MCC950

Cadmium and molybdenum co-induce pyroptosis via ROS/PTEN/PI3K/ AKT axis in duck renal tubular epithelial cells*

a b s t r a c t
Cadmium (Cd) and excess molybdenum (Mo) are harmful to animals, but the combined nephrotoxic mechanism of Cd and Mo in duck remains poorly elucidated. To assess joint effects of Cd and Mo on pyroptosis via ROS/PTEN/PI3K/AKT axis in duck renal tubular epithelial cells, cells were cultured with 3CdSO4$8H2O (4.0 mM), (NH4)6Mo7O24$4H2O (500.0 mM), MCC950 (10.0 mM), BHA (100.0 mM) and
combination of Cd and Mo or Cd, Mo and MCC950 or Cd, Mo and BHA for 12 h, and the joint cytotoxicity was explored. The results manifested that toxicity of non-equitoxic binary mixtures of Mo and Cd exhibited synergic interaction. Mo or/and Cd elevated ROS level, PTEN mRNA and protein levels, and decreased PI3K, AKT and p-AKT expression levels. Simultaneously, Mo or/and Cd upregulated ASC, NLRP3, NEK7, Caspase-1, GSDMA, GSDME, IL-18 and IL-1b mRNA levels and Caspase-1 p20, NLRP3, ASC, GSDMD protein levels, increased the percentage of pyroptotic cells, LDH, NO, IL-18 and IL-1b releases as well as relative conductivity. Moreover, NLRP3 inhibitor MCC950 and ROS scavenger BHA could ameliorate the above changed factors induced by Mo and Cd co-exposure. Collectively, our results reveal that combination of Mo and Cd synergistically cause oxidative stress and trigger pyroptosis via ROS/ PTEN/PI3K/AKT axis in duck tubular epithelial cells.

1.Introduction
As an essential trace element, molybdenum (Mo) widely exists in plants, animals and humans, and can facilitate growth and development (Anke et al., 2007; Berse´nyi et al., 2008). It is now unequivocally established that Mo is the component of many bio- logical enzymes including xanthine oxidoreductases (XOD), sul- phite oxidase (SOD), nitrate reductase (NR) and nitrogenase (Novotny and Turnlund, 2007; R.K, 1987; Schwarz et al., 2009). Additionally, Mo is also worldwide applied abroad in industrial production. It is reported that Mo production amounts to 10 million tons around the world, of which 800 tons are burned into the environment every year (Zhuang et al., 2019). Environmental Mo can enter the animal body via the food chain (Huang et al., 2011). It has been proved that excess Mo is greatly harmful to animals and human beings (Zhai et al., 2013). Since it takes a few weeks for the kidney to completely eliminate Mo, excess Mo ingestion and inadequate excretion cause vast retention of Mo in kidney, liver and other organs, eventually leading to abnormal function and struc- ture of these organs (Xu et al., 2018). Environmental Mo pollution mainly comes from the smelting of Mo, and the mining of tungsten ore which is often accompanied by cadmium (Cd) pollution (Wu et al., 2016; Cui et al., 2018; Pipoyan et al., 2019). As a vital envi- ronmental heavy metal pollutant, Cd exhibits biological toxicity, such as teratogenicity, carcinogenicity and mutagenicity (Bertin and Averbeck, 2006; Wu et al., 2016; Liu et al., 2018; Cai et al., 2019). The world produces about 13,000 tons Cd per year (Zhao et al., 2021). Environmental Cd remains in water, soil and sedi- ment (Waisberg et al., 2003; Xiao et al., 2019).

With persistent migration of flowing water and soil, Cd has remarkedly effects on environmental pollutant, and destroys the ecosystem (Clemens et al., 2013; Clemens and Ma, 2016). Cd-contaminated food, water and dust can enter the body, ultimately causing multiple organs dysfunction and structural changes in animals (Yang and Shu, 2015; Pal et al., 2017). The world is rich in mineral resources, especially tungsten ore, which is associated with Cd and Mo. During the mining and screening of tungsten ore, the tailings containing Cd and Mo are discharged into environment, which leads to joint pollution of Cd and Mo, and ultimately posing threat to the health of animals and human beings through food chain (Cao et al., 2016). Notably, many documents showed that the highest doses of Mo and Cd accumulation were found in the kidney (Anke et al., 2007; Berse´nyi et al., 2008; Gao et al., 2020). In other words, the kidney is the key target organ of Mo and Cd toxicity. Several experiments have illustrated that Cd and Mo are ascribed to kidney injury mainly via oxidative stress (Nigam et al., 1999; Morales et al., 2006; Xia et al., 2015a). The mechanism of oxidative damage co-induced by Cd and Mo has been confirmed in duck renal tubular epithelial cells, which was mediated by overproduction of reactive oxygen species (ROS) (Wang et al., 2020). It is well accepted that oxidative stress can trigger necrosis, apoptosis and pyroptosis (Wang et al., 2010, 2019; Li et al., 2015; Lan et al., 2020).

Until now, most studies on Mo- and Cd-induced nephrotoxicity have been mainly concentrated on apoptosis and necrosis. The apoptotic and necrotic death of kidney are vital mechanisms of Mo- and Cd-induced nephrotoxicity. In recent years, a unique type of pro-inflammatory programmed cell death has been revealed to be different from apoptosis and necrosis in morphology and mecha- nism, which is called pyroptosis (Fink et al., 2009; Xu et al., 2014). Pyroptosis is regarded as Caspase-1-dependent cell death (Jianjin et al., 2017). Its main features are Caspase-1 dependence, rupture of membrane, and intracellular substances release (Chen et al., 2016). Caspase-1 is activated by upstream inflammasome NLRP3 (Miao et al., 2011; Liu et al., 2018). Once activated, Caspase-1 con- tributes to the maturation of interleukin-1b (IL-1b) and interleukin- 18 (IL-18), and cleaves and activates gasdermin D (GSDMD), ulti- mately liberating N terminal domain, which causes membrane pore opening. Subsequently, intracellular contents such as IL-18, IL-1b, lactate dehydrogenase (LDH) and nitric oxide (NO) are released into the extracellular, eventually leading to the occurrence of pyroptosis (Strowig et al., 2012; Shi et al., 2017).

It has been reported that activation of Caspase-1-dependent pyroptosis signaling pathway is closely related to regulation of ROS-mediated PTEN/PI3K/AKT pathway, and further activation of PI3K/AKT participates in inflammatory response (Papaiahgari et al., 2007). Research showed that PTEN/PI3K/AKT pathway had effects on the maturation and releases of pro-inflammatory such as pro-IL- 1b and pro-IL-18, when PI3K and its downstream target AKT were inhibited (Cao et al., 2013a). ROS/PTEN/PI3K/AKT axis means that ROS serves as chemical messenger for cell signaling to mediate PTEN/PI3K/AKT pathway, which regulates cellular defense system against oxidative injury as well as cell proliferation, self-survival and pyroptosis (Xu et al., 2019). Previous study confirmed that excess ROS could up-regulate PTEN expression to various extents (Wang et al., 2016a). The PTEN lies in the upstream of PI3K/AKT pathway, and has an obviously role in inhibiting PI3K/AKT expression, which makes important effects on cell physiological and toxicological injury (Eun-Mi et al., 2016). When PTEN expres- sion is upregulated, its negative regulation markedly decreases PI3K/AKT expression level, eventually triggering cell damage. In a word, pyroptosis is involved in toxic damage process through ROS/ PTEN/PI3K/AKT signaling pathway.

Nowadays, it is urgent to assess the combined toxicity of heavy metals because animals and human beings are facing the reality of their combined pollution. Oxidative stress adjusted PTEN/PI3K/AKT pathway acts an important role in participating in pyroptosis. Our previous study has manifested that co-exposure to Cd and Mo can cause ROS accumulation and oxidative damage (Kang et al., 2019; Wang et al., 2020). Nevertheless, the mechanism of pyroptosis co- induced by Mo and Cd via ROS/PTEN/PI3K/AKT-mediated toxic damage in duck renal tubular epithelial cells remains still poorly understood. Therefore, in the study, based on the establishment of the model of Cd and Mo co-exposure in duck renal tubular epithelial cells, we investigated the combined effects of Mo and Cd on pyroptosis via ROS/PTEN/PI3K/AKT axis. The impact of Mo and Cd co-exposure on cell viability was assessed by the methyl tetra- zolium (MTT) method. The occurrence of oxidative stress was evaluated by flow cytometry assaying ROS level in cells. The IL-18, IL-1b contents, LDH and NO releases as well as relative conductivity were measured by enayme-linked immunosorbent assay (ELISA), icroplater reader instrument, the Griess reagent and DDSJ-308A conductivity meter, respectively. The PTEN/PI3K/AKT pathway and pyroptosis pathway (NLRP3/Caspase-1)-related factors expression levels were detected by real-time quantitative poly- merase chain reaction (RT-qPCR) and western blotting. The aim of the study was to explore the combined nephrotoxic mechanism of Mo and Cd on duck from the insight of ROS/PTEN/PI3K/AKT axis. Our experimental results provide theoretical basis for co-exposure to Mo and Cd nephrotoxic researches on waterfowl.

2.Materials and methods
All the experimental procedures involving animals were approved by the Ethics Committee of Jiangxi Agricultural University (JXAULL-2020-31). Duck tubular epithelial cells isolation and cul- ture followed our previous documents (Zhuang et al., 2019; Wang et al., 2020). In brief, 10e15 days old ducks were selected to remove the kidneys from back in sterile fume hood, and then renal connective tissue membranes and blood vessels were further removed in Dulbecco’s Hanks (D’Hanks) solution (Solarbio Biotechnology, China) containing 3% double antibody. The kidneys were broken and rinsed with D’Hanks solution, 0.1% collagenase I was added for digestion. Then, removed the collagenase completely, and kidney tissue was added D Hanks solution for rinsing. A 200-mesh screen and 4 layers of gauze (sterile) were used to filter kidney tissue. The filtrate was centrifuged for 10 min at 2500 g and washed 3 times. Finally, the collected cells werecultured in DMEM/F12 added streptomycin, penicillin and FBS in the presence of 5% CO2 at 37 ◦C, and culture medium was changedin 24 h for the first time, and then replaced every 24 h.Our previous results showed that half-maximal inhibitory con- centration (IC50) of 3CdSO4$8H2O (Aladdin,China) and (NH4)6Mo7O24$4H2O (Aladdin, China) on duck renal tubular epithelial cells for 12 h exposure were 23.27 mM Cd and 1773.64 mM Mo, respectively (Wang et al., 2020).

Toxicity of non-equitoxic bi- nary mixtures of 3CdSO4$8H2O and (NH4)6Mo7O24$4H2O pre- scriptions were evaluated to confirm the toxic interactions (antagonism, additivity or synergy) ubiquitous. Toxicity of the mixtures was assessed in line with the toxic unit (TU) concept (Sprague, 1971; Wang et al., 2015a), which randomly arranges a value of 1TU to a content of toxic that cause a special response, in the case of the study 50% mortality.Two series of experiments were carried out, six groups were arranged in each series. Concretely, Mo formulation content was fixed at 1TU in all treated groups while the content of Cdformulation was 1.25 mM, 2.5 mM, 5 mM, 10 mM, 20 mM or 40 mM, respectively. Conversely, Cd formulation content was set at 1TU, but the content of Mo formulation was 320 mM, 480 mM, 720 mM, 1080 mM, 1620 mM or 2430 mM, respectively. The cells were treated with these twelve mixtures for 12 h. A control group was exposed to 0 mM Cd and 0 mM Mo in the experiment. Based on obtained inhibition ratio values, a TU-response relationship was established for each series.MTT (TransGen Biotech, China) was applied to assess cell viability in line with the manufacturer’s protocols. Cells were seeded on 96-well plates at a density of 1 × 105 cells and incubated with Cd (1TU) and various concentrations of Mo (320, 480, 720, 1080, 1620, 2430 mМ) or Mo (1TU) and various concentrations of Cd (1.25, 2.5, 5, 10, 20, 40 mМ) co-exposure on cell culture medium at37 ◦C for 12 h to compute cell viability.To further assess the nephrotoxic mechanism of co-exposure to Mo and Cd on duck renal tubular epithelial cells from the insight of ROS/PTEN/PI3K/AKT axis, Mo concentration (500.0 mM) and Cd concentration (4.0 mM) were selected in line with IC50, respectively. Afterward, cells were treated with a series of MCC950 (0 mM, 2.5 mM, 5 mM, 10 mM, 20 mM, 40 mM and 80 mM) for 12 h, respec- tively. The most favorable protective content of MCC950 (10 mM) was screened (date not shown).

And cells were treated with a series of BHA (Sigma-Aldrich, USA) (0 mM, 10 mM, 50 mM, 100 mM, 150 mM and 200 mM) for 12 h, respectively. The most favorable protective content of BHA (100 mM) was screened (date not shown). Cells were treated as follows: 0 mM Cd and 0 mM Mo (control group), 500 mM Mo and 0 mM Cd (Mo group), 4 mM Cd and 0 mM Mo (Cd group), 500 mM Mo and 4 mM Cd (Mo + Cd group), 10 mM MCC950 (MCC950group), 10 mM MCC950, 500 mM Mo and 4 mM Cd(MCC950+Mo + Cd group), 100 mM BHA (BHA group), 100 mM BHA,500 mM Mo and 4 mM Cd (BHA + Mo + Cd group) for 12 h, respectively. After cells were treated, the culture supernatant liquid was collected and examined by DDSJ-308A conductivity meter (Shanghai precision scientific instrument co., Ltd, China). The ddH2O was applied to zero.After cells were treated, cell supernatant fluid was collected. The releases of NO and LDH were determined by NO and LDH analysis kit (Jiancheng Bioengineering Institute, China), respectively.After treatment, cells were resuspended by trypsin and collected by centrifugation. The ROS level was assayed using ROS kit (Beyotime Biotechnology Shanghai, China) in line with the direc- tion. Then cell analysis was performed by flow cytometry (C6 Plus flow cytometry, BD, USA).All procedures strictly followed the instructions of FAM-FLICA® Caspase-1 (YVAD) assay kit (ImmunoChemistry Technologies, Bloomington, MN, USA). Briefly, after treatment, the cells were suspended with 290 mL PBS, and 10 mL of the working solution of FAM-YVAD-FMK were added to these cell suspensions.

The cellswere incubated for 1 h at 37 ◦C under 5% CO2 protected from thelight. After this incubation, 2 ml of the wash buffer was added and washing twice. The cell pellet was then resuspended in 1 ml of thewash buffer containing 1 mg of PI. The samples were measured during the next 15 min by the flow cytometry (C6 Plus flow cytometry, BD, USA).After cells were treated, cell culture medium was collected. According to operating instructions of ELISA kits (Mlbio, China), IL- 18 and IL-1b contents were assayed at 450 nm by a microtiter plate reader.The extraction method of total RNA was the same as the study (Dong et al., 2020), reverse transcription kit (Takara, Japan) was used for reverse transcription into cDNA, and then RT-qPCR was used to quantitatively detect mRNA relative level on ABI Quant Studio7 Flex PCR instrument. As described in Table 1, Primers for the amplification of Caspase-1, NLRP3, ASC, NEK7, GSDMA, GSDME, IL-1b, IL-18, PTEN, PI3K, AKT and b-actin were designed by using oligo 7 software. The fold change in the relative expression of eachmRNA were assessed using the 2-△△CT method and b-actin wasused as the reference gene.After treatment, cells were collected and lysised, then the con- tent of protein was assayed by BCA protein kit (Solarbio Biotech- nology Beijing, China).

The protein was separated by SDS-PAGE (TransGen Biotech Beijing, China) and blotted onto PVDF mem- branes (0.45 mm) (Beyotime Biotechnology Shanghai, China). The PVDF membrane was sealed with 5% skimmed milk powder for 1 h at room temperature, and the membrane was incubated over- night with diluted primary antibodies anti-ASC, anti-GSDMD, anti- NLRP3, anti-Caspase-1 p20, anti-PTEN, anti-PI3K, anti-p-AKT andanti-AKT (Wanleibio, China) at 4 ◦C. Secondary antibodies wereconjugated to horseradish peroxidase at room temperature for 1 h. Membranes were treated with ECL reagents (Applygen) and further was detected with biorad ChemiDoc Touch imager (bioradChemiDoc Touch. America). Then, normalized the density of each band to its respective loading control (GAPDH) (Proteintech, 60004-1-lg).Each experiment was repeated at least 3 times. Data are showed as the mean ± standard and analysis of variance (ANOVA) using SPSS 25.0 software, and differences analysis was executed by the least significant difference (LSD) test. P values < 0.05 were regarded statistically significant. Finally, data were presented applying GraphPad Prism 8.3 software. 3.Results MTT colorimetric method was used to determine cell viability under twelve combined concentrations in two non-equitoxic mixture series. The results of the analysis were showed in Fig. 1A and B. With the increase of Mo and Cd combined concentrations, cell viability was remarkedly reduced (P < 0.001) and presented a concentration-dependent manner. According to linear relationship between inhibition ratio I (%) and TUmix, the results exhibited that TUmix value was less than 1 TU, which indicated synergic interac- tion between non-equitoxic mixture of Mo and Cd (Fig. 1C and D).Morphologic analysis was shown in Fig. 4A, normal cellmorphology was displayed in control group and BHA group, cells grew well and were tightly connected, and cell morphology was intact with clear boundaries. Cells were shrinking and a few of cells were rounded circles, cell vacuolation and density decrease were presented in alone Mo group and Cd group. Under Mo and Cd co- stress, a large number of nucleus ruptured or even disappeared, the cells were severely shrunk, the nucleolus disappeared, chro- matin was concentrated under the nuclear membrane, the nucleus was ring-shaped, and the cells experienced severe pathological changes. BHA could reduce cell damage.We measured PTEN/PI3K/AKT pathway expression levels in cells exposed to Mo or/and Cd. PTEN mRNA and protein levels were markedly increased (P < 0.01 or P < 0.001) (Fig. 2A and E). Conversely, the expression levels of PI3K, p-AKT and AKT were dramatically decreased (P < 0.05 or P < 0.01 or P < 0.001) (Fig. 2A and E). The results revealed that PTEN negatively regulated PI3K/ AKT pathway. Meanwhile, BHA could markedly improve the ten- dency (Fig. 6A and E).To assess the effects of Mo or/and Cd exposure on pyroptosis, the expression levels of factors related to NLRP3/Caspase-1 pathway were detected. As described in Fig. 2B, the mRNA levels of NLRP3, Caspase-1, NEK7, ASC, GSDMA, GSDME, IL-1b and IL-18 wereremarkably upregulated (P < 0.05 or P < 0.01 or P < 0.001) under Mo or/and Cd stress, they in Mo + Cd group were significantly increased (P < 0.05 or P < 0.01 or P < 0.001) in comparison with Moand Cd alone groups. The mRNA levels of the above gene in MCC950+Mo + Cd group (Fig. 5A) and BHA + Mo + Cd group (Fig. 6B) were markedly downregulated (P < 0.01 or P < 0.001) in comparison with Mo + Cd group. Moreover, the mRNA levels of the above genes were presented with heatmap analysis (Figs. 2C and6C). Simultaneously, the changes of Caspase-1 p20, NLRP3, ASC and GSDMD protein levels were the same (P < 0.01 or P < 0.001) (Fig. 2DeG, 5BeC, 6D-G). FAM-FLICA® Caspase-1 (YVAD) assay kit was used to analyze the effects of Mo and Cd on the pyroptosis in duck renal tubular epithelial cell. As shown in Fig. 2H and I, compared to control group, the percentage of pyroptotic cells significantly increased (P < 0.001) in Mo or/and Cd treated groups. Both MCC950 and BHA could markedly attenuate (P < 0.001) the percentage of pyroptotic cells (Fig. 5D and E, 6HeI).As shown in Fig. 4, the IL-18, IL-1b contents, NO, LDH releasesand relative electrical conductivity were dramatically elevated (P < 0.05 or P < 0.01 or P < 0.001) under Mo or/and Cd stress conditions. The change of above indicators in Mo + Cd group were remarkedly increased (P < 0.01 or P < 0.001) compared to Cd andMo alone groups. However, MCC950 and BHA could notably ameliorate the tendency.Heavy metals are stressors. To evaluate oxidative stress in cells, the ROS level was detected. As described in Fig. 3, flow cytometric analysis indicated that ROS levels were dramatically elevated (P < 0.05 or P < 0.01 or P < 0.001) under Mo or/and Cd stress. ROS fluorescence intensity of Mo + Cd group was markedly increased(P < 0.001) in comparison with Cd and Mo groups (Fig. 3A and B).Moreover, ROS fluorescence intensity of Mo + Cd + BHA group was remarkedly reduced (P < 0.001) in comparison with Mo + Cd group (Fig. 3C and D).Correlation analysis indicated that ROS level was significantly correlated with pyroptotic genes and PTEN/PI3K/AKT signaling pathway mRNA expression levels. As shown in Fig. 7A, there was a positive correlation between ROS level and these genes (PTEN, NLRP3, Caspase-1, NEK7, ASC, GSDMA, GSDME, IL-18 and IL-1b)mRNA expression levels. However, there was a negative correlation between ROS level and PI3K/AKT mRNA expression levels. 4.Discussion At present, environmental pollution of heavy metals has become a hot issue of global concern. Once absorbed, heavy metals are accumulated in many vital organs with a long biological half-life (Clemens and Ma, 2016; Jarup, 2003; Li et al., 2018a). As the main excretory organ, the kidney plays a major role in excreting excess heavy metals from the body (Carvalho et al., 2017; Van Ael et al., 2017). Meanwhile, the kidney is also an important target organ for numerous poisons. It is well known that nephrotoxicity is the main toxicity of excessive Mo and Cd exposure. The most important target sites for Mo and Cd accumulation are renal tubular epithelial cells (Timofeev et al., 2018). Nevertheless, the mechanism of nephrotoxicity co-induced by Mo and Cd in duck is still unclear. Heavy metals are stressors that can break the balance of the oxidant/antioxidant system and lead to oxidative damage to the body. Numerous studies indicated that excess Mo and Cd could cause oxidative stress due to ROS overproduction (Shaikh et al., 1999; Xia et al., 2015b; Wang et al., 2020). ROS could mediate PTEN/PI3K/AKT signaling pathway involved in cell toxic process (Lu et al., 2020). Thus, the model of duck renal tubular epithelial cells was established to evaluate nephrotoxicity co-induced by Mo and Cd in the study, mainly concentrating on the correlation between oxidative stress and pyroptosis through PTEN/PI3K/AKT pathway. Primarily, events Mo or/and Cd exposure for 12 h were selected to assess the toxic effects. In this study, the results showed that toxicity of non-equitoxic binary mixtures of Mo and Cd exhibited synergic interaction, simultaneously, co-exposure to Mo and Cd could induce excessive production of ROS and trigger oxidative stress, upregulate PTEN expression level, downregulate PI3K/AKT expression levels, and thereby stimulating pyroptosis in duck renal tubular epithelial cells.Animals are usually exposed to multiple heavy metals instead of single heavy metal under natural environment conditions. The study assessed the toxicity and interaction ubiquitous in non- equitoxic mixtures of Mo and Cd, since it is extremely impossible that heavy metals in a mixture exist in an equitoxic form in natural environment. According to the TU method criteria (Sprague, 1971;Wang et al., 2015b), our results manifested that the non-equitoxic Mo and Cd co-exposure had a dose-dependent relationship and a synergistic effect on duck renal tubular epithelial cells for 12 h treatment.It is well known that pyroptosis, a programmed cell death pathway, is executed by inflammatory caspases. The typical char- acteristics of pyroptosis are the cell membranes rupture and strong inflammatory response. A series of genes and proteins are involved in the process of pyroptosis, such as NLRP3, ASC, NEK7, Caspase-1, GSDMA, GSDMD, GSDME, IL-18, IL-1b and so on. In classical pyroptosis pathway, under the stimulation of endogenous or exogenous signals, NLRP3-ASC inflammasome activates Caspase-1 to mediate maturation and releases of IL-18 and IL-1b, thereby recruiting more inflammatory cells and further amplifying theinflammatory response to induce pyroptosis (Dinarello et al.; Joosten et al., 2013; Orning et al., 2019). ASC contributes to the assembly of inflammasome which bridges pro-caspase-1 and NLRP3 (Sun et al., 2019). NEK7 is a member of the NIMA-related serine-threonine kinase family, which can regulate pyroptosis with NLRP3 (Chen et al., 2019). NLRP3 binding to NEK7 form a component of the NLRP3 inflammasome and promote inflamma- some assembly (He et al., 2016; Shi et al., 2016). In addition, the activation of Caspase-1, and IL-1b and IL-18 releases are hindered in the absence of NEK7. In other words, NEK7 is a vital part of in- flammatory process and its expression is upregulated. When Cas- pase-1-dependent pyroptosis was inhibited, cells damages were attenuated, and NLRP3, Caspase-1, NEK7, IL-18 and IL-1b mRNA levels and Caspase-1 protein level were downregulated (Jianzhao et al., 2019). Gasdermin is a lately confirmed family of pore- forming proteins which consists of Gasdermin A (GSDMA), Gas- dermin D (GSDMD), Gasdermin E (GSDME) and so on, and their N- terminal domains have all been shown to form pores (Man et al., 2017). Numerous researches indicated that GSDMA, GSDMD and GSDME mRNA levels were significantly increased through trig- gering pyroptosis (Shi et al., 2015; Schneider et al., 2017; Lee et al., 2018). Various documents reported that NLRP3, Caspase-1, ASC, GSDMD, IL-1b and IL-18 expression levels were dramatically increased when pyroptosis happened (He et al., 2015; Joshi et al.,2015; Chen et al., 2016; Wu et al., 2018; Chen et al., 2019; Sun et al., 2019). In the study, our results unveiled that the mRNA levels of NLRP3, NEK7, ASC, Caspase-1, GSDMA, GSDME, IL-1b and IL-18, and NLRP3, ASC, Caspase-1 p20 and GSDMD protein levels, and the percentage of pyroptotic cells were markedly increased in cells exposed to Cd or/and Mo. Simultaneously, Mo or/and Cd treatment also remarkedly increased IL-18 and IL-1b contents, LDH and NO releases as well as relative conductivity in cell supernatant, which is consistent with the above results. In addition, changes of above the indicators were more obvious in joint group. A diarylsulfonylurea-based compound MCC950 is a potent and spe- cific inhibitor of the NLRP3 inflammasome (Zhang et al., 2017b), which can block canonical and non-canonical NLRP3 activation (Zhang et al., 2017a; Fan et al., 2018). Many studies showed that MCC950 could inhibit NLRP3 inflammasome activation as evi- denced by suppressing cleavage of Caspase 1 and decreasing mature IL-1b and IL-18 releases, expression levels of NEK7, ASC and GSDMD (Coll et al., 2019; Tapia-Abella´n et al., 2019; Zhang et al., 2019). In the study, as expected, MCC950 could effectively decrease duck renal tubular epithelial cell pyroptosis by inhibiting NLRP3 inflammasome activation, which accompanied by down- regulating mRNA levels of NLRP3, NEK7, ASC, Caspase-1, GSDMA, GSDME, IL-1b and IL-18, and NLRP3, ASC, Caspase-1 p20 and GSDMD protein levels, and decreasing the percentage of pyroptoticcells, IL-1b and IL-18 contents and LDH release in cell supernatant. These results suggest that Mo or/and Cd can induce duck renal tubular epithelial cell pyroptosis, and Mo and Cd exhibited synergic interaction.The PTEN/PI3K/AKT pathway adjusts behaviors and cell growth, and has a key role in participating in the process of cell damage (Pan et al., 2018). It is well known that PTEN lies in the upstream of the pathway, and it can negatively adjust PI3K/AKT pathway (Zheng et al., 2020a,b). The PI3K/AKT pathway is regarded as a protective pathway for cell growth and metabolism (Liang and Slingerland, 2003). Study showed that suppression of PI3K could markedly decrease AKT and p-AKT expression levels (Cao et al., 2013b). At the same time, the PI3K/AKT pathway lies in the upstream of NLRP3 and Caspase-1. The suppression of PI3K/AKT expression levels triggers pyroptosis by upregulating NLRP3 and Caspase-1 expres- sion levels (Diao et al., 2020). Furthermore, previous research revealed that inhibiting pyroptosis could downregulate NLRP3, ASC, Caspase-1 expression levels and IL-1b, IL-18 concentrations as well as returning PI3K and AKT normal expression levels (Li et al., 2018a,b). In the study, our results illustrated that Mo and Cd co- treatment upregulated PTEN expression level, and downregulated PI3K/AKT and p-AKT expression levels through negative regulationrelationship PTEN and PI3K/AKT pathway. Afterward, the NLRP3- Caspase-1-dependent pyroptosis pathway was activated，which is consistent with the above results. These results demonstrate thatMo or/and Cd can trigger pyroptosis via PTEN/PI3K/AKT signaling pathway in duck renal tubular epithelial cells.ROS plays a key role in regulating cell physiological functions and cell signaling (Milkovic et al., 2019). It is found that ROS also acts a pivotal part in regulating PTEN/PI3K/AKT signaling pathway (Liu and Tie, 2019). Extensive studies have indicated that ROS can regulate PTEN expression level, with the increase of ROS level, the expression level of PTEN can be remarkedly elevated (Noh et al., 2016; Wang et al., 2016a,b). Simultaneously, ROS can decrease PI3K and AKT expression levels by increasing PTEN expression level (Lin et al., 2020). Namely, when the body produces excessive ROS, PTEN expression level is induced and PI3K/AKT pathway expression levels are inhibited. In the study, the ROS levels were markedly increased under Mo or/and Cd stress. Meanwhile, the ROS levels were significantly positively correlated with PTEN and pyroptosis-related genes mRNA levels, especially inflammasome related genes (NLRP3, ASC, NEK7), and negatively correlated with PI3K and AKT mRNA levels. Thus, it can be speculated that ROS might be involved in pyroptosis co-induced by Mo and Cd via PTEN/PI3K/AKT pathway. To explore the connection of ROS and pyroptosis, several studies used various antioxidants or free radical scavengers (Wu et al., 2018). BHA, a synthetic phenolic antioxidant, can reduce free radical and ROS generation (HUANG et al., 2011; Zhai et al., 2013; Zhuang et al., 2019). Earlier studies have demonstrated that BHA acts as a free radical inhibitor, which releases hydrogen atoms to bond with free radicals after encountering peroxide, subse- quently forms inactive oxidative radicals, which eventually in- terrupts free radicals chain response (Rychen et al., 2018). Thus, BHA was used to reveal whether ROS was involved in pyroptosis co- induced by Mo and Cd in the study. Our results manifested that BHA addition could remarkedly decrease ROS level and PTEN expression level, up-regulate PI3K AKT and p-AKT expression levels under Mo and Cd co-stress. At the same time, BHA could also markedly improve the changes of pyroptosis-related factors (Caspase-1, NLRP3, ASC, NEK7, GSDMD, GSDMA, GSDME, IL-18 and IL-1bexpression levels, IL-18, IL-1b, LDH and NO releases, relative con-ductivity, the percentage of pyroptotic cells) co-induced by Mo and Cd. It might be speculated that PTEN/PI3K/AKT pathway can be activated by excessive ROS co-induced by Mo and Cd, and then resulting in duck tubular epithelial cell pyroptosis. Conclusion In short, the results manifest that toxicity of non-equitoxic bi- nary mixtures of Mo and Cd exhibits synergic interaction, simul- taneously, Mo and Cd co-exposure can cause oxidative stress and trigger MCC950 pyroptosis via ROS/PTEN/PI3K/AKT axis in duck tubular epithelial cells.