Calpain activation and disturbance of autophagy are induced in cortical neurons in vitro by exposure to HA/β-Ga₂O₃:Cr³⁺ nanoparticles

Introduction

Nanotechnologies have been steadily growing and are used for the delivery of therapeutic material, leading to new important applications in health and life-science (Andrews & Weiss, 2012; Lu et al., 2014a; Lu et al., 2014b; Huang et al., 2013). Nanoscience allows for new approaches in medical intervention, and it may revolutionize established clinical practices (Wagner et al., 2006). For instance, nanoparticles (NPs) represent a promising drug delivery system for treating brain disease (Cardoso et al., 2010; Wang et al., 2016a; Wang et al., 2016b; Ying et al., 2014). However, safety issues related to NPs exposure continue to be debated.

The use of engineered NPs in neurology involves innovative pharmacological strategies, and the blood-brain barrier (BBB) is a defense mechanism against potentially neurotoxic molecules and structures (Barbu et al., 2009). The application of NPs to the central nervous system (CNS) is at an early stage, and the toxic effects of NPs observed in in vitro models of neural cells have been scarce (Landsiedel et al., 2012). Additionally, Fe₄O₃ NPs showed promise for in vitro applications but were subsequently discarded after exhibiting toxicity when utilized in vivo. Therefore, these NPs are now limited to in vitro diagnostics (Win-Shwe & Fujimaki, 2011; Zhang et al., 2016; Bao et al., 2014; Haute & Berlin, 2017). Another type of NPs, gold ultrafine particles (UFPs; <100 nm), exhibited neurotoxicity by causing transient microglia activation and the induction of Toll-like receptor 2 (TLR-2) promoter activity in transgenic mice (Elder et al., 2006; Hutter et al., 2010). The neurotoxicity of NPs generally includes an inflammation response, DNA damage, cell death (Shin et al., 2010; Yang & Klionsky, 2010). Examples of neuronal damage observed in in vivo experiments include damage to the nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB) and inducible nitric oxide synthases (iNOS) in the cortex exposed to air with high concentrated metal NPs. BBB injury has also been observed (Calderon-Garciduenas et al., 2008; Castranova, 2004; Tian & Lawrence, 1996). For these reasons, the safety of NPs intended for biomedical applications should be carefully evaluated for each new type of engineered NPs (Rivet et al., 2012).

Gallium oxide (Ga₂O₃) is a wide band gap semiconductor that is applied in optical components and optical catalysis, for example, catalytic reactions involving CO, H₂ and methanol (Reddy, Ko & Yu, 2015). Ga₂O₃ has multiple crystal forms, including α, β, γ, ε, δ. The β crystal form is the most stable among them, and adding a chromium ion (Cr³⁺) when preparing Ga₂O₃ (β-Ga₂O₃: Cr³⁺) NPs enables the NPs to emit fluorescence and near-infrared persistent luminescence (Wang et al., 2015a; Wang et al., 2015b). β-Ga₂O₃:Cr³⁺ NPs modified with hyaluronic acid (HA) could effectively absorb antineoplastic drugs, and they were shown to bind with CD44, which is highly expressed in tumor cells and is primarily distributed at the tumor site (Ganguly et al., 2016). HA/β-Ga₂O₃:Cr³⁺ NPs have been used as drug carriers to deliver anti-cancer drugs, and it was found that fabricated HA/β-Ga₂O₃: Cr³⁺/DOX could be taken up by HeLa and MCF-7 cells (Wang et al., 2015a; Wang et al., 2015b). In this study, the cell viabilities of both MCF-7 and Hela were all above 80% even when the concentration of HA/β-Ga₂O₃:Cr³⁺ NPs was up to 1,000 µg/ml. And the final concentration of NPs used as a drug carrier is 40 µg/ml. Nevertheless, for the successful translation from bench to bedside, a promising drug delivery system should have low neuronal toxicity.

We here determined the impact of HA/β-Ga₂O₃:Cr³⁺ NPs on neurons. Using the SH-SY5Y cell line and primary cortical neurons in combination with pharmacological methods, in an effort to investigate the cell injury-related signaling following the HA/β-Ga₂O₃:Cr³⁺ NPs exposure. Our results provide evidence to implicate the role of calpain and autophagy in HA/β-Ga₂O₃:Cr³⁺ NPs-induced neurotoxicity.

Materials and Methods

Results

HA/β-Ga₂O₃:Cr³⁺ NPs-induced changes in cell morphology and reduced cell viability

First, the stability of HA/β-Ga₂O₃:Cr³⁺ NPs in different kinds of culture media was detected. As shown in Fig. S1, no significant difference was observed in the size distribution of NPs after incubated in medium with or without FBS for 12 h. SH-SY5Y cells were observed with an inverted fluorescence microscope after incubation with various concentrations (0, 1, 5, 25, 50, 100 µg/mL) of HA/β-Ga₂O₃:Cr³⁺ NPs for 12 h. The results showed that cell morphology of neurons was normal at concentrations lower than 5 µg/mL, but cell deformation was observed at concentrations higher than 25 µg/mL (Fig. 1A). The cell viability relative to the control group after exposure to 1, 5, 25, 50, 100 µg/ml HA/β-Ga₂O₃:Cr³⁺ NPs for 12 h was 93.18%, 88.15%, 83.94%, 78.95%, 71.86% respectively, indicating that the cell viability was decreased in a concentration-dependent manner after treatment with HA/β-Ga₂O₃:Cr³⁺ NPs (Fig. 1B, **P < 0.01, ***P < 0.001). The results of NPs toxicity were further confirmed using the TUNEL assay (Fig. 1C). Consistently, the number of TUNEL⁺ cells relative to DAPI⁺ cells was increased in a dose-dependent manner after exposure to HA/β-Ga₂O₃:Cr³⁺ NPs.

HA/β-Ga₂O₃:Cr³⁺ NPs decreased the phosphorylation of CaMKII in neurons

We next investigate the neuronal toxicity of HA/β-Ga₂O₃:Cr³⁺ NPs using primary cortical neurons from wild-type mice. CaMKII is enriched at synapses where it plays a critical role in regulating synaptic transmission (Li et al., 2017; Liu et al., 2007). Here, the phosphorylation of GluR 1 (Ser831), CaMKII (Thr286) and Synapsin I (Ser603) were assessed by Western blotting in primary cortical neurons following HA/β-Ga₂O₃:Cr³⁺ NPs treatment for 12 h. Phospho-CaMKII (Thr286) was significantly decreased after treated with 25 and 50 µg/mL HA/β-Ga₂O₃:Cr³⁺ NPs (Fig. 2A). Consistently, the expression level of phospho-Synapsin I (Ser603) was markedly decreased in the 25 and 50 µg/mL NPs treatment groups. There was no significant change in the phosphorylation of the postsynaptic AMPA receptor subtype GluR 1 after exposure to various concentrations of NPs. These data suggested that exposure to HA/β-Ga₂O₃:Cr³⁺ NPs may interfere with the functions of cortical neurons.

HA/β-Ga₂O₃:Cr³⁺ NPs disturbed autophagy signaling

Dysfunction of autophagic pathways plays a major role in the pathogenic process of cell injury (Han et al., 2011a; Jiang et al., 2017). We found that the expression of mature Cathepsin B was increased in primary cortical neurons after treatment with HA/β-Ga₂O₃:Cr³⁺ NPs, compared to the control group (Figs. 3A and 3B, P < 0.01). As shown in Figs. 3C–3E, the expression of type II LC3 (16 kDa) was significantly elevated, as well as the level of p62 (P < 0.05). Taken together, these results indicated that the fusion of autophagosomes and lysosomes in the autophagy flux may be blocked in primary cortical neurons after treatment with HA/β-Ga₂O₃:Cr³⁺ NPs for 12 h, thereby resulting in the aberrant accumulation of LC3-II and p62. Consistently, the fluorescence intensity of Cathepsin B (green) was significantly enhanced in a concentration-dependent manner following treatment with HA/ β-Ga₂O₃:Cr³⁺ NPs (Fig. 4). We also used 2′, 7′-dichlorofluorescein diacetateto detect ROS levels in the cells after incubation with NPs. As shown in Fig. S2, an increase in ROS was observed at a dose as low as 1 µg/mL HA/β-Ga₂O₃:Cr³⁺ NPs.

Discussion

The increasing applications for nanomaterials have led to a growing concern about the bioavailability and toxicity of nano-sized materials (Igarashi, 2008). Herein, we evaluated the molecular mechanism underlying HA/β-Ga₂O₃:Cr³⁺ NPs-induced neurotoxicity. Our experimental results showed that HA/β-Ga₂O₃:Cr³⁺ NPs at a certain concentration could decrease cell viability and damage the synaptic functions by inducing calpain activation and excessive autophagy.

β-Ga₂O₃ has been widely applied in optoelectronic devices by being doped with foreign impurities, which can radiate various photoluminescence spectra from visible to near-infrared light (Nogales et al., 2007; Liu et al., 2008). β-Ga₂O₃ doped with Cr³⁺ has been used not only as a fluorescent probe for bio-imaging but also as a drug carrier for biomedicine (Wang et al., 2015a; Wang et al., 2015b). The viability of both MCF-7 and HeLa cells was greater than 80%, even when the concentration of HA/β-Ga₂O₃:Cr³⁺ NPs was as high as 1,000 µg/mL (Wang et al., 2015a; Wang et al., 2015b). Pyramidal neurons in the brain are responsible for brain functions, and they are the most sensitive and delicate cells in bio-organisms, which means that they are among the most vulnerable cells to nanoparticles (Paterson-Brown et al., 1986). To our knowledge, studies addressing whether the HA/β-Ga₂O₃:Cr³⁺ NPs nanoparticles could induce neuronal toxic effects have not been reported previously. In this study, the cellular morphology, cell viability and TUNEL were examined in the SH-SY5Y cell line and in primary cortical neurons to assess the neuronal toxicity of HA/β-Ga₂O₃:Cr³⁺ NPs. The data showed that cell viability relative to the control group after exposure to 1, 5 or 25 µg/mL HA/β-Ga₂O₃:Cr³⁺ NPs for 12 h was 93.18%, 88.15%, 83.94%, respectively, indicating that HA/β-Ga₂O₃:Cr³⁺ NPs can induce neuronal toxicity even at a low concentration (1 µg/mL). The final concentration of NPs we used as a drug carrier was 40 µg/mL (Wang et al., 2015a; Wang et al., 2015b). From these results, we conclude that neurons are very sensitive to HA/ β-Ga₂O₃:Cr³⁺ NPs, and this makes them a prime target for the neurotoxic effects of exogenous HA/β-Ga₂O₃:Cr³⁺ NPs.

CaMKII is enriched at the synapse and plays important roles in controlling synaptic strength and plasticity, which are mediated through substrate binding and the intramolecular phosphorylation of holoenzyme subunits (Li et al., 2017; Liu et al., 2007; Navakkode et al., 2017). Paralleled with the down-regulation of phospho-CaMKII (Thr286), phospho-synapsin I (Ser603) was also significantly decreased in the cultured primary cortical neurons following exposure to higher concentrations of HA/β-Ga₂O₃:Cr³⁺ NPs (25 and 50 µg/mL) for 12 h, whereas the expression level of total synapsin I did not change. Interestingly, we also observed that phospho-synapsin I (Ser603) levels were transiently elevated following exposure to HA/β-Ga₂O₃:Cr³⁺ NPs at 5 µg/mL, and the same phenomenon was also observed for the phosphorylation of CaMKII. This turnover may be interpreted as a compensatory mechanism of the neurons. The activation of CaMKII is closely related to changes in phosphorylation of its post-synaptic and pre-synaptic substrates, such as synapsin I and GluR1 (Liao et al., 2013). However, there was no significant change observed in the phosphorylation of GluR1 following NPs treatment; this may due to the different sensitivity of presynaptic and postsynaptic structures following exposure to HA/β-Ga₂O₃:Cr³⁺ NPs.

Accumulating evidence has shown that intracellular Ca²⁺ increases after cells are exposed to a variety of nanoparticles (Dubes et al., 2017; Huang et al., 2009), and this increase has been associated with the inhibition or activation of various intracellular signaling pathways, leading to a cytotoxic effect. These alterations in Ca²⁺ homeostasis lead to persistent, pathologic activation of calpain, which is indicative of cell cytoskeletal injury (Lu et al., 2016). Our results revealed the pathologic activation of calpain following exposure to HA/β-Ga₂O₃:Cr³⁺ NPs, detected by increased levels of cleaved fragments of spectrin and calcineurin. The hyperactivation of calpain is implicated in the breakdown of cytoskeletal molecules, such as spectrin, microtubule subunits, microtubule-associated proteins, and neurofilaments (Lu et al., 2016; Wu et al., 2004; Ding et al., 2016). Taken together, these data confirm the role of the calpain pathway as a regulatory mechanism involved in HA/β-Ga₂O₃:Cr³⁺ NPs-induced neuronal toxicity.

The homeostasis of autophagy-lysosome signaling plays a critical role in protecting against neurovascular injury (Lu et al., 2014a; Lu et al., 2014b; Han et al., 2011a; Jiang et al., 2017). Based on our results, we could conclude that the disturbance of the autophagy process was also involved in HA/β-Ga₂O₃:Cr³⁺ NPs-induced neurotoxicity. It has been reported that NPs-induced autophagy may be an adaptive cellular response, aiding in the degradation and clearance of the nanomaterial; however, overstimulation of autophagy can induce harmful cellular dysfunction (Johnson-Lyles et al., 2010; Li et al., 2010; Neun & Stern, 2011). During the autophagy process, cathepsin B is released into the cytoplasm from lysosomes and is subsequently activated. Therefore it is involved in the activation of inflammatory processes, which result in an increased inflammation response (Han et al., 2011b; Nakashima et al., 2003). Here, the expression of cathepsin B, LC3 II and SQSTM/p62 was significantly increased in primary cortical neurons following incubation with 10 µg/mL HA/β-Ga₂O₃:Cr³⁺ NPs for 12 h. The immunocytochemistry results also confirmed the elevation of cathepsin B after exposure to HA/β-Ga₂O₃:Cr³⁺ NPs, which was in agreement with the increased uptake of NPs. Since several mechanisms can induce the dysfunction of autophagy flux (Nakashima et al., 2003), we also examined the ROS levels in cells following HA/β-Ga₂O₃:Cr³⁺ NPs treatment. Our fluorescence images showed that HA/β-Ga₂O₃:Cr³⁺ NPs exposure increased the generation of ROS. Indeed, the accumulation of ROS induced by exposure to NPs can disturb the autophagy flux, ultimately leading to cell death, and it can change neuronal function (Wang et al., 2016a; Wang et al., 2016b). Indeed, biological effects of metal ions (Ga³⁺ or Cr³⁺) are poorly understood and their potential toxicity vary depending on the exposure conditions (Catelas et al., 2003; Huk et al., 2004; Wang et al., 2015a; Wang et al., 2015b; Pourshahrestani et al., 2017). Besides the data for stability of HA/β-Ga₂O₃:Cr³⁺ NPs, an important limitation of the present study was that lack of extensive characterization of the solubility and release of the Ga³⁺ or Cr³⁺ over the time course of the experiment. Therefore, a good safety margin should be further studied in an animal model to investigate the neuronal toxicity of HA/β-Ga₂O₃:Cr³⁺ NPs in the context of an intact brain with or without risk factors.

Conclusions

In summary, our results suggested that the exposure of neurons to HA/β-Ga₂O₃:Cr³⁺ NPs could induce calpain activation and the disturbance of autophagy signaling, which may result in neuronal damage.

Supplemental Information