CASP9 and CL-CASP3 increased in the plasmid Atg4B transfected cells compared with the empty vector cells (P<0.01), whereas the level of Bcl-2 protein had a evident decrease in plasmid Atg4B transfected cells as shown in Fig. 6C, D (P<0.05). The release of cytochrome c was increased in the plasmid Atg4B transfected cells compared with the empty vector cells (P<0.05). These results demonstrated that Atg4B played an important role in Cd-induced apoptosis in A549 cells. The role of Atg4B in Cd-induced apoptosis was further examined by TUNEL assay and Hoechst 33342 staining. As demonstrated in Fig. 6E, F, the number of TUNEL-positive cells increased significantly in plasmid Atg4B group compared with empty vector group (P<0.05). Hoechst 33342 staining also showed that the number of pyknotic nuclei in plasmid Atg4B group was significantly increased compared with empty vector group (P<0.01) (Fig. 6G, H). Taken together, these data demonstrated that Atg4B played a critical role in Cd-induced apoptosis in A549 cells.

23

Fig.6. The involvement of Atg4B in Cd-induced apoptosis in A549 cells. (A) Cells were pretreated with Atg4B siRNA or Control siRNA, and then exposed to Cd for 24 hrs. Western blots were performed on the total protein and cytosolic proteins of treated and untreated cells. (B) 24

GAPDH was used as control. Relative expression of these proteins was expressed as a percentage of GAPDH. Each bar represents mean± SD from three independent experiments (*P<0.05 vs. Si Control; #P<0.05 vs. Si Control+Cd, ##P<0.01 vs. Si Control+Cd). (C) Cells were tranfected with plasmid Atg4B or empty vector. The protein level of apoptosis related proteins was analyzed by Western blot. (D) GAPDH was used as control. Relative expression of these proteins was expressed as a percentage of GAPDH. Each bar represents mean± SD from three independent experiments (*P<0.05 vs. Empty vector,

**

P<0.01 vs. Empty vector). (E) TUNEL staining was

used for observing apoptosis in A549 cells. (F) Quantification of the apoptotic index in A549 cells was done. Each bar represents mean± SD from three independent experiments. Scale bar: 20 μm (*P<0.05 vs. Empty vector). (G) Apoptotic morphological changes were observed in A549 cells by Hoechst 33342 staining. (H) The apoptotic index was calculated as the percentage of apoptotic nuclei per total nuclei number per field. Each bar represents mean± SD from three independent experiments. Scale bar: 20 μm (**P<0.01 vs. Empty vector).

Bcl-2 has been well known to localize to the inner mitochondrial and smooth ER membranes [33]. Through localizing to the inner mitochondrial membranes, Bcl-2 retains cytochrome c in the mitochondria and plays an anti-apoptotic role. So, the effect of Atg4B on the localization of Bcl-2 protein was further analyzed. As shown in Fig. 7A, both Cd treatment and overexpression of Atg4B significantly induced Atg4B protein level, and inhibited the expression of Bcl-2 protein in the mitochondria (P<0.05 and P<0.01), while in the cytoplasm Atg4B protein level increased significantly, but Bcl-2 protein level had no significant change (Fig. 7B). Additionally, Atg4B siRNA decreased both mitochondrial and cytoplasm Atg4B protein level, 25

while induced Bcl-2 localization in mitochondria rather than in the cytoplasm in Cd-treated Atg4B-siRNA cells compared with the only Cd-treated cells (P<0.05 and P<0.01) (Fig. 7C, D). The results of immunofluorescence staining in Fig. 7E and 9F also showed that overexpression of Atg4B or treatment of Cd markedly decreased overlap coefficient for Bcl-2 fluorescence and mitochondria. Meanwhile, the percentage of Bcl-2 puncta in mitochondria was dramatically increased in Cd-treated Atg4B-siRNA cells compared with the only Cd-treated cells (P<0.05 and P<0.01). However, under the same condition as above, there was no significant change of the fluorescence of Bcl-2 in the ER (Fig. S1A, B). All these data demonstrated that Atg4B played an important role in Cd-induced apoptosis through decreasing the localization of Bcl-2 in the mitochondria.

26

Fig. 7. Effects of Atg4B on mitochondrial localization of Bcl-2. (A) (B) A549 cells were tranfected with Atg4B plasmid or empty vector, and then treated with or without Cd (2μM) for 24hrs. Mitochondrial and cytosolic proteins were collected separately and the protein level of Bcl-2 and

27

Atg4B was measured by Western blot. VDAC1 and β-actin were used as controls of mitochondrial and cytosolic fractions, respectively. Relative expression of these proteins was expressed as a

percentage of VDAC1 or β-actin. Each bar represents mean ± SD from three independent experiments (*P<0.05 vs. Control,

**

P<0.01 vs. Control). (C) (D) Cells were pretreated with

Atg4B siRNA or Control siRNA, and then exposed to Cd for 24 hrs. Western blots were performed on the mitochondrial and cytosolic protein fractions. Relative expression of these proteins was expressed as a percentage of VDAC1 or β-actin. Each bar represents mean ± SD from three independent experiments (*P<0.05 vs. Si Control; #P<0.05 vs. Si Control + Cd, ##

P<0.01 vs. Si Control + Cd). (E) (F) Immunofluorescence staining was performed to further

determine the effect of Atg4B on mitochondrial localization Bcl-2 in A549 cells. A549 cells transfected with Atg4B siRNA or Atg4B plasmids were treated with or without Cd, and then double stained with Mito-Tracker Green and antibody against anti-Bcl-2 (red), and then with DAPI solution for the *

nucleus staining. Co-localization analysis was measured using the Image-Pro Plus 7.0 ( P<0.05 vs.

Control, **P<0.01 vs. Control, ##P<0.01 vs. Cd). Scale bar: 40 μm.

3.5. Mitochondrial localization of Atg4B and the relationship between MMP and Atg4B in A549 cells To further determine the relationship between mitochondrial dysfunction and Atg4B, the mitochondrial localization of Atg4B and depolarization of MMP were measured. First, the results of immunofluorescence staining in Fig. 8A and 8B showed that the level of Atg4B protein in mitochondria significantly increased when A549 cells were treated with Cd (2μM) for 24hrs, 48hrs, CCCP or tranfected with Atg4B plasmid (P<0.05 or P<0.01). Secondly, the relationship between depolarization of 28

MMP and Atg4B was examined using JC-1 staining. After A549 cells were treated with different concentrations (0.1 and 0.5μM) of CCCP for 24 hrs, the expression of Atg4B was significantly increased at the concentration of 0.5μM compared with that of the control cells (P<0.01) (Fig. 8C, D). This result suggested that the depolarization of MMP had a strong effect on the expression of Atg4B protein. To determine the role of Atg4B on depolarization of MMP, A549 cells were transfected with Atg4B siRNA or plasmid Atg4B. Interestingly, MMP did not change significantly in both Atg4B siRNA and plasmid Atg4B transfected cells compared to control cells (Fig. 8E, F). This result seemed not consistent with the above effects of Atg4B on localization of Bcl-2 and cytochrome c release, but consistent with the attenuated effect of Atg4B-induced autophagy on apoptosis.

29

Fig.8. Mitochondrial localization of Atg4B and the relationship between MMP and Atg4B in A549 cells. (A) A549 cells were treated with or without Cd (2μM) for 24hrs, 48hrs, CCCP or tranfected with Atg4B plasmid. Immunofluorescence staining (40 × 10) was performed to determine

30

mitochondrial localization of Atg4B using Mito-Tracker Green, antibody against anti-Atg4B (red) and DAPI solution for the nucleus staining. (B) Co-localization analysis was measured using the Image-Pro Plus 7.0 (*P<0.05 vs. Control, **P<0.01 vs. Control). Scale bar: 40 μm. (C) The effect of CCCP on

Atg4B expression in A549 cells. (D) GAPDH was used as control. Relative expression of these proteins was expressed as a percentage of GAPDH. Each bar represents mean± SD from three independent experiments (**P<0.01 vs. Control). (E) Cells were exposed to Cd or CCCP, or transfected with Atg4B siRNA or plasmid Atg4B to explicit the relationship between depolarization of MMP and Atg4B. JC-1 was used for observing MMP alteration in A549 cells. (F) Quantitation of the ratio of red and green fluorescence intensity in A549 cells was done. Each bar represents mean± SD from three independent experiments. Scale bar: 20 μm (**P<0.01 vs. Control).

3.6. Bcl-2 played a crucial role in Cd-induced autophagy in A549 cells The Bcl-2 protein is known to inhibit apoptosis by regulating ΔΨm and cytochrome c release needed for activation of caspase-9 [34]. Bcl-2 has also been reported to induce autophagy through blocking Bcl-2-Beclin1 interaction to upregulate Beclin1 [25]. So, in this study, the role of Bcl-2 in Cd-induced autophagy was investigated using Bcl-2 siRNA to knock down Bcl-2 expression and plasmid transfection to over-express Bcl-2. As shown in Fig. 9A, B, Bcl-2 knockdown significantly increased Cd-induced LC3-II and Beclin1, and decreased Cd-reduced P62 protein expression further compared to only Cd-treated cells (P<0.05 and P<0.01). Whereas Bcl-2 over-expression by Bcl-2 plasmid transfection in A549 cells decreased the relative amount of LC3-II and Beclin1, and increased P62 protein 31

expression significantly compared with the empty vector cells (P<0.05) (Fig. 9C, D). These results indicated that Bcl-2 played a negative role in Cd-induced autophagy. 3.7. Crosstalk between Atg4B and Bcl-2 switched apoptosis to autophagy in A549 cells From the above results, both Atg4B and Bcl-2 were involved in Cd-induced apoptosis and autophagy. To verify the regulative mechanism underlying the crosstalk between apoptosis and autophagy induced by Cd, the interaction of Atg4B and Bcl-2 was investigated. Plasmids Atg4B and Bcl-2 were transfected separately or simultaneously in A549 cells. As shown in Fig. 9E, F, over-expression of Bcl-2 reduced Atg4B-transfection-induced LC3, Beclin1, CL-CASP9 and CL-CASP3 (P<0.05 and P<0.01). This result demonstrated that Bcl-2 played a critical role in both Atg4B-induced autophagy and apoptosis. From all above, these results further demonstrated that there was a crosstalk between Atg4B and Bcl-2 to switch apoptosis to autophagy in Cd-treated A549 cells. 3.8. Cd induced interaction of Atg4B and Bcl-2 interrupting the combination of Bcl-2 and beclin1 in A549 cells To further illustrate the regulative relationship between Atg4B and Bcl-2, Co-IP experiment was performed. As shown in Fig. 9G, IP of Bcl-2 with specific anti-Bcl-2 pulled down Atg4B more from Cd-treated A549 cells than from control cells. This indicated that Bcl-2 and Atg4B were bound to each other, and more importantly, Cd treatment promoted the binding of them in A549 cells. IP of Bcl-2 could also pull down Beclin1. Moreover, Bcl-2 pulled down Beclin1 more from control cells than 32

from Cd-treated cells, indicating that treatment with Cd restrained the Bcl-2-Beclin1 interaction. Our data showed for the first time that Atg4B could bind to Bcl-2 interrupting the interactions between Bcl-2 and Beclin1 thus releasing protein Beclin1, which in turn triggered the autophagic cascade.

Fig.9. The involvement of interaction between Atg4B and Bcl-2 in Cd-induced autophagy in 33

A549 cells. (A) Cells were transfected with Bcl-2 siRNA or Control siRNA, and then exposed to Cd for 48 hrs. Western blots were performed on the total protein of untreated and treated cells. (B) GAPDH was used as control. Relative expression of these proteins was expressed as a percentage of GAPDH. Each bar represents mean± SD from three independent experiments (*P<0.05 vs. Si Control, **P<0.01 vs. Control, #P<0.05 vs. Si Control+Cd, ##P<0.01 vs. Si Control+Cd). (C) Cells were transfected with plasmid Bcl-2 or empty vector. The protein level of autophagy related proteins was analyzed by Western blot. (D) GAPDH was used as control. Relative expression of these proteins was expressed as a percentage of GAPDH. Each bar represents mean± SD from three independent experiments (*P<0.05 vs. Empty vector). (E) To confirm the role of Atg4B and Bcl-2 in Cd-induced apoptosis and autophagy, the plasmids of Atg4B and Bcl-2 were transfected together in A549 cells. The protein level of LC3, Beclin1, CL-CASP9 and CL-CASP3 was analyzed by Western blot. (F) GAPDH was used as control. Relative expression of these proteins was expressed as a percentage of GAPDH. Each bar represents mean± SD from three independent experiments (*P<0.05 vs. Control; **P<0.01 vs. Control; $P<0.05 vs. Plasmid Atg4B; $$P<0.01 vs. Plasmid Atg4B). (G) To confirm the effect of Cd on the interactions of Atg4B, Bcl-2 and Beclin1, Co-IP experiment was performed. A549 cells were treated with Cd for 24 and 48 hrs. Whole-cell lysate was generated and precleared with Protein A/G PLUS-Agarose and then mixed with Bcl-2 antibodies or control IgG. The immunoprecipitates were captured on protein A/G PLUS-Agarose and analyzed by Western blot with antibodies against Bcl-2, Atg4B and Beclin1, respectively.

4. Discussion Our results presented here showed for the first time that Atg4B could regulate Bcl-2 protein level and its localization on the mitochondria. The direct binding of 34

Atg4B to Bcl-2 caused the dissociation between Bcl-2 and Beclin1, initiating the crosstalk of apoptosis and autophagy in Cd-treated A549 cells. It had been reported that Cd could induce apoptosis in LLC-PK1 cells [35] and human bronchial epithelial cells [36]. The effect of Cd-induced autophagy has also been detected in human hepatoma cells [37] and PC-12 cells [38]. In this study, we discovered that Cd induced both apoptosis and autophagy in A549 cells, and apoptosis preceded autophagy. Growing evidence confirmed that there was a cooperative relationship between apoptosis and autophagy, and the crosstalk between apoptosis and autophagy is complex [39]. Autophagy plays a dual role in cell survival. Under different conditions, autophagy could delay apoptosis or promote the occurrence of apoptosis. On one hand, persistent stimulate promoted autophagy, leading to cell death [40]. On the other hand, autophagy can promote cell survival and avoid apoptosis [41]. Apoptosis have been evidenced to induce autophagy and autophagy is involved in apoptosis prevention [42]. Our results using apoptosis inhibitor, Z-VAD-FMK and autophagy inhibitor 3MA showed that there was a crosstalk between Cd-induced apoptosis and autophagy. Apoptosis stress initiated autophagy and autophagy attenuated Cd-induced apoptosis and protected A549 cells from apoptotic death. Apoptosis and autophagy are two important processes with complex protein networks. Several inducers of apoptosis are able to activate autophagy, such as p53 and Bcl-2. P53 activation is known to inhibit the activity of mTOR through activation of AMPK [20]. Phosphorylation of Bcl-2 by JNK1 plays major roles in the 35

dissociation of Bcl-2 from Beclin1 and induces autophagy [19]. However, the nature of the relationship between apoptosis and autophagy remains unknown. In our previous study, we found that Atg4B played an important role in Cd-induced autophagy in A549 cells and Cd-induced Atg4B expression was dependent on ROS formation [26]. Mitochondria are well known to be the main source of ROS generation [43, 44] and ROS produced by mitochondria can directly modulate Atg4B [45]. In this study, the expression of Atg4B was significantly increased after CCCP treatment, indicating that depolarization of MMP played a role in the expression of Atg4B. Ni et al. recently reported that the phosphorylation of ATG4B at Ser34 inhibited mitochondrial function, which possibly resulted from the Ser34 phosphorylation-induced enrichment of mitochondrial ATG4B and the repression of F1Fo-ATP synthase activity [46]. Our results demonstrated that Cd increased the level of both mitochondrial ATG4B and cytoplasmic Atg4B, and Atg4B decreased the Bcl-2 protein level, especially in mitochondria but not in cytoplasm. This result suggested that mitochondria-dependent Atg4B expression affect mitochondrial function conversely by decreasing distribution of Bcl-2 in mitochondria. Bcl-2 is a crucial regulator in mitochondria-mediated apoptosis and cancer cells often depend on Bcl-2 to protect themselves from apoptosis [47]. Bcl-2 is believed to be localized in the outer mitochondrial membrane and controls the mitochondrial outer membrane permeabilization (MOMP), retains cytochrome c in the mitochondria through localizing to mitochondria [48]. Apart from this, Bcl-2 is also recognized in ER, the Golgi apparatus, peroxisomes and the nucleus [49]. It is well known that the 36

distribution of Bcl-2 proteins is a dynamic process which determines the cell fate [49]. Many investigations demonstrated that Bcl-2 acts as an anti-apoptotic regulator by localizing mainly to the inner mitochondrial membranes and smooth ER [33]. In this study, we did not found that Atg4B had significant effect on Bcl-2 localizing on ER. Mitochondria-dependent pathway is the main pathway leading to apoptosis. Therefore, the role of ATG4B in Cd-induced apoptosis was investigated. Our results revealed that Atg4B enhanced the release of cytochrome c from the mitochondria to the cytosol, and played a critical role in Cd-induced apoptosis in A549 cells. Interestingly, when A549 cells were transfected with Atg4B siRNA or plasmid Atg4B, the change of MMP was not significant. The reason might be that the Atg4B-induced autophagy rescued Atg4B-induced mitochondrial dysfunction. The results in which the level of MMP decreased significantly after Cd treatment for 12 hrs and then significantly rescued during 36 to 48 hrs further supported that Atg4B-induced autophagy might impose a protective effect on Atg4B-induced damage to MMP. Atg4B is a cysteine protease to reveal a C-terminal glycine which is necessary for conjugation of LC3 proteins to PE and its insertion to autophagosome precursor membranes [50]. Through autophagy regulation, ATG4B participates in physiological and pathological processes including tumorigenesis. However, a recent study has demonstrated that ATG4B can enhance cell proliferation independent of its role in autophagy in colon cancer cells [29]. The regulative effects of Atg4B on Bcl-2 promoted us to hypothesize that the interaction between Atg4B and Bcl-2 underlies the mechanism of Cd-induced switch 37

of apoptosis to autophagy. In this study, we found that Bcl-2 played a downstream role in Cd-induced autophagy in A549 cells. In fact, Bcl-2 has been identified as a possible Atg4D-interacting partner [51]. Our results of immunofluorescence staining demonstrated that both Atg4B and Bcl-2 protein in mitochondria increased after treatments with Cd. Co-IP assay showed that Bcl-2 and Atg4B could bind to each other. Cd treatment promoted the binding of them and interrupted the interactions between Bcl-2 and Beclin1, thus releasing protein Beclin1, which in turn triggered the autophagic cascade. Several studies suggested that the members of the Bcl-2 family could regulate autophagy in addition to their classical function in controlling the pathways of caspase activation [52, 53]. It has been reported that Bcl-2 bound to Beclin1 and thus prevented the interaction between Beclin1 and PI3KC3, the class III PI3K complex to inhibit autophagy [22, 54]. The dissociation of Bcl-2 with Beclin1 is often essential for autophagy in response to stresses [55]. It has been reported that the alteration of Bcl-2 expression could affect the induction of Beclin1 and subsequent autophagy [56]. Therefore, Bcl-2 represents one of the significantly important points of intersection of apoptosis and autophagy. It is also known that Bcl-2 is regulated by a number of post-translational modifications, including phosphorylation, proteolytic cleavage, ubiquitination and proteasomal degradation [57]. These post-translational modifications could destabilize Bcl-2 and result its proteolytic destruction, explaining its downregulations. PPARα was reported to induce Bcl-2 ubiquitination and degradation by the physical 38

interaction of with Bcl2 PPARα [58]. So, our results that Atg4B down regulated Bcl2 might be due to its binding to Bcl2. This would be further investigated in the future. Taken together, in this study we demonstrated for the first time that Atg4B directly bound to Bcl-2 to induce dissociation of Bcl-2 and Beclin1 to realized Cd-induced switch of apoptosis to autophagy in A549 cells. Cd-induced Atg4B serves as a regulator to decrease Bcl-2 protein level and its distribution in mitochondria, which enhanced cell apoptosis. Autophagy becomes evident through the release of Beclin1 from dissociation of Bcl-2-Beclin1 induced by interaction between Atg4B and Bcl-2. Furthermore, Cd-induced apoptosis initiated autophagy and autophagy prevented Cd-induced apoptosis to ensure A549 cell growth and proliferation under Cd stress. Acknowledgments This work was supported by Liaoning Provincial Science Program (2018053000 2) and the National Key Research and Development Program of China (2017YFC17 02006). Conflict of interests The authors declare that there are no conflicts of interest in the present work.

39

References [1] Xia, L.; Chen, S.; Dahms, H. U.; Ying, X.; Peng, X. Cadmium induced oxidative damage and apoptosis in the hepatopancreas of Meretrix meretrix. Ecotoxicology 25:959-969; 2016. [2] Choong, G.; Liu, Y.; Templeton, D. M. Interplay of calcium and cadmium in mediating cadmium toxicity. Chem Biol Interact 211:54-65; 2014. [3] Huff, J.; Lunn, R. M.; Waalkes, M. P.; Tomatis, L.; Infante, P. F. Cadmium-induced cancers in animals and in humans. Int J Occup Environ Health 13:202-212; 2007. [4] Ye, J. L.; Mao, W. P.; Wu, A. L.; Zhang, N. N.; Zhang, C.; Yu, Y. J.; Zhou, L.; Wei, C. J. Cadmium-induced apoptosis in human normal liver L-02 cells by acting on mitochondria and regulating Ca(2+) signals. Environ Toxicol Pharmacol 24:45-54; 2007. [5] Lecoeur, S.; Huynh-Delerme, C.; Blais, A.; Duche, A.; Tome, D.; Kolf-Clauw, M. Implication of distinct proteins in cadmium uptake and transport by intestinal cells HT-29. Cell Biol Toxicol 18:409-423; 2002. [6] Nawrot, T. S.; Martens, D. S.; Hara, A.; Plusquin, M.; Vangronsveld, J.; Roels, H. A.; Staessen, J. A. Association of total cancer and lung cancer with environmental exposure to cadmium: the meta-analytical evidence. Cancer Causes Control 26:1281-1288; 2015. [7] Son, Y. O.; Pratheeshkumar, P.; Roy, R. V.; Hitron, J. A.; Wang, L.; Zhang, Z.; Shi, X. Nrf2/p62 signaling in apoptosis resistance and its role in cadmium-induced carcinogenesis. J Biol Chem 289:28660-28675; 2014. [8] Brama, M.; Politi, L.; Santini, P.; Migliaccio, S.; Scandurra, R. Cadmium-induced apoptosis and necrosis in human osteoblasts: role of caspases and mitogen-activated protein kinases pathways. J Endocrinol Invest 35:198-208; 2012. [9] Olszowski, T.; Baranowska-Bosiacka, I.; Gutowska, I.; Piotrowska, K.; Mierzejewska, K.; Korbecki, J.; Kurzawski, M.; Tarnowski, M.; Chlubek, D. The Effects of Cadmium at Low Environmental Concentrations on THP-1 Macrophage Apoptosis. Int J Mol Sci 16:21410-21427; 2015. [10] Zhang, H.; Li, L.; Wang, Y.; Dong, F.; Chen, X.; Liu, F.; Xu, D.; Fan, Y.; Kapron, C. M.; Ju, L. NF-κB signaling maintains the survival of cadmium-exposed human renal glomerular endothelial cells. Int J Mol Med 38:417-422; 2016. 40

[11] Levine, B.; Yuan, J. Autophagy in cell death: an innocent convict? J Clin Invest 115:2679-2688; 2005. [12] Green, D. R.; Kroemer, G. The pathophysiology of mitochondrial cell death. Science 305:626-629; 2004. [13] Lee, J. Y.; Tokumoto, M.; Fujiwara, Y.; Hasegawa, T.; Seko, Y.; Shimada, A.; Satoh, M. Accumulation of p53 via down-regulation of UBE2D family genes is a critical pathway for cadmium-induced renal toxicity. Sci Rep 6: 21968; 2016. [14] Liu, W.; Dai, N.; Wang, Y.; Xu, C.; Zhao, H.; Xia, P.; Gu, J.; Liu, X.; Bian, J.; Yuan, Y.; Zhu, J.; Liu, Z. Role of autophagy in cadmium-induced apoptosis of primary rat osteoblasts. Sci Rep 6:20404; 2016. [15] Son, Y. O.; Wang, X.; Hitron, J. A.; Zhang, Z.; Cheng, S.; Budhraja, A.; Ding, S.; Lee, J. C.; Shi, X. Cadmium induces autophagy through ROS-dependent activation of the LKB1-AMPK signaling in skin epidermal cells. Toxicol Appl Pharmacol 255:287-296; 2011. [16] Muthukumar, K.; Nachiappan, V. Phosphatidylethanolamine from Phosphatidylserine Decarboxylase2 is Essential for Autophagy Under Cadmium Stress in Saccharomyces cerevisiae. Cell Biochem Biophys 67:1353-1363; 2013. [17] Levine, B.; Kroemer, G. Autophagy in the pathogenesis of disease. Cell 132:27-42; 2008. [18] Rubinstein, Assaf D.; Eisenstein, M.; Ber, Y.; Bialik, S.; Kimchi, A. The Autophagy Protein Atg12 Associates with Antiapoptotic Bcl-2 Family Members to Promote Mitochondrial Apoptosis. Mol Cell 44:698-709; 2011. [19] Pattingre, S.; Bauvy, C.; Carpentier, S.; Levade, T.; Levine, B.; Codogno, P. Role of JNK1-dependent Bcl-2 phosphorylation in ceramide-induced macroautophagy. J Biol Chem 284:2719-2728; 2009. [20] Feng, Z.; Zhang, H.; Levine, A. J.; Jin, S. The coordinate regulation of the p53 and mTOR pathways in cells. Proc Natl Acad Sci U S A 102:8204-8209; 2005. [21] Germain, M.; Nguyen, A. P.; Grand, J. N. L.; Arbour, N.; Vanderluit, J. L.; Park, D. S.; Opferman, J. T.; Slack, R. S. MCL-1 is a stress sensor that regulates autophagy in a developmentally regulated manner. Embo J 30:395-407; 2014.

41

[22] Pattingre, S.; Tassa, A.; Qu, X.; Garuti, R.; Liang, X. H.; Mizushima, N.; Packer, M.; Schneider, M. D.; Levine, B. Bcl-2 antiapoptotic proteins inhibit Beclin 1-dependent autophagy. Cell 122:927-939; 2005. [23] Sinha, S.; Levine, B. The autophagy effector Beclin 1: a novel BH3-only protein. Oncogene 27:137-148; 2008. [24] Pedro, B. S.; Wei, Y.; Sica, V.; Maiuri, M. C.; Zou, Z.; Kroemer, G.; Levine, B. BAX and BAK1 are dispensable for ABT-737-induced dissociation of the BCL2-BECN1 complex and autophagy. Autophagy 11:452-459; 2015. [25] Lian, J.; Wu, X.; He, F.; Karnak, D.; Tang, W.; Meng, Y.; Xiang, D.; Ji, M.; Lawrence, T. S.; Xu, L. A natural BH3 mimetic induces autophagy in apoptosis-resistant prostate cancer via modulating Bcl-2–Beclin1 interaction at endoplasmic reticulum. Cell Death Differ 18:60-71; 2010. [26] Lv, W.; Sui, L.; Yan, X.; Xie, H.; Jiang, L.; Geng, C.; Li, Q.; Yao, X.; Kong, Y.; Cao, J. ROS-dependent Atg4 upregulation mediated autophagy plays an important role in Cd-induced proliferation and invasion in A549 cells. Chem Biol Interact 279:136-144; 2017. [27] Jain, M. V.; Paczulla, A. M.; Klonisch, T.; Dimgba, F. N.; Rao, S. B.; Roberg, K.; Schweizer, F.; Lengerke, C.; Davoodpour, P.; Palicharla, V. R. Interconnections between apoptotic, autophagic and necrotic pathways: implications for cancer therapy development. J Cell Mol Med 17:12-29; 2013. [28] Vezenkov, L.; Honson, N. S.; Kumar, N. S.; Bosc, D.; Kovacic, S.; Nguyen, T. G.; Pfeifer, T. A.; Young, R. N. Development of fluorescent peptide substrates and assays for the key autophagy-initiating cysteine protease enzyme, ATG4B. Bioorgan Med Chem 23:3237-3247; 2015. [29] Liu, P. F.; Leung, C. M.; Chang, Y. H.; Cheng, J. S.; Chen, J. J.; Weng, C. J.; Tsai, K. W.; Hsu, C. J.; Liu, Y. C.; Hsu, P. C. ATG4B promotes colorectal cancer growth independent of autophagic flux. Autophagy 10:1454-1465; 2014. [30] Wu, S.; Su, J.; Qian, H.; Guo, T. SLC27A4 regulate ATG4B activity and control reactions to chemotherapeutics-induced autophagy in human lung cancer cells. Tumor Biol 37:6943-6952; 2016.

42

[31] Akin, D.; Wang, S. K.; Habibzadegahtari, P.; Law, B.; Ostrov, D.; Li, M.; Yin, X. M.; Kim, J. S.; Horenstein, N.; Jr, D. W. A novel ATG4B antagonist inhibits autophagy and has a negative impact on osteosarcoma tumors. Autophagy 10:2021-2035; 2014. [32] Klionsky, D. J.; Abdalla, F. C.; Abeliovich, H.; et. al. Guidelines for the use and interpretation of assays for monitoring autophagy. Autophagy 8:445-544; 2012. [33] Gotow, T.; Shibata, M.; Kanamori, S.; Tokuno, O.; Ohsawa, Y.; Sato, N.; Isahara, K.; Yayoi, Y.; Watanabe, T.; Leterrier, J. F.; Linden, M.; Kominami, E.; Uchiyama, Y. Selective localization of Bcl-2 to the inner mitochondrial and smooth endoplasmic reticulum membranes in mammalian cells. Cell Death Differ 7:666-674; 2000. [34] Lindsay, J.; Esposti, M. D.; Gilmore, A. P. Bcl-2 proteins and mitochondria--specificity in membrane targeting for death. BBA - Mol Cell Res 1813:532-539; 2011. [35] Chen, Z.; Gu, D.; Zhou, M.; Shi, H.; Yan, S.; Cai, Y. Regulatory role of miR-125a/b in the suppression by selenium of cadmium-induced apoptosis via the mitochondrial pathway in LLC-PK1 cells. Chem Biol Interact 243:35-44; 2016. [36] Xu, Y. M.; Wu, D. D.; Zheng, W.; Yu, F. Y.; Yang, F.; Yao, Y.; Zhou, Y.; Ching, Y. P.; Lau, A. T. Proteome profiling of cadmium-induced apoptosis by antibody array analyses in human bronchial epithelial cells. Oncotarget 7:6146-6158; 2016. [37] So, K.-Y.; Oh, S.-H. Prolyl isomerase Pin1 regulates cadmium-induced autophagy via ubiquitin-mediated

post-translational

stabilization

of

phospho-Ser

GSK3αβ

in

human

hepatocellular carcinoma cells. Biochem Pharmacol 98:511-521; 2015. [38] Wang, Q.; Zhu, J.; Zhang, K.; Jiang, C.; Wang, Y.; Yuan, Y.; Bian, J.; Liu, X.; Gu, J.; Liu, Z. Induction of cytoprotective autophagy in PC-12 cells by cadmium. Biochem Biophys Res Commun 438:186-192; 2013. [39] Mariño, G.; Nisosantano, M.; Baehrecke, E. H.; Kroemer, G. Self-consumption: the interplay of autophagy and apoptosis. Nat Rev Mol Cell Biol 15:81-94; 2014. [40] Notte, A.; Leclere, L.; Michiels, C. Autophagy as a mediator of chemotherapy-induced cell death in cancer. Biochem pharmacol 82:427-434; 2011. [41] Bhutia, S. K.; Kegelman, T. P.; Das, S. K.; Azab, B.; Su, Z. Z.; Lee, S. G.; Sarkar, D.; Fisher, P. B. Astrocyte elevated gene-1 induces protective autophagy. Autophagy 107:22243-22248; 2011.

43

[42] Duan, W. J.; Liu, F. L.; He, R. R.; Yuan, W. L.; Li, Y. F.; Tsoi, B.; Su, W. W.; Yao, X. S.; Kurihara, H. Autophagy is involved in the effects of resveratrol on prevention of splenocyte apoptosis caused by oxidative stress in restrained mice. Mol Nutr Food Res 57:1145-1157; 2013. [43] Yang, Y.; Karakhanova, S.; Hartwig, W.; D'Haese, J. G.; Philippov, P. P.; Werner, J.; Bazhin, A. V. Mitochondria and Mitochondrial ROS in Cancer: Novel Targets for Anticancer Therapy. J Cell Physiol 231:2570-2581; 2016. [44] Janda, J.; Nfonsam, V.; Calienes, F.; Sligh, J. E.; Jandova, J. Modulation of ROS levels in fibroblasts by altering mitochondria regulates the process of wound healing. Arch Dermatol Res 308:239-248; 2016. [45] Qiao, S.; Dennis, M.; Song, X.; Vadysirisack, D. D.; Salunke, D.; Nash, Z.; Yang, Z.; Liesa, M.; Yoshioka, J.; Matsuzawa, S. A REDD1/TXNIP pro-oxidant complex regulates ATG4B activity to control stress-induced autophagy and sustain exercise capacity. Nat Commun 6:7014-7027; 2015. [46] Ni, Z.; He, J.; Wu, Y.; Hu, C.; Dai, X.; Yan, X.; Li, B.; Li, X.; Xiong, H.; Li, Y. AKT-mediated phosphorylation of ATG4B impairs mitochondrial activity and enhances the Warburg effect in hepatocellular carcinoma cells. Autophagy 14:685-701; 2018. [47] Singh, K.; Briggs, J. M. Functional Implications of the spectrum of BCL2 mutations in Lymphoma. Mutat Res Rev Mutat Res 769:1-18; 2016. [48] Lithgow, T.; Van, D. R.; Bertram, J. F.; Strasser, A. The protein product of the oncogene bcl-2 is a component of the nuclear envelope, the endoplasmic reticulum, and the outer mitochondrial membrane. Cell Growth Differ 5:411-417; 1994. [49] Popgeorgiev, N.; Jabbour, L.; Gillet, G. Subcellular Localization and Dynamics of the Bcl-2 Family of Proteins. Front Cell Dev Biol 6: 13; 2018. [50] Scherz‐Shouval, R.; Shvets, E.; Fass, E.; Shorer, H.; Gil, L.; Elazar, Z. Reactive oxygen species are essential for autophagy and specifically regulate the activity of Atg4. Embo J 26:1749-1760; 2007. [51] Betin, V. M.; Lane, J. D. Atg4D at the interface between autophagy and apoptosis. Autophagy 5:1057-1059; 2009.

44

[52] He, C.; Zhu, H.; Li, H.; Zou, M. H.; Xie, Z. Dissociation of Bcl-2-Beclin1 complex by activated AMPK enhances cardiac autophagy and protects against cardiomyocyte apoptosis in diabetes. Diabetes 62:1270-1281; 2013. [53] Salminen, A.; Kaarniranta, K.; Kauppinen, A. Beclin 1 interactome controls the crosstalk between apoptosis, autophagy and inflammasome activation: impact on the aging process. Ageing Res Rev 12:520-534; 2013. [54] Kang, R.; Zeh, H. J.; Lotze, M. T.; Tang, D. The Beclin 1 network regulates autophagy and apoptosis. Cell Death Differ 18:571-580; 2011. [55] Yang, J.; Yao, S. JNK-Bcl-2/Bcl-xL-Bax/Bak Pathway Mediates the Crosstalk between Matrine-Induced Autophagy and Apoptosis via Interplay with Beclin 1. Int J Mol Sci 16:25744-25758; 2015. [56] Du, Y.; Ji, X. Bcl-2 down-regulation by small interfering RNA induces Beclin1-dependent autophagy in human SGC-7901 cells. Cell Biol Int 38:1155-1162; 2014. [57] Chitta, K.; Paulus, A.; Caulfield, T. R.; Akhtar, S.; Blake, M. K.; Ailawadhi, S.; Knight, J.; Heckman, M. G.; Pinkerton, A.; Chanan-Khan, A. Nimbolide targets BCL2 and induces apoptosis in preclinical models of Waldenströms macroglobulinemia. Blood Cancer J 4:e260-270; 2014. [58] Gao, J.; Liu, Q.; Xu, Y.; Gong, X.; Zhang, R.; Zhou, C.; Su, Z.; Jin, J.; Shi, H.; Shi, J.; Hou, Y. PPARalpha induces cell apoptosis by destructing Bcl2. Oncotarget 6:44635-44642; 2015.

45

Highlight 

CdCl2 induced both apoptosis and autophagy in A549 cells, and apoptosis preceded autophagy.



Atg4B was involved in both CdCl2-induced apoptosis and autophagy.



Atg4B could bind to Bcl-2 and subsequently promote disassociation of Bcl-2-Beclin1, releasing Beclin1 to activate autophagic pathway

46