Exploring the effects of Mt on the eye is important for assessing human health risks and potential ocular applications. Although Mt is widely used, their toxic potential has not been well studied so far, especially its corneal toxicity, which is still unclear. Here, a comprehensive study of five types of Mt (Na-Mt, H-Na-Mt, C-H-Na-Mt, Ca-Mt, and MMt) and their corneal toxicity in vitro and in vivo was performed, and the underlying mechanism of Mt-induced corneal toxicity experimentally explored in detail.
Previous studies reported that Mt exerts toxic effects on kidney cells [13], hepatoma cells [24], bronchial epithelial cells [25], and mice lung tissue [26]. To our best knowledge, this is the first study to show that corneal toxicity can arise from exposure to different types of Mt. Notably, both the unmodified Mt and modified Mt have been used as drug delivery vectors in studies of ocular diseases; however, the toxicity of these Mt types in the eye has gone overlooked. For example, bromonidine, an α2-adrenergic receptor agonist, is inserted into the interlamellar space of the Na-Mt lattice to increase the preocular surface retention time of the drug and to reduce intraocular pressure in glaucoma patients [21]. In addition to pristine Na-Mt, studies have shown that some organically modified Mts, such as H-Na-Mt [17, 22, 23] and C-H-Na-Mt [35], can also deliver eye drops to delay drug release and increase drug bioavailability for use in glaucoma therapy. The eye is a highly special and sensitive organ. Although Mt present in wastewater and food packaging does not directly contact the eye, there is still a risk of damage to it from long-term exposure to the presence of Mt. Both Ca-Mt [36] and MMt [11] have been studied for use in the field of environmental governance. Accordingly, in our study, Na-Mt, H-Na-Mt, C-H-Na-Mt, Ca-Mt, and MMt were chosen for their corneal toxicity assessment, and human HCEC-B4G12 corneal cells were used in the in vitro assays.
Since the safety profile of Mt may be changed by their organic modification [26, 37], we first characterized the different types of Mt to compare the differences between their physicochemical properties and to verify the successful Mt modifications (Fig. 1). The SEM and TEM enabled us to analyze the surface morphology of Na-Mt, H-Na-Mt, C-H-Na-Mt, Ca-Mt, and MMt. Overall, Ca-Mt appears to have less cracking and smoother surfaces relative to Na-Mt. The many spherical nanoparticles appearing on the surface of Mt after its Fe3O4 treatment confirmed the successful preparation of MMt. Acid washing of Na-Mt using H2SO4 increased wrinkles to the edge of the thin layer; subsequent introduction of chitosan into H-Na-Mt significantly reduced the presence of nanosheets. Furthermore, the successful preparation of H-Na-Mt and C-H-Na-Mt was verified by XRD and EDS mapping. After successfully obtaining the two pristine Mts (i.e., Na-Mt and Ca-Mt) and the three organically modified Mts (i.e., H-Na-Mt, C-H-Na-Mt, and MMt), we were in a sound position to investigate their toxicity to cornea cells, both in vitro and in vivo.
To confirm whether Mt and its derivatives are capable of corneal toxicity, we performed extensive cytotoxicity tests using HCEC-B4G12 cells. In this respect, ATP content, cell viability and LDH leakage are widely used indicators in cytotoxicity assays [28, 33]. All five tested Mt types induced different magnitudes of cytotoxicity in the HCEC-B4G12 cells. With a greater Mt concentration and longer exposure time, ATP content and cell viability of HCEC-B4G12 cells gradually decreased, while LDH leakage gradually increased, suggesting that enhanced Mt cytotoxicity was time- and concentration dependent (Fig. 2). Except for MMt, among the Mts, the other four types exhibited high cytotoxicity, of which Na-Mt was clearly the most toxic; hence, H-Na-Mt, C-H-Na-Mt, and MMt as modified based on Na-Mt decreased the cytotoxicity of HCEC-B4G12 cells to varying degrees, respectively. This finding is similar to that reported in a recent study [26], in which modifying the Mt with quaternary ammonium compounds reduced its pulmonary toxicity to exposed mice. By contrast, other research reported that the organic modification of Mt could increase its hazard potential. For example, Cloisite 30B, an organically modified Mt nanoclay, induced higher toxic effects relative to the original nanoclay in human lung cells [25]. In other work, Connolly et al. [37] prepared two organoclays, by introducing hexadecyl trimethyl ammonium bromide and octadecyl trimethyl ammonium chloride into bentonite, respectively; both increased cytotoxicity in different cell models, including HaCaT skin cells, C3A liver cells, and J774.1 macrophage-like cells.
When cells are treated with nanomaterials and inorganic materials with large particle sizes, most of these materials are taken up by cells and they can even enter key organelles such as mitochondria, thus producing toxic effects in these cells [38, 39, 40]. Here we employed TEM to visualize the distribution of different types of Mt in cells and the negative intracellular changes induced by their treatment with various Mts (Fig. 3). TEM imaging confirms that all five types of Mt are taken up by HCEC-B4G12 cells within 24 h and localized within the cytoplasm; the mitochondrial structure of cells was destroyed, which implies the potential mechanism of cytotoxicity induced by Mts may be ascribed to greater apoptosis generated by augmented ROS production and mitochondria destruction. An in vitro study using J774.1 macrophages demonstrated that cells can uptake organoclays, which are then encased in multiple intracellular vesicles in the cytoplasm [37]. Similarly, in our study, intracellular vesicles containing different types of Mt were clearly detected in the cytoplasm of HCEC-B4G12 cells and some of these cells were undoubtedly in a state of Mt phagocytosis. Notably, a large amount of Na-Mt is attached to the cell membrane but not so in the cytoplasm, which could explain why Na-Mt is more toxic. Moreover, several studies have shown that certain nanomaterials, including SiO2 and CeO2 nanoparticles (NPs), can alter the morphology of cells [28, 41, 42]. Thus, we further investigated whether the cytotoxicity and the cellular uptake of Na-Mt, H-Na-Mt, C-H-Na-Mt, Ca-Mt, and MMt are capable of affecting the morphology of HCEC-B4G12 cells. As seen in Fig. 4, HCEC-B4G12 cells without an Mt treatment were evenly distributed, having a regular shape and smooth surface, whereas the density of cells exposed to the five types of Mt decreased in a time- and dose-dependent manner, and their cell membrane was damaged and the cell contour was ambiguous. This result is consistent with another study that found Fe3+-Mt damaged the cell membrane and changed the morphology of A549 cells [43]. That Na-Mt caused the most severe morphological changes, whereas MMt could only elicit significant morphological changes at a high concentration, is consistent with their cytotoxicity testing results (Fig. 2).
As mentioned above, we selected Na-Mt (pristine Mt) and C-H-Na-Mt (organically modified Mt) as representatives for the follow-up in-depth toxicity tests, because of their distinctive toxic effects and known use in ocular applications. Guided by published experimental methods [42, 44], we then examined the in vivo injuries and apoptosis to the cornea of rats exposed to Na-Mt or C-H-Na-Mt exposure through 1-week repeated exposure experiments (Fig. 5). Consistent with in vitro cytotoxicity, all these latter results showed concentration-dependent corneal toxicity of Na-Mt and C-H-Na-Mt, being higher from Na-Mt.
Oxidative stress is generally the most commonly studied factor after toxic effects in toxicology studies. The excessive increase in ROS generation due to mitochondrial damage disrupts the balance between oxidative pressure and the natural antioxidant defense system, resulting in oxidative stress [45, 46]. Previous studies have found that oxidative stress may be an important toxicological mechanism responsible for the adverse health effects of inorganic nanomaterials and air pollution particles [28, 38, 47, 48, 49]. Work by Maisanaba et al. [24] demonstrated that Mt (Cloisite®Na+) at relatively high concentrations can induce oxidative stress in HepG2 cells that is accompanied by increased ROS production. An earlier study obtained similar results, in which Mt (Cloisite Na+®) induced a significant increase in ROS production in HepG2 cells; however, ROS production did not occur in cells treated with organoclay (Cloisite 93A®) [50]. The excessive production of ROS, as an upstream event, can trigger oxidative stress that ultimately leads to cell death [33]. Our experimental data prove, for the first time, whether in vivo or in vitro, that exposure to Mt with different properties drives ROS overproduction in corneal cells (Fig. 6). Crucially, in vitro, ROS production induced by Mt in HCEC-B4G12 cells was time- and concentration-dependent; likewise, ROS production in corneal tissues exposed to Mt in vivo was also concentration-dependent. It is worth noting that, in the in vitro experiments, the trend in ROS production induced by the parent Na-Mt is of decreasing after attaining its peak value, whereas that induced by modified C-H-Na-Mt is one of continual increase. A reason for this disparity could be that the Na-Mt treatment quickly leads to the disappearance of mitochondria in cells or even the death of many cells, so that ROS can no longer be produced, in contrast to the C-H-Na-Mt treatment under the same conditions, which causes far less damage to cells. Furthermore, the inhibition of ROS significantly lessened the rate of cellular ATP depletion induced by the different types of Mt (Fig. 6), suggesting that ROS accumulation is the upstream event that triggers cytotoxicity, this being consistent with our previous findings [28, 33, 41]. Additionally, we found that Mt activated the MAPK signaling pathway both in vitro and in vivo (Fig. 7). MAPK signaling entails the involvement of p38 MAPK, JNK, and ERK, which participate in various physiological and pathological processes [33, 51]. Interestingly, ROS is known to mediate the activation of p38, JNK, and ERK1/2 signaling molecules in the MAPK signaling pathway [52]. In our study, pretreatment with NAC, an antioxidant that scours for ROS, attenuated the Mt-induced increase of p-p38, revealing that p-p38 activation was in fact ROS-mediated (Fig. 8C). The use of a p38 MAPK inhibitor also provided evidence that p38 is paramount for Mt-induced cell death (Fig. 8A, B). However, ERK1/2 and JNK were phosphorylated under Mt exposure (Fig. 7), but the NAC treatment did not affect their activation (Fig. 8C), which suggests these processes may not be directly related to ROS. Meanwhile, pretreatment with ERK1/2 and JNK inhibitor did change the ATP content compared with treatment Mt alone (Supplementary Fig. 1). We should emphasize that ROS inhibition only partially alleviated the cytotoxicity caused by Mt. The further causes of Mt-induced toxicity remain to be investigated.