Molecular Switch Role of Akt in Polygonatum odoratum Lectin-Induced Apoptosis and Autophagy in Human Non-Small Cell Lung Cancer A549 Cells

Polygonatum odoratum lectin (POL), isolated from traditional Chinese medicine herb (Mill.) Druce, has drawn rising attention due to its wide biological activities. In the present study, anti-tumor effects, including apoptosis- and autophagy-inducing properties of POL, were determined by a series of cell biology methods such as MTT, cellular morphology observation, flow cytometry, immunoblotting. Herein, we found that POL could simultaneously induce apoptosis and autophagy in human non-small cell lung cancer A549 cells. POL initiated apoptosis through inhibiting Akt-NF-κB pathway, while POL triggered autophagy via suppressing Akt-mTOR pathway, suggesting the molecular switch role of Akt in regulating between POL-induced apoptosis and autophagy. Moreover, ROS was involved in POL-induced inhibition of Akt expression, and might therefore mediate both apoptosis and autophagy in A549 cells. In addition, POL displayed no significant cytotoxicity toward normal human embryonic lung fibroblast HELF cells. Due to the anti-tumor activities, POL might become a potent anti-cancer drug in future therapy, which might pave the way for exploring GNA-related lectins into effective drugs in cancer treatment.


Introduction
Lectins are designated as carbohydrate-binding proteins that widely exist in animals, plants and microorganisms, and they could bind carbohydrates, agglutinate cells or precipitate polysaccharides and glycoconjugates [1]. Rapid progress has been achieved in isolation and characterization of plant lectins, and recently the classification of lectins has been emended from 7 families into 12 families [2][3][4].
Of note, cancer is associated with programmed cell death (PCD), which plays important roles in homeostasis preservation, cellular differentiation, growth control, cell defense and etc., jointly sealing the ultimate fate of cancer cells [5]. Generally, there are two main types of PCD, referring to apoptosis and autophagy. Autophagy is an evolutionarily conserved cellular mechanism for clearance of damaged or superfluous macro-complexes and organelles in eukaryotic cells, which leads to either pro-survival or pro-death effects [6]. Apoptosis, which can be regulated by numerous molecular signaling pathways, is mainly targeted for tumor suicide. Autophagy and apoptosis bear distinct morpholog-ical characteristics and physiological process, however, there still exist intricate interrelationships between them [7].
Polygonatum odoratum (Mill.) Druce, a typical representative of the Liliaceae family, is an important traditional Chinese herbal medicine owing to its wide varieties of biologically active compounds. Polygonatum odoratum lectin (POL), isolated from Polygonatum odoratum (Mill.) Druce, is a mannose-binding specific Galanthus nivalis agglutinin (GNA)-related family lectin, and exerts remarkable growth-inhibition effects against A375 cells [8]. Previous report has demonstrated that POL induced L929 cell apoptosis through both death-receptor and mitochondrial pathways, as well as amplified TNFa-induced L929 cell apoptosis [9]. However, whether POL could simultaneously induce apoptosis and autophagy in cancer cells are still in their infancy. And hitherto, only Polygonatum cyrtonema lectin (PCL) can simultaneously induce both apoptosis and autophagy in human melanoma A375 cells [10] and L929 cells [11]. Therefore, the apoptosis-and autophagy-inducing effects of POL needs to be explored.

Cell culture
Lung adenocarcinoma A549 cell lines were purchased from American Type Culture Collection (ATCC, Manassas, USA), while normal human embryonic lung fibroblast HELF cell lines were purchased from cell bank (Chinese Academy of Sciences, Shanghai, China). Human non-small cell lung cancer A549 cells were cultured in RPMI-1640 medium (Gibco), containing 10% FBS, 100 mg/mL streptomycin (Invitrogen), 100 U/mL penicillin (Invitrogen), and maintained at 37uC with 5% CO 2 at a humidified atmosphere. And human embryonic lung fibroblast HELF cells, used as corresponding control group, were cultured in Dulbecco minimal essential medium (Gibco) containing the same materials.

Growth inhibition assay
A549 cells were seeded into 96-well plates and cultured for indicated time, and the cytotoxic effects of various concentration of POL were performed as previously described [14]. Absorbance at 570 nm was measured with a spectrophotometer (Model 3550 Microplate Reader, Bio-Rad).

Mannose inhibition assay
MTT assay was determined as mentioned above except that lectins were preincubated at 37uC for 30 min with different types of mannoses, such as mannose a (1,3), mannose a (1,6) and mannose a (1,3:1,6), which can completely inhibit the hemagglutinating activity at different concentrations of POL.

Apoptosis assay
A549 and HELF cells were treated with PBS and 23 mg/mL POL for 12 h, 24 h, 36 h and 48 h. Cell morphology was imaged by phase contrast microscopy (Leica, Wetzlar, Germany) as well as fluorescence microscopy (Olympus, Tokyo, Japan) with hoechst 33342 staining. Then, total of 6610 5 A549 cells/well were plated in 6-well plates and incubated for 24 h, and treated with PBS or 23 mg/mL POL for another 12 h, 24 h, 36 h or 48 h incubation. Harvested cells were analyzed by flow cytometry assay as described before [15]. In order to further classify early and late apoptosis in POL-treated A549 cells, Annexin V-FITC/PI double staining assay was performed by Annexin V-fluorescein isothiocyanate (FITC) apoptosis detection kit according to the manufacturer's instructions (BD Pharmingen, San Diego, CA, USA).

Autophagy assay
After incubation with 23 mg/mL POL for indicated time, A549 and HELF cells were cultured with 0.05 mM MDC at 37uC for 60 min. The fluorescence intensity of cells was analyzed by flow cytometry. In addition, A549 cells incubated with PBS or 23 mg/ mL POL for indicated time were stained with acridine orange (10 ng/mL) for 15 min and subjected to flow cytometry. Cells were respectively transfected with Beclin1 siRNA, LC3 siRNA and control siRNA at 33 nM using Lipofectamine 2000 (Invitrogen) according to the manufacturer's instructions. The transfected cells were used for subsequent experiments 24 h later.

Caspase assay
Apoptosis was assessed by measuring the activity of caspase-3 by using the caspase-3 activity kit (Beyotime Institute of Biotechnology, Haimen, China) following the manufacturer's instructions.

Detection of mitochondrial membrane potential
Mitochondrial membrane potential was measured using the fluorescent dye rhodamine-123 as described previously [15]. The fluorescence intensity of A549 cells interfered with 23 mg/mL POL for 12 h, 24 h, 36 h and 48 h was analyzed by flow cytometry.

Determination of intracellular ROS
After treatment with various concentration of POL for the indicated time periods, A549 cells were incubated with 10 mM 29,79-dichlorofluorescein diacetate (DCFH-DA) for 60 min at 37uC. The fluorescence intensity of A549 cells was measured by flow cytometry with an excitation wavelength of 480 nm and an emission wavelength of 525 nm. Intracellular GSH content and GPX enzyme activity were measured according to the manufacturer's instructions.

Transfections
A549 Cells were plated in 60-mm dishes with RPMI 1640 containing 10% FBS and 1% penicillin-streptomycin reached to a density of 1610 6 cells/dish. Then A549 cells continuously cultured in PRMI 1640 without 10% FBS and 1% penicillin-streptomycin for another 2 h. In order to form plasmid-Lipofectamine LTX complex, plasmids were mixed with 25X diluent Lipofectamine LTX, and incubated for 20 min in room temperature. Then transfections were respectively conducted in OptiMEM reducedserum medium (Invitrogen) containing 1 mg/mL DNA, 0.1% PLUS reagent and 0.3% Lipofectamine LTX transfection reagent, and added the medium with plasmid in 60-mm dishes for another 48 h culture of A549 cells. Finally, the transfection efficiency were detected by Western Blot analysis.

Western blot analysis
A549 cells were treated with 23 mg/mL POL for indicated time, and subsequently both adherent as well as floating cells were collected. Western blot analysis of cell lysates was performed as previously mentioned [16,17].

Statistical analysis of the data
All data were expressed as mean6SEM from at least three independent experiments. Statistical comparisons were made by two-way ANOVA and one-way ANOVA with Bonferroni post hoc test. A p value#0.05 was considered as significant.

Molecular modeling of POL
Lectins from GNA-related family bear different amino acid sequence constitution, but they all exhibited similar threedimensional structure. GNA-related family lectins bear strict mannose-binding activity, and conserved amino acid motif of 'QXDXNXVXY' plays important roles in mannose recognition. POL possessed three conserved 'QXDXNXVXY' subdomains, and therefore, POL was strict mannose-binding lectin. As shown in Fig. 1A, the three-dimensional structure of POL was built, and the monomer of POL exhibited a b-barrel structure comprising three subdomains (I, II, III), each subdomain is made up of three or four strands of antiparallel b sheet interacting by V loop.

Cytotoxic effects of POL on A549 cells
The growth inhibitory effects of POL were evaluated by MTT assay on human cancer cells, including colon carcinoma Caco-2 cells, cervix carcinoma HeLa cells, non-small cell lung cancer A549 cells, hepatocellular carcinoma HepG 2 and 7721 cells, breast carcinoma MCF-7 cells, melanoma A375 cells, and the IC 50 values were 47 mg/mL, 30 mg/mL, 23 mg/mL, 42 mg/mL, 38 mg/mL, 50 mg/mL and 26 mg/mL, respectively (Fig. 1B). This result showed that POL was more sensitive to non-small cell lung cancer A549 cells, and hence, A549 cells were chosen for further exploration.
POL induced A549 cell death in a dose-and time-dependent manner, and 0 mg/mL to 100 mg/mL POL exerted potent inhibitory effects on the growth of A549 cells (Fig. 1C). After 24 h incubation with 23 mg/mL POL, the inhibitory rate reached nearly to 50%.

The correlation between sugar-binding and antiproliferative activity of POL
The mannose inhibition assay was performed to determine the relationship between sugar-binding activity of POL and its antiproliferative property. Different concentration of POL was preincubated with mannose a (1,3), mannose a (1,6) and mannose a (1,3:1,6) for 30 min, respectively. Due to the mannose-binding activity of POL, all three types of mannoses completely inhibited the hemagglutinating activity of POL. As shown in Fig. 1D, the growth inhibitory effects of various concentration of POL were lost at the corresponding level with the lose of mannose-binding activity, indicating that there may be a close link between the hemagglutinating activity and anti-proliferative property of POL.

POL induces apoptosis in A549 cells
Marked apoptotic morphological changes as membrane blebbing, cell volume reduction and rounding, were observed in 23 mg/mL POL-treated A549 cells by phase contrast microscopy ( Fig. 2A). Besides, considerable DNA fragmentation in nuclei was observed in POL-induced A549 cells, while the PBS-treated control group showed normal, round and homogeneously stained nuclei (Fig. 2B). However, 23 mg/mL POL did not result in apparent apoptotic morphology in non-cancerous HELF cells after 24 h or 48 h treatment ( Fig. 2C and Fig. 2D). These results suggested that POL might selectively induce A549 cells apoptosis but could not trigger apoptosis of normal HELF cells. In addition, as shown in Fig

POL triggers autophagy in A549 cells
In order to detect the formation of autophagic vacuoles, MDC fluorescent intensity of 23 mg/mL POL-treated A549 and HELF cells for 24 h or 48 h were analyzed by flow cytometry. As shown in Fig. 3A, autophagic vacuoles were formed in A549 cells that was featured by increased MDC fluorescent intensity, indicating the presence of autophagic cell death. However, in Fig. 3B, MDC fluorescent intensity of HELF cells treated with POL was not increased neither after 24 h nor 48 h, suggesting autophagic vacuoles were not formed in POL-induced HELF cells. We could include that POL might selectively induce autophagic cell death in A549 cells but not in HELF cells. By further using acridine orange staining, the formation of acidic vesicular organelles (AVO) in POL-triggered A549 cells was demonstrated (Fig. 3C). Since Beclin1 enhancement and transformation of LC3I to LC3II were the characteristics of autophagy, the expressions of Beclin1 and LC3 in POL-induced A549 cells were detected. As shown in Fig. 3D, POL induced the time-dependent enhancement of Beclin 1 level and accumulation of the LC3-II in A549 cells. Subsequently, the expressions of Beclin1 and LC3 were knocked down by using specific siRNAs (Fig. 3E). Transfection with Beclin1 or LC3 siRNA decreased POL-induced cell growth inhibition (Fig. 3F). All these results indicated that POL induced autophagy in A549 cells.

POL Induces caspase-dependent apoptosis in A549 cells
As shown in Fig. 4A, after autophagic inhibitor 3-MA treatment, 23 mg/mL POL significantly enhanced caspase-3 activity in a time-dependent manner, indicating POL-induced apoptosis occurs through the activation of common apoptotic executors such as caspase-3. Moreover, Western blot analyses after autophagy was inhibited showed that after treatment with 23 mg/ mL POL for indicated time, caspase-3 and -9 were cleaved, since procaspase-3 and -9 were decreased while cleaved caspase-3 and -9 were increased (Fig. 4B a). Also, up-regulation of pro-apoptotic Bax and Bid proteins as well as down-regulation of anti-apoptotic Bcl-2 and Bcl-X L proteins were observed after POL treatment. Besides, 23 mg/mL POL led to cleavage of PARP from an 116 kDa band to an 89 kDa fragment with increasing time (Fig. 4B b), and cytochrome c was released from mitochondria into cytoplasm after POL treatment (Fig. 4B c). As shown in Fig. 4C and Fig. 4D, when autophagy was suppressed by 3-MA, POL decreased the rhodamine 123 fluorescence intensity in a timedependent manner. Furthermore, cleavage of caspase-3 and -9 could be rescued in the presence of 10 mM z-VAD-fmk in POLinduced A549 cells (Fig. 4E). And, Sub-G 1 cell proportion analysis and Annexin V/PI double staining were also demonstrated that caspase inhibitor z-VAD-fmk could protect A549 cells from apoptotic cell death ( Fig. 4F and Fig. 4G). All these results clearly indicate that POL induced apoptosis through mitochondrialmediated pathway in A549 cells.

Inhibition of autophagy partially reduces POL-induced apoptosis
As shown in Fig. 5A, after autophagy was suppressed, POLinduced cell growth-inhibition was significantly decreased. Inhibition of autophagy by 3MA or knockdown of Beclin1 by pretreatment with Beclin1 siRNA (Fig. 5B) decreased the ratio of MDC positive cells (Fig. 5C a), indicating that autophagy was suppressed by 3MA and Beclin1 siRNA. Besides, depressing of POL-induced autophagy inhibited POL-induced apoptosis in A549 cells, which were featured by reduction of sub-G 1 phage DNA content (Fig. 5C b), as well as early and late apoptotic portion (Fig. 5C c). Consistently, suppression of autophagy by either 3MA or Beclin1 siRNA also led to decrease in PARP cleavage and LC3-II expression in POL treated A549 cells (Fig. 5D and Fig. 5E). All these results indicated that inhibition of autophagy partially reduces POL-induced apoptosis in A549 cells.
To further verify whether these proteins are participated in POL-induced apoptosis or autophagy, morphology detection and flow cytometry assay were applied. As shown in Fig. 6C, overexpression of Akt reversed POL-induced cell death, while overexpression of NF-kB and mTOR still partially maintained POLinduced cell death. It is indicated that Akt might be in charge of POL-induced both apoptosis and autophagy in A549 cells. Furthermore, Akt over-expression resulted in the decrease of SubG 1 portion (Fig. 6D a), early apoptotic (Annexin V+/PI2) and late apoptotic (Annexin V+/PI+) cells (Fig. 6D b), MDC fluorescent intensity (Fig. 6D c) as well as acidic vesicular organelles (AVO) formation (Fig. 6D d). These results demonstrated that over-expression of Akt could reverse POL-induced apoptosis and autophagy simultaneously, indicating Akt protein might involve in both POL-induced apoptosis and autophagy. However, over-expression of NF-kB only resulted in suppression of apoptotic features, including significantly decreased portions of SubG 1 (Fig. 6D a) as well as early and late apoptotic cells (Fig. 6D  b). While, over-expression of mTOR only reversed POL-induced autophagy, as characterized by reduced fluorescent intensity of MDC (Fig. 6D c) and AO (Fig. 6D d). Taken together, we can conclude that POL induced apoptosis through suppressing Akt-NF-kB pathway while initiated autophagy via inhibiting Akt-mTOR pathway. Akt drives the switch between autophagy and apoptosis in POL-treated A549 cells.

POL-induced molecular switch role of Akt is ROSdependent
ROS levels were detected using ROS-detecting fluorescent dye DCFH-DA. It is demonstrated that ROS accumulation was enhanced in POL-induced A549 cells in a dose-and timedependent manner (Fig. 7A). Furthermore, incubation of A549 cells with 23 mg/mL POL led to the reduction of GSH content and GPX enzyme activity in a time-dependent manner, suggesting that POL abrogated the GSH antioxidant system and finally resulted in massive ROS accumulation in A549 cells (Fig. 7B). And, this 23 mg/mL POL-induced increase of ROS could be inhibited by pretreatment with the ROS scavenger, NAC (Fig. 7C). Moreover, this increase could also be inhibited by pretreatment with apoptosis inhibitor z-VAD-fmk or autophagy inhibitor 3-MA (Fig. 7D). These results demonstrated that ROS might be implicated in POL-induced both apoptosis and autophagy.
Complement with these findings, ROS inhibitor NAC could also reverse POL-induced cytotoxicity (Fig. 7E), and suppress POL-induced reduction of phosphorylated-Akt, phosphorylated-NF-kB and phosphorylated-mTOR levels, the cleavage of caspase 3, and the accumulation of LC3II (Fig. 7F). From mentionedabove results we can inferred that ROS was involved in POLinduced inhibition of Akt expression, and might therefore mediate both apoptosis and autophagy.
In order to further detect whether POL-induced ROS generation was associated with POL-induced apoptosis and autophagy, SubG 1 portion (Fig. 7G a), early apoptotic (Annexin V+/PI2) and late apoptotic (Annexin V+/PI+) cell portion (Fig. 7G b), MDC fluorescent intensity (Fig. 7G c) were determined by flow cytometry. As expected, POL-induced ROS generation was not only implicated in POL-induced apoptosis, but also in POL-induced autophagy.

Discussion
In the present study, POL bears different anti-neoplastic effects toward different cancer cells, of which POL was more sensitive to non-small cell lung cancer A549 cells with IC 50 value at 23 mg/mL (24 h). However, POL exhibited neither apoptosis-nor autophagy-inducing effects toward normal HELF cells, suggesting POL bears a greater susceptibility to the malignant cells. These results indicated that POL bears the potential to be used in the treatment of different types of cancer, but further in vivo xenograft study is still needed to evaluate its anti-neoplastic effects.
Plant lectins bear affinity toward different carbohydrates, herein, three types of mannoses completely suppressed growth inhibitory effect of POL. These results suggested that there may be a close link between the hemagglutinating activity and antiproliferative activity of POL. Like other lectins, POL might exert its death-inducing property through binding with high content of mannose-or glucose-containing moieties on the cell membrane receptors [18].
With the increasing culturing time, the percentages of early as well as late apoptotic cells were enhanced, while the percentage of alive cells was decreased in POL-treated A549 cells. POL treatment resulted in activation of caspase3/9, release of cytochrome c, reduction of mitochondrial membrane potential, up-regulation of pro-apoptotic Bax and Bid, down-regulation of anti-apoptotic Bcl-2 and Bcl-X L , as well as cleavage of PARP, all of which are the features of caspase-and mitochondrial-dependent apoptosis. In addition to apoptosis, POL also induced autophagy in A549 cells, as indicated by a time-dependent increase in the formation of acidic vesicular organelles (AVO) in macrophages,  Recently, a viewpoint has been widely accepted that there might exist positive or negative correlation between apoptosis and autophagy. Autophagy might promote or regulate apoptotic cell death, but under certain circumstances, autophagy could be initiated only when apoptosis was restrained [19]. Unlike apoptosis, autophagy plays not only pro-survival mechanism against nutrient shortage, but also paradoxically acts as a route to cancer cell death [20]. However, whether cells undergo survival or death mainly rely on cell type, drugs as well as concentration and duration. Herein, we reported that inhibition of autophagy could reduce POL-induced apoptosis. Suppression of POLinduced autophagy resulted in decreased growth inhibitory ratio, and also deprive of autophagy by either 3MA or Beclin1 siRNA led to decrease in the ratio of apoptotic cells and the cleavage of PARP in POL-induced A549 cells.
Cancerous growth is one of the most difficult diseases to cure, and nowadays researchers accepted that targeting only one form of PCD does not generally produce quantifiable improvement. And also, these two kinds of PCD might share some common factors [21], suggesting the existence of 'molecular switch' between apoptosis and autophagy [22]. Herein, POL was further demonstrated to induce decreased expressions of phosphorylated-NF-kB (p65), phosphorylated-Akt(Ser473/Thr308), phosphorylated-mTOR, phosphorylated-70S6K and phosphorylated-4E-BP1. Through consecutive activation of NF-kB, Akt or mTOR by genetic approach, we demonstrated that POL induced apoptosis via inhibiting Akt-NF-kB pathway while triggered autophagy via suppressing Akt-mTOR pathway. Also, after POL treatment of A549 cells, the levels of ROS were enhanced, and inhibition of ROS through inhibitor NAC resulted in the decrement of phosphorylated-Akt (Ser473/Thr308) expressions as well as portion of apoptotic and autophagic cells.
Therefore, we could infer that Akt might exert the 'molecular switch' role in regulating between POL-induced apoptosis and autophagy. After POL treatment, cells might undergo apoptotic cell death if Akt-NF-kB pathway was inhibited, while cells might also exhibit pro-death autophagy when Akt-mTOR signaling was suppressed. Akt is frequently activated in non-small cell lung cancer cells, and usually, nuclear factor kappa-light-chainenhancer of activated B cells (NF-kB) acts as downstream of Akt [23]. Activation of NF-kB is dependent on the phosphorylation of IkB kinase (IKKa), which is under the control of Akt kinase [24]. Akt-mTOR pathway is essential in the regulation of autophagic cell death, and usually, mTOR interacting with Raptor forms rapamycin-sensitive TORC1 complex, which controls phosphorylation of p70S6K (70 kDa ribosomal S6 kinase) [25]. Also, mTOR binding with Rictor forms rapamycin-insensitive complex, TORC2, which in turn modulates phosphorylation of Akt [26]. After POL treatment, ROS was gradually accumulated in A549 cells, and ROS is in charge of the dephosphorylation of Akt, inferring that ROS might regulate the molecular switch role of Akt. Intracellular ROS are exquisitely balanced by nonenzymatic antioxidants (like glutathione (GSH)) and antioxidant enzymes, and if the balance toward the oxidant side, causes damage to biomolecules such as DNA, proteins, and lipids [27]. Herein, we demonstrated that nonenzymatic antioxidants GSH and its partner GPX were decreased with culturing time, indicating that POL might abrogate the GSH antioxidant system and result in massive ROS accumulation in A549 cells.
However, the regulation of tumor cell death might be considered as a 'dynamic balance' between apoptosis and autophagy [30]. Still, there exist a complex and not fully defined relationship between them [31], and the exact 'molecular switch' role of Akt needs to be further verified. Whether there exist precedence between POL-induced apoptosis and autophagy? How did Akt protein decide between POL-induced apoptosis and autophagy? Why POL-induced cell death resulted in ROS accumulation? How did ROS exert regulatory effect on Akt expression and POL-induced cell death? All above-mentioned questions are still hanging in doubt, and more strict and deep experiments should be further performed. And we have summarized all the pathways in Fig. 8.
In summary, all above-mentioned findings would provide new evidence for further exploring more PCD mechanisms of GNArelated lectins in cancer research. With the complex molecular mechanisms of GNA-related lectin-induced PCD becoming better elucidated, there is no doubt that new therapeutic strategies would be developed into targeting apoptotic and autophagic cell death pathways for cancer therapeutics.