Autophagy Regulates Colistin-Induced Apoptosis in PC-12 Cells

Abstract

Colistin is a cyclic cationic polypeptide antibiotic with activity against multidrug-resistant Gram-negative bacteria. Our recent study demonstrated that colistin induces apoptosis in primary chick cortex neurons and PC-12 cells. Although apoptosis and autophagy have different impacts on cell fate, there is a complex interaction between them. Autophagy plays an important role as a homeostasis regulator by removing excessive or unnecessary proteins and damaged organelles. The aim of the present study was to investigate the modulation of autophagy and apoptosis regulation in PC-12 cells in response to colistin treatment. PC-12 cells were exposed to colistin (125 to 250 μg/ml), and autophagy was detected by visualization of monodansylcadaverine (MDC)-labeled vacuoles, LC3 (microtubule-associated protein 1 light chain 3) immunofluorescence microscopic examination, and Western blotting. Apoptosis was measured by flow cytometry, Hoechst 33258 staining, and Western blotting. Autophagosomes were observed after treatment with colistin for 12 h, and the levels of LC3-II gene expression were determined; observation and protein levels both indicated that colistin induced a high level of autophagy. Colistin treatment also led to apoptosis in PC-12 cells, and the level of caspase-3 expression increased over the 24-h period. Pretreatment of cells with 3-methyladenine (3-MA) increased colistin toxicity in PC-12 cells remarkably. However, rapamycin treatment significantly increased the expression levels of LC3-II and beclin 1 and decreased the rate of apoptosis of PC-12 cells. Our results demonstrate that colistin induced autophagy and apoptosis in PC-12 cells and that the latter was affected by the regulation of autophagy. It is very likely that autophagy plays a protective role in the reduction of colistin-induced cytotoxicity in neurons.

INTRODUCTION

Colistin, a cyclic cationic polypeptide antibiotic, has been used as the last-line therapy against multidrug-resistant Gram-negative bacteria which can cause life-threatening infections (1–5). However, optimization of its clinical use is limited by its nephrotoxicity and neurotoxicity (6). Recent studies showed that colistin induces apoptosis in primary chick cortex neurons and a tumor cell line, PC-12 cells (7, 8). In cell culture studies, colistin treatment activates caspase-3 and leads to elevated intracellular concentrations of calcium (8, 9). It has been discovered in a mouse model that autophagy is involved in colistin-induced nephrotoxicity (10). However, whether colistin induces autophagy in neurons and the interplay between autophagy and apoptosis remain unknown.

Autophagy is a catabolic process involving the degradation of dysfunctional cellular components by lysosomal systems (11–13). It plays a key role in cell fate as a homeostasis regulator and enables cells to survive stresses, pathogen infection, and hypoxia (14, 15). The current evidence also suggests that defective autophagy promotes neurodegenerative disorders, cancer, liver disease, and aging, while massive autophagy can deplete cellular organelles and proteins and kill severely damaged cells (16). Both autophagy and apoptosis are forms of programmed cell death and play important roles in homeostasis and diseases (17). Recent studies have suggested that autophagy may defer or promote the activation of apoptosis under certain circumstances (e.g., SIRT1 protects against apoptosis by promoting autophagy and oridonin phosphate-induced autophagy effectively enhances cell apoptosis) (18, 19). Autophagy and apoptosis have a very complex relationship, and the precise mechanism remains to be determined.

The PC-12 cell line is derived from a pheochromocytoma in a rat adrenal medulla and contains both neuroblastic and eosinophilic cells (20, 21). It is commonly employed as a model system for neuronal differentiation and neurosecretion and is one of the most widely used neuronal cell lines for examining mechanisms associated with neurotoxicity and neurodegenerative disorders (20, 21). In this study, we employed PC-12 cells to investigate whether colistin treatment causes autophagy and its potential neuroprotective effect against colistin-induced neurotoxicity.

MATERIALS AND METHODS

Reagents and drugs.

Fetal bovine serum (FBS) was obtained from Gibco BRL (Gaithersburg, MD). Colistin sulfate (20,195 U/mg) (lot number 095K1048; Sigma-Aldrich, St. Louis, MO) and 3-methyladenine (3-MA) (SKU [stock-keeping unit] number M9281; Sigma-Aldrich) were dissolved in distilled water. Rapamycin (product number R117; Sigma-Aldrich) was prepared in dimethyl sulfoxide (DMSO). Monodansylcadaverine (MDC), annexin V-fluorescein isothiocyanate (FITC), propidium iodide (PI), 3-(4,5-dimethyl-2-thiazolyl)-2,5-diphnyl-2H-tetrazolium bromide (MTT), and 4,6-dianmidino-2-phenylindole (DAPI) were purchased from Sigma Chemical Co. (St. Louis, Missouri, USA). A bicinchoninic acid (BCA) protein assay kit was obtained from Wuhan Boster Bio-engineering Limited Co. (Wuhan, Hubei, China).

Primary antibodies against LC3-II/I (microtubule-associated protein 1 light chain 3), beclin 1, and caspase-3 were purchased from Cell Signaling Technology (Beverly, MA). Anti-β-actin rabbit monoclonal antibody (MAb) and secondary antibodies (horseradish peroxidase [HRP]-labeled goat anti-rabbit IgG) were obtained from Beijing Zhongshan Golden Bridge Biotechnology Co. Ltd. (Beijing, China).

Cell culture.

PC-12 cells were purchased from the Cell Bank of Type Culture Collection, Shanghai Institute of Cell Biology, Chinese Academy of Sciences. The cells were maintained in Dulbecco modified Eagle medium (DMEM) containing 10% FBS in 5% CO₂ atmosphere at 37°C.

Cell viability determination by MTT assay.

The MTT assay is based on the colorimetric conversion of yellow 3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl-2H-tetrazolium bromide to the purple formazan product (22) and was employed to assess cell viability in the present study. PC-12 cells were seeded in 96-well plates with a density of 0.5 × 10⁵ cells/well. Twenty-four hours after cell seeding, cells were pretreated with 3-MA (5 mM) and rapamycin (50 nM) for 1 h, followed by colistin (125 μg/ml) for 24 h (23). Subsequently, 20 μl of 5 mg/ml of MTT working solution was added to the culture medium and the plate was incubated for 2 h at 37°C. The culture supernatant was removed and the formazan crystals were dissolved in 100 μl DMSO. The absorbance of each well was measured at 540 nm with an enzyme-linked immunosorbent assay (ELISA) reader (Thermo Fisher Scientific, United States).

Visualization of MDC-labeled vacuoles.

Cells were plated in 24-well tissue culture plates for 24 h and incubated with colistin (125 μg/ml) for 1.5, 3, 6, 12, and 24 h or with different concentrations of colistin (0 to 250 μg/ml) for 12 h. Autophagic vacuoles were stained with MDC (0.05 mmol/liter) (24) in DMEM for 20 min at 37°C in a dark environment. Then, the cells were washed three times with phosphate-buffered saline (PBS) and examined using a fluorescence microscope (Eclipse TE2000U; Nikon, Japan) that was equipped with a filter system (barrier filter, 450 nm, and V-2A excitation filter, 380 to 420 nm) (25).

Nuclear morphology assessment by Hoechst 33258 staining.

Cells were plated in 24-well tissue culture plates for 24 h and incubated with colistin (125 μg/ml) for 1.5, 3, 6, 12, and 24 h or with different concentrations of colistin (0 to 250 μg/ml) for 24 h. The PC-12 cells were washed twice with PBS. Cells were fixed with 4% paraformaldehyde for 10 min at 4°C and incubated with Hoechst 33258 (5 μg/ml) (Beyotime, Shanghai, China) at 37°C for 15 min in the dark. After being washed two times with PBS, cells were examined with a fluorescence microscope (Nikon Eclipse TE 2000U). The apoptotic cells containing condensed and fragmented nuclei produced bright blue fluorescence due to staining with Hoechst 33258 (25).

LC3 immunofluorescence microscopic examination.

Cells were preincubated with 5 mM 3-MA or 50 nM rapamycin for 1 h and then incubated with colistin (125 μg/ml) for 12 h. Subsequently, cells were fixed with 4% paraformaldehyde for 15 min and treated with 0.2% Triton X-100 on ice for 10 min. After blocking with 5% bovine serum albumin (BSA) for 1 h, cells were treated with LC3 antibody (1:400 dilution) overnight and then with FITC-conjugated goat anti-rabbit IgG (1:400 dilution) for 1 h and 2.5 μg/ml of DAPI solution for 20 min. Images of stained neurons were obtained under a fluorescence microscope (Nikon Eclipse TE 2000U) (25).

Apoptosis assay.

To investigate the interplay between colistin-induced autophagy and apoptosis, 3-MA and rapamycin were employed to inhibit and stimulate autophagy, respectively, and apoptosis was examined. Cells were preincubated with 5 mM 3-MA or 50 nM rapamycin for 1 h and then incubated with colistin (125 μg/ml) for 24 h. Apoptosis was detected by phosphatidylserine externalization using flow cytometry (26). In brief, after treatment, cells (5 × 10⁵) were suspended in PBS and then in 100 μl of binding buffer containing annexin V-FITC (5 μl) and PI (5 μl) labeling solution for 20 min at room temperature in the dark. Apoptotic cells were quantitated by flow cytometry (Coulter Epics XL, Beckman Coulter, United States). The percentage of cells that were apoptotic was calculated as the sum of annexin V-positive (annexin V⁺)/PI-negative (PI⁻) (early apoptosis) and annexin V⁺/PI⁺ (late apoptosis) cells. Annexin V⁻/PI⁺ cells in the upper left quadrant indicated necrosis.

Observation of autophagy by electron microscopy.

Electron microscopy analysis was conducted to observe autophagy in colistin-treated PC-12 cells (27). After being treated with drugs for 12 h and 24 h, cells were washed with PBS and fixed with 3.5% glutaraldehyde (pH 7.4) at 4°C for 12 h. Cells were then postfixed in 1% osmium tetroxide at 4°C for 30 min and embedded in epoxy resins. The ultrathin sections were treated with uranyl acetate and lead citrate for contrast and then examined with a GEM-1200ES transmission electron microscope (JEOL Ltd., Tokyo, Japan).

Western blotting.

Cells were washed twice with cold PBS after treatment, and total proteins were prepared in radioimmunoprecipitation assay (RIPA) lysis buffer followed by centrifugation at 12,000 × g for 30 min at 4°C. Equal amounts of protein, estimated using the BCA protein assay kit (28), were separated by SDS-polyacrylamide gel electrophoresis (SDS-PAGE) (10 to 15%) and transferred to nitrocellulose membranes (Millipore, Bedford, MA). The blots were blocked and then immunolabeled with primary antibodies for LC3 (1:1,000 dilution), beclin 1 (1:1,000 dilution), caspase-3 (1:1,000 dilution), and β-actin (1:1,000 dilution) at room temperature for 1.5 h. The membrane was incubated with diluted enzyme-linked secondary antibody (1:2,000 dilution) at room temperature for 2 h. The immunoblots were visualized by enhanced chemiluminescence (ECL) reagent following exposure of the filters to X-Omat Kodak films. The autoradiogram was scanned, and the protein bands were analyzed by densitometry using Image J version 1.42 software.

Statistical analysis.

Data were obtained from three or five independent experiments and are presented as mean results ± standard deviations (SD). Statistical analysis was conducted with one-way analysis of variance (ANOVA) using SPSS version 13.0 (SPSS, Chicago, IL). Differences at a P value of <0.05 compared with the results for the control were considered statistically significant.

RESULTS

Effects of drugs on cell viability.

In order to select appropriate doses for 3-MA and rapamycin in this study, PC-12 cells were treated for 24 h with a range of concentrations. The MTT assay data show that cell viability decreased in a dose-dependent manner after treatment with ≥10 mM 3-MA (Fig. 1A). Therefore, 5 mM 3-MA was used in the following experiments. Similarly, 50 nM was chosen for rapamycin for the studies described below (Fig. 1B).

FIG 1.

Effects of different drug exposures on PC-12 cell viability using the MTT assay (n = 5). (A) 3-MA alone at the indicated concentrations. (B) Rapamycin alone at the indicated concentrations at 24 h. (C) Colistin plus 3-MA or rapamycin at the indicated times. Error bars represent standard deviations of the means (SD). **, P < 0.01 versus the results for the control; ##, P < 0.01 for comparison between treatments.

The MTT assay results also show that colistin treatment significantly decreased cell viability in a time-dependent manner (Fig. 1C). Interestingly, rapamycin treatment significantly improved cell viability in the presence of 125 μg/ml colistin; cell viability did not decrease until 6 h after colistin treatment (Fig. 1C). In contrast, 3-MA pretreatment significantly reduced the viability of cells treated with 125 μg/ml colistin at 6, 12, and 24 h. From 6 h to 24 h, significant differences in cell viability were evident among cells that received the three treatments described above.

Autophagy induced by colistin.

The autophagy-related proteins LC3-II/I (microtubule-associated protein 1 light chain 3) and beclin 1 were examined after exposure to colistin (31.25 to 250 μg/ml) for up to 24 h. As autophagosomes are formed, cytosolic LC3 changes from LC3-I to LC3-II and is recruited to the membrane of autophagosomes (29). As shown by the results in Fig. 2, colistin induced LC3-I conversion in a concentration- and time-dependent manner. The expression of LC3-II reached its peak at 12 h, and the expression of beclin 1 had a trend similar to that of LC3-II. These data indicate that colistin may induce autophagy by upregulating the expression of LC3-II and beclin 1. The expression of caspase-3 after exposure to colistin was also concentration and time dependent (Fig. 2).

FIG 2.

Colistin induces autophagy and apoptosis in PC-12 cells. (A) Western blots assaying levels of LC3, beclin 1, and activated caspase-3 (a) and quantifications of the protein levels (b to d) following exposure to colistin during a time course of up to 24 h. Samples containing 30 μg of protein were loaded onto 15% (for LC3 and caspase-3) or 10% (for beclin 1) SDS-PAGE gels, and the blots were probed with antibodies to LC3, beclin 1, and caspase-3. The β-actin level was used as a loading control. (B) Effects of different concentrations of colistin on the levels of LC3, beclin 1, and activated caspase-3 following 12 h or 24 h of exposure. Panels b to d show the quantification of the expression levels in the Western blots shown in panel a. (C) Autophagy was activated after colistin treatment. PC-12 cells were incubated with different concentrations of colistin (31.25 to 250 μg/ml) for 12 h and stained with MDC (0.05 mmol/liter). Fluorescent dots indicate late autophagic vacuoles (white arrows). (D) PC-12 cells were incubated with 125 μg/ml colistin for the times indicated, up to 24 h. (E) Effects of different concentrations of colistin (31.25 to 250 μg/ml) on the nuclear morphology of PC-12 cells following 24 h of exposure. Changes in nuclear morphology were observed by Hoechst 33258 staining. Fragmentation of the nucleus into oligonucleosomes and chromatin condensation were examined by fluorescence microscopy. Arrows indicate apoptotic cells. (F) Effect of 125 μg/ml colistin on the nuclear morphology of PC-12 cells at the times indicated, up to 24 h.

The lysosomotropic agent MDC preferentially labels autophagic vacuoles (24). In the present study, the effect of colistin on PC-12 cells was examined with fluorescence microscopy after MDC staining. The results showed that colistin induced autophagy in a time- and concentration-dependent manner (Fig. 2C and D). The formation of autophagic vacuoles was evident at 12 h for both 125 and 250 μg/ml colistin; therefore, a concentration of 125 μg/ml was used for colistin in the following studies. The nuclei of control cells stained homogenously using Hoechst 33258, although they were less bright than the nuclei in the colistin-treated cells (exposure to colistin at 125 and 250 μg/ml for 24 h); in the treated cells, hypercondensed nuclei and chromatin condensation were observed (Fig. 2E and F).

Protective effects of autophagy on colistin-induced apoptosis in PC-12 cells.

To examine the role that autophagy plays in colistin-induced apoptosis, 3-MA (as an autophagy inhibitor) and rapamycin (as an autophagy inducer) were employed. After treatment with colistin combined with 3-MA, cells started to float in the culture plate wells at 12 h and did not attach to the culture surface (data not shown), whereas cells treated with colistin alone or in combination with rapamycin did not show this effect. The results of the fluorescence microscopy experiment revealed no LC3 dots in the control cells (Fig. 3A). After treatment with colistin for 12 h, different sizes and stages of autophagic vacuoles were visible by immunofluorescence staining with LC3 antibody. However, compared with the results for the colistin-treated cells, LC3 dots were not observed in the cells treated with 3-MA (5 mM), while 50 nM rapamycin pretreatment substantially increased the number of LC3 dots intracellularly (Fig. 3A).

FIG 3.

Effects of 3-MA and rapamycin on colistin-induced autophagy and apoptosis. (A) Fluorescence images of cells treated with colistin alone (125 μg/ml) or preincubated with 3-MA (5 mM) or rapamycin (50 nM) after immunofluorescence staining with LC3 antibody (green) and DAPI staining (blue). The appearance of punctuated staining (white arrows) indicates autophagosome-associated LC3-II. (B) Changes in nuclear morphology of cells treated with colistin alone (125 μg/ml) or pretreated with 3-MA (5 mM) or rapamycin (50 nM) shown by Hoechst 33258 staining. Cells that received different treatments displayed different degrees of karyopyknosis or karyorrhexis (white arrows). (C) Cells were treated with colistin alone (125 μg/ml) or pretreated with 3-MA (5 mM) or rapamycin (50 nM) for 1 h and then with colistin for 12 or 24 h. Levels of apoptosis and necrosis were analyzed using fluorescence-activated cell sorting (FACS). The abscissa and ordinate data represent the fluorescence intensities of annexin V and PI, respectively. The data shown are representative of the results of three independent experiments, and the respective quantitative results are presented in panel D. (D) Quantitative analysis of the effects of 3-MA and rapamycin on apoptosis induced by colistin in PC-12 cells. Bars represent the mean percentages of apoptotic cells ± SD (n = 3). **, P < 0.01 versus the results for the control; ##, P < 0.01 for the results for different treatments. (E) Electron microscopy observations of the control cells and cells treated with colistin alone (125 μg/ml) or pretreated with 3-MA (5 mM) or rapamycin (50 nM) before treatment with colistin. Autophagosomes are marked by black arrows in the images of cells treated with colistin (12 h, 24 h) and colistin plus rapamycin (12 h, 24 h), whereas apoptotic changes, such as chromatin condensation (white arrows) and cytoplasmic vacuolization (yellow arrows), are marked in the images of cells treated with colistin (24 h) and colistin plus 3-MA (12 h, 24 h). (F and G) Western blot assays of LC3, beclin 1, and caspase-3 expression. Samples containing 30 μg of protein were loaded onto 15% (for LC3 and caspase-3) or 10% (for beclin 1) SDS-PAGE gels. The β-actin level was used as a loading control. Panels b and c show the quantification of the expression levels in the Western blots shown in panels a.

Hoechst 33258 was employed in the present study for visualizing changes in nuclear morphology. Pretreatment of cells with 3-MA increased the number of cells with chromatin condensation and nuclear fragmentation, while rapamycin treatment decreased the damage to the nuclei (Fig. 3B). Staining with annexin V-FITC and PI was employed for measuring apoptosis after exposure to colistin with 3-MA or rapamycin. As revealed by the results in Fig. 3C and D, 3.3% (12 h) and 4.4% (24 h) of the control PC-12 cells were annexin V⁺/PI⁻, representing early apoptosis. The percentage of apoptotic cells (26.1% ± 1.7%) increased significantly when PC-12 cells were treated with 125 μg/ml colistin for 24 h. Compared with the results for colistin treatment, cotreatment with 3-MA significantly increased the percentage of apoptosis (P < 0.01), while the percentage decreased significantly after exposure to the combination of colistin and rapamycin (P < 0.01) (Fig. 3D).

The formation of autophagic vacuoles is regarded as the most important morphological feature of autophagic cells (30). Identification of autophagic vacuoles using conventional electron microscopy remains the gold standard for assessing autophagy in cultured cells (27). As shown by the results in Fig. 3E, there were a large number of autophagosomes with characteristic double or multiple membranes intracellularly. After 12 h of treatment with colistin, different sizes and stages of autophagic vacuoles were evident in the electron microscopy images (Fig. 3E), and they predominantly contained cytoplasm organelles and autophagosomes. In the colistin-treated cells (24 h), nuclear chromatin concentration, edge accumulation, and cytoplasmic vacuolization were observed. In the cells treated with colistin plus 3-MA for 12 h or 24 h, autophagosomes were not obvious (Fig. 3E). Those cells displayed shrunken nuclei with condensed chromatin, and cytoplasmic vacuolization was evident (Fig. 3E). In the cells treated with colistin plus rapamycin for 12 h, more autophagosomes appeared in the cells, while in the cells treated with colistin plus rapamycin for 24 h, the cell cytoplasm was homogeneously stained and cytoplasmic vacuolization was not as evident as in the cells treated with colistin for 24 h (Fig. 3E).

To investigate the interplay between autophagy and apoptosis due to colistin treatment, the expression of LC3, beclin 1, and caspase-3 in cells pretreated with 3-MA or rapamycin was examined with Western blotting. In the cells exposed to 125 μg/ml colistin for 12 h and 24 h, the protein levels of LC3-II/I, beclin 1, and caspase-3 increased significantly compared to their levels in the control cells (Fig. 3F and G). However, compared with the results for cells treated with colistin alone, 3-MA (5 mM) treatment significantly decreased the expression of LC3-II/I and beclin 1, and the increased expression of caspase-3 was very likely due to the inhibition of autophagy. In contrast, pretreatment of cells with rapamycin (50 nM) significantly increased the protein levels of LC3-II/I and beclin 1 compared with the results for treatment with colistin alone. Interestingly, the expression of caspase-3 decreased significantly, very likely due to the protective effect of the autophagy.

DISCUSSION

Recent pharmacological data indicate that the possibility of increasing the currently recommended dosage regimens of colistin and polymyxin B is limited due to their nephrotoxicity and neurotoxicity (1, 3, 5). As polymyxin B is only available in North America, South America, and Southeast Asia and is much less widely used in the clinic (31), colistin was examined in the present study. There is no information in the literature on the effects of different concentrations of colistin in nervous tissue in animals or humans. Hence, in the present mechanistic study, we first investigated the autophagy and apoptosis of PC-12 cells with a wide range of concentrations of colistin (31.25 to 500 μg/ml; Fig. 2). The lower end of the colistin concentrations examined is achievable in plasma in mice after subcutaneous administration (32). Nevertheless, caution is required when directly extrapolating the results obtained in the present study to animals and humans, as drug exposure in nervous tissue and allometric scaling need to be considered.

Autophagy plays a crucial role in removing damaged organelles and proteins in the axons of neurons (33). In contrast to apoptosis, autophagy is protective against acute neural injury induced by chlorpyrifos and in Parkinson's disease (23, 34). Although autophagy and apoptosis have different impacts on cell fate, their functions are not completely independent (17, 35). The interaction between autophagy and apoptosis in colistin-exposed nerve cells was never examined prior to the present study. Hence, our objective here was to investigate whether the regulation of autophagy modulates colistin-induced apoptosis in a neuronal model cell line, PC-12 cells.

LC3 (also known as autophagy-related protein 8) and beclin 1 (a novel BH3-only protein) are specific biochemical markers for autophagy (36, 37). There are two different forms of LC3, the soluble 18-kDa LC3-I and the lipidated 16-kDa LC3-II. After the induction of autophagy, LC3-II is converted from LC3-I and then binds tightly to the membranes of autophagosomes, thereby leading to the formation of ring-shaped structures in the cytosol (29, 38). In general, the level of LC3-II is directly correlated with the number of autophagosomes (29). Beclin 1, also known as autophagy-related protein 6, is a major component of a type III phosphatidylinositol-3 kinase (type III PI3K) complex involved in the initial formation of autophagosomes (39, 40). It has been reported that autophagosomes can be induced by beclin 1 in HEK293 cells (41), and beclin 1 is important in the regulation of autophagy and apoptosis (42). Caspase-3 cleavage of beclin 1 is able to induce inactivation of autophagy, which leads to apoptosis in HeLa cells (42). In general, induction of autophagy causes increased intracellular concentrations of LC3-II and beclin 1 (42, 43). Our present study demonstrated that the levels of LC3-II and beclin 1 increased significantly after exposure to colistin (31.25 to 250 μg/ml), indicating that colistin treatment led to increased numbers of autophagosomes. Our result is in keeping with previous reports that autophagy preceded apoptosis under certain stress conditions (44–46). The evidence currently accumulating in the literature demonstrates that toxins and chemicals can activate autophagy and apoptosis and that the early autophagic response provides potential protection against the toxic effects (47–49).

To further investigate the interplay between autophagy and apoptosis caused by colistin treatment, we pretreated PC-12 cells with 3-MA or rapamycin. 3-MA is commonly used for investigating the mechanism of autophagy (e.g., lysosomal self-degradation). 3-MA blocks autophagosome formation via inhibition of type III PI3K and inhibits autophagy. Type III PI3K is crucial for vesicle nucleation and the formation of the preautophagosome, which is a very early event of autophagy (50). In our present study, pretreatment of cells with 3-MA before colistin treatment significantly decreased the viability of cells and promoted chromatin condensation and cytoplasmic vacuolization. Importantly, 3-MA pretreatment decreased the expression of LC3-II, beclin 1, and caspase-3 and accelerated apoptosis. These data implied that autophagy is involved in colistin-induced apoptosis in PC-12 cells.

A number of autophagic proteins, including beclin 1, can be cleaved by caspases, which eventually inactivates their autophagic function (51). In apoptosis, the caspase-cleaved fragments of beclin 1 protein are not able to induce autophagy but cause the release of cytochrome c in response to apoptotic signals (52). At the later stages of autophagy, activation of caspases is the major cause of reduced autophagosomes (53), and our immunohistochemical, biochemical, and imaging results (Fig. 2 and 3) support the idea that autophagy was inhibited by caspase-3 at 24 h after colistin treatment. It is very likely that autophagy is a prosurvival mechanism in PC-12 cells in response to the toxic effect of colistin.

Rapamycin inhibits the activity of a serine/threonine kinase, mTOR (mechanistic target of rapamycin) and promotes autophagy (54). It has been shown that rapamycin reduces injury in different models of neurodegenerative disorders through the increase in autophagy (55, 56). In neuron cells, rapamycin enhances the clearance of aggregated or misfolded proteins and/or dysfunctional mitochondria, thereby providing protection against proapoptotic insults (54). Previous studies have reported that autophagy induced by rapamycin prevents apoptosis (54, 57) and that inhibition of autophagy exaggerates cell apoptosis (58–60). In our present study, pretreatment with rapamycin significantly increased cell viability and delayed the occurrence of cell apoptosis in PC-12 cells (Fig. 1C and 3C and D). Our finding is supported by the decreased expression of caspase-3 due to rapamycin pretreatment (Fig. 3G); this is consistent with the literature reporting its inhibition of caspase-3 activation, a hallmark of apoptosis (23, 57, 61). It is very likely that rapamycin has a neuroprotective effect against colistin-induced apoptosis, in which autophagy and apoptosis interplay with each other.

The mechanisms of autophagy and apoptosis are very different and involve different pathways and distinct executioner and regulatory molecules (62, 63). The interplay between autophagy and apoptosis is very complex, and a previous study showed that autophagy can contribute to the alleviation of neuronal apoptosis in rats (64). Our results revealed that autophagosomes and two autophagy proteins, LC-3 and beclin 1, were all involved in the colistin-induced apoptosis in PC-12 cells (Fig. 3). Further studies are warranted to elucidate the detailed mechanism of the interplay between autophagy and apoptosis induced by colistin in neuron cells.

Conclusions.

Our study is the first to demonstrate, using an in vitro cell culture system, that colistin induces autophagy and apoptosis in neuron cells. That autophagy is protective against colistin-induced apoptosis may provide a novel approach for attenuating the neurotoxicity caused by this last-line antibiotic.