The Involvement of Oxidative Stress, Neuronal Lesions, Neurotransmission Impairment, and Neuroinammation in Acrylamide-Induced Neurotoxicity

Acrylamide (ACR) is a typically contaminate during environmental industry and poses potential health hazards that have been attracting increasing attention. Its neurotoxicity is known to cause signicant damage to health. However, the mechanisms of ACR-induced neurotoxicity require further clarication. This study explores how ACR-induced oxidative stress, neuronal lesions, neurotransmission impairment, and neuroinammation mutually contribute to neurotoxicity using a mouse model. According to the results, oxidative stress was indicated by the presence of a distinct increase in cellular reactive oxygen species levels, malondialdehyde, and 8-hydroxy-2-deoxyguanosine content, as well as a signicant decrease in the glutathione content after ACR exposure. Moreover, ACR caused neurological defects associated with gait abnormality and neuronal loss while suppressing the levels of acetylcholine and dopamine and increasing the protein expression of α-syn, further inhibiting cholinergic and dopaminergic neuronal function. Additionally, ACR treatment caused an inammation response via NF-κB activation and increased the protein expression of NLRP3. Consequently, ACR activated the NLRP3 inammasome constituents, including Caspase-1, ASC, N-GSDMD, IL-1β, and IL-18. The results revealed the underlying molecular mechanism of ACR-induced neurotoxicity via oxidative stress, neurotransmission impairment, and neuroinammation-related signal cascade. This information will further improve the development of an alternative outcome pathway strategy for investigating the risk posed by ACR.


Introduction
Acrylamide (ACR) is a white crystal-chemical used to produce polyacrylamide (Friedman, 2003). It has been traditionally used as an industrial raw material in soil conditioning, wastewater treatment, the textile industry, and more (Erkekoglu and Baydar, 2014). Reports have indicated that workers can be heavily exposed to ACR, showing symptoms and signs of peripheral nervous dysfunction (Hagmar et al., 2001).
In addition, ACR has been revealed as a typical contaminant originating from the heat processing of food (Mottram et al., 2002;Stadler et al., 2002;Tareke et al., 2002). Since the presence of ACR has been revealed in food, ways to reduce its content has been continually sought, but existing methods cannot completely inhibit its formation (Dias et  Studies have shown that ACR displays neurotoxicity, genotoxicity, reproductive toxicity, and potential carcinogenicity (Exon, 2006). Of the ACR-induced injuries, damage to the nervous system is the most serious (Guan et al., 2018;LoPachin et al., 2003). During the early 1990s, researchers observed subchronic neurotoxic symptoms in workers exposed to ACR for an extended time (Abelli et al., 1991).
Repeated exposure to ACR at levels between 0.5 mg/kg/day and 50 mg/kg/day induced similar neurotoxic phenotypes in several species of laboratory animals, such as rodents, rabbits, guinea pigs, cats, and dogs (Barber et al., 2007;Erkekoglu and Baydar, 2014;LoPachin, 2004). Chemical level analysis indicated that this might be due to the formation of a covalent adduct between the ACR and highly nucleophilic cysteine at the active presynaptic neuron site, leading to neuron inactivation and affecting the neurotransmitter transmission process, nally resulting in neurotoxicity (LoPachin and Gavin, 2012).
From a physiological and biochemical perspective, oxidative stress, which acts as an activation signal, is associated with direct or indirect ACR-induced neurotoxicity (Zamani et al., 2017;Zhao et al., 2017b).
Oxidative stress is caused by an imbalance between the production and elimination of oxygen species (ROS) (Cook and Petrucelli, 2012), while the central nervous system actively participates in oxygen metabolism (Patel, 2016). However, it is highly susceptible to oxidative damage since the enzymatic antioxidant activity in this region is lower than in other tissues (Salim, 2017). Both in vivo and in vitro experiments have shown that ACR can induce oxidative stress in neurocytes (Pan et al., 2017;Yousef and El-Demerdash, 2006), further inducing mitochondrial-dependent apoptosis that leads to neurotoxicity (Chen et al., 2013;Lee et al., 2014). In addition, cholinergic and dopaminergic dysfunction reportedly play a role in various neurodegenerative diseases, as well as other pathologies of the brain, such as Parkinson's disease and Alzheimer's disease, and are associated with oxidative stress (Pepeu and Grazia Giovannini, 2017;Qamar et al., 2017). Furthermore, ACR reduces acetylcholinesterase (AChE) activity in the brain (Elblehi et al., 2020). However, the relationship between ACR-induced oxidative stress and the dysfunction of neurotransmission requires further investigation.
Additionally, in ammation can also activate an immune response and enhance ACR-induced neurotoxicity (Santhanasabapathy et al., 2015). A previous study found that ACR induces in ammation via NF-κB translocation, releasing the downstream cytokine, IL-1β, to enhance the neurotoxic effect (Zhao et al., 2017a). The NOD-like receptor family pyrin domain containing 3 (NLRP3) refers to a downstream site regulated by NF-κB that is involved in the IL-1β release pathway (Zhong et al., 2016b). NLRP3 in ammasome activation is essential in triggering the in ammatory responses in the central nervous system (Song et al., 2017;Zhou et al., 2016). Oxidative stress can activate and assemble NLRP3 in ammasome (Abderrazak et al., 2015), playing a crucial role in neurodegenerative disease development, such as cerebral arteriosclerosis and Parkinson's disease (Heneka et al., 2014;Libby and Everett, 2019;Pirzada et al., 2020;Sarkar et al., 2017). We hypothesized that the shared target during NLRP3 in ammasome activation, ACR-induced neurotoxicity, and neurodegenerative factors suggest that NLRP3 in ammasome activation may lead to ACR neurotoxicity and may ultimately contribute to pathological neurodegenerative processing.
Gait abnormality are considered classical morphological characteristics of the neurotoxicity resulting from ACR (Tan et al., 2019). Several studies have examined cerebellum damage and/or peripheral motor nerve injury to explain ACR-induced gait abnormality (Lebda et al., 2015; Yan et al., 2018). Locomotion is a highly complex process involving afferent sensory input, central neuron system processing, efferent motor commands and skeletal muscle coordination, which can be compromised due to failure at multiple levels (LoPachin et al., 2002b). Thus, both the cerebellum and cerebrum are important targets of ACRrelated neurotic injury and degeneration (He et al., 2017;Lehning et al., 2002). However, the systemic analysis of ACR-induced dysfunction in both the cerebellum and cerebrum remains lacking. Furthermore, the synergistic contribution of oxidative stress, neurotransmission impairment, and neuroin ammation to ACR-induced neuronal lesions in the cerebellum and cerebrum requires further clari cation.
This study aims to determine the connection between ACR-induced neurotic toxicological signi cance and the molecular mechanism in the central neuron system. First, a mouse model exposed to low, medium, and high-ACR doses is established to present the different neurotoxic processes. Then, the ACRmediated neurotoxic and neurodegenerative effects are assayed. Follow-up analyses revealed that ACRinduced neurotoxicity stems from oxidative stress, neurotransmission impairment, and neuroin ammation in the central nervous system. Moreover, the signal cascade of NLRP3 in ammasome prime and activation is analyzed. This study aims to establish a more complete theoretical basis to reveal the mechanism of ACR toxicity and improve the adverse outcome pathway, provide a novel approach to the nal targeted search of its toxic intervention, and promote the coordinated development of health safety.

Materials And Methods
Chemicals ACR (purity > 99%) was purchased from Titan (Shanghai, China) and diluted with physiological saline for use in mice.

Animal and experimental design
The Animal Care and Use Committee of Laboratory Animals (Chinese Academy of Sciences; license number: SYXK (HU) 2019-0013) provided research ethics approval for the use of animals. The SLAC Laboratory Animal Center (Shanghai, China, license number: SCXK (HU) 2017-0008) provided the wildtype C57BL/6 male mice.
The mice were accommodated separately in a humidity-and temperature-controlled room, with a 12-h light/dark cycle. The animals received tap water and a standard diet. Before the experiment, they were allowed one week to acclimatize to the environment. At six weeks of age, the mice were randomly divided into six groups according to their weight (three negative control groups and three ACR treatment groups at different doses; n = 10 in each group). ACR neurotoxicity is progressive, while the rate of progression is dose-dependent (LoPachin, 2004). Considering that the dose-rate determined the onset time and development of neurotoxicity, 5 mg/kg BW ACR was selected as the low dose, referring to the lowest observed adverse effect level (LOAEL) of ACR exposure LoPachin, 2004). Furthermore, 25 mg/kg ACR was selected as the medium dose and 50 mg/kg ACR as the high dose, to observed obvious nerve terminal damage caused by ACR (LoPachin et al., 2006). Table 1 shows the experimental details. The negative control groups were fed a powdered standard diet only and given normal saline daily via oral administration. All ACR treatment groups had their negative control groups throughout the experimental progression. The low-dose group received ACR once daily via oral administration at a dose of 5 mg/kg BW for 60 successive days (sub-chronic toxicity). The medium-dose group received ACR once daily via oral administration at a dose of 25 mg/kg BW for 30 successive days (subacute toxicity). The high-dose group received ACR once daily via oral administration at a dose of 50 mg/kg BW for 15 successive days (acute toxicity). Then, 18 h after the last ACR administration, the mice were anesthetized, euthanized, and dissected immediately. The cerebrums, cerebellums, livers, and spleens were immediately removed and weighed. For the biochemical analysis, the brains of ve mice from each group were removed and immersed in liquid nitrogen for later use. For the histopathological examination, the cerebrums and cerebellums of ve mice from each group were xed with a 4% paraformaldehyde (PFA) solution. This process was followed by the dehydration, embedding, and sectioning of the brains, after which they were stained with H&E and Nissl. The samples were photographed, followed by histopathological analysis.
Gait score Ataxia and hind-limb muscle weakness are the primary neurological defects associated with the induction of distal axonopathy by chemicals such as ACR (Spencer and Schaumburg, 1974).
Gait score observations represent a relatively sensitive measure of the onset and progression of neurological changes during ACR exposure (LoPachin et al., 2002b). Here, gait scores were determined every 3 d as indices of developing neurotoxicity, which was assessed by an independent examiner not involved in the experiments using a previously described method (Yan et al., 2018). The abnormal gaits of the mice, such as hind-limb spread, foot splay, and balance disorder, were observed and recorded while the animals were permitted to move freely in an unobstructed environment for 3 min. The scores were divided into four grades ranging from 1 to 4 according to severity. Here, 1 indicated a normal gait, 2 indicated a slightly affected gait, 3 denoted a moderately affected gait, and 4 referred to a severely affected gait. Furthermore, to allow for the occurrence of random errors, three consecutive assessments were conducted.

Landing foot splay
Landing foot splay was used as a simple and sensitive assessment method to determine the degree of disability and peripheral neuropathy caused by ACR-induced neurotoxicity (Edwards and Parker, 1977).
The landing foot splay was measured according to previously described methods to assess athletic muscle ability (Yan et al., 2018b. ). Brie y, at the end of each exposure period, the rear limbs of each mouse were painted with ink. They were dropped from a height of 30 cm onto a soft surface, after which measurements were taken of the distance between the central points of the right and left rear limbs. Three measurements were performed for each mouse at 30 min intervals. The landing foot splay was represented by the average value between the three measurements.

Histochemical analysis H&E staining
The xed tissues were subjected to a routine follow-up process for 48 h. The samples were dehydrated and shaped into para n blocks, which were cut into sections of 4 µm thick for observation on slides. After H&E staining, the histological modi cations were observed using light microscopy (Olympus BX51, Japan), while the pathological analysis of the brain was performed as described previously (Yan et al., 2018). Different colored arrows represented the histological results. The black arrow denoted the nucleus condensation and pycnosis, while the blue arrow indicated changes in the neuron morphology.

Nissl staining
The para n sections were subjected to Nissl staining to evaluate the number and morphological characteristics of the neurons using a light microscope (Olympus BX51, Japan). The evaluation criteria were established according to the toxicological pathology (Kaufmann et al., 2012). Moreover, three sections were assessed for each sample, while the number of Nissl body and the size of the neurons were quanti ed via Analyze Particles using Image J software (Wayne Rasband, NIH).

Biochemical assays
The oxidative stress-related indexes, including the ROS, MDA, 8-OHdG adduct levels, and GSH content of the complete brains of the mice, were measured after tissue homogenization using commercial kits according to standard protocols (Nanjing Jiancheng Bioengineering Institute, Nanjing, China).
The neurotransmission-related enzymes and metabolites, such as dopamine (DA), acetylcholine (Ach), acetylcholinesterase (AChE), and choline acetyltransferase (ChAT), were measured after homogenization of all the brain tissue, using commercial kits according to standard protocols (Nanjing Jiancheng Bioengineering Institute, Nanjing, China).
Measurements of the TNF-α and IL-6 cytokines were obtained using an ELISA kit and the homogenized brains of the mice according to standard protocols (Thermo Fisher Scienti c, USA).

Immunohistochemical staining
Sections of tissue were cut from the para n blocks and positioned on positively charged slides for immunohistochemical staining. These sections were placed in an oven for 1 h, after which they were dehydrated and depara nized by being submerged in a xylol-alcohol series. Then, phosphate buffer solution (PBS) was used to wash the tissue sections, followed by treatment for 10 min with a 3% hydrogen peroxide (H 2 O 2 ) solution. Using an antigen retrieval solution, microwave treatment was performed twice at 500 W for 5 min to clarify the endogenous peroxidase-inactivated antigens in the tissues, which were then cooled at room temperature and washed with PBS employing a protein block to prevent the binding of non-speci c antibodies. After the protein block was removed, the tissues were incubated with anti-NF-κB-p-p65 (1:200; Cell Signaling Technology, USA) and anti-TH (1:250; Abclonal, China) antibodies, followed by a PBS washing process and treatment with secondary antibodies. The tissues were subjected to 3-3′ diaminobenzidine (DAB) chromogen, after which they were washed with PBS for the last time and rinsed with tap water. Mayer's hematoxylin was then used for counterstaining. The TH-positive neurons and NF-κB-p-p65-labeled cells in the cerebrums and cerebellums were observed and counted using Image J software (Wayne Rasband, NIH).

Statistical analysis
The Graph Pad Prism statistical software package (Version 7.0, Graph Pad Software, CA) was used for statistical analysis, curve tting, and graph plotting. The data were expressed as the mean ± standard deviation (SD). Furthermore, the statistical comparisons between the experimental groups were evaluated via one-way ANOVA with Duncan's multiple comparison test using the SPSS software program or twoway ANOVA with Sidak's multiple comparison test using the Graph Pad Prism statistical software. Consequently, P < 0.05 values were regarded as statistically signi cant and were denoted by different letters. Figure 1A illustrates the ow chart of the treatment administered to the animals. Due to the different experimental periods, ACR treatment at low, medium, and high doses had negative control groups throughout the experimental process. The behavior of the mice was monitored daily, while their body weights and gait scores were recorded every 3 d. The landing splay was measured on the last day of every exposure period, i.e., day 15 for the high-dose group and its control group (tan arrow); day 30 for the medium-dose group and its control group (blue arrow); and day 60 for the low-dose group and its control group (purple arrow).

ACR-induced abnormalities in gait in mice
No notable abnormal performances were observed in any of the negative control groups. At an ACR dosage of 5 mg/kg, splay foot evidence started to appear during the nal 6 d of the 60 d period (Fig. 1E). All the mice in the 25 mg/kg/day ACR group started displaying symptoms of hypokinesia and lethargy after 21 d (Fig. 1E). The high-dose, 50 mg/kg ACR group exhibited tremors, walking di culties, and weakness after 10 d (Fig. 1E). These results indicated that both the dosage and time in uenced the gait score. In the 25 mg/kg/day and 50 mg/kg/day ACR groups, the distance between the two rear limbs was the widest of the three groups and increased signi cantly during the landing foot splay tests (P < 0.05, Fig. 1B and Fig. 1F). Therefore, the ACR exposure of the 25 mg/kg and 50 mg/kg dosage groups caused substantial gait abnormalities and weakness in the hind limbs (P < 0.05). Higher doses reduced the time required for ACR toxicity to induce gait abnormality. Figure 1C shows that the average body weight declined slightly in the 25 mg/kg/day and 50 mg/kg/day ACR groups, indicating that medium-and highdose ACR treatment negatively affected the growth of the mice. However, no mortality or clinical signs were observed in the primary organs of the mice in any of the groups. Figure 1D shows that there were no statistical differences between the control group and the ACR treatment groups regarding organ coe cients, including the liver, spleen, cerebrum, and cerebellum (P > 0.05), indicating that ACR did not signi cantly affect the development of the organs of mice.

ACR-induced neuron loss and neuron lesions
The neurotoxic effect of ACR-induced brain injury was further con rmed by H&E staining and histological assessment. As shown in Fig. 2A and Fig. 2B, normal neurocytes in the hippocampus and dentate nucleus displayed a clear nucleus, a uniform cytoplasm, and a well-arranged structure. In contrast, obviously abnormal histo-architectures of hippocampus and dentate nucleus were observed after ACR exposure. All ACR treatments caused neurocytes degeneration, with vague cellular boundaries and neuron pyknosis, denoted by black arrows. Moreover, the neuron morphology was changed (blue arrow), and in ammatory cell in ltration was found after the 50 mg/kg high-dose ACR treatment ( Fig. 2A and  Fig. 2B).
ACR-induced neuronal lesions were observed after Nissl staining. In the hippocampus and cerebellum control groups, the granular cell layer displayed a uniform color and regular arrangement (Fig. 2C and  Fig. 2D). As shown in Fig. 2E and Fig. 2G, the number of Nissl bodies in the hippocampal DG regions and the Purkinje cells in the dentate nucleus were decreased signi cantly after medium-and high-dose ACR treatment, respectively (P < 0.05). Moreover, a distinct decrease in staining intensity and vague cellular boundaries were evident after high-dose ACR treatment (Fig. 2D). It is worth noting that the size of the neurons and Purkinje cells signi cantly increased with obvious cell swelling in the hippocampal DG regions and dentate nucleus after all ACR treatment doses (Fig. 2F and Fig. 2H), exhibiting similar characteristics to ACR-induced cell pyroptosis. In summary, ACR exposure caused signi cant neuron loss and neuron lesions in the cerebrum hippocampus and cerebellum dentate nucleus of the mice.

ACR-induced oxidative stress in mice
Oxidative stress typically causes pathophysiological events in neurodegenerative diseases, as well as during ACR-induced neurotoxic progress (LoPachin and Gavin, 2008). The ROS accumulation, GSH content, and oxidative stress products, such as MDA and 8-OHdG, were measured in the brain homogenate after exposure to all ACR doses. Figure 3A shows signi cant ROS accumulation (P < 0.05), while the GSH content decreased substantially after all ACR treatments in a dose-dependent manner (P < 0.05, Fig. 3D). The levels of the oxidative biomarkers, MDA (Fig. 3B) and 8-OHdG (Fig. 3C) were visibly higher than in the control groups. In the other groups, the MDA level in the 50 mg/kg/day ACR group showed a more signi cant increase (P < 0.05). These results suggest that ACR causes oxidative damage, and the accumulated ROS probably induces a downstream signal cascade.

ACR-induced neurotransmission impairment
Neurotransmitters are essential for animals to transmit nerve impulses (Grant, 2015). Of these, ACh and DA are particularly important for transmitting nerve impulses between synapses (Faure et al., 2014). The results showed that ACR decreased the ACh content (P < 0.05, Fig. 4C) and ChAT activity (P < 0.05, Fig. 4A) in a dose-dependent manner, while ACR exposure increased the activity of AChE. More signi cant AChE activity was promoted only in the high-dose group (P < 0.05, Fig. 4B). These ndings showed that the ACh-related neurotransmitter was inhibited after ACR treatment and may cause gait abnormality.
An increase in the α-syn levels led to abnormal aggregation, causing neuronal degeneration (Ottolini et al., 2017). The expression levels of α-syn were examined via western blot analysis to further clarify the possible mechanisms involved in gait abnormality. The results indicated that ACR caused a signi cant increase in α-syn expression after all ACR treatments (Fig. 4G).
The substantia nigra (SN) is rich in dopaminergic neurons, and is a region of the brain that controls movement which is closely related to neurodegenerative diseases (Patil et al., 2014). TH is the ratelimiting enzyme responsible for conversion of L-DOPA to dopamine (DA) (Daubner et al., 2011). To assess the impact of ACR on the functionality of dopaminergic neurons in the SN, the DA content, TH protein expression, and TH-positive cell levels were investigated (Fig. 4E and Fig. 4G). Compared with the control groups, ACR treatment signi cantly reduced the DA level in all exposure groups, while the lowest concentration of DA was found in the high-dose exposure group (P < 0.05, Fig. 4F). The expression and distribution of TH-positive cells were markedly decreased, indicating that dopaminergic neurons were lost in the SN after ACR treatment (Fig. 4E). These results showed that ACR caused damage to the dopaminergic and cholinergic neurons, leading to neurotransmission impairment and gait abnormality.
ACR increased the NLRP3 priming, promoted neuroin ammation, and activated the NLRP3 in ammasome-related pathway NF-κB represents the upstream signaling of the NLRP3-related in ammasome components (Zhong et al., 2016a). The activation of NF-κB (p-p65) was measured using the immunohistochemical method. The ACR stimuli signi cantly activated NF-κB in the cerebrum and cerebellum, while the levels of p-p65 visibly increased ( Fig. 5A and Fig. 5B). The protein expression levels of NLRP3 also increased signi cantly in all ACR treatment groups ( Fig. 6A and Fig. 7A), indicating NLRP3 priming. Furthermore, the NF-κB pathway participated in the in ammatory reaction, regulating various other cytokines, such as TNF-α and IL-6 (Bonizzi and Karin, 2004). The protein levels of TNF-α and IL-6 increased signi cantly in the cerebrum and cerebellum at all treatment doses (P < 0.05, Fig. 5C and Fig. 5D).

Discussion
Neurotoxicity resulting from ACR causes neuronal degeneration, neuronal energy inactivation, learning and memory alterations, and peripheral motor nerve injury (LoPachin et al., 2002a; Murray et al., 2020; Zhang et al., 2017). However, the underlying mechanism of ACR-induced neurotoxic phenotypes requires further clari cation (Zong et al., 2019). This study indicates that neuroin ammation and neurotransmission impairment in the central neuron system contribute to ACR-mediated neurotoxicity at different intoxication periods in a time-and dose-dependent manner. Subsequently, it was con rmed that an increase in oxidative stress, as well as in ammatory cytokine release, and the activation of NLRP3caspase 1-GSDMD related pathways represent essential responses to ACR-induced neurotoxicity. Additionally, the results reveal the biomarker of ACR-induced gait abnormality and neuronal survival, including α-syn aggression, a decrease in TH + cell distribution and DA content, and the disruption of the ACh-related metabolism. This indicated that ACR-induced neurotoxicity was closely associated with the initiation and progression of neurodegenerative disease. Overall, these results enhance the understanding of the potential mechanisms of ACR-induced neurotoxicity. The cellular responses could present a more systematic insight for establishing alternative pathway strategies for assessing the risk associated with ACR.
In the central nervous system, ACR-induced damage is a systematic response that can be attributed to injury to the cerebellum and cerebrum. Gait abnormality, weakness of the skeletal muscles, and hind limb numbness are typical symptoms of ACR toxicity and are closely related to the dysfunction of the cerebellum and axon damage (Erkekoglu and Baydar, 2014). In this study, representative phenotypes, including an increase in the gait score and foot landing splay, started 21 d and 10 d after 25 mg/kg and 50 mg/kg ACR exposure, respectively (Fig. 1E). At the end of the intonation period (54th day), a slight gait abnormality was observed in the 5 mg/kg ACR exposure group. The histological results also con rmed ACR-induced neuron loss, Purkinje cells pyknosis, and in ammatory cell in ltration in cerebellum ( Fig. 2G and Fig. 2H). In hippocampus, histological staining showed a loose, disordered cellular arrangement of pyramidal cells in the CA3 regions ( Fig. 2A), while the number of Nissl bodies was visibly reduced in the DG regions of the medium-and high-dose ACR-exposed mice (Fig. 2E). CA3 is crucial for working memory processes, as well as retrieving and consolidating short-term memory, while DG is essential for spatial memory encoding (Denny et al., 2014). In line with our results, the ACR-induced injury of the CA3 and DG regions resulted in learning and memory damage during the Morris water maze test conducted by Liu et al. . Thus, neurological de cits of cerebellum and cerebrum could contribute to ACRinduced gait abnormality.
Evidence has shown that gait abnormality-related neurodegenerative alterations are related to decreased neuron function (Jafarian et al., 2015). The neurobiological markers of cholinergic and dopaminergic neurons were also investigated in this study, showing that ChAT and AChE were the key enzymes involved in the synthesis and metabolism of ACh (Vijayaraghavan et al., 2013). The data showed that ACR decreased the ACh level and inhibited the ChAT activity in a dose-dependent manner while signi cantly increasing AChE activity (Fig. 4A and Fig. 4B). Furthermore, the rate-limiting enzyme, TH, is responsible for the synthesis of DA, while TH-positive cells represent dopaminergic neurons (Daubner et al., 2011). Immunocytochemistry and Western blot analysis were used to assess the TH-positive cells and the expression of TH protein. Results showed that TH protein expression was signi cantly decreased, while a loss in dopaminergic neurons was evident in the striatum of the mice subjected to ACR treatment ( Fig. 4G and Fig. 4E). Subsequently, all ACR doses reduced the DA levels in the brains of the mice (Fig. 4D), indicating that the dopaminergic neurons represent a primary site for ACR activity. Additionally, Barber and LoPachin (Barber and LoPachin, 2004) indicated that the neurological imperfections a liated with ACR exposure are conciliated by damaged peripheral and central synaptic neurotransmission. α-syn is essential for adequately supplying synaptic vesicles in the presynaptic terminals in physiological conditions (Lautenschläger et al., 2017). Aggregated α-syn could induce brils and accumulate the pathological hallmark in neurodegenerative diseases (Stefanis, 2012). A high expression of aggregated αsyn was found in the brains of ACR-exposed mice (Fig. 4G). The mechanism by which α-syn aggregation induces neuronal toxicity may occur via α-syn and TH interaction, decreasing TH activity and releasing DA (Pan et al., 2012). Therefore, it is inferred that ACR-induced neurotoxicity occurs due to a decrease in the DA and ACh levels, reduced ChAT activity and TH expression, as well as an increase in AChE activity and α-syn aggregation, ultimately suppressing cholinergic and dopaminergic neuronal functionality.
Oxidative stress and in ammatory response are present during the entire neurological process of ACR intoxication. Exposure to ACR at 5 mg/kg, 25 mg/kg, and 50 mg/kg was associated with signi cant upregulation in the levels of ROS, MDA, and 8-OHdG, while downregulating the GSH levels in the brains of the mice (P < 0.05, Fig. 3). Unbalanced redox status and pro-in ammatory cytokine release are crucial mediators of neuroin ammation, further contributing to acute and chronic ACR-induced neurodegeneration in the central nervous system. Moreover, this study highlighted the involvement of the NLRP3 in ammasome pathway in ACR-induced neuroin ammation. In both the cerebellum and cerebrum, ACR-induced NF-κB-related NLRP3 priming allowed the assembly of the NLRP3 in ammasome, activating downstream signaling cascades, which included ASC, cleaved caspase-1, N-GSDMD, IL-1β, and IL-18 ( Fig. 6 and Fig. 7). These ndings were consistent with previous work, which reported the release of IL-1β in vitro (Zhao et al., 2017a, b). The results suggest that the NLRP3-caspase-1-GSDMD enrolled neuroin ammation, which occurred in both the cerebellum and cerebrum, possibly contributed to the pathogenesis of the gait abnormality and neurological de cit induced by ACR.
In recent years, the involvement of NLRP3-related pathways in exogenous chemically-induced neurotoxicity has received signi cant attention and include cadmium, arsenic trioxide, and molybdenum (Pei et  that directly caused neuronal damage in the cerebrum of rats . However, in previous research, the hippocampus and frontal cortex in the cerebrum were mainly considered to be associated with ACR-induced neurotoxicity . The cerebellum, skeletal muscle, and peripheral nerves are also vulnerable targets and are closely related to exogenous stimuli-induced gait abnormality and neurological injury (De La Monte and Kril, 2014). Here, our study fully demonstrated that neuronal damage, neurotransmission impairment, and neuroin ammation in both of the cerebellums and cerebrums of mice mutually contributed to ACR-induced gait abnormality. Additionally, the endpoints at different times in conjunction with exposure to the low, medium, and high ACR dosages represent acute, subacute, and sub-chronic neurotoxic responses in this study. Therefore, the causal relationship among behavioral phenomena, such as gait abnormality, and the potential mechanism of oxidative stress, in ammation response, neurotransmission impairment, and neuroin ammation, is more clearly evident in our study.

Conclusion
In conclusion, this study demonstrates that neuronal damage, neurotransmission disorder, oxidative stress, and neuroin ammation contribute to ACR-mediated neurotoxicity in a time-and dose-dependent manner. The potential ACR mechanism responsible for inducing gait abnormality and neurological de cit can be correlated with oxidative stress, neurotransmission impairment, and in ammation response-NLRP3 in ammasome signaling cascade activation. These ndings may provide a more systematic insight for assessing the risk associated with ACR.
In future research, we will mainly focus on the correlation between oxidative stress and neuroin ammation. In particular, the potential biomarkers like NLRP3 and GSDMD, should be further con rmed in vitro. Due to the different functions of microglia, astrocytes, and neurons, further research should compare the different responses in different in vitro neurocyte models to con rm the neuroin ammation-related pathway and investigate its role in ACR-induced neurotoxicity. Then, an additional intriguing question relates to realizing cell-cell communication between glia and neurons, especially regarding the microglia-astrocyte-neuron effect on ACR-induced neuroin ammation. A cocultural model may be established to explain the interaction. Additionally, an appropriate transgenic model should be used to further con rm the NLRP3 related biomarkers and explore the mechanism. Abbreviations ASC, apoptosis-associated speck-like protein containing CARD; Caspase-1, cysteinyl aspartate speci c proteinase 1;  results are expressed as mean ± SD (n =10). Different lowercase letters denote signi cant differences (P < 0.05) between the groups after multiple comparisons. and dentate nucleus (D); (E-G) The number of Nissl bodies in the hippocampus (E) and the number of Purkinje cells in the dentate nucleus (G); (F-H) The size of neurons in the hippocampus (F) and the size of the Purkinje cells in the dentate nucleus (H). The exposed mice received low-dose (5 mg/kg, 60 d), medium-dose (25 mg/kg, 30 d), and high-dose (50 mg/kg, 15 d) ACR treatment. The control group received physiological saline that served as a control. The results are presented as mean ± SD (n = 5), Scale bar = 100 μm, * P < 0.05.

Figure 3
Exposure to ACR caused oxidative stress in the brain tissues of the mice. Oxidative stress markers of (A) ROS, (B) MDA, (C) 8-OHDG, and (D) GSH in the brain tissue were detected using oxidative stress kits. The exposed mice received low-dose (5 mg/kg, 60 d), medium-dose (25 mg/kg, 30 d), and high-dose (50 mg/kg, 15 d) ACR treatment, while the control group received physiological saline that served as a control. The results are presented as mean ± SD (n = 5). Different lowercase letters denote signi cant differences (P<0.05) between the groups after multiple comparisons. results are presented as mean ± SD (n = 5). Different lowercase letters denote signi cant differences (P < 0.05) between the groups after multiple comparisons. Exposure to ACR promoted the activation of the NF-κB pathway and the release of in ammatory cytokines in the cerebrum and cerebellum. Immunohistochemical staining of p-p65 (tan) in (A) the cerebrum and (B) the cerebellum zone region. In ammatory markers, including (C) IL-6 and TNF-α in the cerebrum, and (D) IL-6 and TNF-α in the cerebellum, were detected using ELISA. The results are presented as mean ± SD (n = 5). Different lowercase letters denote signi cant differences (P < 0.05) between the groups after multiple comparisons.