*Article* **Evidence for Nanoparticle-Induced Lysosomal Dysfunction in Lung Adenocarcinoma (A549) Cells**

**Arnold Sipos 1,2,\*, Kwang-Jin Kim 1,2,3,4,5, Constantinos Sioutas <sup>6</sup> and Edward D. Crandall 1,2,7,8**


Received: 29 September 2019; Accepted: 21 October 2019; Published: 23 October 2019

**Abstract:** Background: Polystyrene nanoparticles (PNP) are taken up by primary rat alveolar epithelial cell monolayers (RAECM) in a time-, dose-, and size-dependent manner without involving endocytosis. Internalized PNP in RAECM activate autophagy, are delivered to lysosomes, and undergo [Ca2<sup>+</sup>]-dependent exocytosis. In this study, we explored nanoparticle (NP) interactions with A549 cells. Methods: After exposure to PNP or ambient pollution particles (PM0.2), live single A549 cells were studied using confocal laser scanning microscopy. PNP uptake and egress were investigated and activation of autophagy was confirmed by immunolabeling with LC3-II and LC3-GFP transduction/colocalization with PNP. Mitochondrial membrane potential, mitophagy, and lysosomal membrane permeability (LMP) were assessed in the presence/absence of apical nanoparticle (NP) exposure. Results: PNP uptake into A549 cells decreased in the presence of cytochalasin D, an inhibitor of macropinocytosis. PNP egress was not affected by increased cytosolic [Ca2+]. Autophagy activation was indicated by increased LC3 expression and LC3-GFP colocalization with PNP. Increased LMP was observed following PNP or PM0.2 exposure. Mitochondrial membrane potential was unchanged and mitophagy was not detected after NP exposure. Conclusions: Interactions between NP and A549 cells involve complex cellular processes leading to lysosomal dysfunction, which may provide opportunities for improved nanoparticle-based therapeutic approaches to lung cancer management.

**Keywords:** autophagy; lysosome; lysosomal membrane permeability; mitochondria; pneumocyte

#### **1. Introduction**

Nanoparticle (NP) exposure has been reported to induce stress in various cells and tissues [1–4]. The cellular response to NP exposure is dependent on the chemical (e.g., components and surface charge) and physical (e.g., shape and size) characteristics of NP [5–9]. Due to the heterogeneity of the physicochemical properties of NP, it is difficult to generalize their effects in cells. However, it has been reported that NP exposure of cells and tissues affects functions of various intracellular organelles (e.g., mitochondria and endoplasmic reticulum (ER)) [4,10–12]. The involvement of lysosomes in cellular stress (especially in lung alveolar epithelial cells (AEC)) in response to inhaled NP may be especially important in the pathogenesis of chronic lung disease.

Following internalization of foreign materials (e.g., proteins, bacteria, virus and (nano)particles), autophagy is activated as part of cellular defense mechanism(s) [13]. Autophagy is a catabolic process that helps maintain cellular homeostasis by removing excess, harmful, damaged, or foreign cellular components [14,15], although autophagy may play different roles in cancer versus normal cells, leading to either cell survival or death [16–18]. During autophagic processing, autophagosomes are formed to separate the target cytosolic component(s) from the remainder of the cell [19]. The autophagosome then delivers its cargo to the lysosome via autophagosome-lysosome fusion. Intracellular NP have been reported to activate autophagy [20], and NP localized inside (auto)lysosomes were shown to induce alteration(s) in lysosomal function [21]. Inhalation of airborne pollution particles (including those overlapping in size with nanoparticles) may contribute to chronic diseases (e.g., chronic obstructive pulmonary disease and pulmonary fibrosis) [22], although the mechanisms by which this occurs are not well understood [23,24].

A549 is a continuous cell line derived from a human pulmonary adenocarcinoma that is widely used as a model of mammalian lung alveolar epithelial type II cells. Various ultrastructural characteristics of A549 cells are similar to those in type II pneumocytes. Phospholipid composition of the A549 cell line has been shown to be similar to that of primary isolates of type II cells [25].

In this study, ambient air pollution particles (PM0.2, diameter <0.2 μm) and polystyrene nanoparticles (PNP) were used to investigate their intracellular handling/fate and their effects on A549 cells. PM0.2 contain nanoparticles (defined as particles whose size at least in one dimension is -100 nm) that may contribute to cellular health effects (including development of chronic lung diseases). PNP are engineered nanoparticles that are relatively nontoxic and non-metabolizable, making them suitable for studies of nanoparticle interactions/kinetics. They are useful especially for live cell imaging by taking advantage of their fluorescent labels. Utilizing NP (i.e., PM0.2 and 20 nm carboxylated PNP) and A549 as a model type II pneumocyte, we investigated in this study the internalization mechanisms, egress characteristics, and intracellular NP fate/handling/effects. We found that apical NP exposure of A549 cells leads to activation of autophagy and increased lysosomal membrane permeability (LMP) without mitochondrial dysfunction.

#### **2. Results**

#### *2.1. Live Cell Imaging of Intracellular PNP in A549 Cells*

At 24 h of apical PNP exposure of A549 cells, PNP (in red) accumulated in intracellular vesicles (Figure 1), while diffuse cytosolic PNP distribution was also seen. Plasma membranes of A549 cells were labeled by Dylight 488-conjugated tomato lectin (in green).

**Figure 1.** Intracellular accumulation of polystyrene nanoparticles (PNP) in A549 cells. Apical exposure of A549 cells to PNP at 80 μg/mL for 24 h led to accumulation of PNP (red) in intracellular vesicles, while diffuse distribution of PNP in cytosol was also seen. Plasma membranes of A549 cells were labeled by Dylight 488-conjugated tomato lectin (green). Scale bar is 10 μm.

### *2.2. Mechanism(s) of PNP Entry into A549 Cells*

Nocodazole (an inhibitor of microtubule polymerization) decreased intracellular PNP content by 17% and cytochalasin D (CCD) decreased intracellular PNP content by 57%, indicating that macropinocytosis played a role in PNP entry into A549 cells (Figure 2). Clathrin-mediated endocytosis did not appear to be involved in PNP internalization, as monodansylcadaverine (MDC) failed to decrease intracellular PNP content (Figure 2). Further evidence for endocytic internalization of PNP into A549 cells was provided by colocalization of PNP with early endosomes that were pre-transduced with a Rab5-GFP vector (Figure 3).

**Figure 2.** Relative changes in intracellular PNP content in A549 cells in the presence or absence (control) of endocytosis inhibitors after 24 h of apical PNP exposure. Monodansylcadaverine (MDC, an inhibitor of clathrin-mediated endocytosis) failed to decrease intracellular PNP content, whereas cytochalasin D (CCD, an inhibitor of macropinocytosis) and nocodazole (an inhibitor of microtubule polymerization) decreased intracellular PNP content by 60% and 17%, respectively. \* *p* < 0.05 compared to control.

**Figure 3.** Colocalization of early endosome marker Rab5a-GFP with PNP in A549 cells. A549 cells were transduced for 2 h with an early endosome marker (Rab5a-GFP, green) and apically exposed thereafter to PNP (red) for 24 h. Colocalization (arrowheads, yellow) of PNP with Rab5a-GFP-positive vesicles was observed in some of the vesicles. Contours of cells were added (dotted lines) on the basis of the cell plasma membrane marker Dylight 405-conjugated tomato lectin (blue). Images are representative of 4–5 observations. Scale bar is 10 μm.

#### *2.3. PNP Egress from A549 Cells*

A549 cells were apically exposed to PNP (80 μg/mL) for 12 h, followed by washing with fresh cell culture fluid. Intracellular PNP content was assessed over time for up to 24 h thereafter. Intracellular PNP content of A549 cells decreased ~90% over 24 h (Figure 4). The egress profile in the continued presence of 10 μM apical ATP was not significantly different from that without ATP (Figure 4a), despite repeated elevations in cytosolic [Ca2<sup>+</sup>] due to brief (2.5 min) ATP stimulation (Figure 4b).

**Figure 4.** PNP egress from A549 cells. (**a**) A549 cells were apically exposed to PNP for 12 h, followed by washing with fresh culture fluid and assessing intracellular PNP content at designated time points for up to 24 h thereafter. When 10 μM ATP was applied apically to A549 cells at time zero and remained present throughout the entire experiment, no difference in PNP egress kinetics between control (no stimulation) and ATP-treated A549 cells during egress was observed. *n* = 4–6 for each time point. (**b**) Representative recording of oscillations in intracellular [Ca2<sup>+</sup>] detected upon 2.5 min presence of 10 μM ATP in the apical bathing fluid of A549 cells. Different colors represent intracellular [Ca2<sup>+</sup>] observed in two different A549 cells.

#### *2.4. Intracellular NP Processing in A549 Cells*

We investigated the involvement of autophagy in intracellular processing of NP. A549 cells were preincubated with an inhibitor (e.g., 40 μM chloroquine) of fusion of autophagosomes with lysosomes for 30 min prior to apical NP (PNP at 80 μg/mL or PM0.2 at 1 μg/mL) exposure, followed by exposure to NP (PNP or PM0.2) for 24 h in the continued presence of chloroquine. Immunolabeling for LC3-I/II of NP-exposed and chloroquine-treated A549 cells showed that the intracellular presence of NP led to activation of autophagy (Figure 5). This finding was confirmed in live LC3-GFP-transduced A549 cells (subsequently treated with chloroquine as well), where colocalization of PNP with LC3-GFP-positive intracellular vesicles (i.e., autophagosomes) was found (Figure 6).

**Figure 5.** Apical nanoparticle (NP) exposure induced activation of autophagy in A549 cells. A549 cells were preincubated with chloroquine (40 μM, 30 min) and exposed thereafter to NP (PNP or ambient air pollution particles (PM0.2)) for 24 h in the continued presence of chloroquine, followed by assessment of LC3 expression by immunolabeling. LC3 expression (red) was detected in NP-exposed A549 cells. No or very low level of LC3 expression was found in control cells not exposed to NP. Plasma membranes of A549 cells were labeled by Dylight 488-conjugated tomato lectin (green), whereas nuclei were labeled by Hoechst 33342 (blue). Images are representative of 4–5 observations. Scale bars are 25 μm.

**Figure 6.** Colocalization of PNP with LC3-GFP in A549 cells. Following transduction of A549 cells with the autophagosome marker LC3-GFP construct for 2 h, cells were preincubated with chloroquine (40 μM) and apically exposed thereafter to PNP for 24 h in the continued presence of chloroquine. Colocalization of PNP (red) with LC3-GFP-positive vesicles (green) was observed. Contour of cell was added (dotted line) on the basis of the cell plasma membrane marker Dylight 405-conjugated tomato lectin (blue). Images are representative of 4–5 observations. Scale bar is 10 μm.

Since autophagic processing of intracellular NP might affect the intracellular content of NP, we assessed intracellular PNP content in A549 cells in the presence of pharmacological inhibitors of autophagy. When autophagosome formation was inhibited by 3-methyladenine (3-MA), intracellular PNP content decreased by 38%, whereas impaired autophagic flux (in the presence of bafilomycin) resulted in a 64% decrease in intracellular PNP content, compared to that in control A549 cells at 24 h post-exposure (Figure 7).

**Figure 7.** Effects of autophagy inhibitors on intracellular PNP content in A549 cells at 24 h post-exposure to PNP. Intracellular PNP content in A549 cells was reduced by 38% and 64%, respectively, when autophagosome formation was inhibited with 3-methyladenine (3-MA) or autophagosome-lysosome fusion was inhibited with bafilomycin. Data are normalized to control. \* *p* < 0.05 compared to control.

#### *2.5. Assessment of NP Exposure-Induced Lysosomal Dysfunction*

Colocalization of PNP (red) and lysosomes (green) was observed (Figure 8). PNP were also seen in A549 cells without colocalization with Lysotracker Green. PNP is non-metabolizable and likely to accumulate in lysosomes if intact lysosomal membranes are maintained. Acridine orange (AO) is known to accumulate in lysosomes and can be released upon lysosomal injury (e.g., derangement of lysosomal membrane integrity). When A549 cells were apically exposed to PNP or PM0.2 for 24 h, increased presence of AO in cytosol and nucleus was seen, indicating that LMP was increased (Figure 9).

**Figure 8.** Colocalization of PNP with lysosomes in A549 cells. At 24 h post exposure to PNP, lysosomes in A549 cells were labeled with Lysotracker Green (green). Colocalization (yellow; arrowheads) can be seen between PNP (red) and Lysotracker Green in some of the vesicles in the perinuclear area. Plasma membrane of A549 cells was labeled by Dylight 405-conjugated tomato lectin (blue or shown with dotted line). Images are representative of 4–5 observations. Scale bar is 10 μm.

**Figure 9.** Detection of increased lysosomal membrane permeability (LMP) using acridine orange (AO) in A549 cells. (**a**) Increased LMP was detected when AO (green; ex/em: 488/500–525 nm) intensity in lysosomes was decreased along with increased AO intensity in cytoplasm and nucleus. In "no NP" panels, A549 cells showed virtually no detectable nuclear AO signal. A549 cells pretreated with ciprofloxacin (CPX, 150 μM for 24 h, without NP exposure) as positive control exhibited AO accumulation in both nucleus (arrow) and cytoplasm (arrowhead). NP (PNP or PM0.2) exposure also led to AO accumulation in cytoplasm and nucleus. Contours of cells were added (dotted line) on the basis of the cell plasma membrane marker Dylight 405-conjugated tomato lectin (blue). Scale bars are 20 μm. (**b**) When A549 cells were apically exposed to either PNP or PM0.2 for 24 h, increased LMP index (AO green fluorescence intensity in cytoplasm and nucleus/total cellular AO green fluorescence intensity) was found to be similar to that of positive control (CPX). *n* = 4–6. \* *p* < 0.05 compared to control.

#### *2.6. NP Exposure and Mitochondrial Function in A549 Cells*

Mitochondria were labeled with the mitochondrial membrane potential sensitive fluorescent dye tetramethylrhodamine methyl ester (TMRM). When A549 cells were apically exposed to PNP or PM0.2 for 24 h, no change in mitochondrial membrane potential was observed (Figure 10). To confirm these data, we also used the Mtphagy dye, which is capable of staining damaged mitochondria (undergoing mitophagy and being delivered to lysosomes). No colocalization of the Mtphagy dye and LI lysosome marker dye was found, indicating that 24 h of apical exposure of A549 cells to PNP or PM0.2 did not lead to cellular stress involving mitochondrial dysfunction (Figure 11).

**Figure 10.** Effect of apical NP exposure for 24 h on mitochondrial membrane potential in A549 cells. Mitochondrial membrane potentials were normalized to the corresponding controls (i.e., not exposed to either NP or carbonyl cyanide 4-(trifluoromethoxy)phenylhydrazone (FCCP)). Mitochondrial membrane potential was nearly abolished in the presence of FCCP (1 μM, positive control), whereas it did not decrease following 24 h of apical exposure to PNP or PM0.2. \* *p* < 0.05 compared to control.

**Figure 11.** Absence of mitophagy in NP-exposed A549 cells. Mitophagy is detected on the basis of an increase in the fluorescent signal of the Mtphagy dye (in comparison to that of control; red) and colocalization of the Mtphagy dye with lysosomal indicator (LI; green) dye. In control (no exposure to PNP or PM0.2), the Mtphagy dye had weak fluorescence and minimal colocalization with lysosomes. After FCCP exposure (positive control), the Mtphagy dye exhibited strong fluorescence with extensive colocalization with the LI lysosomal marker dye. Exposure of A549 cells to either PNP or PM0.2 yielded weak fluorescence with the Mtphagy dye and showed minimal lysosomal colocalization of the Mtphagy dye, consistent with a low level (or relative absence) of mitophagy. Contours of cells were added (dotted line) on the basis of the cell plasma membrane marker Dylight 405-conjugated tomato lectin. Scale bars (25 μm) are shown in right panels only.

#### **3. Discussion**

Using a live cell imaging approach, we demonstrated that PNP are taken up into A549 cells, at least in part by macropinocytosis. Most (>90%) intracellular PNP content is released from A549 cells over 24 h, whereas mobilization of intracellular Ca2<sup>+</sup> by ATP did not speed up PNP egress from A549 cells. Internalized NP (PNP or PM0.2) activated autophagy, delivering NP to lysosomes. The presence of NP in lysosomes led to increased lysosomal membrane permeability. Mitochondrial membrane potential was not affected and mitophagy was not observed in NP-exposed A549 cells.

### *3.1. Uptake of NP into A549 Cells*

We previously reported that primary rat alveolar epithelial cell monolayers (RAECM) do not appear to utilize endocytic process(es) for PNP uptake, which are internalized mainly by diffusion-like entry into RAECM [26]. A549 cells, a lung adenocarcinoma cell line, differ from primary AEC in many ways, including the inability to form a barrier with high transepithelial resistance and utilization of macropinocytosis for NP uptake. PNP internalization that is not inhibited by cytochalasin D may take place via non-endocytic process(es) (e.g., diffusion across and/or poration by PNP of apical cell plasma membranes [27–29]). In addition, clathrin-mediated endocytosis is not involved in PNP uptake into A549 cells (Figure 2). The decrease in PNP uptake in the presence of nocodazole suggests that suppression of vesicle movement on an intracellular microtubular network is insufficient to entirely inhibit endocytosis. When A549 cells were transduced with the early endosome marker Rab5a-GFP, colocalization between early endosomes and PNP was found at 24 h post exposure to PNP, consistent with the presence of PNP endocytosis into A549 cells.
