*3.2. Egress of NP from A549 Cells*

Intracellular PNP content decreased 90% over 24 h (Figure 4a) in egress experiments. Intracellular PNP was compartmentalized in A549 cells, as shown in Figure 1, with some PNP diffusely distributed in the cytoplasm. It has been reported that epithelial cells are capable of releasing the content of intracellular vesicles (e.g., lysosomes) [30,31]. For example, release of PNP localized in lysosomes of primary RAECM was dependent on elevations in cytosolic [Ca2<sup>+</sup>] [26]. By contrast, A549 cells stimulated with ATP led to oscillatory elevations in cytosolic [Ca2+] (Figure 4b) without altering the kinetics of PNP egress over 24 h (Figure 4a). In A549 cells, it is possible that lysosomal PNP leaked into cytosol and may not be available to exit primarily via lysosomal exocytosis.

#### *3.3. Activation of Autophagy in Intracellular Processing of NP in A549 Cells*

Autophagy is a defense mechanism aiming to maintain cellular homeostasis. The presence of NP in AEC has been reported to activate autophagy in primary RAECM [26]. In order to identify the involvement of autophagy in intracellular fate of internalized NP in A549 cells, we sought to detect the expression of microtubule-associated protein 1A/1B light chain 3B (LC3) protein. LC3 is localized to the phagophore membrane, followed by localization in inner and outer membranes of autophagosomes. Upon fusion of autophagosomes with lysosomes (i.e., autophagic flux), LC3 localized at the inner membrane of (auto)lysosomes is degraded by lysosomal proteolysis. LC3-GFP fusion protein expression is used widely to identify autophagosomes, with the caveat that GFP loses its fluorescence in the lysosomal acidic environment [32,33]. Because the autophagy process is relatively fast [34], there is only a small detection window (before autophagosomes fuse with lysosomes) for observing LC3-GFP, necessitating inhibition of autophagic flux (e.g., with chloroquine or bafilomycin to block the fusion of autophagosomes with lysosomes). In the presence of chloroquine, we observed colocalization of PNP with LC3-GFP after 24 h of apical PNP exposure, indicating that autophagy plays a role in cellular handling of NP in A549 cells (Figure 6). Further evidence for autophagy activation in NP-exposed A549 cells was obtained by immunolabeling of LC3-positive vesicles in NP-exposed A549

cells (Figure 5). PM0.2-exposed primary RAECM exhibited autophagy activation (i.e., expression of LC3) as well.

We observed that interference with autophagy led to decreased intracellular PNP content in A549 cells (Figure 7). When autophagic flux was blocked, autophagosomal content could not be delivered to lysosomes for further processing, as discussed below. Therefore, it is conceivable that the observed lower intracellular PNP content might be related to the decrease in lysosomal PNP localization. It cannot be ruled out that yet-unknown feedback mechanism(s) between autophagy and endocytic process(es) might also play a role. When we performed PNP uptake experiments in primary RAECM in which autophagy was inhibited with 3-MA or bafilomycin, intracellular PNP content was decreased by ~80 and ~50%, respectively [26]. In contrast to primary RAECM, the inhibition in A549 cells of autophagic flux resulted in greater loss in intracellular PNP content (64%) compared to inhibition of autophagosome formation (38%). Although we do not have a clear explanation for this difference between primary RAECM and A549 cells, it may indicate that A549 cell autophagy is regulated differently (especially by 3-MA [35]) from autophagy in primary AEC [36,37]. Lysotracker Green-negative intracellular vesicles that contain PNP may include amphisomes [20,38]. It is possible that the lysosomal dysfunction found in NP-exposed A549 cells (discussed below) also contributed to the difference in intracellular PNP content found in the presence of pharmacological inhibitors of autophagy.

#### *3.4. Lysosomal Dysfunction in NP-Exposed A549 Cells*

We found relatively low PNP content in lysosomes of A549 cells at 24 h post exposure (Figure 8). Most PNP resided in Lysotracker Green-negative vesicles, indicating that the PNP-filled vesicles were primarily non-lysosomal vesicles. This finding might explain why ATP stimulation failed to speed up the kinetics of PNP egress from A549 cells (Figure 4), as most intracellular PNP are not localized to lysosomes. Acridine orange (AO) is a nucleic acid-sensitive cationic dye often used to label DNA in cells [39,40]. However, when used in nanomolar concentrations, it labels primarily acidic intracellular organelles (e.g., lysosomes) without labeling other intracellular components (e.g., nuclei) (Figure 9a). On the basis of this characteristic, we used 7 nM AO for labeling acidic cellular organelles (mostly lysosomes), followed by detection of increased LMP index in NP-exposed A549 cells (Figure 9b). Apical NP (PNP or PM0.2) exposure caused the release of AO from lysosomes and accumulation in cytosol and nucleus. Interestingly, vesicular staining of AO did not fully disappear following NP exposure of A549 cells, indicating perhaps that the increase in LMP index was not permanent. The magnitude of NP exposure-induced increase in LMP index was comparable to that achieved by the positive control ciprofloxacin (CPX). PNP exposure in HeLa cells has been reported to result in lysosomal dysfunction via the activation of autophagy [41].

#### *3.5. Absence of Mitochondrial Dysfunction in NP-Exposed A549 Cells*

Mitochondria are often reported as damaged or dysfunctional in the presence of cellular stress (e.g., mild to severe loss in mitochondrial function) following exposure of cells to various NP [42]. One of the most sensitive approaches to test mitochondrial integrity is to monitor changes in fluorescence of the mitochondrial membrane potential sensitive dye TMRM in live cells. We found no appreciable changes in TMRM fluorescence intensity upon apical exposure to NP, although the positive control (i.e., FCCP) lowered TMRM fluorescence intensity by >90%. This finding might suggest that the NP dose we used in this study was too low to induce major mitochondrial impairment and that lysosomes are more sensitive than mitochondria to PNP or PM0.2 exposure at the concentration of NP used to expose A549 cells in this study.

In addition to the mitochondrial membrane potential measurement, we also utilized the novel Mtphagy dye, capable of reporting mitophagy in real time, for assessment of mitochondrial (dys)function. When mitophagy takes place (i.e., damaged mitochondria are delivered to lysosomes), the fluorescence intensity of the Mtphagy dye increases due to acidic lysosomal pH. As a positive

control, FCCP treatment of A549 cells led to enhanced fluorescence intensity from the Mtphagy dye with increased colocalization of the Mtphagy dye with the LI lysosomal marker dye. By contrast, NP exposure of A549 cells failed to cause any discernible increase in the Mtphagy dye fluorescence intensity and no measurable colocalization of the Mtphagy dye with the LI lysosome marker dye, indicating no or minimal impairment in mitochondrial function.
