**Assessment of Toxic Effects of Ochratoxin A in Human Embryonic Stem Cells**

**Slaven Erceg 1,\*, Eva María Mateo 2, Iván Zipancic 3, Francisco Javier Rodríguez Jiménez 1, María Amparo Pérez Aragó 1, Misericordia Jiménez 2, José Miguel Soria <sup>4</sup> and Mª Ángeles Garcia-Esparza 5,\***


Received: 21 February 2019; Accepted: 4 April 2019; Published: 10 April 2019

**Abstract:** Ochratoxin A (OTA) is a mycotoxin produced by different *Aspergillus* and *Penicillium* species, and it is considered a common contaminant in food and animal feed worldwide. On the other hand, human embryonic stem cells (hESCs) have been suggested as a valuable model for evaluating drug embryotoxicity. In this study, we have evaluated potentially toxic effects of OTA in hESCs. By using in vitro culture techniques, specific cellular markers, and molecular biology procedures, we found that OTA produces mild cytotoxic effects in hESCs by inhibiting cell attachment, survival, and proliferation in a dose-dependent manner. Thus, we suggest that hESCs provide a valuable human and cellular model for toxicological studies regarding preimplantation stage of human fetal development.

**Keywords:** Ochratoxin A (OTA); human Stem Cells; mycotoxins; cells; cytotoxicity; cell culture

**Key Contribution:** OTA has a great impact in early stages of development. In vitro cell culture of hESCs in the presence of OTA at different concentrations reduced the viability, decreased cellular hESC proliferation, induced apoptosis, and increased the expression of oxygen stress markers. This work may contribute in elucidating the mechanisms underlying OTA embryotoxicity.

#### **1. Introduction**

Human pluripotent stem cells (hPSCs) represent heterogeneous populations, including induced pluripotent stem cells (iPSCs), endogenous plastic somatic cells, and embryonic stem cells (ESCs). Human ESCs (hESCs) are derived from the inner cell mass of the blastocyst, characterized by the ability to self-renew indefinitely and to give rise to all cell types of embryonic lineage (pluripotency) under the guidance of the appropriate chemical, mechanical, and environmental cues [1].

There are high expectations regarding the use of hESCs for treating injuries and degenerative diseases, for modelling complex illnesses and developments, for screening and testing of pharmacological products, and for examining toxicity, mutagenicity, teratogenicity, and potential carcinogenic effects of a variety of environmental factors, including mycotoxins [2,3].

Ochratoxin A (OTA) is the most abundant and toxic member of the ochratoxins, a group of secondary metabolites produced by fungi belonging to the genera *Aspergillus* and *Penicillium* [4–7].

OTA can contaminate a wide variety of foods because of fungal infection in crops, in fields during growth, at harvest, or during storage and shipment. Besides cereals and cereal products, OTA is also found in a range of other food commodities, including coffee, cocoa, wine, beer, pulses, spices, dried fruits, grape juice, pig kidney, and other meat and meat products from non-ruminant animals exposed to foodstuffs contaminated with this mycotoxin [8].

Research into the toxicity of this mycotoxin is mostly centered on its teratogenic [9], nephrotoxic [10], immunotoxic [10], neurotoxic [11–13], and carcinogenic [14] effects that result from exposure to a range of different food types, particularly of plant origin, that may be contaminated by OTA [15,16]. The kidney has been considered as the key target organ of OTA toxicity in most of the mammalian species [17]. Additionally, in humans OTA has been found in blood plasma [5,18], and frequent exposure to OTA is attributed to its nephrotoxic effects, especially in children [19]. Several studies have highlighted OTA as a possible causative agent of Balkan endemic nephropathy, an endemic, severe, progressive, and fatal kidney disease found in the Balkan countries [14,19,20].

Furthermore, investigations in animal models showed OTA as a neurotoxic agent [21,22]. In addition, different studies in vitro have demonstrated a direct relationship between some environmental products and prenatal development [23]. Thus, although OTA appears to exert multiple biological actions, and is cytotoxic, few studies conducted to date have explored whether OTA negatively affects embryonic development [24,25].

During normal embryogenesis, the process of apoptosis removes abnormal or redundant cells from pre-implantation embryos [26]. Induction of apoptosis during early stages of embryogenesis (i.e., following exposure to a teratogen) compromises embryonic development [27,28]. The main methods to study teratogens are either through epidemiological studies in human populations or by controlled exposure in animal models. Previous studies found that OTA induced apoptosis in mammalian cells, including monkey and human kidney epithelial cells, porcine kidney PK15 cells, and human OK cells [29–31]. Although these methods are still essential, more reliable and indicative human-based toxicity tests are needed to represent toxicity effects in humans. Due to the ethical issues regarding teratogenic effect assessment of OTA in human embryos, in this study we have used hESCs as an in vitro model for teratogen screening in a human developmental setting using physiologically relevant doses. There is clear evidence that hESCs represent faithful in vitro toxicity models, as a wide range of chemicals were tested and showed adverse effects in these cells [32–35] with no toxicity in animal models, such as in the case of thalidomide [36]. As hESCs are cells derived from the blastocyst stage, toxicity assays with hESCs can provide toxicity information at a very early stage after fertilization. Having unique proliferation and differentiation capacities toward a wide range of cells in the human body, hESCs closely mimic human embryogenesis [37], thus they offer a unique cellular, developmental, functional, and reproductive human in vitro model for toxicological testing.

The purpose of this study was to assess and determine toxicity of OTA using hESCs as a model for preimplantation embryos. Our data show that (1) hESCs can be used to measure toxicity of food contaminants such as OTA, and (2) OTA exerts its effect through possible mechanisms of apoptosis and oxidative stress.

#### **2. Results**

#### *2.1. Ochratoxin A Reduces the Viability and Decreases the Cellular Proliferation of Human Embryonic Stem Cells (hESCs)*

OTA treatment (1–100 ppm) reduced the viability of hESCs in a dose-dependent manner. Evident toxic effects of OTA were observed after 8 h when approximately 60% of cells survived at a concentration of 10 ppm. Similar effects were observed with a concentration of 50 ppm of OTA, and this was considered the 50% effective concentration (EC50) (Figure 1A,B). In all treatments, the

percentage of colonies that underwent shrinkage during exposure exponentially increased (data not shown).

**Figure 1.** (**A**) Dose-dependent survival rate (MTS assay) of human embryonic stem cells (hESC) at 8 h shows decrease of cell survival to 60% at doses of 10 ppm (*n* = 6). (**B**) Representative bright field micrographs of hESC colonies treated with vehicle (ethanol) or OTA (5 and 10 ppm). White arrows indicate surviving cells and black arrows indicate dead cells. Micrographs show the suitable aspect and shape of surviving cells. White asterisks indicate the area from which the photographs were taken. (**C**) Number of attached cells after 6 and 24 h in two experimental groups compared to control. Magnification times: 20×.

Since the major OTA effects on cell death were observed with concentrations of 5 and 10 ppm, further experiments were performed using these concentrations. Thus, to determine how OTA solutions affected proliferation of hESCs, the surviving cells after 5 and 10 ppm treatments were detached, seeded, and used in subsequent experiments to assess cell attachment and growth during the following 24 h (Figure 1C). Throughout this process, video data of colonies in each group were collected using the IncuCyte System. All videos were first analyzed to determine whether colonies grew, shrunk, or died during incubation. During the 24 h, an evident decrease of cell growth and attachment was observed in comparison with non-treated cells (Figure 1C).

In order to investigate the role of OTA on cell death and apoptotic processes, the cells were stained with nucleic acid IncuCyte®Cytotox Red Reagent for counting necrotic cells, which was able to penetrate and dye compromised cell membranes associated with dead or dying cells, followed by imaging with the IncuCyte ZOOM every 4 h over a 24 h period. A dose-dependent increase in cell death when treated with OTA was observed across all hESCs (Figure 2A, (A1, A3 and A5) and Figure 2B. The same results regarding cell death were observed over 24 h.

**Figure 2.** Ochratoxin A (OTA) increases necrotic cells and apoptosis in hESCs at 5 and 10 ppm. (**A**) Representative fluorescent micrographs of hESC colonies treated with vehicle (ethanol in A1, A2), OTA (5 ppm in A3, A4), and (10 ppm in A5, A6) captured after 8 h with the IncuCyte ZOOM. IncuCyte®Cytotox Red Reagent was used for counting necrotic cells (red) and caspase-mediated apoptosis using a kinetic caspase 3/7 reagent (Essen Bioscience) (green). (**B**) The number of "objects" per well was calculated using IncuCyte software and graphed, showing a significant increase in the number of cells undergoing necrosis (red) or caspase-mediated apoptosis when treated with OTA (green) compared to ethanol. (*n* = 3; \*, † = *p* ≤ 0.05). Scale bar: 300 μm. Magnification times: 20×.

#### *2.2. Ochratoxin A (OTA) Induces Caspase-Mediated Apoptosis in hESCs*

In addition to cytotoxicity quantification, hESCs were also stained and imaged for caspasemediated apoptosis using a kinetic caspase 3/7 reagent (Essen BioScience, 300 West Morgan Road, Ann Arbor, MI, USA). The number of "caspase-3 objects" per well was calculated using IncuCyte integrated analysis software and graphed, showing a significant increase in the number of cells undergoing caspase-mediated apoptosis when treated with OTA compared to vehicle treated cells (Figure 2A, (A2, A4 and A6) and Figure 2B). The trend in caspase 3/7 activation in hESCs correlated with their EC50 value.

#### *2.3. OTA Increases the Expression of Oxygen Stress Markers in hESCs*

After apoptosis and cytotoxicity assays, the cells were collected, and RT-PCR for main oxidative markers was performed to determine the role of oxidative stress in OTA cytotoxicity. In cells treated with OTA, analysis of reactive oxygen stress markers showed a significant but not dose-dependent increase of the expression of glutathione synthetase (gss), superoxide dismutases 1 (sod1), superoxide dismutases 2 (sod2) and activating transcription factor 3 (atf3) in reactive oxygen species (ROS) for both doses: 5 and 10 ppm (Figure 3). A greater fold-change compared to vehicle control was observed in hESCs, strongly suggesting OTA-induced cell death.

**Figure 3.** OTA significantly increases the expression of oxygen stress markers in hESCs. Analysis of reactive oxygen stress markers showed a dose-dependent increase of the expression of glutathione synthetase (GSS), superoxide dismutases 1 (SOD1), superoxide dismutases 1 (SOD2), and activating transcription factor 3 (ATF3) as main markers involved in oxidative stress. (*n* = 3; \* = *p* ≤ 0.05).

#### **3. Discussion**

OTA exposure studies have been developed on different cell lines of human and animal models, especially describing the mechanisms associated with increased levels of oxidative stress, DNA, and lipid and protein damage [38]. Embryos are generally more sensitive to chemicals than adults are, and for this reason it is essential to develop faithful human cell assays for preimplantation stages of human development when possible [39]. OTA is a mycotoxin commonly found in food, which can produce serious toxic effects in the organism and, specifically, in the developing brain [40,41]. In this study, we evaluated the impact of OTA exposure in hESCs as a model for pre- and post-implantation of human embryos. OTA showed toxic, dose-dependent effects only after 4 h of treatment. The mechanism through which OTA induces toxicity in vitro is mainly attributed to multiple effects on various subcellular structures, such as loss of membrane integrity, confirmed by LDH leakage assay in other cells [42,43]. In the present study, OTA exhibited cell toxicity via cell mortality, confirmed by MTS, and through mechanisms of apoptosis and oxidative stress. Our results are in line with earlier studies, which have demonstrated that OTA-induced oxidative stress leads to cytotoxicity and apoptosis in Neuro-2a cells [43], highlighting that elevated ROS is a principle event in oxidative stress in cells treated with OTA. Our results corroborate other findings where OTA was confirmed as a potent ROS inducer [44–46]. Results of our study suggest that OTA cytotoxicity is mediated by oxidative stress in a dose-dependent manner. Although the oxidative species were not measured, the significant increase of expression of main oxidative stress markers, such as GSS, SOD 1, SOD2, and ATF3, strongly indicate that oxidative stress is one of the underlying mechanisms for OTA-induced loss of cell viability and DNA damage. Since mitochondria events are the major generator of ROS, mitochondria could play a crucial role in toxicity of OTA [47]. Generation of free radicals and other oxidative species triggers lipid peroxidation and permeability of the mitochondrial membrane, which produces apoptotic cell death [47,48]. Indeed, it was previously shown that OTA treatment leads to loss of mitochondrial membrane potential and DNA damage in a dose-dependent manner [43]. Our study confirms results obtained by Sava et al. [11,22], in which the authors tested neural stem/progenitor cells (NSCs) prepared from the hippocampus of an adult mouse brain for their vulnerability to OTA in vitro. In that study the authors observed that OTA caused a dose- and time-dependent decrease in viability of both proliferating and differentiating NSCs. Along with decreased viability, OTA elicited pronounced oxidative stress, evidenced by a robust increase in total and mitochondrial SOD activity. This study concluded that greater vulnerability to the toxin exhibited proliferating number of NSCs compared to differentiated, more mature neurons, despite robust DNA repair and antioxidant responses. Further studies need to be performed in order to clarify whether the same mechanisms of oxidative stress

are triggered in hESC by OTA. Our results are in line with previous studies, which have reported OTA as a trigger for the caspase-9 and caspase-3 activation with potential mitochondrial membrane loss in different human primary cells [49–51]. To our knowledge, the present study is the first one describing the effects of OTA in a prenatal human cellular model, demonstrating the importance of assessing toxicity in early stages of development. In this context, it is crucial to develop new simple and faithful in vitro assays that are able to screen the effects of environmental and food chemicals in various stages of the developing fetus. To this purpose, many assays, such as explants of rodent embryos [52,53] or embryonic bodies derived from mESC to model post-implantation development, have been developed [54]. The approach used in this study represents a quick and simple human in vitro method for assessing environmental toxicants in hESCs, a model for the inner cell mass of preimplantation embryos already used for other environmental pollutants [55] such as tobacco smoke [35] or thalidomide [56]. This study may contribute to elucidating the mechanisms underlying OTA teratogenicity in the early days of human fetal development.

#### **4. Materials and Methods**

#### *4.1. Undifferentiated hESC Line Maintenance*

WA09 hESCs were obtained from the WiCell Research Institute (Madison, WI, USA) and were maintained in feeder-free conditions using mTeSR1 media (StemCell Technologies, Vancouver, BC, Canada) on hESC-qualified Matrigel (BD Biosciences, San Jose, CA, USA) coated plates. To maintain the undifferentiated stem cell population, differentiated colonies were removed daily through aspiration and medium was replaced. Additionally, the hESCs were only used in experiments up to passage 40 and were karyotyped approximately every 10 passages to minimize and monitor the potential for genetic instability. hESCs were passaged at 90%–95% confluence (approximately every 7 days) using Accutase. Cell cultures were maintained at 37 ◦C under 5% CO2.

#### *4.2. In vitro Culture of hESCs*

All experimental treatments were carried out in 96-well plates coated with Matrigel. To minimize plating variability and increase reproducibility, hESCs were removed from a 6-well plate using TrypLE (Life Technologies). The cells were washed with DMEM/F12 (Dulbecco´s Modified Eagle Medium F-12 Nutrient Mixture (Ham), GIBCO, Paisley, Scotland, UK) and re-suspended in mTeSR1 that contained 10 uM/L Y27632 Rho-associated kinase inhibitor (Merck KGaA/Calbiochem, Darmstadt, Germany). The rho-associated kinase inhibitor was added to the plating media to increase plating efficiency by decreasing dissociation-induced apoptosis. Five thousand hESCs were plated as a single cell suspension and maintained in an undifferentiated state until 80% confluence was reached.

#### *4.3. Analysis of Cellular Viability*

Analysis of cellular viability in the presence of respiratory inhibitors was performed using MTS assay (Promega, G1111, Promega Corporation, Madison, WI, USA) according to the manufacturer's instructions. Briefly, hESCs were seeded at a density of 5000 cells per well in Matrigel-treated 96-well culture microplates in 100 μL of culture media, and they were incubated for 4 h at 37 ◦C. The cells were used when 80% confluence was reached. After 24 h of compound (ethanol) treatment, 20 μL of MTS reagent was applied to each well of a 96-well plate. Absorbance at 490 nm was recorded after 2 h incubation. Sextuplets were prepared for each condition.

#### *4.4. hESC Compound Exposures*

hESCs were treated with OTA at different concentrations equivalent to previous and published in vitro studies [21]. To test OTA exposure, all compound stock solutions were made with ethanol.

#### *4.5. Toxin Preparation*

A standard of OTA was supplied by Sigma-Aldrich (Alcobendas, Spain). The OTA standard was dissolved in ethanol (Ethanol HPLC 99.5%-gradient grade, Burker, Deventer, Netherlands) to give a stock solution of 1000 μg/mL (ppm). OTA solutions of 100, 50, 10, 5, and 1 ppm were prepared by dilution of suitable aliquots of the stock solution with ethanol. An aliquot of these mycotoxin solutions was added to the wells containing the cell cultures to obtain the final concentration of the toxin. Blank controls having no mycotoxin, but the same volume of solvent, were performed in parallel.

#### *4.6. RNA Extraction and Reverse Transcriptase Polymerase Chain Reaction (RT-PCR and qRT-PCR)*

Cells were collected by centrifugation, and total RNA was isolated with the RNeasy Mini Kit (Qiagen, Hilden, Germany) following the manufacturer´s instructions. They were treated with DNase1 to remove any genomic DNA contamination. QuantiTect Reverse Transcription Kit (Qiagen) was used to carry out cDNA synthesis from 1 μg of total RNA according to the manufacturer's instructions. For quantitative real-time PCR (qRT-PCR), the relative quantification analysis was performed using a CFX96 RealTime PCR Detection system and C1000 Thermal Cycler (Bio-Rad, Hercules, CA, USA). The PCR cycling program consisted of denaturing at 95 ◦C for 10 min followed by 40 cycles of 95 ◦C for 15 s and annealing/elongation at 60 ◦C for 1 min. The reactions were done in triplicate using TaqMan Gene Expression Master Mix and the following TaqMan probes (Applied Biosystems, Foster City, CA, USA): SOD1 (Hs00533490\_mL), SOD2 (Hs00167309\_mL), GSS (Hs00609286\_mL), and ATF3 (Hs00231069\_mL). PCR was done in triplicate, and the expression of polymerase 2A (POL2A; Hs00172187\_mL) was used as three endogenous controls to normalize the variations in cDNA quantities from different samples. The results were analyzed using Bio-Rad CFX software (CFX Maestro Software for Bio-Rad CFX Real-Time PCR Systems) and Microsoft Excel (software version 16.16.8 (190312), 2018, Microsoft, Redmond, WA, USA).

#### *4.7. Cytotoxicity and Apoptosis Assays*

IncuCyte ZOOM Live-Cell Imaging system (Essen Bioscience, Ann Arbor, MI, USA) was used for kinetic monitoring of cytotoxicity and apoptotic activity of OTA in hESCs. These two assays were performed at the same time. Five thousand hESC cells were seeded at day 3 in mTESR medium in each of the 96-well plates, in such manner that by day 1 the cell confluence was approximately 30%. Cells were treated with increasing two concentrations (5 and 10 ppm) of OTA in the presence of 5 μM of Caspase 3/7 Apoptosis Assay Reagent (Essen Bioscience). The Caspase 3/7 reagent labeled apoptotic cells yielding green fluorescence. At the same time, IncuCyte®Cytotox Red Reagent for counting dead cells was applied. This reagent labeled dead cells yielding red fluorescence. The plate was scanned, and fluorescent and phase-contrast images were acquired in real time every 4 h from 0 to 48 hours post treatment. Normalized green object count per well at each time point and quantified time-lapse curves were generated by IncuCyte ZOOM software (IncuCyte®ZOOM Live-Cell Analysis Systems, 2018. Essen BioScience, 300 West Morgan Road, Ann Arbor, MI, USA). Ratios of caspase 3/7 level in OTA-treated cells compared to vehicle were plotted in Microsoft Excel. The cells were monitored for confluence. At a confluence of 50% we performed the experiment, monitoring cell growth using the IncuCyte System to capture phase contrast images every 2 h, and analyzed results using the integrated confluence algorithm. Caspase 3/7 diluted reagent at 1:1000 (5 μM final concentration) and Cytotox Red Reagent (final volume of 50 μL/well) or vehicle (ethanol) were added to the wells. Then, the medium was aspirated. The images were captured every 2–3 h (10× or 20×) in the IncuCyte® System.

#### *4.8. Statistical Analysis*

Statistical analyses of qRT-PCR data from at least three biological replicates were calculated using Student's t-test using GraphPad Prism 5.02 (GraphPad Software 2365 Northside 560 San Diego, CA, USA).

**Author Contributions:** Conceptualization, M.A.G.-E., J.M.S., S.E. and M.J.; Data curation, F.J.R.J. and E.M.M.; Formal analysis, I.Z. and M.A.P.A.; Investigation, M.A.G.-E., J.M.S., S.E., M.A.P.A. and E.M.M.; Supervision, M.A.G.-E., J.M.S. and S.E.; Validation, M.A.G.E., J.M.S. and S.E.; Writing original draft, M.A.G.-E., J.M.S., S.E., I.Z. and M.J.; Writing review and editing, M.A.G.-E., J.M.S., S.E.I.Z., M.J. and I.Z. All the authors approved the final version of the manuscript.

**Funding:** This research was funded by: (1) Ayudas a la Consolidación de Indicadores en Investigación Programa Banco Santander Universidad CEU Cardenal Herrera (grant number: INDI 18/43). Principal investigator: José Miguel Soria. (2) Ministry of Economy and Competitiveness, Spanish Government and European Regional Development Fund (grant number AGL2014-53928-C2-1-R). Principal investigator: Misericordia Jiménez. (3) Ayudas a la Consolidación de Indicadores en Investigación Programa Banco Santander Universidad CEU Cardenal Herrera (grant number: INDI 18/17). Principal investigator Mª Ángeles Garcia-Esparza.

**Acknowledgments:** We acknowledge the University CEU Cardenal Herrera, University of Valencia and Research Center Principe Felipe for the help, support and facilities.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## *Article* **Deoxynivalenol Induces Inflammation in IPEC-J2 Cells by Activating P38 Mapk And Erk1**/**2**

#### **Hua Zhang 1, Xiwen Deng 1, Chuang Zhou 2, Wenda Wu 1,3,\* and Haibin Zhang 1,\***


Received: 11 February 2020; Accepted: 4 March 2020; Published: 13 March 2020

**Abstract:** Fusarium-derived mycotoxin deoxynivalenol (DON) usually induces diarrhea, vomiting and gastrointestinal inflammation. We studied the cytotoxic effect of DON on porcine small intestinal epithelium using the intestinal porcine epithelial cell line IPEC-J2. We screened out differentially expressed genes (DEGs) using RNA-seq and identified 320 upregulated genes and 160 downregulated genes. The enrichment pathways of these DEGs focused on immune-related pathways. DON induced proinflammatory gene expression, including cytokines, chemokines and other inflammation-related genes. DON increased IL1A, IL6 and TNF-α release and DON activated the phosphorylation of extracellular signal-regulated kinase-1 and-2 (ERK1/2), JUN N-terminal kinase (JNK) and p38 MAPK. A p38 inhibitor attenuated DON-induced IL6, TNF-α, CXCL2, CXCL8, IL12A, IL1A, CCL20, CCL4 and IL15 production, while an ERK1/2 inhibitor had only a small inhibitory effect on IL15 and IL6. An inhibitor of p38 MAPK decreased the release of IL1A, IL6 and TNF-α and an inhibitor of ERK1/2 partly attenuated protein levels of IL6. These data demonstrate that DON induces proinflammatory factor production in IPEC-J2 cells by activating p38 and ERK1/2.

**Keywords:** deoxynivalenol; IPEC-J2 cells; RNA-seq; inflammation; MAPKs

**Key Contribution:** DON induces proinflammatory gene expression, including cytokines, chemokines and other inflammation-related genes. DON enhances inflammation in IPEC-J2 via p38 and ERK1/2.

#### **1. Introduction**

Deoxynivalenol (DON; vomitoxin) is a type B trichothecene mycotoxin produced by strains of *Fusarium graminearum* and *F. culmorum* [1]. DON mainly contaminates cereal, especially barley, oats, wheat, corn and their subsequent products. In addition, DON accumulation is a potential sign for the occurrence of other mycotoxins [2]. Due to its adverse effects on animals, DON is known as one of the most significant mycotoxins in animal production.

Consumption of DON-contaminated foods and feeds has been associated with a spectrum of adverse effects and the immunotoxic effects of DON are of increasing concern for farm animals, as well as for humans [2,3]. According to the dose, timing of exposure, time and functional immune assay being used, DON may exert immunosuppressive or immunostimulatory effects [4]. Our preliminary experiments indicate that exposure to DON induces the overexpression of cytokines and chemokines, leading to immune stress, which caused immune function damage [5,6].

The intestinal epithelium forms an important physical barrier against external matter and it is highly sensitive to mycotoxins and important for maintaining health [7]. Consuming DON-contaminated

food is related to gastroenteritis flare-ups and DON exposure leads to intestinal lesions in vivo (animals studies), ex vivo (intestinal explants) and in vitro (cell line) [8–10]. Numerous studies have concluded that DON upregulates the expression of cytokines, chemokines and inflammatory genes [11–13]. However, the mechanism underlying DON-induced inflammation in intestinal epithelial cells (IECs) remains unclear.

MAPKs, including p38 MAPK, extracellular signal-regulated kinase-1 and-2 (ERK1/2) and JUN N-terminal kinase (JNK), modulate many cellular processes associated with cell proliferation, differentiation, survival and death [14]. MAPK signaling has basic functions in immunoregulation and immunopathology, including inflammatory responses and enteritis. Recent research has suggested that DON and other trichothecenes induce the activation of MAPKs in IPEC-J2 cells [15–17], which contributes to autophagy, oxidative stress, epithelial tight junction disruption and intestinal barrier dysfunction. However, few correlative studies have investigated the interaction of MAPK signaling with DON-induced inflammation in the intestinal epithelium.

Therefore, the aims of the present study were to use the IPEC-J2 cell line, an in vitro model of porcine small IECs, to investigate the capacity of DON to induce inflammation and relate the immunomodulatory effects of DON to MAPK activation.

#### **2. Results**

#### *2.1. DON Decreases the Viability and Induces Inflammation in IPEC-J2 Cells*

IPEC-J2 cells were treated for different time periods (2, 6, 12 and 24 h) and with different concentrations of DON (0.25, 0.5, 1, 2 and 4 μg/mL). As presented in Figure 1a, DON (≥0.5 μg/mL) significantly reduced IPEC-J2 cell viability in a time- and concentration-dependent manner.

DON at concentrations of 1.0 and 2.0 μg/mL markedly enhanced the gene expression levels of IL6, IL1A and TNF-α at 2 h compared to the control group (Figure 1b,d,f). After treatment with DON at concentrations of 1.0 and 2.0 μg/mL, the expression of IL1A and IL6 was significantly increased at 6 h (Figure 1b,f) and the expression of IL6 was significantly increased at 12 h (Figure 1b). Moreover, IL6, IL1A and TNF-α protein release into the incubation medium was elevated after treatment with DON at concentrations of 1.0 and 2.0 μg/mL (Figure 1c,e,g). To investigate the immunomodulatory effects of DON, IPEC-J2 cells were exposed to 2 μg/mL DON for 2 h in subsequent experiments.

**Figure 1.** *Cont*.

**Figure 1.** Deoxynivalenol (DON) decreases the viability and induces inflammation in IPEC-J2 cells (**a**) Cell viability in IPEC-J2 cells with or without DON. Two-way ANOVA using Holm-Sidak method was used to assess significant differences in cell viability compared with of the control. Symbols: \* indicates difference in cell viability relative to the control at specific time point (*p* < 0.05) and ε indicates difference in cell viability relative to the 2h exposure time at specific dose (*p* < 0.05). Effects of DON on IL6 (**b**), TNF-α (**d**) and IL1A (**f**) gene expression. and IL6 (**c**), TNF-α (**e**) and IL1A (**g**) cytokine release in IPEC-J2 cells. Samples were collected after 2, 6, 12 and 24 h (mRNA) or 12 h (protein release). One-way ANOVA with a Holm-Sidak test was used to assess significant differences in the mRNA and protein release of IL6, TNF-α and IL1A compared with of the control. The data are expressed as the mean ± SEM. \* *p* < 0.05, \*\* *p* < 0.01 and \*\*\* *p* < 0.001 versus control.

#### *2.2. Identification and Functional Enrichment Analysis of Di*ff*erentially Expressed Genes (DEGs)*

Based on the RNA-seq data, we obtained 480 differentially expressed genes (DEGs) with 320 upregulated genes and 160 downregulated genes (Supplementary Materials: Table S1). In Figure 2, the heatmap and volcano plot show that these genes were clearly separated (Figure 2a,b). According to the Gene Ontology (GO) terms (Figure 2c), 71 genes were enriched in the immune system process. Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis revealed that the upregulated DEGs were

mainly enriched in the following immune-related pathways: TNF signaling pathway, cytokine-cytokine receptor interaction, MAPK signaling pathway, NF-kappa B signaling pathway, Jak-STAT signaling pathway, Toll-like receptor signaling pathway and NOD-like receptor signaling pathway (Figure 2d). Table 1 shows several enriched pathway terms and 15 DEGs were enriched in the MAPK signaling pathway. These results suggest that DON-induced inflammation may associate with the MAPK signaling pathway.

**Figure 2.** (**a**) Cluster heatmap. A change in color from blue to red indicates that the expression level of the gene was relatively high. (**b**) Volcano plot of the DEGs. Blue indicates downregulated genes and red indicates upregulated genes. (**c**) Gene ontology (GO) analysis classified the DEGs into 3 groups: molecular function, biological process and cellular component. (**d**) Bubbles of Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways of the DEGs. The coloring indicates higher enrichment in red and lower enrichment in green. The point size indicates the number of DEGs enriched in a certain pathway. Lower q-values indicate more significant enrichment. (**e**) Validation of DEG data by real-time quantitative PCR (RT-qPCR). The x-axis represents the mRNAs and the y-axis is the fold change between the RT-qPCR and sequencing values.


**Table 1.** Pathway enrichment analysis of the differentially expressed genes (DEGs).

#### *2.3. Integration of Protein-Protein Interaction (PPI) Network Analysis*

To further investigate regulatory pathways of DON, a protein-protein interaction (PPI) network was formulated based on the data in the Search Tool for the Retrieval of Interacting Genes/Proteins (STRING) database with a total of 371 nodes and 729 relationship pairs (Figure 3a). The top 10 hub genes were TNF, IL6, JUN, MYC, CXCL8, FOS, EGR1, CSF2, EDN1 and ATF3 and they were key node proteins in the PPI network. To better analyze the interaction of the proteins, we detected two modules using the Cytoscape plugin Molecular Complex Detection (MCODE) with a score >5 and the top module is shown in Figure 3b. Pathway enrichment analysis of the top module showed that it was mainly related to the MAPK signaling pathway, cytokine-cytokine receptor interaction and TNF signaling pathway.

**Figure 3.** Protein-protein interaction (PPI) network of the DEGs (**a**) and the most significant modules (**b**). Purple nodes represent upregulated genes and yellow nodes represent downregulated genes.

#### *2.4. Validation of the Expression Profile Analysis by RT-qPCR*

Ten genes were selected from the significant DEGs for RT-qPCR analysis to validate their expression levels. The transcriptional levels according to the sequencing and RT-qPCR data were consistent (Figure 2e), thus confirming that the sequencing information was reliable.

#### *2.5. DON Promotes the Expression of Inflammatory Factors and Induces Inflammation in IPEC-J2 Cells Through p38 and ERK1*/*2*

We hypothesized that there may be a link between the activation of the MAPK pathway and DON-induced inflammation. We measured the phosphorylated protein levels of p38, ERK1/2 and JNK. DON effectively increased the phosphorylation of p38, ERK1/2 and JNK (Figure 4).

**Figure 4.** DON induces MAPK activation in IPEC-J2 cells. The levels of p-ERK, p-p38 and p-JNK were detected by western blotting. Data analyzed as described in Figure 1b legend. The quantitative data are presented as the mean ± SEM. \* *p* < 0.05, \*\* *p* < 0.01 and \*\*\* *p* < 0.001 versus control.

To gain insight into the mechanism of MAPKs in DON-induced inflammatory factor upregulation, IPEC-J2 cells were pretreated with inhibitors, including U0126 (ERK 1/2 inhibitor, 10 mM), SP600125 (JNK inhibitor, 20 mM) and SB203580 (p38 inhibitor, 10 mM), before DON treatment. As shown in Figure 5, the inhibition of p38 significantly attenuated DON-induced IL6, TNF-α, CXCL2, CXCL8, IL12A, IL1A, CCL20, CCL4 and IL15 production, whereas the inhibition of JNK had no effect. In addition, the inhibition of ERK 1/2 attenuated DON-induced IL15 and IL6 production. In contrast, CCL4, CCL20 and CXCL2 production increased after treatment with the ERK 1/2 and JNK inhibitors. DON treatment did not significantly affect CCL2 production. In addition, the inhibition of p38 significantly attenuated IL6, IL1A and TNF-α protein release and the inhibition of ERK 1/2 partly attenuated DON-induced IL6 protein release (Figure 6). These results suggest that both p38 and ERK 1/2 contribute to DON-induced inflammation.

**Figure 5.** DON promotes the expression of inflammatory factors through p38 and ERK1/2. Data analyzed as described in Figure 1b legend. The data are expressed as the mean ± SEM. \* *p* < 0.05, \*\* *p* < 0.01 and \*\*\* *p* < 0.001 versus control. # *p* < 0.05, ## *p* < 0.01 and ### *p* < 0.001 versus control-DON.

**Figure 6.** DON induces inflammation in IPEC-J2 cells through p38 and ERK1/2. Data analyzed as described in Figure 1b legend. The data are expressed as the mean ± SEM. \* *p* < 0.05, \*\* *p* < 0.01 and \*\*\* *p* < 0.001 versus control. # *p* < 0.05, ## *p* < 0.01 and ### *p* < 0.001 versus control-DON.

#### **3. Discussion**

The mycotoxin DON is a frequent contaminant of cereals and co-products. The intestine, which serves as the first barrier against food contaminants, shows high sensitivity to DON and related mycotoxins [7,13,18]. After pigs are exposed to DON, most absorption occurs in jejunal epithelial cells. DON mainly causes oxidative stress, disrupts epithelial tight junctions and induces intestinal barrier dysfunction [15,17]. However, the mechanism underlying DON-induced inflammation in IECs is not completely clear. To gain insight into the genes and pathways related to DON in IPEC-J2 cells, we conducted RNA-seq analysis to identify the top inflammatory factors and molecular pathways following DON treatment.

DON robustly upregulates proinflammatory gene expression [4]. DON increased the expression of genes and proteins associated with inflammation, such as TNF-α and IL6 in IPEC-J2 cells, which was consistent with a previous study [19]. TNF-α and interleukins are classic proinflammatory factors that are quickly secreted and cause inflammation when the body is exposed to exogenous stimulation [20]. Overabundant production of TNF-α causes excess secretion of other inflammatory factors, such as IL1β, IL2 and IL8, thereby inducing intestinal mucosal injury [20–22]. Accordingly, inflammatory factors play roles in intestinal immunity. Our data showed that DON significantly upregulated the levels of proinflammatory factors in a concentration-dependent manner in IPEC-J2 cells, indicating that DON enhances the production of inflammatory mediators.

Apart from proinflammatory cytokine upregulation, DON upregulates the transcription levels of several chemokines, including CXCL2, CCL2 and CCL20 [6,23,24]. In our study, DON upregulated the chemokines CXCL2, CXCL8, CCL4 and CCL20. The chemokine CXCL2 is a cytokine secreted by IPEC-J2 cells and a chemotactic for polymorphonuclear leukocytes [25]. CXCL8 is a proinflammatory chemokine that acts as a strong chemoattractant but can create tissue injury with long-term exposure [26]. CCL4 serves as a chemoattractant for monocytes, natural killer cells and a variety of other immune cells [27] and CCL20 is strongly chemotactic for lymphocytes [28]. DON induces the release of CXCL8 in several intestinal epithelial cell lines [29,30]. These previous results are in agreement with our study. Thereby, the inflammation effects of DON may, in part, be influenced by the leukocyte chemotaxis induced by chemokine dysfunction.

The KEGG pathway enrichment analysis showed that the significant DEGs were enriched in immunological pathways and that the MAPK signaling pathway was one of the main signaling pathway enriched in 15 DEGs. Pathway enrichment analysis of the top module showed that it was mainly associated with MAPKs. MAPKs are a type of protein kinase that is pivotal for the development of inflammation [31]. MAPK pathways are activated by kinases, cytoskeletal proteins, transcription factors and other enzymes [32]. The first step to their activation consists of relieving their autoinhibition by a smaller ligand (such as Ras for c-Ra and GADD45 for MEKK4) [33]. DUSPs negatively regulate some MAPKs. DUSP5 and DUSP6 inactivate ERK1/2 and DUSP1 interacts with p38-α, ERK2 and JNK1 [34,35]. MAP3K8, MAP3K5 and MAP3K14 are important MAP3 kinases. The transcription factors JUN, MYC and FOS regulate the expression of inflammation- and immune-related genes [4]. In our study, the upregulation of GADD45B, GADD45G, RASA1, MAP3K8, MAP3K14, MAP3K5, IL1A, MYC, FOS, TNF and JUN, which are related to the MAPK pathway, contributed to MAPK activation and the expression of inflammatory factors. According to the RNA-seq analysis, DON may induce inflammation via the MAPK pathway.

MAPK contributes to DON-induced transactivation and the mRNA stabilization of inflammatory factors [36,37]. To determine whether DON induces porcine intestinal epithelium cell inflammation via the MAPK pathway, MAPK inhibition assays were performed. It has been reported that the MAPK pathway is one of the main pathways for DON to induce inflammation [11,22]. The results in the present study showed that DON induced activation of MAPKs. The p38 inhibitor attenuated DON-induced gene expression levels of IL6, TNF-α, CXCL2, CXCL8, IL12A, IL1A, CCL20, CCL4 and IL15 as well as protein expression levels of IL1A, IL6 and TNF-α. The ERK1/2 inhibitor had only a small inhibitory effect on IL1A and IL6 gene expression levels as well as IL6 protein levels, while the JNK inhibitor had no effect. We demonstrated that DON induced the expression of proinflammatory cytokines and chemokines via the p38 MAPK and ERK1/2 signaling pathways. CXCL8 secretion were upregulated in various human intestinal epithelial cell lines exposed to DON [29,30,38]. In response to DON, dose-dependent increases in IL-8 secretion were observed in Caco-2 cells and this was linked to the ribotoxic-associated activation of PKR, NF-kB and p38 [29,38]. DON elevates CXCL8 generation via ERK1/2 but not p38 in human embryonic epithelial intestine 407 (Int407) cells [30]. This discrepancy may be due to the maturation status of the cells: differentiated mature Caco-2 cells and IPEC-J2 vs. undifferentiated Int407 cells.

In conclusion, the results of the present study indicate that DON induces inflammation in IPEC-J2 cells. This discovery provides a theoretical basis for further exploring the molecular mechanisms of IEC inflammation induced by DON.

#### **4. Materials and Methods**

#### *4.1. Reagents*

DON was obtained from Sigma-Aldrich (St. Louis, MO, USA). Cell culture medium and supplements were purchased from Life Technologies (Grand Island, NY, USA). Anti-phospho-p38 (4511), anti-p38 (8690), anti-phospho-JNK (4668), anti-JNK (9252), anti-phospho-ERK (4370), anti-ERK (4695) and anti-β-actin (4970) antibodies were purchased from Cell Signaling Technology (Beverly, MA, USA). SB203580 was obtained from Promega (Madison, WI, USA). U0126 and SP600125 were acquired from Cayman Chemicals (Ann Arbor, MI, USA).

#### *4.2. Cell Culture and Treatment*

The IPEC-J2 cell line was a gift from Professor Qian Yang, Nanjing Agricultural University, Nanjing, China. Cells were grown in DMEM/F12 medium supplemented with antibiotics and 10% fetal bovine serum. Cells were maintained in the exponential growth phase by passages at intervals of 2–3 days. Compounds were prepared as stock solutions and diluted with the cell culture medium before use. The working concentrations were as follows: DON (0.25, 0.5, 1, 2 and 4 μg/mL), U0126 (10 μM), SP600125 (20 μM) and SB203580 (10 μM). The final concentration of dimethyl sulfoxide (DMSO) was less than 0.1%, which exerted no effect on cell viability. Cells were treated with or without DON and the indicated test compounds for various times according to the experimental protocol.

#### *4.3. Cell Viability Assay*

Cell viability was measured using the MTT (Sigma, M5655) method according to the manufacturer's instructions after DON treatments for 2, 6, 12 and 24 h. The optical density of the control group was considered to be 100% viable.

#### *4.4. Quantitative Real-Time PCR (qRT-PCR) Assay*

Total RNA was isolated using TRIzol reagent (Takara, Dalian, China). cDNA was obtained by reverse transcription using a cDNA transcription kit (Takara, Dalian, China). Real-time PCR was performed in 96-well optical plates on an ABI StepOne Plus Real-time PCR system using SYBR Premix Ex Taq™ (Takara, Dalian, China). The primers used for RT-PCR are shown in Table 2. Analysis of the relative gene expression level was achieved using the 2-ΔΔCT method and gene expression levels were normalized to GAPDH.


#### **Table 2.** Primer sequences of RT-PCR target genes.

#### *4.5. Cytokine Detection by ELISA*

IPEC-J2 cell supernatants were collected after treatment with 2.0 mg/mL DON for 12 h. Porcine IL6, IL1A and TNF-α ELISAs (MEIMIAN, Jiangsu, China) were performed according to the manufacturer's instructions. Samples were analyzed in duplicate.

#### *4.6. RNA-seq Analysis*

After 2 h of exposure to DON, IPEC-J2 cells were collected. Total RNA was extracted using the miRNeasy Mini Kit (Qiagen, Hilden, Germany) following the manufacturer's instructions. cDNA library construction and sequencing with an Illumina HiSeq 2000 sequencer were performed at Shanghai Biotechnology Corporation (Shanghai, China). The resulting RNA-seq reads were mapped onto the reference genome of Sscrofa11.1. The generated RNA-seq data were deposited in the National Center for Biotechnology Information (NCBI) Sequence Read Archive (SRA) repository with accession number PRJNA578240. The expression of transcripts was quantified as fragments per kilobase of exon model per million mapped reads (FPKM). Genes with differential expression levels were identified using edgeR [39]. Differential expression P-values were false discovery rate (FDR)-adjusted using the q-value Bioconductor package. Genes with a q-value ≤0.05 and |fold change| ≥2 were defined as differentially expressed. We analyzed the enrichment of the DEGs using GO functional enrichment analysis and KEGG pathway analysis. ClusterProfiler is a R package applied to perform GO function and KEGG pathway enrichment analyses on DEGs. The terms were considered to be significantly enriched if q-value ≤ 0.05.

#### *4.7. PPI Network Analysis*

The STRING online tool (https://string-db.org/cgi/input.pl) was used to construct a PPI network of the DEGs with a confidence score >0.4 defined as significant [40]. We then imported the interaction data into Cytoscape (version 3.6.0, http://chianti.ucsd.edu/cytoscape-3.6.0/) to map the PPI network [41]. The MCODE plugin for Cytoscape was used to analyze the interaction relationships of the DEGs with their encoded proteins and to screen the hub genes.

#### *4.8. Western Blot Analyses*

After 2 h of exposure to DON, IPEC-J2 cells were collected and lysed in cell lysis buffer (Beyotime, Haimen, China). Protein concentrations were determined using a BCA protein assay kit (Beyotime, China). Proteins were separated by electrophoresis and transferred to PVDF membranes. Anti-phospho-p38 (1:1000), anti-p38 (1:1000), anti-phospho-JNK (1:1000), anti-JNK (1:1000), anti-phospho-ERK (1:2000), anti-ERK (1:1000) and anti-β-actin antibodies were used as primary antibodies. Proteins bound by the primary antibodies were visualized with an appropriate secondary antibody (1:5000) and then detected by an ECL Chemiluminescence kit (Vazyme, E411-05). Protein bands were quantified using NIH ImageJ software (available in the public domain) and detected using a Bio-Rad imaging system (Bio-Rad, Hercules, CA, USA).

#### *4.9. Statistical Analysis*

All data were statistically analyzed using SigmaPlot 11 for Windows (Jandel Scientific; San Rafael, CA, USA). Data of cell viability were analyzed by a two-way ANOVA using the Holm–Sidak method. Other test were assessed by one-way ANOVA with Holm-Sidak tests. Data were considered to be statistically significant difference if *p* < 0.05.

**Supplementary Materials:** The following are available online at http://www.mdpi.com/2072-6651/12/3/180/s1, Table S1: All the identified DEGs.

**Author Contributions:** Conceptualization, H.Z. (Hua Zhang), H.Z. (Haibin Zhang) and W.W.; Methodology, H.Z. (Hua Zhang), X.D., C.Z.; Software, H.Z. (Hua Zhang); Validation, H.Z. (Hua Zhang), C.Z.; Formal Analysis, H.Z. (Hua Zhang), X.D.; Investigation, H.Z. (Hua Zhang), X.D. and C.Z.; Resources, H.Z. (Haibin Zhang); Data Curation, H.Z. (Hua Zhang); Writing – Original Draft Preparation, H.Z. (Hua Zhang); Writing – Review & Editing, W.W.; Visualization, X.D., C.Z.; Supervision, H.Z. (Haibin Zhang), W.W.; Project Administration, H.Z. (Haibin Zhang), W.W.; Funding Acquisition, H.Z. (Haibin Zhang). All authors have read and agreed to the published version of the manuscript.

**Funding:** This work was supported by the National Natural Science Foundation of China (31572576), National Key R & D Program (2016YFD0501207, 2016YFD0501009), China Postdoctoral Science Foundation (2016T90477), PAPD, Excelence project PrF UHK 2212/2019.

**Conflicts of Interest:** The authors declare that there are no conflict of interest.

#### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## *Article* **T-2 Toxin Induces Oxidative Stress, Apoptosis and Cytoprotective Autophagy in Chicken Hepatocytes**

#### **Huadong Yin** †**, Shunshun Han** †**, Yuqi Chen** †**, Yan Wang, Diyan Li and Qing Zhu \***

Farm Animal Genetic Resources Exploration and Innovation Key Laboratory of Sichuan Province, Sichuan Agricultural University, Chengdu 611130, Sichuan, China; yinhuadong@sicau.edu.cn (H.Y.); hanshunshun@stu.sicau.edu.cn (S.H.); chenyuqi@stu.sicau.edu.cn (Y.C.); as519723614@163.com (Y.W.); diyanli@sicau.edu.cn (D.L.)

**\*** Correspondence: zhuqing@sicau.edu.cn; Tel.: +86-028-8629-0991

† These authors contributed equally to this work.

Received: 13 December 2019; Accepted: 27 January 2020; Published: 29 January 2020

**Abstract:** T-2 toxin is type A trichothecenes mycotoxin, which produced by fusarium species in cereal grains. T-2 toxin has been shown to induce a series of toxic effects on the health of human and animal, such as immunosuppression and carcinogenesis. Previous study has proven that T-2 toxin caused hepatotoxicity in chicken, but the regulatory mechanism is unclear. In the present study, we assessed the toxicological effect of T-2 toxin on apoptosis and autophagy in hepatocytes. The total of 120 1-day-old healthy broilers were allocated randomly into four groups and reared for 21 day with complete feed containing 0 mg/kg, 0.5 mg/kg, 1 mg/kg or 2 mg/kg T-2 toxin, respectively. The results showed that the apoptosis rate and pathological changes degree hepatocytes were aggravated with the increase of T-2 toxin. At the molecular mechanism level, T-2 toxin induced mitochondria-mediated apoptosis by producing reactive oxygen species, promoting cytochrome c translocation between the mitochondria and cytoplasm, and thus promoting apoptosomes formation. Meanwhile, the expression of the autophagy-related protein, ATG5, ATG7 and Beclin-1, and the LC3-II/LC3-I ratio were increased, while p62 was downregulated, suggesting T-2 toxin caused autophagy in hepatocytes. Further experiments demonstrated that the PI3K/AKT/mTOR signal may be participated in autophagy induced by T-2 toxin in chicken hepatocytes. These data suggest a possible underlying molecular mechanism for T-2 toxin that induces apoptosis and autophagy in chicken hepatocytes

**Keywords:** T-2 toxin; hepatocyte; apoptosis; autophagy; chicken

**Key Contribution:** T-2 toxin-induced hepatotoxicity was characterized by the induction of mitochondrial-mediated apoptosis and PI3K/AKT/mTOR-mediated autophagy in chicken.

#### **1. Introduction**

Mycotoxins are the main secondary metabolites of molds and lead to widespread contamination on crop plants and fruits. Among the most important mycotoxins, T-2 toxin is a mycotoxin that can cause multiple effects in organisms [1]. T-2 toxin is a type A trichothecene produced by several *Fusarium* species [2], which shows the most potent cytotoxicity [3]. Furthermore, T-2 toxin leads to the effects of cytotoxin radiomimetic, which is due to impaired protein synthesis. T-2 toxin hampers synthesis of DNA and RNA in eukaryotic cells, which ultimately triggers cell apoptosis in vitro and in vivo [4]. Many studies have shown that T-2 toxin induces apoptotic cell death in hematopoietic tissue [5], spleen, liver [6], skin and intestinal crypt in mice [7]. In chickens, apoptosis induced by T-2 toxin was detected in the thymus, bursa of Fabricius and primary hepatocytes [8,9]. Previous studies have demonstrated a crosstalk between autophagy and apoptosis, as apoptosis increases when the autophagic pathway is completely inhibited [10].

T-2 toxin contamination is usually found on cereals, such as maize, wheat and oats, which are the main food and feed resources for human and livestock [11]. The presence of T-2 toxin can be reduced but not completely eliminated. T-2 toxin can cause chronic toxicity in organisms after oral exposure, dermal exposure and inhalation. In livestock, this results in anorexia, reduced body weight and nutritional efficiency, altered neuro-endocrine system, and immune modulation [12]. In addition, residues of the T-2 toxin and its metabolites in animal products are an important human health problem. Poultry is extremely sensitive to the toxic effects of T-2 toxins, leading to yellow cheese-like necrosis at the edge of the septum, hard mucosal mucosa and typical angular cheilitis of the mouth and tongue [13]. In addition, chickens exposed to T-2 toxin show enhanced mortality from *Salmonella* infection and low-resistance titers for Newcastle disease and infectious bursal disease [14,15].

Multiple studies have examined the effects of T-2 toxin in inducing of hepatotoxicity in chickens. However, the relationship between T-2-induced autophagy and apoptosis has not been examined. Here, we investigated the effects of T-2 toxin on hepatocyte apoptosis and autophagy and provide experimental evidence for the potential molecular mechanism of T-2 toxin-induced hepatotoxicity in broiler chickens.

#### **2. Results**

#### *2.1. Pathological Lesions*

To determine the effect of T-2 toxin on chicken livers, we examined the pathomorphological changes in the liver. In the control group, the liver tissue structure was normal, the cell structure was intact, and the cells were arranged neatly (Figure 1A). In the 0.5 mg/kg T-2 toxin treatment group, the liver pathological changes were mild; the hepatocyte volume was increased and mild swelling manifested as blisters, with occasional inflammatory cell infiltration (Figure 1B). In the 1 mg/kg and 2 mg/kg treatment groups, the hepatocytes were swollen and showed balloon-like deformation; the cytoplasm was vacuolated, and the nucleus was located in the center of the vacuole or squeezed on one side. Additionally, hepatic sinus stenosis, a small amount of red blood cell deposits, focal inflammatory cell infiltration and massive proliferation of interlobular bile duct epithelial cells were observed in the 1 mg/kg and 2 mg/kg treatment groups (Figure 1C,D).

#### *2.2. T-2 Triggers Apoptosis in Hepatocytes*

We next performed flow cytometry to determine if T-2 toxin induced apoptosis in hepatocytes from T-2 treated chickens. The amounts of apoptotic cells in the treatment groups were significantly higher (*p* < 0.01) than that in the control, and this difference was dose-dependent (Figure 2A,B). Western blot results showed cleavage of rapamycin (PARP) in the T-2 treatment groups; furthermore, pro-caspase-3 and pro-caspase-9 expressions were reduced in a dose-dependent manner, whereas the cleaved form of caspase-3 and caspase-9 increased (Figure 2C,D). These data further indicate that T-2 toxin induced apoptosis in hepatocytes.

**Figure 1.** Photomicrographs of hematoxylin and eosin stained chicken liver sections of 21 day chicken after treatment of T-2 toxin with different concentration of 0, 0.5, 1 and 2 mg/kg. (**A**) No obvious pathological changes were observed in hepatocytes. (**B**) Hepatocytes with mild steatosis and slight congestion. (**C**) Hepatocytes were slightly swollen, with vacuolar degeneration and lymphocyte neutrophil infiltration. (**D**) The liver showed slight congestion, local vacuolar degeneration was obvious, and the bile duct epithelium and cells demonstrated slight hyperplasia. Red arrow: red blood cell; yellow arrow: bile duct epithelial cell; hematoxylin and eosin (H&E); bar, 20 μm.

#### *2.3. The Mitochondrial Pathway is Activated by T-2 Toxin*

To evaluate whether the mitochondrial pathway participates in the T-2 toxin-induced apoptosis, we first examined the mitochondrial reactive oxygen species (ROS) levels in hepatocytes from T-2 treated chickens by flow cytometry. Low intracellular ROS levels were found in the untreated group, whereas they increased dramatically in the 1 mg/kg and 2 mg/kg T-2 toxin treatment groups (Figure 3A,B). In addition, T-2 toxin significantly suppressed the activity of the antioxidant enzymes GSH-Px, CAT and SOD, but the MDA level was significantly higher in treatment groups than in the control group (Figure 3C). We next evaluated the protein expression of Bax and Bcl-2 and found that Bax protein abundance was upregulated, whereas Bcl-2 abundance was downregulated in a dose-dependent manner, with an increase in Bax/Bcl-2 ratio (Figure 3D). We also examined the mitochondrial release of cytochrome (cyt c) during T-2 toxin-induced apoptosis. The level of mitochondrial cyt c decreased with the increase of T-2 toxin concentration, whereas the level of cytosolic cyt c increased (Figure 3E).

**Figure 2.** Effect of different concentration (0, 0.5, 1 and 2 mg/kg, respectively) of T-2 toxin on hepatocyte apoptosis. (**A**) Scattergram and (**B**) apoptosis rate of apoptotic hepatocytes. (**C**) The protein levels of PARP, caspase-3 and caspase-9, and their cleaved forms in hepatocytes. (**D**) The bar showed the relative protein cleaved level of caspase-3, caspase-9 and PARP. The data are presented as the means ± standard error of the mean (SEM) of three independent experiments. \* *p* < 0.05 and \*\* *p* < 0.01, compared with the control group.

**Figure 3.** T-2 toxin induced hepatocyte apoptosis via activation of the mitochondria-dependent pathway. (**A,B**) Intracellular reactive oxygen species (ROS) levels in hepatocytes from chickens treated with T-2 toxin of different concentration at 0, 0.5, 1 and 2 mg/kg. (**C**) The activity of antioxidant enzymes SOD, CAT, GPX-Sh and MDA content in hepatocytes from chickens treated with T-2 toxin of different concentration at 0, 0.5, 1 and 2 mg/kg. (**D**) The Bax and Bcl-2 mRNA and protein levels in hepatocytes from chickens treated with T-2 toxin of different concentration at 0, 0.5, 1 and 2 mg/kg. (**E**) The cytosolic and mitochondrial cyt c level in hepatocytes from chickens treated with T-2 toxin of different concentration at 0, 0.5, 1 and 2 mg/kg. All the data are presented as means ± SEM of three independent experiments. \* *p* < 0.05 and \*\* *p* < 0.01, compared with the control group.

#### *2.4. T-2 Toxin Triggers Autophagy in Hepatocytes*

To determine if T-2 toxin induces autophagy in hepatocytes from T-2 treated chickens, we measured the transcript levels of autophagy genes including ATG5, ATG7 and Beclin-1 genes (Figure 4A). T-2 toxin treatments induced greater expression levels of ATG5, ATG7 and Beclin-1 genes compared with controls. Furthermore, the ratio of LC3-II/LC3-I increased with the T-2 toxin dosage, while the protein abundance of p62 decreased (Figure 4B). In addition, the cell ultrastructure changed; typical autophagy features were observed and the number of autophagosomes increased in the treatment groups compared with controls (Figure 4C).

**Figure 4.** Effect of T-2 toxin on autophagy in chicken hepatocytes. (**A**) The mRNA levels of Beclin-1, Atg5 and Atg7 in hepatocytes from chickens treated with T-2 toxin of different concentration at 0, 0.5, 1 and 2 mg/kg. (**B**) The protein expression levels of LC3, p62 and Beclin-1 in hepatocytes from chickens treated with T-2 toxin of different concentration at 0, 0.5, 1 and 2 mg/kg. (**C**) Morphological observation of autophagy in hepatocytes from chickens treated with T-2 toxin of different concentration at 0, 0.5, 1 and 2 mg/kg, autophagic vacuoles (red arrows, magnification from left to right: ×1200, ×5000). All the data are presented as means ± SEM of three independent experiments. \* *p* < 0.05 and \*\* *p* < 0.01, compared with the control group.

#### *2.5. Autophagy Protects Apoptosis in T-2 Treated Hepatocytes*

Increased autophagy is considered a protective mechanism against apoptosis as both autophagy and apoptosis share common proteins. To explore the relationship between autophagy and apoptosis, the specific autophagy inhibitor 3-methyladenine (3MA) and autophagy inducer rapamycin (RAP) were used on T-2 toxin-treated hepatocytes. Immunofluorescence showed that T-2 toxin treatment significantly increased the numbers of LC3B puncta, and autophagy flux was further enhanced after the addition of RAP, but autophagy intensity was significantly decreased after the addition of 3MA (Figure 5A,B). When autophagy was inhibited by 3MA, the levels of caspase-3 and caspase-9 cleavage were significantly enhanced after T-2 treatment. Conversely, when autophagy was induced by RAP, the levels of caspase-3 and caspase-9 cleavage were significantly decreased (Figure 5C,D). These results may suggest that autophagy hinders apoptosis in T-2 toxin-treated hepatocytes.

**Figure 5.** Autophagy delays apoptosis in T-2 treated hepatocytes. (**A**) Hepatocytes stained with LC3 (red) antibody using a confocal microscope (600x), Nuclei were stained with 4,6-diamino-2-phenyl indole (DAPI) (blue; bar = 10 μm). (**B**) The bar showed the number of LC3 dots. (**C**) Western blots showed the expression levels of caspase-3 and caspse-9 cleaved in hepatocytes. (**D**) The bar showed the protein level of cleaved caspase-3 and caspase-9. All the data are presented as means ± SEM of three independent experiments. \* *p* < 0.05 and \*\* *p* < 0.01, compared with the control group.

#### *2.6. T-2 Toxin Inhibits the PI3K*/*Akt*/*mTOR Signal Pathway*

To determine if T-2 toxin regulates the PI3K/Akt/mTOR signal pathway in hepatocytes from T-2 treated chickens, we next examined the protein abundance of the tumor suppressor factor, phosphatase and tensin homolog (PTEN), which has a dual-specificity phosphatase activity. PTEN expression level increased in hepatocytes with the increase in T-2 toxin concentration (Figure 6A). In addition, we examined the protein abundance and phosphorylation levels of PI3K, Akt, mTOR and p70S6K, which are key proteins in the PI3K/Akt/mTOR pathway. We found that the protein abundances of PI3K, Akt, mTOR and p70S6K did not differ among the treatment groups, but their phosphorylation levels gradually decreased with the increase in T-2 toxin concentration (Figure 6B,C). These results may suggest that T-2 toxin inhibits the PI3K/Akt/mTOR signal pathway in hepatocytes.

**Figure 6.** Effect of T-2 toxin on PI3K/Akt/mTOR in hepatocytes. (**A**) Protein abundance of phosphatase and tensin homolog (PTEN) in hepatocytes from chickens treated with T-2 toxin of different concentration at 0, 0.5, 1 and 2 mg/kg. (**B**) Representative blots showed the expression abundance of p-PI3K, PI3K, p-Akt, Akt, p-mTOR, mTOR, p-70S6K and 70S6K in hepatocytes from chickens treated with T-2 toxin of different concentration at 0, 0.5, 1 and 2 mg/kg. (**C**) The bar graphs showed the ratio of p-PI3K/PI3K, p-Akt/Akt, p-mTOR/mTOR, and p-p70S6K/p70S6K. All the data are presented as means ± SEM of three independent experiments. \* *p* < 0.05 and \*\* *p* < 0.01, compared with the control group.

#### **3. Discussion**

The T-2 toxin has harmful mutagenic, carcinogenic and teratogenic effects on humans and animals [16–18]. Although various studies have examined hepatocyte apoptosis in broilers treated with T-2 toxin [19,20], no reports have focused on the relationship between autophagy and T-2 toxin-induced apoptosis. Herein, we reported that T-2 toxin-induced hepatotoxicity was characterized by the induction of mitochondrial-mediated apoptosis and PI3K/AKT/mTOR-mediated autophagy in chicken.

The liver is the main organ of metabolism in which foreign substances accumulate and are detoxified. The T-2 toxin suppresses hepatocyte protein synthesis and inhibits metabolic enzyme activity and liver fat peroxidation, which ultimately leads to hepatocyte apoptosis [21–23]. In the current study, histopathological analysis showed that T-2 toxin caused pathological changes in liver tissue, including hepatocyte swelling, volume increase and more granules in the cytoplasm, suggesting that T-2 toxin leads to hepatocyte apoptosis. Our results were consistent with the report by Meissonnier et al. who showed that exposure of pigs to T-2 toxin via diet for 28 days caused liver histopathological changes, excessive hepatic glycogen accumulation and mild interstitial inflammatory cell infiltration [24].

Apoptosis is a crucial physiological cell death process that can be induced by toxic stimuli [25]. Previous studies have shown that T-2 toxin injection can strongly induce cell apoptosis in different tissues, such as thymus, spleen and liver, particularly in the liver [6]. Yang et al. incubated primary chicken hepatocytes with T-2 toxin for 24 h and found that the cell activity was significantly reduced and apoptosis gradually increased in a dose-dependent manner [9], which was similar to our finding that hepatocyte apoptosis gradually increased with the increasing dosage of T-2 toxin.

The mitochondrial pathway has a vital role in the intrinsic apoptosis pathway [26], which depends on the translocation of the apoptogenic protein, cyt c, into the cytoplasm. This occurs via the Bax/Bcl-2 pathway, as their relative levels determine cell destiny by activating death-driving proteolytic proteins known as caspases [27]. In the current study, several findings suggested that T-2 toxin induced the mitochondrial apoptotic pathway in hepatocytes: (1) Bcl-2 was downregulated and Bax was upregulated, thus increasing the Bax/Bcl-2 ratio; and (2) cyt c was released from the mitochondria into the cytosol, followed by apoptosome formation with the apoptotic proteases Apaf-1 and caspase-9. In addition, T-2 toxin treatment lead to an increase in ROS and MDA levels and a decrease in the activities of SOD, CAT, and GSH-Px, resulting in oxidative stress and a concentration-dependent increase in apoptotic cells. Mu et al found that T-2 toxin can induce the ROS accumulation and an increase in mitochondrial mass, which indicated that oxidative stress and mitochondrial enhancement occurred in T-2 toxin-treated primary hepatocytes, which is similar to our result [28]. In addition to our results, other studies have shown apoptosis induced by T-2 toxin via the ROS-mediated mitochondrial pathway in other cells, such as ovarian granulosa cells [29], embryonic stem cells and fibroblast 3T3 cells [30] in mouse.

Autophagy is a crucial homeostasis mechanism that is involved in multiple physiological and pathological processes [31]. Autophagy also shows a complex relationship with apoptosis, as autophagy not only increases caspase-dependent cell death, but also promotes cell survival [32]. In the present study, the increase in gene expression of Atg5, Atg7 and Beclin-1, which are autophagy marker genes, suggested that T-2 toxin induced autophagy in hepatocytes. Moreover, we found an increase in the LC3-II/LC3-I ratio and Beclin-1 protein abundance and a decrease in expression of p62 protein, further suggesting that T-2 toxin induced autophagy in hepatocytes. Bcl-2 and Beclin-1 participate in the regulation of both apoptosis and autophagy [33], and Bcl-2 interacts with Beclin-1 to suppress Beclin-1-dependent autophagy [34]. In our study, we found that Beclin-1 was activated by T-2 toxin, but Bcl-2 was suppressed, and T-2 toxin-induced apoptosis can be delayed by autophagy. Wang et al showed that autophagy may reduce zearalenone-induced cytotoxicity and prevent rat Leydig cell apoptosis [35]. Wu et al found that autophagy plays a role in protecting human cells from T-2 toxin-induced apoptosis, because autophagy may decrease toxic responses induced by T-2 toxin [36]. Our results were consistent with these reports.

The PI3K/AKT/mTOR/p70S6K signaling pathway plays a vital role in autophagy regulation in eukaryotic cells [37]. PI3K induces a signaling cascade and phosphorylates the serine/threonine kinase, mTOR, by activating the serine/threonine kinase, Akt [38]. PTEN has also been proven to suppress the Akt/mTOR signal [39]. As the major upstream modulator, the PI3K pathway regulates autophagy by phosphorylating AKT, which affects the downstream factors p70S6K and 4E-BP1 [40]. Several mycotoxins induce autophagy by inhibiting the PI3K/Akt/mTOR axis, such as zearalenone in donkey granulosa cells [41], aflatoxin B2 in chicken hepatocytes [16] and sterigmatocystin in human gastric epithelium cells [42]. In this study, T-2 toxin inhibited the phosphorylation of PI3K, Akt, mTOR and p70S6K, whereas it activated PTEN, suggesting that the PI3K/AKT/mTOR/p70S6K pathway may be participated in the autophagic process induced by toxicity effect of T-2 toxin. These findings are similar to a previous study that showed that deoxypodophyllotoxin induced cytoprotective autophagy against apoptosis through inhibition of the PI3K/AKT/mTOR pathway in osteosarcoma U2OS cells [42].

In summary, T-2 toxin treatment activates the mitochondrial apoptotic pathway by triggering ROS production and Bcl-2 family protein expression, resulting in hepatocyte apoptosis. In addition, T-2 toxin may involve in the PI3K/AKT/mTOR signal to regulate hepatocellular autophagy. This study provides new insights into the mechanisms underlying the toxicological effect of T-2 toxin in chicken hepatocytes.

#### **4. Materials and Methods**

#### *4.1. Ethics Approval*

All experimental operations were approved by the Animal Ethics Committee of Sichuan Agricultural University, and the approved number was 2018-2121 (21 May 2018). Relevant guidelines and regulations were followed while performing all the methods.

#### *4.2. Animals*

A total of 120 ROSS 308 male chickens at one-day of age were used in this study. After being weighted, chickens were randomly divided into four groups (n = 30 per group); each treatment had six replicates with five chickens. Experimental replicates were raised in separate cages. The four groups were maintained under the same condition and received general nutrient composition and levels that met the requirement of ROSS 308, with T-2 toxin in feed as follows: 0 mg/kg (control), 0.5 mg/kg, 1 mg/kg, and 2 mg/kg. Feed and water were freely available during the whole trial period.

#### *4.3. Exposure of Chickens*

All the feed was made up by the processing-workshop of feedstuff in the Animal Nutrition Institute of Sichuan Agricultural University, which meet the nutritional requirement of ROSS 308. There were no common mycotoxins, such as aflatoxins, deoxynivalenol, ochratoxin A, zearalenone and T-2 toxin, were found in this feed by the ELISA kit (Huaan Mangech Biotech, Beijing, China). Firstly, the T-2 toxin (purity ≥ 98%; Sigma Aldrich, St. Louis, MO, USA) powder was dissolved by 95% ethanol, and mixed in 1 kg feed and dry it. Then, the mixture was added into feed to get the get the target concentration (0 mg/kg, 0.5 mg/kg, 1 mg/kg, and 2 mg/kg, respectively) of T-2 toxin. At last, we used the ELISA kit (Huaan Mangech Biotech) to confirm the final concentration of toxins in the feed.

#### *4.4. Sample Collection and Preparation*

After 21 days of feeding, six chickens (one chicken for every replicate) were randomly selected from the same treatment and euthanized. Livers were collected to determine the pathological histology and hepatocyte apoptosis rate. Fresh livers were dissected, minced, and stored at − 80 ◦C for extracting RNA and protein.

#### *4.5. Pathological Observation*

Liver tissues were fixed overnight in 4% phosphate-buffered paraformaldehyde (Jianke Biotech, Chengdu, Sichuan, China) and then paraffin-embedded blocks were archived. We sliced 5 μm thick tissue sections from paraffin-embedded tumor blocks and mounted the sections onto glass slides. Hematoxylin and eosin (H&E) staining was performed on tissue sections, and pathological examination was performed using an optical microscope (Olympus, Tokyo, Japan).

#### *4.6. Apoptosis Detection*

Livers were minced in pre-cold phosphate-buffered saline (PBS; Beyotime, Shanghai, China), and the suspension was passed through a 300 mesh nylon filter. After filtration, the hepatocyte suspensions were washed in PBS twice. Hepatocytes were re-suspended in 1× binding buffer (BD Pharmingen, Santiago, CA, USA) to obtain a concentration of 1 <sup>×</sup> 106 cells/mL. Next, 100 <sup>μ</sup>L were transferred into a culture tube and 5 μL of propidium iodide (PI; BD Pharmingen, Shanghai China) and 5 μL of Annexin V-FITC (BD Pharmingen, Shanghai, China) were added. After mixing, the cells were incubated at 25 ◦C for 15 min in the dark and then 400 μL of 1× binding buffer (BD Pharmingen, Shanghai, China) was added. Cells were then analyzed by FACSCanto II flow cytometry (BD Bioscience, San Diego, CA, USA).

#### *4.7. Real-Time PCR*

Total RNA of the livers were isolated by Trizol reagent (TaKaRa, Dalian, China). First-strand complementary cDNA was synthesized by PrimeScirptTM RT reagent kit with gDNA eraser (TaKaRa, Dalian, China) following the manufacturer's protocol, and then was stored at − 20 ◦C for RT-PCR. PCR amplifications were performed as follows: 95 ◦C for 5 min and 36 cycles each with 95 ◦C for 10 s, 60 ◦C for 30 s and 72 ◦C for 20 s, then 65 ◦C for 5 s and 95 ◦C for 5 s using the BIO-RAD CFX ConnectTM real time system (Bio-Rad, Hercules, CA, USA). All PCR reactions were performed in triplicate. β-actin was

used as the endogenous reference gene. Specific primers are referenced to Chen et al [16] or designed by the software of Primer Premier 5.0 (Ottawa, Ontario, Canada, 2007), and the primer sequences are listed in Table 1.


**Table 1.** Table: Gene-special primers for RT-PCR.

#### *4.8. Western Blot Analysis*

The refrigerated livers were washed with pre-cold PBS twice and centrifugation at 3000× *g* for 5 min at 4 ◦C, then removed the supernatant. Total protein extracts were obtained by homogenizing liver in RIPA lysis buffer (Sigma Aldrich) supplemented with protease inhibitor cocktail and phosphatase inhibitors. After centrifugation, the supernatant was collected and stored at −80 ◦C. Protein concentration was determined by the BCA protein detection kit (Sangon Biotech, Shanghai, China). Western blot analysis was performed as previously described by Han et al. The primary antibodies were used: caspase-3 (ZenBio, Chengdu, China), caspase-9 (ZenBio), β-actin (Abcam, Cambridge, MA, USA), Bax (ZenBio), Bcl-2 (Santa Cruz, Heidelberg, Germany), LC3B (Sigma), P62 (Santa Cruz), beclin-1 (Sigma), PI3K/Akt/mTOR/70S6K protein and phosphorylated antibody were purchased from Bioss Biotechnology Co. Ltd. (Bioss, Beijing, China). The secondary antibodies used were as follows: mouse anti-rabbit (Sigma), goat anti-rabbit (Sigma), mouse anti-rabbit horseradish peroxidase (HRP) (Zenbio). The enhanced chemiluminescence (ECL) kit (Beyotime, Jiangsu, China) was used to capture the bands via a CanoScan LiDE 100 scanner (Canon, Tokyo, Japan), and western blots were analyzed by Image J software (Bethesda, MD, USA, 2007).

#### *4.9. Cytochrome C Release*

The cytoplasm was first isolated from the mitochondria using the cytochrome C release apoptosis kit (BioVision, Mountain View, CA, USA). After treatment with E2 for 24 h, the cells were lysed by homogenizing in the cytosol extraction solution provided by the kit and then centrifuged at 700× *g* for 10 min. Cells were then centrifuged at 12,000× *g* for 30 min to separate cytoplasmic and mitochondrial components. Determination of cytoplasmic and mitochondrial cytochrome C abundance was performed by western blot using mouse monoclonal antibodies provided in the kit.

#### *4.10. Transmission Electron Microscopy (TEM) Observations*

Hepatocytes were fixed in 2.5% glutaraldehyde phosphate buffer saline (Sigma, St. Louis, MO, USA) and post-fixed in 1% osmium tetroxide (Sigma). The samples were dehydrated in graded ethanol solutions, and cells were embedded in the stimulating resin. Sections (60 nm) were cut using ultramicrobody (Leica Microsystems, Milan, Italy). The divided grid has a saturated solution of uranyl acetate and lead citric acid. Samples were examined by electron microscopy (FEI, Milan, Italy).

#### *4.11. Intracellular Reactive Oxygen Species (ROS) Detection*

Production of intracellular ROS production was measured using the fluorescent dye substrate 2',7'-dichlorofluorescin-diacetate (DCFH-DA; Procell, Wuhan, China) as a substrate. Cells were incubated for 60 min at 37 ◦C with 10 μM DCFH-DA and then harvested and suspended in Hank's Balanced Salt Solution (D-HBSS; Procell). The generation of ROS was analyzed using FACSCanto II flow cytometry (BD Bioscience, New York, NJ, USA).

#### *4.12. Antioxidative Enzymes and Malondialdehyde Detection*

The activities of superoxide dismutase (SOD), glutathione peroxidase (GPX-Px) and catalase (CAT) and malondialdehyde (MDA) level were determined by commercial assay kits (Jiancheng, Nanjing, China) according to the manufacturer's instructions. After mixing the liver cell homogenate with the reagents, the cells were incubated at 37 ◦C overnight for multi-scan spectroscopy detection.

#### *4.13. Immunofluorescence and Confocal Microscopy*

Hepatocytes grown on 24-well plates were fixed with 4% paraformaldehyde (Jianke Biotech, Guangzhou, China) for 10 min. After washing with PBS twice, cells were blocked using 3% bovine serum albumin (BSA; Thermo Fischer Scientific; former Savant, MA, USA) and 0.2% Triton X-100 (Thermo Fischer Scientific, Waltham, MA, USA) in PBS for 10 min at 37 ◦C. The samples were incubated with the relevant antibodies in PBS/10% FSC for 1 h and then stained with the appropriate fluorescent secondary antibody. Fluorescence intensities were captured by an Olympus FluoView FV1000 confocal microscope (Olympus, Melville, NY, USA). To block or induce autophagy, cells were treated with 3-methyladenine (10 mM; Sigma, St. Louis, MO, USA) or rapamycin (4 μM; Sigma), respectively, for 6 h.

#### *4.14. Statistical Analysis*

Statistical analyses were performed using SPSS 19.0 software (SPSS Inc., Chicago, IL, USA, 2000). Data are shown as least squares means ± standard error of the mean (SEM). Differences between groups were assessed using t-test, and values were considered significant difference at *p* < 0.05.

**Author Contributions:** Conceptualization, Y.C.; data curation, S.H.; formal analysis, S.H., and D.L.; funding acquisition, Q.Z. and H.Y.; investigation, S.H., and Y.C.; project administration, Y.W. and H.Y.; resources, H.Y.; supervision, H.Y.; writing—original draft, S.H. and H.Y.; writing—review and editing, Q.Z. All authors read and approved the final manuscript.

**Funding:** This work was financially supported by the China Agriculture Research System (CARS-40), Sichuan Science and Technology Program (2016NYZ0050).

**Acknowledgments:** We thank Liwen Bianji, Edanz Editing China (www.liwenbianji.cn/ac), for polishing the English text of this manuscript.

**Conflicts of Interest:** The authors declare no conflicts of interest.

#### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## *Article* **Comparison of Apoptosis and Autophagy in Human Chondrocytes Induced by the T-2 and HT-2 Toxins**

**Fang-Fang Yu 1, Xia-Lu Lin 2, Xi Wang 2, Zhi-Guang Ping <sup>1</sup> and Xiong Guo 2,\***


Received: 31 March 2019; Accepted: 7 May 2019; Published: 8 May 2019

**Abstract:** In this report, we have investigated the apoptosis and autophagy of chondrocytes induced by the T-2 and HT-2 toxins. The viability of chondrocytes was measured by the MTT assay. Malondialdehyde (MDA) and superoxide dismutase (SOD) kits were used to measure the oxidative stress of chondrocytes. The apoptosis of chondrocytes was measured using flow cytometry. Hoechst 33258 and MDC staining agents were introduced to analyze apoptosis and autophagy induction in chondrocytes, respectively. Protein expression of Bax, caspase-9, caspase-3, and Beclin1 was examined by western blotting analysis. The T-2 and HT-2 toxins significantly decreased the viability of chondrocytes in a time-dependent manner. The level of oxidative stress in chondrocytes induced by the T-2 toxin was significantly higher when compared with that of the HT-2 toxin. The apoptosis rate of chondrocytes induced by the T-2 toxin increased from 3.26 ± 1.03%, 18.38 ± 1.28%, 34.5 ± 1.40% to 49.67 ± 5.31%, whereas apoptosis rate of chondrocytes induced by the HT-2 toxin increased from 3.82 ± 1.03%, 11.61 ± 1.27%, 25.72 ± 2.95% to 36.28 ± 2.81% in 48 h incubation time. Hoechst 33258 staining confirmed that apoptosis of chondrocytes induced by the T-2 toxin was significantly higher than that observed when the chondrocytes were incubated with the HT-2 toxin. MDC staining revealed that the autophagy rate of chondrocytes induced by the T-2 toxin increased from 6.38% to 63.02%, whereas this rate induced by the HT-2 toxin changed from 6.08% to 53.33%. The expression levels of apoptosis and autophagy related proteins, Bax, caspase-9, caspase-3, and Beclin1 in chondrocytes induced by the T-2 toxin were significantly higher when compared with those levels induced by the HT-2 toxin.

**Keywords:** T-2 toxin; HT-2 toxin; apoptosis; autophagy

**Key Contribution:** The T-2 and HT-2 toxins can significantly induce apoptosis and autophagy of chondrocytes; and apoptosis and autophagy of chondrocytes induced by T-2 toxin were much higher when compared with that of the HT-2 toxin.

#### **1. Introduction**

Kashin–Beck disease (KBD) is an endemic, chronic, and deformed osteoarthropathic disease. There are 0.64 million KBD patients distributed from northeast to southwest regions of China. KBD patients suffer from joint pain, morning stiffness, limited motion, and joint enlargement [1,2]. Three underlying risk factors are considered to be responsible for KBD: mycotoxin (T-2 toxin) in grain, selenium deficiency, and organic acid in drinking water [3,4]. Recently, research effort has focused on apoptosis of human chondrocytes induced by the T-2 toxin [5]. Previous epidemiologic studies have confirmed the presence of higher concentrations of the T-2 toxin in grains from endemic areas [6]. The T-2 toxin has been showed to exert various toxin effects on experimental animals and human

chondrocytes, including the dysplasia tibial growth plate cartilage in chicken, and induction of apoptosis of human chondrocytes, which involves p53, Bcl-xL, Bcl-2, Bax, and caspase-3 signaling pathways [7].

T-2 toxin is hydrolyzed to the HT-2 toxin in nature. For example, the T-2 toxin is rapidly metabolized to the HT-2 toxin in microsomes of liver, kidney, and spleen with conversion rates of 80%. The T-2 toxin detected in grains can be rapidly converted in vivo to the HT-2 toxin after consuming the contaminated food. The T-2 toxin rapidly combines with proteins in the blood and is delivered to organs through the mouth, skin, and respiratory tract [8]. The T-2 toxin is metabolized to the HT-2 toxin in the liver after entering enterohepatic circulation. Rats treated with T-2 toxin for 8 h, were found to convert the toxin at different rates in various tissues with conversion rates ranging from 68.20% to 90.70%, and the T-2 and HT-2 toxins were both detected in the skeletal system (thighbone, knee joint, and costal cartilage) [9]. The T-2 toxin in cultured chondrocytes is metabolized into the HT-2 toxin. The concentration of the T-2 toxin in the cell medium was found to decrease from 20 to 6.67 ng/mL during 48 h incubation period, while the concentration of the HT-2 toxin increased from 0 to 6.88 ng/mL over the same period [10]. Metabolism of the T-2 toxin to the HT-2 toxin in the liver and digestive systems can directly affect the human skeletal system. However, the toxic effects of the HT-2 toxin on human chondrocytes remain poorly understood.

In this study, human chondrocytes were cultured in fetal bovine serum (FBS) media. The chondrocytes were exposed to the T-2 and HT-2 toxins for 48 h, and the resulting apoptotic and autophagic effects were monitored.

#### **2. Results**

#### *2.1. Viability of Chondrocytes Induced by the T-2 and HT-2 Toxins*

Chondrocytes were incubated with the T-2 and HT-2 toxins (20 ng/mL) for 48 h. As shown in Figure 1, the T-2 and HT-2 toxins decreased the viability of chondrocytes significantly in a time-dependent manner, and the toxic effect of the T-2 toxin on the viability of chondrocytes was significantly higher than that of the HT-2 toxin after 24 h and 48 h incubation time. Therefore, the toxic effect of the T-2 toxin on the viability of chondrocytes was significantly higher when compared with that of the HT-2 toxin.

**Figure 1.** Effects of same concentration of the T-2 and HT-2 toxins on the cellular viability of chondrocytes were estimated by MTT reduction. \* *p* < 0.05 was considered as significant difference between the two groups.

#### *2.2. Oxidative Stress of Chondrocytes Induced by the T-2 and HT-2 Toxins*

MDA is an important indicator of lipid peroxidation damage in tissue and cells. As shown in Figure 2a, the chondrocytes were treated with the T-2 and HT-2 toxins (20 ng/mL) for 48 h. The MDA content in the chondrocytes increased as the incubation time in the presence of the two toxins increased. At the same dose and incubation period (12 h and 48 h), the MDA content in the chondrocytes induced by the T-2 toxin was significantly higher than the chondrocytes exposed to HT-2 toxin (*p* < 0.05).

**Figure 2.** The malondialdehyde (MDA) (**a**) and superoxide dismutase (SOD) (**b**) content in the chondrocytes were induced by the T-2 and HT-2 toxins. \* *p* < 0.05 was considered as significant difference between the two groups.

SOD is an important antioxidant defense enzyme in humans. As shown in Figure 2b, the SOD content in the chondrocytes decreased as the incubation period of the T-2 and HT-2 toxins increased. At the same dose and incubation period (12 h and 24 h), the SOD content in the chondrocytes incubated with the T-2 toxin was significantly lower when compared with the SOD content present in chondrocytes cultured with the HT-2 toxin (*p* < 0.05).

#### *2.3. Apoptosis of Chondrocytes Induced by the T-2 and HT-2 Toxins*

Flow cytometry was used to analyze apoptosis of chondrocytes induced by the T-2 and HT-2 toxins (20 ng/mL). Apoptosis of chondrocytes increased gradually as the incubation period with the toxins increased (Figure 3); the apoptosis of chondrocytes induced by the T-2 toxin increased in the range 3.26 ± 1.03%, 18.38 ± 1.28%, 34.5 ± 1.40%, and 49.67 ± 5.31% after incubation for 0, 12, 24, and 48 h, respectively, whereas apoptosis of chondrocytes induced by HT-2 toxin increased in the range 3.82 ± 1.03%, 11.61 ± 1.27%, 25.72 ± 2.95%, and 36.28 ± 2.81% over the same time. At the same dose and incubation period, the apoptosis of chondrocytes induced by the T-2 toxin was significantly higher than that induced by the HT-2 toxin, and the difference was statistically significant (*p* < 0.05).

Hoechst 33258 staining was used to analyze apoptosis of chondrocytes induced by the T-2 and HT-2 toxins. Cell nuclei that stained white and thick dense were considered to be positive apoptosis cells. As shown in Figure 4, the apoptosis rate of chondrocytes incubated with the T-2 toxin increased from 3.94% to 60.67%, whereas the apoptosis rate of chondrocytes incubated with the HT-2 toxin increased from 3.74% to 40.75%. Therefore, the apoptosis of chondrocytes induced by the T-2 toxin was significantly higher when compared with that of HT-2 toxin, and the difference was statistically significant between two groups (*p* < 0.05).

**Figure 3.** Apoptosis of chondrocytes was induced by the T-2 and HT-2 toxins using flow cytometry analysis. \* *p* < 0.05 was considered as significant difference between two groups.

**Figure 4.** Apoptosis of chondrocytes was induced by the T-2 and HT-2 toxins using Hoechst 33258 staining. The cell nucleus with a white, thick dense cells were considered to be positive apoptosis cells under the fluorescence microscope (×400). \* *p* < 0.05 was considered as significant difference between two groups.

#### *2.4. Apoptosis-Related Proteins in Chondrocyte Induced by T-2 and HT-2 Toxins*

As shown in Figure 5, chondrocytes incubated with the T-2 and HT-2 toxin (20 ng/mL) for 48 h shown an increased expression level of Bax (Figure 5a), caspase-9 (Figure 5b), and caspase-3 (Figure 5c), and the increases in protein levels were dependent on the incubation period. The relative expression level of Bax, caspase-9, and caspase-3 proteins in chondrocytes induced by the T-2 toxin was 1.29 fold, 0.99 fold, and 1.32 fold, respectively. The relative expression level of Bax, caspase-9, and caspase-3

proteins in chondrocytes induced by the HT-2 toxin was 0.91 fold, 0.68 fold, and 1.12 fold, respectively. The increased expression levels of Bax, caspase-9, and caspase-3 in chondrocytes induced by the T-2 toxin were statistically significant when compared with that of the HT-2 toxin (*p* < 0.05).

**Figure 5.** Apoptosis and autophagy related with proteins (**a**) Bax, (**b**) caspase-9, (**c**) caspase-3, and (**d**) Beclin1 of chondrocytes were induced by the T-2 and HT-2 toxins. The expression levels of Bax, caspase-9, caspase-3, and Beclin1 referred to the GAPDH (load control) were calculated in the T-2 toxin group and HT-2 toxin group. And then the expression levels of Bax, caspase-9, caspase-3, and Beclin1 were compared between two groups. \* *p* < 0.05 was considered as significant difference between two groups.

#### *2.5. Autophagy of Chondrocytes Induced by the T-2 and HT-2 Toxins*

As shown in Figure 6, the MDC kit was used to analyze autophagy of chondrocytes induced by the T-2 and HT-2 toxins (20 ng/mL). Cell nuclei stained cyan-green were positive for an acidic autophagosome. The autophagy rate of chondrocytes induced by the T-2 toxin increased from 6.38% to 63.02%, and the autophagy rate of chondrocytes induced by the HT-2 toxin increased from 6.08% to 53.33%. Therefore, the autophagy rate of chondrocytes induced by the T-2 toxin was significantly higher than that caused by the HT-2 toxin, and the difference was statistically significant (*p* < 0.05).

**Figure 6.** Autophagy of chondrocytes was induced by the T-2 and HT-2 toxins using MDC staining, cell nuclei stained cyan-green were positive for an acidic autophagosome under the fluorescence microscope (×400). \* *p* < 0.05 was considered as significant difference between two groups.

#### *2.6. Autophagy-Related Proteins in Chondrocytes Induced by the T-2 and HT-2 Toxins*

As shown in Figure 5d, the relative expression level of Beclin1 was observed to increase gradually as the incubation period increased. The relative expression level of Beclin1 in chondrocytes induced by the T-2 toxin was increased 1.03 fold, and the relative expression level of Beclin1 in chondrocytes induced by the HT-2 toxin was increased 0.88 fold after an incubation of 48 h. The increased expression level of Beclin1 in chondrocytes induced by the T-2 toxin was significantly higher when compared with that of the HT-2 toxin (24 h and 48 h), and the difference was statistically significant (*p* < 0.05).

#### **3. Discussion**

Currently, research efforts have focused mainly on T-2 toxin contamination in grains for the etiology of KBD. However, it remains unclear that whether the T-2 toxin in grains specifically damages articular cartilage of children KBD, and that the expression levels of apoptotic and autophagic proteins in chondrocytes are exposed to the T-2 toxin during the early stages of the disease. Based on our previous experiments [9,10], the T-2 toxin is metabolized to the HT-2 toxin after entering into the skeletal system of rats. The T-2 toxin levels were observed to decrease in chondrocytes over a 48 h period concomitant with the significant increase in the concentration of the HT-2 toxin. In contrast, there is a paucity of data describing the toxicity of the HT-2 toxin on chondrocytes, and there is no comparative study describing the toxicity of the T-2 and HT-2 toxins toward chondrocytes. Therefore, in this study, chondrocytes were incubated with the same concentration of the T-2 and HT-2 toxins to explore the apoptotic and autophagic affects induced by these toxins.

In this study, chondrocytes were incubated with the T-2 and HT-2 toxins for 48 h. MTT analysis revealed that the toxicity of the toxins on chondrocytes is time-dependent. The T-2 and HT-2 toxins increased oxidative stress in chondrocytes significantly. Flow cytometry analysis showed that the T-2 toxin induced an increase from 3.26% to 49.67% in the apoptosis rate of chondrocytes, whereas the apoptosis rate of chondrocytes induced by the HT-2 toxin increased significantly from 3.82% to 36.28%. Immunofluorescence analysis also confirmed that the apoptosis rate of chondrocytes induced by the two toxins increased significantly. Western blot analysis showed that the relative expression levels of Bax, caspase-3, and caspase-9 in chondrocytes incubated with the T-2 and HT-2 toxins increased significantly. The oxidative stress level, apoptosis rate, and apoptosis-related proteins for chondrocytes induced by the T-2 toxin were significantly higher than those observed when these cells were incubated with the HT-2 toxin. Nonetheless, the oxidative stress of chondrocytes incubated

with both toxins increased significantly, which caused a change to the mitochondrial membrane potential and mitochondrial membrane osmosis, release of mitochondrial pro-apoptotic protein Bax, and the subsequent release of cytochrome C related proteins (caspase-9 and caspase-3). All these factors eventually resulted in apoptosis of the chondrocytes.

Autophagy of the chondrocytes was induced by both toxins, and immunofluorescence analysis showed that the autophagy rate of chondrocytes induced by the T-2 toxin increased from 3.94% to 60.67%, whereas the autophagy rate of chondrocytes induced by the HT-2 toxin increased from 3.74% to 40.75%. Western blot analysis revealed that the expression levels of Beclin1 increased significantly in chondrocytes incubated with the T-2 and HT-2 toxins. Autophagy of chondrocytes induced by the T-2 toxin was also significantly higher than that induced by the HT-2 toxin. Recent studies have reported that the T-2 and HT-2 toxins induce autophagy and apoptosis of porcine and mouse oocytes, rat brain, primary cardiomyocyte, liver cells, and mouse primary leydig cells. Two studies [11,12] showed an increase in the ROS levels of porcine and mouse oocytes when incubated with the HT-2 toxin, indicating an increase in oxidative stress. ROS levels in the treated group were also higher, confirming that the HT-2 toxin caused oxidative stress, which induced apoptosis and autophagy. A previous study [13] reported autophagy in the brain and apoptosis in the pituitary, suggesting that the T-2 toxin may induce different acute reactions in different tissues. Three studies [14–16] also confirmed that incubation of the T-2 toxin with mouse primary leydig cells, liver cells, and primary cardiomyocyte caused up-regulation of LC3-II and Beclin1, suggesting that the T-2 toxin promotes a high level of autophagy. Pretreatment of these cells with chloroquine and rapamycin was shown to increase and decrease the rate of apoptosis, respectively. Therefore, autophagy may prevent apoptosis of cells by reducing T-2 toxin-induced cytotoxicity. The T-2 toxin is also an environmental risk factor of the KBD, and our results showed that the T-2 and HT-2 toxins can significantly induce apoptosis and autophagy of chondrocytes, and these observations were consistent with previous studies [11–17]. The apoptosis and autophagy rates of chondrocytes induced by the T-2 toxin were much higher than those rates induced when the chondrocytes were incubated with the HT-2 toxin, and such an observation has not been reported previously.

The results showed an increase of both apoptosis and autophagy in chondrocytes treated with the T-2 and HT-2 toxins, which is in agreement with previous studies. Autophagy is a normal physiological activity of cells, which was activated in chondrocytes treated with the T-2 and HT-2 toxins, to avoid further cell damage. There is a complex relationship between autophagy and apoptosis. When cells are exposed to low environmental pressures, activation of autophagy can prevent apoptosis and subsequent cell death. When cells are subjected to strong or prolonged environmental stress, the process of autophagy consumes excessive levels of intracellular proteins or organelles, leading to cell survival failure that promotes programmed cell death [18–20].

#### **4. Conclusions**

In conclusion, our results showed that the T-2 and HT-2 toxins induce apoptosis and autophagy of chondrocytes, and that the level of oxidative stress plays an important role in autophagy activation. The activation of autophagy can reduce oxidative damage and therefore functions in protecting chondrocytes from apoptosis through capture, elimination, and degradation of damaged mitochondria.

#### **5. Methods and Materials**

#### *5.1. Reagents and Antibodies*

Fetal bovine serum (FBS), dimethyl sulfoxide (DMSO), and Hoechst 33258 were purchased from Sigma-Aldrich (St. Louis, MO, USA). The T-2 and HT-2 toxins were purchased from J&K Chemical Ltd (Beijing, China). The thiazolyl blue tetrazolium bromide (MTT) was purchased from Amresco (Solon, OH, USA). The malondialdehyde (MDA) kit and the superoxide dismutase (SOD) kit were purchased from the Nanjing Jiancheng Bioengineering Institute (Nanjing, China). The Bicinchoninic Acid (BCA) Protein Assay kit was purchased from TianGen Biotech (Beijing, China). Anti-Bax, anti-Caspase 9, anti-Caspase 3, and anti-Beclin1 antibodies were purchased from Cell Signaling Technology, Inc. (Danvers, MA, USA).

#### *5.2. Cell Culture and Treatment*

The human chondrocytes cell line (C28/I2) was cultured in DMEM/F12 medium with 9% FBS at 37 ◦C and 5% CO2 in a humidified atmosphere. Once the chondrocytes had reached a steady state of the exponential growth phase, these cells were seeded at a density of 1.0 <sup>×</sup> 10<sup>4</sup> per well in 96-well plates and grown overnight. The cells were then cultured in a medium containing either the T-2 toxin or the HT-2 toxin (20 ng/mL) for 0, 12, 24, and 48 h. The T-2 and HT-2 toxins (1 mg) were freshly dissolved in 1 mL DMSO and protected from light. In the T-2 toxin and HT-2 toxin treatment group, the cellular viability, oxidative stress, apoptosis, and autophagy of chondrocyte were determined.

#### *5.3. MTT Assay*

Human chondrocytes in the logarithmic phase were suspended in 0.1% EDTA trypsin. Two hundred microliter cell suspensions were seeded into individual 96-well plates at a density of 1 <sup>×</sup> 10<sup>4</sup> cells per well. The experiments were carried out in the toxin group and control group. Complete medium with either the T-2 toxin or HT-2 toxin (20 ng/mL) was added and the cells were incubated for 0, 12, 24, and 48 h. Twenty microliters of MTT was added into the toxin and control groups to a final concentration of 0.5 mg/mL at each incubation time point. After 4 h at 37 ◦C, the medium containing MTT was aspirated and replaced with 150 μL DMSO and incubated for a further 1 h. Following this incubation, the absorbance was measured using an automatic microplate reader at 510 nm. The calculation of the viability rate at different concentrations and time points is as follows (1):

$$\text{Viability rate (\%)} = \text{[[(control group - blank control group) - \\$rand control group)]} \tag{1}$$

$$- \text{ (toxin group - blank torsion group)} \text{[(control group - blank control group)]} \times 100\% $$

#### *5.4. Oxidative Stress*

Sample pretreatment: The supernatant of cell culture was discarded, and the pellet was digested with 0.25% trypsin for 2 min. Then, the culture medium was added to stop digestion by gentle micropipetting, and transferred into an EP tube and centrifuged at 3500–4000 rpm for 10 min. The supernatant was discarded and the precipitated cells were broken into suspension using ultrasonic wave. Their protein concentration was determined using bicinchoninic acid (BCA) protein assay kit. A volume of 0.2 mL of the suspension in a centrifuge tube (1.5 mL) was used for the assay.

The MDA assay kit was purchased from Nanjing Jiancheng Bioengineering Institute and used to measure oxidative stress damage. The centrifuge tubes were divided into four groups: standard tubes (0.2 mL 10 nmon/mL standards + 0.2 mL reagent 1 + 1.5 mL reagent 2 + 1.5 mL reagent 3), standard blank tubes (0.2 mL absolute ethyl alcohol + 0.2 mL reagent 1 + 1.5 mL reagent 2 + 1.5 mL reagent 3), measure tubes (0.2 mL measure sample + 0.2 mL reagent 1 + 1.5 mL reagent 2 + 1.5 mL reagent 3), and measure blank tubes (0.2 mL measure sample + 0.2 mL reagent 1 + 1.5 mL reagent 2 + 1.5 mL 50% glacial acetic acid). A spiral vortex mixer was used to mix samples in the standard tubes, standard blank tubes, measure tubes, and measure blank tubes. Test tubes were placed in a water bath for 40 min at 95 ◦C, cooled with a water cooling tube, and centrifuged at 3500–4000 rpm for 10 min. The supernatant was collected and the absorbance value (OD) of samples in each tube was measured at 532 nm with 1 cm optical path (2).

MDA (nmol/mgprot) = [(OD measure tube − OD measure blank tube)/(OD standard tube − OD standard blank tube)] × concentration of standard sample (10 nmol/mL) ÷ protein concentration of measure sample (mgprot/mL) (2)

The SOD assay kit was also purchased from Nanjing Jiancheng Bioengineering Institute and used to determine the activity of SOD using the WST-1 method. The centrifuge tubes were divided into four groups: control tubes (20 μL double distilled water + 20 μL enzyme working solution + 200 μL substrate application solution), control blank tubes (20 μL double distilled water + 20 μL enzyme diluents solution + 200 μL substrate application solution), measure tubes (20 μL measure sample + 20 μL enzyme working solution + 200 μL substrate application solution), and measure blank tubes (20 μL measure sample + 20 μL enzyme diluents solution + 200 μL substrate application solution). A spiral vortex mixer was used to mix samples in the control tubes, control blank tubes, measuring tubes, and measuring blank tubes, and then samples were incubated at 37 ◦C for 20 min. The absorbance value (OD) of samples was measured at 450 nm. The SOD activity is then measured by the degree of inhibition of this reaction. One unit of SOD was defined as the amount of enzyme needed to produce 50% dismutation of superoxide radical. The calculation of SOD activity is as below (3):

(U/mgprot) = inhibition rate of SOD (%) ÷ 50% × [reaction system (0.24 mL)/ dilution ratio (0.02 mL)] <sup>÷</sup> protein concentration of measuring sample (mgprot/mL) (3)

#### *5.5. Flow Cytometry of AV*/*PI*

The apoptosis assay kit was used to measure apoptosis using Annexin V and PI double staining. Chondrocytes were incubated with either the T-2 or HT-2 toxin (20 ng/mL) for 0, 12, 24, and 48 h. Chondrocytes were washed with PBS twice and 250 μL binding buffer was added to resuspend chondrocytes at a density of 1.0 <sup>×</sup> 106/mL. The cell suspension (100 <sup>μ</sup>L), PI solution (10 mL, 20 <sup>μ</sup>g/mL), and Annexin V/FITC (5 μL) were added to the 5 mL flow tube. The flow tube was mixed and incubated for 15 min in the dark at room temperature. Then 400 μL PBS was added to the reaction tube for flow cytometry analysis. As Annexin V and PI double staining were used to measure the apoptosis rate, automatic compensation regulation was used to avoid overlapping of two fluorescein wavelengths in the flow cytometry. When obtaining data from flow cytometric analysis, the gating was set using the combination of Forward Scatter (FSC) and Side Scatter (SSC), to establish FSC versus SSC dot diagram. By setting FSC threshold according to the size and granularity of chondrocytes, it can distinguish different cell populations, and remove from cell fragments, dead cells, and adhesion cells. The early apoptotic cells had been quantified using the Gated% data, the calculation of early apoptosis rate was Lower Right/(Upper Left + Upper Right + Lower Left + Lower Right).

#### *5.6. Fluorescence Intensity Analysis*

Hoechst 33258 was used to detect apoptosis of the chondrocytes. Chondrocytes were cultured in a 12-well plate. After the cells absorbed to the plate, the supernatant was discarded and the chondrocytes were fixedwith 4% formaldehyde for 30 min. The fixed chondrocytes were stained using a Hoechst 33258 working solution for 1 h at 37 ◦C with 5% CO2 in a humidified atmosphere. The maximum excitation and emission wavelength of Hoechst-DNA were 352 and 461 nm, respectively. Nuclei of normal chondrocytes fluoresced blue under the fluorescence microscope, whereas pale, dense, and hyperchromatic nuclei represented apoptotic cells.

Monodansylcadaverine (MDC) can be used to specifically mark the formation of autophagosomes. The chondrocytes were treated with the T-2 and HT-2 toxins in 24-well plates, and the medium was absorbed. One hundred microliters of the MDC staining solution was added to each well and staining

was carried out for 30 min at room temperature in the presence of light. The culture medium was discarded, and the cells were washed three times with 1× wash buffer (300 μL). The cell slide was covered with the collection buffer (100 μL). The wavelengths of stimulation and blocking filters of the fluorescence microscope were of 355 and 512 nm, respectively. Cell nuclei stained cyan-green were positive for an acidic autophagosome. When the cells were photographed using the fluorescence microscope, we identified four microscope fields of every replication microscopic picture (×400) and the cells were counted. Then the apoptosis rate was calculated as number of apoptotic cells/number of apoptotic cells and normal cells. Finally, the results were presented by mean ± standard deviations.

#### *5.7. Protein Extraction and Western Blot Analysis*

The chondrocytes were lysed using RIPA (Trizol method) and total protein in cell lysates was harvested by centrifugal separation according to the manufacturer's instructions. The concentration of extracted protein was quantified by the BCA assay kit (Beijing Tiangen Biotech Company, Beijing, China). Equal amounts (50 μg) of extracted protein were subjected to 10% (w/v) sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and electrophoretically transferred onto PVDF membrane. They were pre-incubated in blocking buffer containing 5% (w/v) non-fat milk with Tween 20 for 60 min at room temperature, rinsed three times with TBST for 5 min. The membranes were incubated with different primary antibodies against Bax, caspase-9, caspase-3, and Beclin1 (Cell Signaling Technology, Boston, MA, USA) overnight at 4 ◦C; all primary antibodies were used at a 1:1000 dilution. After being washed three times in TBS, the membrane was incubated with an appropriately diluted horseradish peroxidase-labeled secondary antibody (1:5000) in blotting buffer for 30 min. The blots were visualized by enhanced chemiluminescent (ECL). Western blot signals were exposed to X-ray films and the bands were quantified by Quantity One software. The protein levels were standardized by comparison with anti-GAPDH antibody.

#### *5.8. Statistical Analysis*

All experiments were performed in three independent trials, each of which included three replications. Experimental data were presented as the mean and standard deviations. SPSS18.0 software (IBM, Armonk, NY, USA) was used to analyze the experimental data, and the *t*-test was used to compare the differences between two groups. *p* < 0.05 was considered to be statistically significant between two groups.

**Author Contributions:** The conception and design of the study: F.-F.Y., X.-L.L. Collection, analysis and interpretation of the data entry as described above: F.-F.Y., X.-L.L. and X.W. Draft of the article and critical revision of the article for important intellectual content: F.-F.Y., Z.-G.P. and X.G.

**Funding:** This research was funded by the National Natural Scientific Foundation of China (81620108026), General Financial Grant from the China Postdoctoral Science Foundation (2019M652595) and Cultivating grand for youth key teacher in Higher Education Institutions of Henan province (2017GGJS012).

**Acknowledgments:** We thank Liwen Bianji, Edanz Editing China for editing the English text of a draft of our manuscript.

**Conflicts of Interest:** The authors declare no competing financial interests.

#### **References**


© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

*Article*
