iASPP facilitates tumor growth by promoting mTOR-dependent autophagy in human non-small-cell lung cancer

Autophagy serves a critical function in the pathogenesis, response to therapy and clinical outcome in cancers. Although a recent report showed a role of iASPP in suppressing autophagy, its potential activity as a regulator of autophagy has not been investigated in lung cancer. Here we investigated the potential function and molecular mechanism of iASPP in mediating autophagy in human non-small-cell lung cancer. Our data suggested that forced expression of iASPP triggered autophagic flux, while inhibition of iASPP suppressed autophagy at the autophagsome formation stage in vitro. Furthermore, in vivo overexpression of iASPP in SCID/NOD mice promoted tumorigenesis and autophagy, with an increase in the conversion from LC3-I to LC3-II. The effects of iASPP were mediated through activation of mTOR pathway. Finally, cytoplasmic iASPP expression was upregulated in lung cancer patients, and was identified as an independent poor prognostic factor for lung cancer-specific death in patient samples. Taken together, our data showed that iASPP could promote tumor growth by increasing autophagic flux, and iASPP could serve as a poor prognostic factor and a potential therapeutic target in lung cancer.

Non-small-cell lung cancer (NSCLC) accounts for~85-90% of all lung cancers, which remains the leading cause of cancerrelated death worldwide. 1 Given its high prevalence and poor 5-year survival rate of 18% in all stages, 1 it is important to detect new prognostic and predictive markers, as well as novel and more effective treatment options.
The apoptosis-stimulating protein of p53 family (ASPP) stimulates the apoptotic function of p53 upon DNA damage and functions as tumor suppressors. ASPP1 and ASPP2, two members of the ASPP family, can bind to p53 protein and aid the transactivation function of p53 on pro-apoptotic genes, thereby modifying p53-dependent apoptosis responses. 2 In addition, they can also enhance the p53-independent apoptosis by binding to two other members of the family, p63 and p73, without altering the cell cycle. 3 The inhibitor of ASPP (iASPP, encoded by PPP1R13L) functions as oncogene by binding to apoptosis-regulating proteins such as p53, p63, p73, Bcl-2, NF-κB, p65 and so on, resulting in an inhibition of apoptotic cell death typically after DNA damage. 2 Aberrant expression of ASPP1, ASPP2 and iASPP has been detected in many kinds of cancer cells. High expression of iASPP was identified to facilitate prostate cancer progression, 4 resist chemoradiotherapy 5 and be associate with worse disease status and poor survival. 6,7 One recent study for the first time reported that iASPP acts as an autophagy inhibitor in non-transformed keratinocytes. 8 However, little is known about whether and how iASPP affects autophagy activity in cancer cells. As an emerging hallmark of cancer, aberrant autophagy, serves a critical function in the pathogenesis, outcome and response to therapy in a number of cancers. 9 The role of autophagy in cancer has been controversial, with some reports indicating that autophagy suppresses tumor development, whereas other reports suggesting that autophagy contributes to tumor growth and progression by attenuating cellular metabolic stress and resisting therapeutic agent-induced cell death. 10 A better understanding of the molecular mechanism and regulation of autophagy may provide novel therapeutic targets.
Here we explored the role of iASPP in lung cancer. We showed that expression of iASPP was important for modulating autophagy metabolic process in lung cancer cells. In addition, silencing iASPP expression could not only promote the apoptotic signaling but also prevent autophagy by activating mammalian target of rapamycin (mTOR) and promoting autophagosome formation. Moreover, high cytoplasmic expression of iASPP was an independent prognostic factor and was associated with poor survival of patients with NSCLC. Taken together, our results indicate that iASPP may serve as a poor prognostic factor and a potential therapeutic target in NSCLC.

Results
Forced expression of iASPP increases cellular autophagy in H1975 cells. To select appropriate cell lines for iASPP knockdown and iASPP overexpression, we first analyzed the iASPP expression in seven NSCLC cell lines. As shown in Supplementary Figure S1A, among the seven cell lines, H1975 displayed the lowest iASPP expression. Then, we decide to overexpress iASPP in the H1975 cells. Autophagy has pro-or anti-survival effects on cancer cells depending on tumor type or context. To evaluated whether autophagy was induced by iASPP in lung cancer cells, we analyzed the localization and distribution of endogenous LC3 in lung cancer cells by immunofluorescence. We found an increase of LC3 dots, a measure for autophagy, in iASPPoverexpressed H1975 cells compared with the diffuse pattern in control cells (Figures 1a and b, left). Autophagy inducer, Earle's balanced salt solution (EBSS) alone, or late-phase autophagy inhibitor (also cause an increase in LC3-II levels) chloroquine (CQ) alone, or EBSS together with CQ also caused accumulation of the LC3 dots in cells, but the abundance of LC3 dots in iASPP-overexpressed cells was much more than that in the control cells (Figures 1a and b). Furthermore, we monitored the alteration of autophagic flux mediated by iASPP overexpression using immunoblotbased LC3 flux assay. The expression level of LC3-II (the cleaved and lipidated form of microtubule-associated protein 1 light chain 3), an early marker for the formation of autophagosomes, was analyzed. The conversion of LC3-I to LC3-II was augmented in the iASPP-overexpressing cells, but not in the control cells ( Figure 1c, middle panel, lane 2 versus lane 1 and Figure 1d). EBSS alone, CQ alone or EBSS together with CQ caused accumulation of LC3-II in both of the iASPP-overexpressing and the control cells, but the increase was much more significant in the former (Figure 1c  Knockdown of iASPP suppresses cellular autophagy. To further assess the role of iASPP in autophagy, we knocked down iASPP expression by short-hairpin RNA (shRNA) in A549 cells. The expression of iASPP was inhibited efficiently by shiASPP (Figure 2g, upper panel). Immunofluorescence study showed that knockdown of iASPP led to a decrease in LC3 dots as compared with scramble cells (Figures 2a and b, left). In addition, the accumulation of LC3 dots in iASPPknockdown cells resulted from EBSS, CQ or EBSS together with CQ, was much less than that in scramble cells (Figures 2a and b). The cytosolic form of LC3 (LC3-I, 18 kDa) and the cleaved, lipidated and autophagosomebound form (LC3-II, 16 kDa) was identified by western blot. Compared with scramble cells, the conversion of LC3-I to LC3-II was abrogated after iASPP knockdown in all types of cells including those with no treatment, treated with EBSS alone, treated with CQ alone, or treated with EBSS and CQ  Control  +  -+  -+  -+  -iASPP  -+  -+  -+  -+  EBSS  --+  +  --+  +  CQ  ----+  +  +   1  2  3  4  5  We then performed a rescue assay by co-transfecting iASPP expression plasmid with iASPP shRNA followed by EBSS treatment or CQ treatment. Western blot analysis showed that the conversion of endogenous LC3-I to LC3-II was significantly increased in cells co-transfected with iASPP expression plasmid and iASPP shRNA compared with those cells transfected with iASPP shRNA alone (Figure 2e, middle Scramble + - . These data show that iASPP rescued the shiASPP-induced inhibition of autophagy, and suggest that endogenous iASPP has an essential role in cellular autophagy.

Knockdown of iASPP impairs autophagosome formation.
To further investigate the stage at which the autophagy process is inhibited in iASPP-silenced cells, we tested omegasome formation, also known as the phagophore assembly site, which can be labeled by exogenous GFP-DFCP1. A549 cells were co-transfected with GFP-DFCP1 and shiASPP or scramble vectors, and then observed under a fluorescence microscope. Compared with scramble cells, shiASPP-transfected cells displayed abnormally enlarged GFP-DFCP1 puncta ( Figure 2i, lower panel), and the percentage of this kind of cells in the iASPP-silenced cells was much higher than that in the scramble control cells treated with or without EBSS (Figures 2i and j, ***Po0.001), suggesting the accumulation of omegasomes. Collectively, these results indicated that iASPP might regulate the autophagy at the stage of autophagosome formation.  Furthermore, we detected the influence of iASPP on the expression level of crucial transcription factors in the maintenance of self-renewal and pluripotency in stem cells, including OCT4, Nanog and SOX2. Clearly, the expression level of OCT4, Nanog and SOX2 in iASPP-overexpressing cells were increased to 9 times, 15 times and 8 times, respectively, as compared with control cells (Figures 4l and m). Consistently, the expression level of these transcription factors in iASPP knockdown cells were significantly decreased (Figures 4n and o). Collectively, iASPP, as an oncogene, can promote cell proliferation and tumor growth partially by activating crucial transcription factors related with stem cells.
Cytoplasmic iASPP expression is increased in patients with NSCLC and is associated with poor clinical outcome. We assessed the level of iASPP expression in radical resection samples from 134 patients with NSCLC and more than 10 years' clinical follow-up by immunohistochemistry. Similar to our previous study, an increase in cytoplasmic iASPP expression was observed in NSCLC samples compared with adjacent lung epithelium ( Figure 5a and Table 1). Lack of iASSP expression was seen in only 27.6% (37/134) of lung cancer tissues in contrast to 65.7% (88/134) in adjacent non-cancerous tissues ( Table 1, Fisher's exact test for an R × C table P = 0.01).
We further investigated whether increased iASPP expression in NSCLC patients has a prognostic significance. Univariate Kaplan-Meier survival analysis revealed that increased cytoplasmic iASPP expression was associated with an increased risk of lung cancer-specific death in patients undergoing radical surgery (Figure 5b). Overall survival time for patients with high cytoplasmic iASPP expression was significantly shorter than those with low cytoplasmic iASPP expression (Figure 5b, median survival time was 53.1 versus 87.7 months, P = 0.043). Furthermore, cytoplasmic iASPP expression in lung cancer patients was identified as an independent prognostic factor for cancer-specific death by Cox regression multivariate analysis ( Table 2, HR = 1.714, P = 0.019). These results suggest that increased expression of cytoplasmic iASPP has clinical significance by conferring an increased risk of lung cancer-specific death.

Discussion
iASPP is an inhibitory member of the ASPP family, which has been reported to be overexpressed as an oncogene in several cancers, such as breast cancer and hepatocellular cancer. [11][12][13] We have shown previously that iASPP was overexpressed in human NSCLC and that downregulation of iASPP inhibited the proliferation and colony formation of lung cancer cells by activating p53/p21/PUMA pathway in vitro. 14 Here we have reinforced the notion of a constitutive cancerpromoting function of iASPP in vivo. Moreover, for the first time we showed that iASPP could regulate cell autophagy in lung cancer cells.
Autophagy, a highly conserved catabolic process involving the delivery of cytoplasmic components to lysosomes for degradation, has a pivotal role in the maintenance of the cellular environment. The role of autophagy in cancer has been controversial. It has been shown that iASPP-silenced keratinocytes displayed activation of autophage and protection from apoptosis. 8 However, our findings suggest that overexpression of iASPP triggered autophagic flux followed by is totally opposite to that in cancer cells. Non-transformed cells treated with either iASPP-specific siRNA or shRNA reduced levels of apoptosis, 8,15 whereas silencing of iASPP in several cancer cell lines promoted apoptosis. 12,13,[16][17][18][19][20] Therefore, unlike keratinocytes, knockdown of iASPP did not induce but rather prevented autophagy in cancer cells. Autophagy has context-dependent functions in different stages of cancers. As a suppressor pathway, autophagy prevents tumor initiation; but as a survival pathway, autophagy contributes to tumor growth and progression by attenuating cellular metabolic stress and resisting therapeutic agent-induced cell death. 10 Obviously, our data show that autophagy induced by forced expression of iASPP contributed to cancer promotion in NSCLC. mTOR is a serine/threonine kinase that has an important role in regulating cellular homeostasis and metabolism. The mTOR protein consists of at least two distinct multi-protein complexes, namely mTOR complex 1 (mTORC1) and mTOR complex 2 (mTORC2). 21 The mTORC1 integrates stimulating signals and then phosphorylates p70 S6 kinase 1 (S6K1) and eukaryotic initiation factor 4E-binding protein 1, which results in cell growth, activation of translation and ribosome biosynthesis. Up to date, mTORC1 is the best-characterized regulator of autophagy. It has been shown that rapamycin and its analogs could inhibit the activity of mTORC1. 22 For the first time, our study revealed phospho-mTORC1 (Ser-2448) and its downstream effector ribosomal S6 kinase (p-p70S6K ) protein were increased after depletion of iASPP, indicating that increased mTORC1 activity might contribute to the antiautophagic effect of iASPP silencing. Rapamycin treatment could rescue mTOR activation, and at least partially resist antiautophagy and anti-proliferation effect in shiASPP cells. Furthermore, we observed the accumulation of LC3-labeled structures, enlarged DFCP1 structures and LC3 lipidationrelated proteins, including Atg3, Atg7 and Atg12-Atg5 conjugate, in iASPP-silenced cells. Thus, depletion of iASPP function may impair autophagosome formation. Collectively, knockdown of iASPP might block autophagy by  Autophagy and apoptosis have significant roles in cell metabolism and survival. However, the interaction between these two processes is still far from being completely understood. Zhou et al 23 found that autophagy postponed apoptotic cell death through Bad-Beclin1 interaction. In our previous study, iASPP depletion suppressed the proliferation of lung cancer cells by stimulating apoptosis through activation of p53/p21/PUMA pathway. 14 Here our data revealed that higher expression of phosphorylated Bad (Ser112) in shiASPP cells than that in control cells (Supplementary Figure S1C). In addition, increased levels of activated caspase-3, a crucial enzyme in the apoptotic process, and PARP cleavage by caspases, a marker of apoptotic cell death, 21 were both identified in iASPP-silencing cells, compared with control cells (Supplementary Figure S1D). Taken together, our data indicated that cell apoptosis by iASPP silencing may be associated with autophagy.
Furthermore, we showed that iASPP could activate OCT4, Nanog and SOX2, which are crucial transcription factors in the maintenance of self-renewal and pluripotency in stem cells. To our best knowledge, this is the first report that iASPP could activate the expression of these stem cell markers. These data are in keeping with the reports that autophagy contributes to the enrichment and survival of cancer stem cells. 24,25 It has been reported that mTORC1 signaling is essential for the regulation of hematopoietic stem cell quiescence, 26 but the effect of iASPP on cancer stem cell properties in NSCLC still needs to be further elucidated.
Moreover, we found that iASPP was identified as one of the independent prognosis factors in patients with NSCLC. Activation of autophagy is a hallmark in tumor cells treated with chemotherapy. Autophagy is a protective mechanism of tumor cells induced by stress to avoid low nutrition, ionizing radiation and chemotherapy. Ren et al 27 found that upregulation of autophagy had a major role in cisplatin resistance of lung adenocarcinoma. Here we show that iASPP was upregulated in patients with NSCLC and was associated with increased autophagy and poor clinical outcome. In addition, we found that overexpression of iASPP induced higher levels of expression of stem cell-related transcription factors, suggesting iASPP may have an important role in chemotherapy resistance of lung cancer cells, which may at least partially explain its association with a poor prognosis.
In summary, our results show that iASPP induces autophagy and facilitates tumor growth in lung cancer. iASPP regulates the autophagosome formation through mTOR signaling. We also found that iASPP expression is increased in patients with NSCLC and is associated with poor prognosis. Our findings provide new insights into the function of iASPP and suggest that iASPP may serve as a new therapeutic target for patients with NSCLC.

Materials and Methods
Cell lines and human tissue samples. Human NSCLC cell lines, A549, H1975 and H1299, were originated from American Type Culture Collection (Manassas, VA, USA) and cultivated in RPMI 1640 medium supplemented with 10% fetal bovine serum, 100 U/ml penicillin and 100 mg/ml streptomycin (Invitrogen, Grand Island, NY, USA) in a humidified atmosphere of 5% CO 2 at 37°C.
Lung cancer specimens (n = 134) and matched adjacent normal tissues were obtained from patients who were treated at Peking University Cancer Hospital from 2000 to 2007. Prior written and informed consent was obtained from all patients or their families. The study was approved by the Research and Ethical Committee of Peking University Cancer Hospital. Clinicopathologic characteristics of the tumors were defined according to the tumor-node-metastasis staging system of the Union for International Cancer Control (Supplementary Table S1). As quality control, we performed Kaplan-Meier survival analysis based on lymph node metastasis status or tumor thrombus, which is commonly considered to be associated with survival of patients with lung cancer. As shown in Supplementary Figure S1E, the median survival of patients with lymph node metastasis or tumor thrombus was significantly shorter than those without lymph node metastasis or tumor thrombus.
RNA isolation and quantitative real-time PCR. Total RNA was isolated from cells using the Trizol reagent (Invitrogen, Carlsbad, CA, USA). First-strand cDNA was synthesized from total RNA using oligodT (15) primers using moloney murine leukemia virus reverse transcriptase (Invitrogen). Quantitative real-time PCR was performed using SYBR Green PCR Master Mix (Toyobo, Osaka, Japan) on ABI7500 System (Applied Biosystems, Foster City, CA, USA). The expression of the target gene was normalized to that of glyceraldehyde 3-phosphate dehydrogenase (GAPDH). The expression level of target gene was calculated by the 2 − ΔCt method, where ΔCt = Ct (target gene)-Ct (GAPDH). The sequences of primers were listed in Supplementary Table S2.
Plasmid construction and cell transfection. The iASPPoverexpressing vector was purchased from OriGene (RC227121, Rockville, MD, USA) and is named as pCMV-iASPP. Cells were transfected with pCMV-iASPP plasmid followed by 500 mg/ml G418 screening to obtain cells overexpressing iASPP. The lentivirus RNAi vector containing the sequence of shRNA for the iASPP target gene, pGCL-GFP/U6-shiASPP, was purchased from Genchem Company (Shanghai, China). The target sequence of iASPP was 5′-AACACATGGATCTGAAGCAGA-3′, and the sequence of 5′-AATGTACTGCGCGTGGAGA-3′ was used as scramble shRNA control. The lentiviral shuttle vectors were transfected into 293T cells together with lentiviral helper plasmids pHelper1.0 and pHelper2.0 to generate lentiviruses and then infect cancer cells. GFP-LC3 and GFP-DFCP1 plasmids were kindly provided by Professor Yingyu Chen from Peking University Health Science Center. All plasmids were confirmed by sequencing.
Induction and inhibition of autophagy. Cellular autophagy was induced by nutrient deprivation through incubation in EBSS (Sigma, St. Louis, MO, USA). Autophagy inhibition was achieved by treating cells with 10 μM of CQ (Sigma), a lysosomal inhibitor. Rapamycin (RAPA, a specific mTOR inhibitor, Sigma) was used at 0.1 nM concentration to block mTOR activity.
Immunofluorescence. Cells treated with different reagents were seeded on glass coverslips and grown to confluence. After 24 h treatment, cells were fixed with formaldehyde for 5 min, permeabilized in 0.5% Triton X-100 for 10 min and then incubated with primary antibody at 37°C for 1 h. To observe the endogenous LC3 puncta, cells were exposed to LC3 antibody (Sigma). An appropriate fluorescence-labeled secondary antibody with DAPI (1 μg/ml) was then added and incubated at 37°C for 1 h. Cells transfected with GFP-DFCP1 plasmids were directly observed by fluorescence microscopy. The number of LC3 puncta per cell and the percentage of enlarged GFP-DFCP1 structures were counted in five non-overlapping fields in three independent experiments. Glass coverslips were mounted in 90% glycerol medium and photographed with laser scanning confocal microscope (Leica).
Western blot analysis. Cells were lysed with RIPA buffer containing protease inhibitor cocktail (Roche, Basel, Switzerland). Protein concentration was assessed with a Bio-Rad Protein Assay Kit (Bio-Rad Laboratories, Hercules, CA, USA). Equal amounts of proteins from each sample (30 μg) were separated by 10 or 12% SDS-PAGE and transferred to polyvinylidene difluoride membrane (Merck Millipore, Billerica, MA, USA). The membrane was blocked for 1 h in PBS containing 5% nonfat milk. For detection with primary antibodies (anti-iASPP: Rockland; anti-GAPDH: Santa Cruz Biotechnology, Dallas, TX, USA; anti-p-mTOR and anti-ATG3: Cell Signaling Technology, Danvers, MA, USA; anti-Beclin1 and anti-ATG7: Abcam; anti-LC3B, anti-GFP, anti-mTOR, anti-p70S6K and anti-p-p70S6K: Sigma), the membranes were incubated with the appropriate antibody at 4°C overnight, followed by incubation with HRP-conjugated secondary antibodies, and then developed using ECL western blot detection reagents (PerkinElmer, Waltham, MA, USA).
Cell viability assay. In all, 5 × 10 3 cells/well were seeded in 96-well plates containing 100 μl of culture medium. After rapamycin treatment, 10 μl of solution from Cell Counting Kit-8 (Dojindo Molecular Technologies, Kumamoto, Japan) was added to each well. These plates were continuously incubated for 2 h in a humidified CO 2 incubator at 37°C and then the intensity was measured at a wavelength of 450 nm. Absorbance values were presented as the mean ± S.E. M. of the mean value from three independent experiments for each treatment. The absorbance of untreated cells was considered to be 100%.
Statistical analysis. For continuous variables, data were presented as mean ± S.E.M. of the mean value from three independent experiments and analyzed by the Student's t-tests to determine the difference between two independent groups. Otherwise, data were analyzed using methods as indicated in the text. A two-sided P-value of 0.05 was considered to be statistically significant. All statistical data were performed by SPSS 13.0 (SPSS Inc., Chicago, IL, USA).

Conflict of Interest
The authors declare no conflict of interest.