Aged Zinc Oxide Nanoparticles Did Not Induce Cytotoxicity Through Apoptosis Signaling Pathway as Fresh NPs

Zinc oxide nanoparticles (ZnO NPs) are being used in a wide range of applications including industry, commercial products and medicine eld. Numerous mechanistic studies for ZnO NPs’ toxicity are performed on pristine (fresh) NPs. However, the cytotoxicity induced by the transformed (aged) ZnO NPs and the underlying mechanisms remain unclear. Here, we rstly conrmed the physicochemical transformation of ZnO NPs underwent over time and compared the cytotoxicity induced by fresh and aged NPs. Then, we found that fresh NPs induced higher apoptosis levelthan aged NPs. Accordingly, RNA sequencing data from aged ZnO NP-treated human-hamster hybrid (A L ) cells showed that p53, PI3k-Akt, FoXO, Glutathione, ErbB, HIF-1, Oxytocin and Jak-STAT signaling pathway were enriched but no apoptosis pathway. Quantitative PCR results conrmed the signicantly higher mRNA level of IL1B and CD69 in fresh NP-treated groups compared to that of aged ZnO NP- and zinc chloride-treated groups. Our data indicated that the lower cytotoxicity of aged ZnO NPs is closely related to the low level of apoptosis induced by it and that the transcriptional regulation of the multiple pathways activated by aged NPs helps to build the cellular homeostasis. Our results highlight the aging (environmental transformation) process to the toxicity and safety assessment of ZnO NPs. different cytotoxicities between fresh and aged ZnO NPs. ZnO NPs with two different particle sizes (20 nm and 90–200 nm) were applied systematically. The cytotoxicity assays demonstrated that aged ZnO NPs induced less pronounced morphological abnormalities and relatively higher cell viabilities than their fresh counterparts. RNA sequencing data revealed that apoptotic genes were enriched in fresh ZnO NP-treated cells, whereas these genes were much less affected by aged ZnO NP-treatments. In addition, the cells exposed to aged ZnO NPs showed reduced level of cleaved Caspase-3 protein, further indicating the higher potency of fresh ZnO NPs in eliciting apoptosis in cultured cells. Combined with our previous results, this study suggested that the decreased cytotoxicity of aged ZnO NPs is attributed to their attenuatedability in triggering cell apoptosis. protein Caspase3, key executor in cell apoptosis.


Introduction
With the rapid development of nanotechnology over the last decades, nanoparticles (NPs) have been applied in various elds, including industry, human daily life and nanomedicine [1,2]. The Nanotechnology Consumer Product Inventory (CPI) shows a 30-fold increase between 2005 and 2015 in the numbers of nano-products, including 762 health ( tness) products, 72 food (beverages), and 23 baby products [2].The growing application of NPs in consumer products and various elds increased the possibility of NPs entering into the environment, and raise safety concerns with regard to their potential adverse impacts.
Zinc oxide (ZnO) NPs are among the most commonly-used NPs, with a wide range of applications including, but not limited to, biomedical imaging, drug/ gene delivery, biosensing, and antibacterial/ antifungal applications [3]. The global annual output of ZnO NPs has reached nearly 3,400 tons [4].
Evidences have indicated that some substances previously considered as biologically inert could become toxic in their nanoparticulate state. An increasing number of studies have shown that ZnO NPs may pose signi cant risks to mammalian cells and animals by inducing signi cant toxicity [5][6][7][8].
Various strategies including coating, surface functionalization and oxidation state modi cation have been used to design NPs that could be biologically and environmentally safer. Each of these methods can target and prevent toxic mechanisms by modifying the physical and chemical properties of NPs (such as the dissolution, agglomeration and perturbation of cell membranes). These modi cations ameliorate and in certain materials prevent toxic effects obviously. However, there are reports that that these design to increase the safety of NPswere not always effective under certain exposure conditions and environments [8][9][10]. Actually, many kinds of NPs are not stable and inclined to undergo "aging" or "environmental transformation" after being intentionally or unintentionally released into the natural environment [10][11][12][13].
In recent years, many researchers have carried out lots of work on the environmental transformation process of NPs, however, the research on the toxic effects of "transformed (aged)" NPs is still very limited, let alone their toxic mechanisms.
As a typical representative of non-persistent NPs, ZnO NPs have very high reactivity, and are prone to transform in physical and chemical properties and occurrence state after being released into the environment or ingested by animals and this change would signi cantly affect their toxicological effects [13,14]. For example, studies have shown that the sul dation process of ZnO NPs changed the charge, hydrophobicity, and aggregation state, resulting in the adsorption of sul de state NPs in human saliva, sweat, and bronchoalveolar lavage uid. And the protein adsorbed by ZnO NPs forms a special protein crown, which affects its biological effect [15]. Phosphates in physiological solutions could convert ZnO NPs into metastable ZnHPO 4 and Zn 3 (PO 4 ) 2 within about 5-10 h [16].The occurrence of complete transformation of ZnO NPs (≤ 3 µg/mL) in the in vitro exposure system to human T lymphocytes (37 °C, cell culture medium RPMI1640 containing 10% FBS for 24 h) was investigated by using synchrotron radiation X-ray absorption near-edge structure spectroscopy (XANES) [17]. The above studies suggest that it is not enough to study pristine (fresh) NPs to assess their environmental and health risks. It is necessary to fully consider the aging and environmental transformation processes of NPs [18].
Our previous study unveiled that ZnO NPs aged for 40 to 120 days in ultrapure water underwent physicochemical transformation and turned into Zn 5 (CO 3 ) 2 (OH) 6 , Zn (OH) 2 , and Zn 2+ [19]. Interestingly, aged NPs exhibited lower cytotoxicity than fresh ZnO NPs [19], yet the toxicity mechanisms of such kind of variation are unclear. In the present study, we set out to explore the underlying reasons of different cytotoxicities between fresh and aged ZnO NPs. ZnO NPs with two different particle sizes (20 nm and 90-200 nm) were applied systematically. The cytotoxicity assays demonstrated that aged ZnO NPs induced less pronounced morphological abnormalities and relatively higher cell viabilities than their fresh counterparts. RNA sequencing data revealed that apoptotic genes were enriched in fresh ZnO NP-treated cells, whereas these genes were much less affected by aged ZnO NP-treatments. In addition, the cells exposed to aged ZnO NPs showed reduced level of cleaved Caspase-3 protein, further indicating the higher potency of fresh ZnO NPs in eliciting apoptosis in cultured cells. Combined with our previous results, this study suggested that the decreased cytotoxicity of aged ZnO NPs is attributed to their attenuatedability in triggering cell apoptosis.

Nanoparticles and Reagents
The commercially available ZnO nanopowders (ZnO NPs), with manufacturer's reported average size 20 nm (99.5% purity, nearly spherical) and 90-200 nm (99.9% purity, irregular morphology), were purchasedfrom Nanostructured & Amorphous Materials (Houston, TX). Except for otherwise noted, all the reagents and chemicals used in this study were purchased from Sigma-Aldrich (Shanghai, China).

Nanoparticle Dispersion, Aging and Characterization
ZnO NPs stock suspensions (1 mg/mL) were prepared by suspending dry nanopowders in Milli-Q H 2 O (Millipore, 18 MΩ•cm), and sterilized by autoclaving (120°C, 30 min), then stored at 25°C for natural aging period ranging from 0 to 60 days. The 0-and 60-days' naturally transformed ZnO NPs were designated as fresh and aged NPs, respectively. To ensure proper dispersion, the fresh and aged suspensions were sonicated (100 W) for 30 min in an ultrasonic bath before characterization or incubation with cells. The morphology, particle size and aggregation of fresh and aged ZnO NPs were characterized by using transmission electron microscopy (TEM, JEOL JEM-2010, Tokyo, Japan). The crystal structure of fresh and aged ZnO NPs were determined using power X-ray diffraction (XRD, PANalytical B. V., Shanghai, China) by comparing to authentic standards.The details of natural aging process and characterization on ZnO NPs have been described previously [19].
Cell Culture and Treatment with ZnO NPs A L cell line, a kind of human-hamster hybrid cells formed by fusion of the gly2A mutant of Chinese hamster ovary (CHO) and human broblasts, was used in this study. These hybrid cells contained a standard set of CHO-K1 chromosomes and a single copy of human chromosome 11 , and were cultured in Ham's F12 medium (Hyclone, Grand Island, NY) supplemented with fetal bovine serum (8%, Hyclone, Grand Island, NY), gentamicin (25g/mL) and glycine (2×10 -4 M) at 37°C in a humidi ed 5% CO 2 incubator [20].The stock suspensions of fresh and aged ZnO NPs were dispersed b y30 min of ultrasonication (100W) to prevent agglomeration, subsequently diluted to appropriate concentrations with cell culturemedia for the exposure of cells. Cells maintained in cell culture media without NPs were served as control in each experiment.

Assay for Detecting the Cytotoxicity
A L cells at a logarithmic phase of growth were cultured on glass slides in 35-mm Petri dishes (6 × 10 4 cells/dish) for 24 h before stimulation, followed by treatment with 2 mL of culture medium containing 1, 5, 10, 12, 15 and 20 µg/mL fresh or aged ZnO NPs 72 h. After the completion of treatment time, the images of cell morphology were obtained using a Leica DM4B microscope (Leica, Germany). ZnCl 2 was included as zinc ions reference for comparing the cytotoxicity with ZnO NPs.
The cell counting kit (CCK)-8 (APExBIO, Shanghai, China) was used for detecting the cell viability. In details, A L cells were seeded into 96-well plates (4 × 10 3 cells/well) with cell culture media for 24 h, and treated with medium containing various concentrations of ZnCl 2 , fresh and aged ZnO NPs for 24, 48 and 72 h, respectively. For working solution, the volume of added NPs from the stock suspension was less than 5% of the total volume of the culture medium in each well. After the completion of treatment time, the culture medium was aspirated, and the cells were incubated with 100 µL CCK-8 working solution for 2 h at 37°C following the manufacturer's instructions. Then, the absorbance was recorded at 450 nm using a Spetra Max M2 uorescence reader (Molecular Devices, Wokingham, Berks, UK).Cell viability was calculated as a percentage of absorbance in wells, with each concentration of NPs normalized to the absorbance of control cells (100%).

RNA Extraction, Reverse Transcription and Quantitative PCR
A L cells at a logarithmic phase of growth were seeded into 35-mm diameter Petri dish (6 × 10 4 cells/dish) with cell culture media for 24 h. Then the medium was replaced with 2 mL of culture medium containing 12 µg/mL ZnCl 2 , fresh and aged ZnO NPs for 72 h. After the completion of treatment time, the culture medium was aspirated, and cells were washed 3 times with PBS. Subsequently, 1mL of Trizol reagent (Invitrogen, Carlsbad, CA, USA) was added to each dish to extract total RNA according to the manufacturer's instructions. Concentration and purity of total RNA obtained after the extraction were quanti ed usinga Q5000UV-Vis Spectrophotometer (Quawell, USA). After quanti cation, reverse transcription was performed using TransGene RT-PCR kit (TransGene Biotech, Beijing, China)to obtain cDNA from the RNA template according to the manufacturer's protocols.The resulting cDNA samples were quanti ed by usingthe Q5000 UV-Vis Spectrophotometer, and then analyzed using SYBR-Green as uorescence dye (TransGene Biotech, Beijing, China) on Roche RT-PCR system (Applied Biosystems).
The housekeeping gene encoding Glyceraldehyde-3-phosphate Dehydrogenase (GAPDH) was used as internal control for evaluating Il-1α, Il-1β, Caspase 3, CD69, Jun and MT1 mRNA expression. The results were expressed as the relative expression ratio between the targeted gene and Gapdh. The primer sequences used in this study are provided in Table 1.

Western Blotting
A L cells at a logarithmic phase of growth were seeded into 60-mm diameter Petri dish (1.5 × 10 5 cells/dish) with cell culture media for 24 h. Then the medium was replaced with 4 mL of culture medium containing 12 µg/mL fresh or aged ZnO NPs for 24 h. At the end of exposure period, the culture medium was aspirated, and then cells were washed 3 times with PBS and lysed on ice with RIPA lysis buffer (Beyotime, China) to collect cellular proteins. Equal amounts of cellular proteins were separated on 12% SDS-PAGE gels, and then transferred to a polyvinylidene uoride (PVDF) membrane (Roche, Swiss). Brie y, after 2 h blocking with 5% nonfat milk in TBST at 25°C, the membranes were subsequently incubated with primary antibody at appropriate dilutions (according to the manufacturer's protocols) at 4°C overnight, followed by incubating with HRP-conjugated secondary antibodies (1:5000, Promega, Madison, USA) for 2 h at 25°C.Finally, immunolabeling was detected using an enhanced chemiluminescence (ECL) (BOSTER, China) solution.The primary antibodies of anti-pro/cleaved Caspase-3 and anti-Actin were purchased from Cell Signaling Technology and ImmunoWay, respectively.

Statistics
Statistical analysis was compiled on the means ofthe results obtained from at least three independent experiments. All Data were presented as means ± standard deviation (SD), and statistically compared using one-way analysis of variance (ANOVA). In all plots p values< 0.05 were showed as * and considered to be statistically signi cant.

Characterization of ZnO NPs
To determine the differences in detailed physicochemical characteristics between fresh and aged ZnO NPs, we rst observed the morphology of NPs using a TEM (Fig. 1A). Our results indicated that 20 nm fresh ZnO NPs were nearly spherical crystals and 90-200 nm fresh ZnO NPs were irregularly rodlike/cubical crystals. The single particle size was consistent with the size provided by the manufacturer. Obviously, both 20 nm & 90-200 nm ZnO NPs were inclined to aggregate in ultrapure water. Also, regardless of the shape and size of the original NPs, both 20 nm and 90-200 nm ZnO NPs' microstructure were dramatically changed from a clear crystal structure to an amorphous or sheet/needle-like state after aged for 60 days. Furthermore, the crystalline nature and phase purity of both fresh and aged NPs were determined by using X-ray diffraction (XRD)withCu Kα radiation (λ = 0.15418 nm) approach at 25 °C and the obtained patterns were presented in Fig. 1B. The XRD pattern of fresh ZnO NPs indicated that the samples were comprised of crystalline wurtzite structure and no characteristic impurity peaks were identi ed, suggesting a high quality of fresh NPs. While the XRD pattern of aged NPs showed the neoformation of Zn 5 (CO 3 ) 2 (OH) 6 (card number 00-011-0287) and Zn

Morphological Observation of A L cells Exposed to Fresh and Aged ZnO NPs
NPs' treatment results in a noticeable change in cellular shape, or morphology, in vitro [21]. Therefore, A L cells exposed to fresh or aged ZnO NPs at 10&15 µg/mL for 72 h were examined under a stereoscopic microscope. As shown in Fig. 2

Aged ZnO NPs Induced Lower Cytotoxicity than Fresh NPs
To further investigate the difference in cytotoxicity between fresh and aged ZnO NPs, we examined the cell viability by using CCK-8 kits. As shown in Fig. 3, incubation A L cells with graded doses of fresh and aged ZnO NPs (ranging from 0 to 20 µg/mL, 20 nm and 90-200 nm) for 24 h, 48 h, or 72 h showed a dose-dependent decrease of cell viability. At the dosages of ZnO NPs ≤ 10 mg/mL, there was no observed cytotoxicity. But, when the dosage of fresh and aged ZnO NPs went up to 12 and15 mg/mL, the cell viability showed a time-dependent decrease tendency. Obviously, the cell viability in aged NP-treated groups was signi cantly higher than fresh NP-treated groups. In addition, the cell viability of ZnCl 2 -treated groups also showed a dose-and time-dependent decrease tendency. But, its cytotoxicity here was less than that of both fresh and aged ZnO NPs.

Fresh ZnO NPs' Treatment Activated Apoptosis Pathways and Up-regulated the Expression of Apoptotic Genes
To unveil the underlying mechanisms leading to the lower cytotoxicity of aged NPs, we analyzed RNA sequencing data from both fresh and aged ZnO NPs. As shown in Fig. 4A&B, after treatment with fresh ZnO NPs, apoptosis pathway was signi cantly enriched in Jurkat cells (p = 0.017) and HMDM cells (p = 0.041). The apoptosis genes: ANXA1, CYLD, TNFSF10, IER3, CDKN1A, JUN, SAT1, PMAIP1, CD38, and ISG20 were signi cantly enriched in fresh ZnO NP-treated Jurkat cells . The apoptosis genes: CD38,  TNFRSF12A, CCNA1, BMP2, PPP2R5B, EREG, IFNGR1, CD44, CD14, GNA15, GCH1, TIMP1, BTG2, IL1B,  IL1A, BTG3, BCL2L11, SC5D, and SPTAN1 were signi cantly enriched in fresh ZnO NP-treated HMDM cells (Fig. 4C & D). Since Jurket cells (peripheral blood T lymphocyte cells) and HMDM cells (human monocyte-derived macrophages) belong to different kinds of cells, the way they trigger apoptosis might be different. Accordingly, the difference in activation of apoptosis pathway resulted in different apoptosis genes were activated. These results showed that fresh ZnO NPs' exposure could activate different apoptosis pathways in various kinds of cells.
Aged ZnO NPs did not Up-regulate the Expression of Apoptotic Genes as Fresh ZnO NPs Our RNA sequencing data from aged ZnO NP-treated A L cells showed that p53, PI3k-Akt, FoXO, Glutathione, ErbB, HIF-1, Oxytocin, Jak-STAT signaling pathway were enriched (Fig. 5A). The apoptosis genes enriched in Jurket and HMDM cells were not signi cantly changed in expression in the aged ZnO NP-treated cells (Fig. 5B). To further con rm the ndings, we tested the expression of related genes by real time PCR. We found that some of the apoptosis genes: BMP2, PMAIP1, IL1α, CD69, CCNA1, CD38, and IL1β were undetectable in aged ZnO NPs-treated A L cells (data not shown), since most of these genes are expressed in immune system cells. The other apoptosis genes (IL1α, IL1β, and CD59) with obviously increased expression level after fresh ZnO NPs' treatment were not signi cantly changed in expression in aged ZnO NP-treated A L cells. The MT1, serve as a positive control, was signi cantly increased in expression as same as RNA-seq data. Also, the expression of Caspase 3 was not signi cantly changed (Fig. 5C).These data showed that fresh ZnO NPs activates apoptosis pathway genes in A L cells but not aged ZnO NPs.

Fresh ZnO NPs (but not Aged NPs) Increased the Expression Level of the Activated Caspase3 Protein
The change of expression for Caspase-3 at mRNA level could not directly re ect the activation of apoptosis pathway. To further analysis whether ZnO NPs' treatment could change apoptotic proteins' level, the expression of cleaved Caspase 3 protein was examined by western blotting analysis, considering the close relation between the activation of Caspase3 and the apoptotic signaling pathways [22]. As shown by Fig. 6, compared to the control group, fresh ZnO NPs (20 nm) treatment increased the cellular level of cleaved Caspase 3 protein by 1.31 ± 0.023 -fold, which was signi cantly higher than that of aged 20 nm ZnO NPs-treated group (1.12 ± 0.039 -fold). When the particle size of fresh ZnO NPs was increased to 90-200 nm, the expression of cleaved Caspase 3 protein induced by fresh NPs was increased by 1.46 ± 0.078-fold, signi cantly greater than that of aged NPs (1.07 ± 0.075-fold).These data further illustrated the higher potency of fresh ZnO NPs in inducing cell apoptosis but not aged ZnO NPs.

Discussion
ZnO NPs was reported to underwent physicochemical transformation into Zn 5 (CO 3 ) 2 (OH) 6 with the release of Zn 2+ during the natural aging process [19,23]. However, the cytotoxicity induced by the transformed (aged) ZnO NPs and the underlying possible mechanisms are unclear. Here, to unveil the mechanism of diverse cytotoxicity between fresh and aged ZnO NPs, RNA sequencing analysis and RT-PCR test were conducted. Also, western blotting was applied to examine the protein level of Caspase3, the key executor in cell apoptosis.
Our data showed that aged ZnO NPs induced much less cytotoxicity than fresh ZnO NPs in A L cells. The LC 100 of both fresh ZnO NPs (90-200 nm and 20 nm) in our present study was lower than 15 mg/mL ( Fig. 3), which was consistent with previous reporting that the LC 100 of ZnO NPs with 19-36 nm to NIH-3T3 or MSTO cell is about 15 mg/mL [24]. We con rmed that the environmental transformations of physicochemical properties in NPs are expected to dramatically alter their toxicity. It has been reported that the sul dation process of ZnO NPs changed their charge, hydrophobicity, and aggregation state, resulting in the adsorption of sul de state NPs in human saliva, sweat, and bronchoalveolar lavage uid. And the protein adsorbed by ZnO NPs formed a special protein crown, which affected its biological effect [25]. Phosphates widely present in physiological solutions (such as saliva) could convert ZnO NPs into metastable ZnHPO 4 and Zn 3 (PO 4 ) 2 within about 5-10 h, and showed cytotoxicity to digestive tract epithelial cells [16]. Ivasket al. proved the occurrence of complete transformation of ZnO NPs (≤ 3 µg/mL) in the in vitro exposure system to human T lymphocytes (37 °C, cell culture medium RPMI1640 containing 10% FBS for 24 h) using synchrotron radiation X-ray absorption near-edge structure spectroscopy (XANES). The spectrum and cytotoxicity of the transformation products were consistent with those of ZnSO 4 [17]. Our results in this study showed that ZnCl 2 is toxic to A L cells and the cell viability showed a dose-and time-dependent decrease tendency, but its cytotoxicity seems much lower than fresh and aged ZnO NPs (Fig. 3). It was consistent with the nding that Zn 2+ released from ZnO NPs could not fully explain the toxicity fresh ZnO NPs to cells [26].
Our previous study also showed that aged ZnO NPs have an increase in size and ROS generation but signi cantly decreased cell death rate compared to fresh ZnO NPs [19]. Aged ZnO NPs caused less cytotoxicity and could be easier to adapt for the mammalian cells. The present study of RNA sequencing data illustrated that apoptotic genes have been up-regulated in fresh ZnO NP-treated cells, where they were much less affected in aged NP-treated groups. IL1α and IL1β are members of the interleukin 1 cytokine family. The release of IL1α and IL1β activates Caspase 8 partially dependent apoptosis [27]. CD69 encodes a member of the calcium-dependent lectin superfamily of type II transmembrane receptors. Increased CD69 expression was associated with an increased expression of the apoptosis annexin V and CD95 (Fas) marker [28]. JUN is an AP-1 Transcription Factor Subunit. Increased JUN activity proteolytically cleavage alpha-fodrin, a substrate of the interleukin 1beta-converting enzyme (ICE), and CED-3 family of cysteine proteases, which further causes programmed cell death [29]. The increased expression of these apoptotic genes revealed that fresh NPs trigger apoptosis in several different ways. After the elevation of these apoptotic gene expressions, apoptosis processes are eventually executed by apoptotic proteins (Fig. 7). Caspase 3 is the core protease for various apoptotic scenarios, cleavage of this protein is necessary to activate both extrinsic and intrinsic apoptotic pathways [30,31]. Therefore, detection of cleaved caspase3 is a common method for identifying apoptosis induced by a wide variety of apoptotic signals [32]. Our western blotting data revealed that, for both 20 nm and 90-200 nm ZnO NPs, sublethal exposure did not alter the level of Pro caspase 3 in all treatment groups. In contrast, cleaved Caspase 3 was signi cantly elevated by fresh NPs treatment, where aged NPs showed few (if any) effects on the level of cleaved caspase 3 (Fig. 6). Combined with RNA expression analysis, our results clearly elucidated the higher potency of fresh ZnO NPs in inducing cell apoptosis.

Conclusions
In the present study, the natural physicochemical transformation of ZnO NPs in ultrapure water was con rmed, and variations in cytotoxicity induced by fresh & aged NPs were investigated. We focused on RNA sequencing data from our aged ZnO NP-treated A L cells and that of fresh NPs from database. We compared those signaling pathway speci cally enriched in aged NP-treated group, which are different from that of fresh NP-or ZnCl 2 -treated groups. Our data indicated that the lower cytotoxicity of aged ZnO NPs is closely related to the low level of apoptosis induced by it and that the transcriptional regulation of the multiple pathways activated by NPs help the mammalian cells to build the cellular homeostasis. Declarations