Aflatoxin B1 affects apoptosis and expression of death receptor and endoplasmic reticulum molecules in chicken spleen

Aflatoxin B1 (AFB1) is a natural product of the Aspergillus genus of molds, which grow on several foodstuffs stored in hot moist conditions, and is among the most potent hepatocarcinogens and immunosuppression presently known. The latter was related to the up-regulated apoptosis of immune organs. However, the effect of expression of death receptor and endoplasmic reticulum molecules in AFB1-induced apoptosis of chicken splenocytes was largely unknown. The objective of this study was to investigate this unknown field. One hundred and forty four one-day-old chickens were randomly divided into control group (0 mg/kg AFB1) and AFB1 group (0.6 mg/kg AFB1), respectively and fed with AFB1 for 21 days. Histological observation demonstrated that AFB1 caused slight congestion and lymphocytic depletion in the spleen. TUNEL and flow cytometry assays showed the excessive apoptosis of splenocytes provoked by AFB1. Moreover, quantitative real-time PCR analysis revealed that AFB1 induced the elevated mRNA expression of Fas, FasL, TNF-α, TNF-R1, Caspase-3, Caspase-8, Caspase-10, Grp78 and Grp94 in the spleen. These findings suggested that AFB1 could lead the excessive apoptosis and alter the expression of death receptor and endoplasmic reticulum molecules in chicken spleen.

Secondary to the effect on liver, the immuno suppressive nature of AFB 1 is the best documented area of its toxicity [4]. AFB 1 inhibited the development of bursa of Fabricius [8], thymus [9], affected the weight and function of immune organs [10], decreased the percentages of T cell subsets, reduced the Igcontaining cell number [11] and the counts of splenic plasma cells [12], depressed the mitosis of B cells [13] and immunoglobulin as well as antibody production [14,15]. In addition, AFB 1 caused oxidative stress in lymphoid tissue [16,17], cell cycle arrest [18], and mitochondria injury in the lymphoid organs [19].
Apoptosis is associated with the normal development and homeostasis of animal tissues, and also involves in the pathogenesis [20]. Available information revealed that AFB 1 caused excessive apoptosis of several poultry and mammal cells such as hepatocytes [21], thymocytes [19,22], splenocytes [17], bursa of Fabricius cells [8], bronchial epithelial cells [23], jejunal mucosal cells [24], bone marrow cells [25], and renal Research Paper cells [26]. Furthermore, our early researches have shown the possible link between mitochondrial molecules and apoptosis of hepatocytes [21], thymocytes [19], and bursa of Fabricius cells provoked by AFB 1 [27]. However, some death receptor molecules may not be related to AFB 1 induced excessive cell death in the bursal cells [28], and the endoplasmic reticulum molecules may not be connected with the thymocyte apoptosis caused by AFB 1 [19].
The spleen, as one of the peripheral immune organs, is the largest lymphoid organ of the body [29]. With a large number of T and B lymphocytes, it is the center of cellular and humoral immunity [30]. In peripheral lymphoid organs, apoptosis is linked with the proliferation and maturation of lymphocytes after antigen recognition [31]. However, the effect of death receptor and endoplasmic reticulum molecules on AFB 1 induced splenic apoptosis remains practically unknown. Thus, we conducted this study in order to explore the alteration of death receptor and endoplasmic reticulum molecule expression in AFB 1 induced apoptosis in splenocytes of chicken by histopathological observation, flow cytometry, immunohistochemistry and relative realtime fluorescent quantitative PCR (RTqPCR) analysis. The outcomes from the present study could provide a reference for the further study of apoptosis mechanism caused by AFB 1 in human and other animals in the future.

Growth performance
The effects of dietary AFB1 on growth performance of chickens are shown in Figure 1. Compared with the control group, consumption of the AFB 1 diet reduced body weight and caused poor feed conversion rate at 14 and 21 days (p < 0.05 or p < 0.01). Meanwhile, feed intake in the AFB 1 group was not different from that in the control group (p > 0.05).

Absolute weight and relative weight of spleen
The absolute weight and relative weight of spleen in the AFB 1 group were significantly lower than those in the control group on day 14 and 21 (p < 0.05 or p < 0.01) ( Figure 2).

Histopathological observation
The parenchyma of chicken spleen was classified as white and red pulps. The former was subdivided into the splenic nodule, periarterial lymphoid tissue, and periellipsoidal lymphoid tissue. Compared with mammals, the chicken spleen had indistinct red pulp and white pulp, rich periellipsoidal lymphoid tissue, and few splenic nodules. The slight congestion was seen in some places of the red pulp, and the lymphocyte density was chiefly decreased in the white pulp in the AFB 1 group in comparison to the control group ( Figure 3).

Splenocyte apoptosis by TUNEL and flow cytometer analysis
The nuclei of TUNELpositive cells were stained brown. Under microscope, more positive cells in the AFB 1 were observed than those in the control group during the experiment ( Figure 4A-4D). Moreover, microscopic quantitative analysis also demonstrated the elevated number of positive cells in the AFB 1 group (p < 0.01) when compared with the control group ( Figure 4E). Flow cytometry assay showed the increased percentages of apoptotic splenocytes in the AFB 1 group (p < 0.05 or p < 0.01) during the experiment ( Figure 4F-4H).

Expression levels of apoptosis associated genes by qRT-PCR
The results of expression levels of apoptosis associated genes by qRTPCR are shown in the Figure 5. Compared with the control group, the expression of Fas, FasL, TNFα, TNFR 1 , Caspase10, Caspase8, Caspase3, Grp78 and Grp94 mRNA in the AFB 1 group was significantly raised (p < 0.05 or p < 0.01).

DISCUSSION
Our present study revealed that AFB 1 significantly decreased body weight and affected feed conversion rate, suggesting that dietary AFB 1 (0.6 mg/kg) affected the growth performance of the chicken, similar to previous reports [4].
As the the largest secondary immune organ, spleen takes part in activating the immune response to antigens, and in screening foreign substances [29]. The splenic size and gross morphology vary duo to different species and distension, but, the splenic weight is crucial in its functional evaluation [29]. The relative weight of spleen keeps relatively stable irrespective of age [29]. This study demonstrated that AFB 1 could reduce the splenic absolute weight and relative weight along with lymphocytic depletion, similar to the report by Chen et al. [11], suggesting that this toxin had the detrimental effects on the development and immune function of spleen. Similarly, several reports also revealed that AFB 1 could decrease the relative weights of central immune organs of chicken [8,28,32]. However, contrary reports also existed. For instance, Ortatatli et al. [33] reported that no statistical difference was found in the relative weight of spleen between the aflatoxinstreated broiler chicks and control ones. Peng et al. [34] demonstrated that aflatoxincontaminated corn intake significantly increased the relative weight of chicken spleen. This discrepancy might partially be associated to the types of toxin because the aflatoxincontamined diet in Ortatatli and Peng's researches contained different kinds of mycotoxins including AFB 1 [33,34].
Apoptosis has an important role in deveolpment, differentiation, proliferation and homeostasis of cell, tissue and organ [35]. AFB 1 directly or indirectly activated apoptotic process [36,37], inducing apoptosis of several poultry and mammal cells [8,9,17,19,[21][22][23][24][25][26]. The apoptotic cells could be evaluated by Flow cytometry and TUNEL assay [38,39]. Our present research revealed an increased apoptosis in the AFB 1 group demonstrated by TUNEL and flow cytometry, suggesting that AFB 1 could lead excess apoptosis in the chickens' splenocytes, in line with previous researches in thymocytes [19,22], bursa of Fabricius cells [27] and renal cells [26]. Lymphocytes are the main components within the lymphoid organs, and severe lymphocyte depletion in the lymphoid organs was due to apoptosis [40]. Therefore, it is tempting to speculate that the increased apoptosis of splenocytes provoked by AFB 1 might lead to the lymphocyte depletion, which may partly responsible for the declined splenic absolute and relative weights demonstrated in this study. Moreover, excessive apoptosis of lymphocytes was associated to immunosuppression in various circumstances [31]. Thus, our present results indicated that excessive apoptosis of spleen might cause immunosuppression in broilers exposed to AFB 1 . In addition, our present study revealed that the apoptotic percentage of the control group rose substantially by 21 days based on the flow cytometry assay, indicating that the normal apoptosis of the chicken's spleen showed with the means ± standard deviation (n = 6), * p < 0.05 and ** p < 0.01, compared with the control group. an increased changing pattern. Early researches also demonstrated that apoptsis of spleen was very obvious and presented developmentrelated changes [41,42]. Therefore, the present result that the apoptosis of AFB 1 treated samples increased from 7 to 14 to 21 days may not be due to this increasing baseline, rather than from compounding effects of AFB 1 .
The signal pathways of apoptosis are complex and different under apoptosis induced factor stimulating. The death receptor and endoplasmic reticulum molecules are important molecules related to cell apoptosis, of which caspases are the the final executioners [43]. To provide a reference for the further study of apoptosis mechanism caused by AFB 1 , we explored the alteration of Fas, FasL, TNFR 1 , Caspase3, Caspase8, Caspase10 Grp94 and Grp78 expression in AFB 1 induced apoptosis in splenocytes of chicken.
After bound with FasL or TNFR 1 and TNFR 2, respectively, Fas and TNFα were activated, then Caspase10 and Caspase8 were recruited and these Caspases including Caspase3 were activated, leading the cell apoptosis [44][45][46][47]. The present study demonstrated that AFB 1 diet led to the elevated expression of Fas, FasL, TNFR 1 , Caspase3, Caspase8 and Caspase10 mRNA expression in the spleen, which was consistent with earlier reports on the AFB 1 induced apoptosis of thymocytes and hepatocytes [19,21]. However, Yuan et al. reported that the mRNA expression of Fas, FasL, FADD, Caspase8 and Caspase10 in the bursa of Fabricius cells of the AFB 1 treated chickens were not significant different from those of the control ones, suggesting that the excessive apoptosis of the bursal cells caused by AFB 1 was not attributed to death receptor molecules [28]. Therefore, the effect of molecules involved in the AFB 1 caused apoptosis was different and complicated, and may vary depending on different tissues. The endoplasmic reticulum (ER) pathway is initiated by ER stress which causes the accumulation of misfolded or unfolded proteins in the ER [48,49]. This accumulation activates the expression of Grp94 and Grp78 and enhance the protein folding machinery [49,50]. If the unfolded protein reaction is unable to control the unfolded and misfolded protein levels in the ER, the apoptotic signaling provoked by ER is triggered by activating Caspase12, and Caspase3, and ultimately induces cell death [43,49]. Our present result demonstrated that AFB 1 could lead the elevated expression of Grp78 and Grp94 mRNA. This is consistent to previous report in the bursa of Fabricius cells [28]. However, contradictory result showed that the expression of Grp94 and Grp78 mRNA were not significant different between the AFB 1 treated group and control group of thymocytes, suggesting that the ER molecules may not involve in the AFB 1 induced apoptosis of thymocytes [19]. This is also confirmed that the effect of molecules related to the AFB 1 induced apoptosis may be different duo to various cell types.
Our study demonstrated that dietary AFB 1 (0.6 mg/kg) could cause the decline in the absolute weight and relative weight of chickens' spleen along with mild congestion and lymphocytic depletion, and induce splenocyte apoptosis accompanied by the upregulation of Fas, FasL, TNFα, TNFR 1 , Caspase3, Caspase8, Caspase10, Grp78 and Grp94 mRNA expression.

Animals and diets
The animal protocols used in this work and all procedures of the experiment were performed in compliance with the laws and guidelines of Sichuan Agricultural University Animal Care and Use Committee (Approval No: 2012-024). One hundred and forty four onedayold healthy Cobb male chickens were purchased from a commercial rearing farm (Wenjiang poultry farm, Sichuan province, China), and randomly divided into two equal groups, namely control group (0 mg/kg AFB 1 ) and AFB 1 group (0.6 mg/kg AFB 1 ). All of the chickens were put into cages with three replicates per group and 24 birds per replicate. The basal diet, namely the control diet, was formulated according to National Research Council (NRC, 1994) [51] and Chinese Feeding Standard of Chicken (NY/T332004). AFB 1 was purchased from SigmaAldrich (USA, A6636). The AFB 1 contaminated diet was made, similarly to the method described by Kaoud [52]. Briefly, 27 mg AFB 1 farinose solid was dissolved into 30 mL methanol completely, and then the 30 mL mixture was mixed into 45 kg cornsoybean basal diet to formulate AFB 1 diet of experimental groups containing 0.6mg/kg AFB 1 . The equivalent methanol was added into cornsoybean basal diet to formulate the control diet. Then the methanol of diets was evaporated at 98°F (37°C). The AFB 1 concentrations were analyzed by HPLC (Waters, Milford, MA, USA) with fluorescence detection (Waters, Model 2475, Milford, MA, USA), and the AFB 1 concentration was determined as < 0.001 mg/kg and 0.601 mg/kg in the control diet and AFB 1 contaminated diet, respectively. Chickens were fed in cages with electrically heated units and provided with water as well as aforementioned diet ad libitum for 21 days.

Growth performance
Body weight, feed intake, and feed conversion rate per cage were recorded weekly from 7 to 21 days of age.

The absolute and relative weights of spleen
At 7, 14, and 21 days of age, six chickens in each group were randomly chosen, weighted, and humanely euthanized. And spleens were removed, and weighted. The relative weight of spleen was calculated by the following formula: Relative weight of spleen = absolute weight of spleen (g)/body weight (kg)

Histopathological observation
At 7, 14, and 21 days of age, the splenic tissue samples from six chickens in each group were collected and fixed in 4% paraformaldehyde (PFA), and then were dehydrated and embedded in paraffin wax. Blocks were cut into 5 µm sections with a microtome (Leica, Germany, RM2135) for haematoxylin and eosin (HE.) staining and TUNEL assay. The histological structure of the tissues was observed under light microscope and photographed with a digital camera (Nikon DSRi1, Japan).

Apoptosis detection by TUNEL assay
TUNEL assay was performed with an apoptosis detection Kit (MK1020 Boster, Wuhan, China) according to the manufacturer's instructions. Briefly, tissue sections were dewaxed with 100% xylene, and rehydrated in successive changes of 100%, 95%, 85% and 75% ethanol. After endogenous peroxidase activity was quenched for 10 min in 3% H 2 O 2 with distilled water at 37°C, the sections were incubated with proteinase K diluted 1:200 in TBS at 37°C for 5-10 min in a humidified chamber. A labeling mixture containing digoxindUTP in TdT (Terminal deoxynucleotidyl Transferase) enzyme buffer was added to the sections and incubated at 37°C for 2 h. After three successive washings with TBS for 2 min, sections were covered with antidigoxinbiotin conjugate diluted 1:100 in blocking regent and incubated for 30 min at 37°C. The tissues were then incubated for 1 h at 37°C with strept avidinbiotincomplex (SABC) diluted 1:100 in TBS. Labeling was visualized with 3′3′diaminobenzidene. The sections were then counterstained with haematoxylin. For the negative control, representative sections were processed in the same way but incubation with TdT enzyme buffer was omitted.
The number of TUNELpositive cells in the spleen was evaluated by ImagePro Plus5.1 (USA) image analysis software. Briefly, photographs of TUNEL staining were taken with a digital microscope camera system (Nikon DSRi1, Japan). For each section, five fields of 0.064 mm 2 (corresponding to five fields at 400 × magnification) were analyzed. By selecting "colourchosen target" in the option bar of the morphologic analysis system, all TUNELpositive cells in the field were marked in colour. Then, "calculating" in the option bar was selected to automatically calculate the number of TUNELpositive cells. Results were expressed as the average of TUNEL positive cells per 0.064 mm 2 area.

Apoptosis detection by flow cytometry
At 7, 14, and 21 days of the experiment, six chickens in each group were euthanized, and spleens were sampled to determine the percentage of apoptotic cells by flow cytometer, using the method by Chen et al. [53]. Briefly, the dissected spleens were thereupon homogenized to form a cell suspension and filtered, and then the cells were washed and resuspended in phosphate buffer at a concentration of 1 × 10 6 cells/mL. 5 μL Annexin VFluorescein isothiocyanate (VFITC) and 5 μL propidium iodide (PI) were added into 100 μL cell suspension, and incubated at 25°C for 15 min in the dark. 400 μL 1 × Annexin binding buffer was added to the mixture, and then the apoptotic cells were assayed by flow cytometer (BD FACSCalibur) within 1 h. The annexin VFITC Kit was purchase from BD Pharmingen (USA, 556547).

Expression levels of apoptotic regulator mRNAs by quantitative real-time PCR
Quantitative realtime PCR (qRTPCR) assay was carried out as reported by Chen et al. [22]. Briefly, the spleens from six chickens in each groups at 7, 14, and 21 days of the experiment were obtained and stored in liquid nitrogen, respectively. Adding liquid nitrogen, the samples were crushed with pestle to homogenize until powdery, respectively. Total RNA was extracted from the powdery of samples using RNAiso Plus (9108/9109, Takara, Otsu, Japan). The mRNA was then reverse transcribed into cDNA using PrimScriptTM RT reagent Kit with gDNA Eraser (RR047A, Takara, Otsu, Japan). The cDNA was used as a template for qRTPCR analysis.
For qRTPCR reactions, 25 μL mixtures were made by using SYBR ® Premix Ex Taq TM II (DRR820A, Takara, Otsu, Japan), containing 12.5 μL Tli RNaseH Plus, 1.0 μL of forward and 1.0 μL of reverse primer, 8.5 μL RNAase free water and 2 μL cDNA. Reaction conditions were set to 3 min at 95°C (first segment, one cycle), 10 s at 95°C and 30 s at Tm of a specific primer pair (second segment, 44 cycles) followed by 10 s at 95°C, and 72°C for 10 s (dissociation curve segment) using Thermal Cycler (C1000, BIO RAD, CA, USA). The mRNA expression of Fas, FasL, TNFα, TNFR 1 , Caspase10, Caspase8, Caspase3, Grp78 and Grp94 was analyzed. βactin was used as an internal control gene. Sequence of primers was obtained from GenBank of NCBI. Primers were designed with Primer 5, and synthesized by BGI Tech (Shenzhen, China) ( Table 1). The qRTPCR data were analyzed and fold change in expressions were calculated using the 2 ΔΔCT method [54].

Statistical analysis
The significance of difference between two groups was analyzed by variance analysis, and the results were expressed by mean ± standard deviation. The analyses were performed using the independent sample test of SPSS 20.0 software (IBM Corp, Armonk, NY, USA) for windows. Statistical significant differences were considered at p < 0.05 and markedly significant differences were considered at p < 0.01.

ACKNOWLEDGMENTS
This work was supported by the program for Changjiang scholars, the University Innovative Research Team (IRT 0848), the Education Department of Sichuan Province (2012FZ0066 and 2013FZ0072) and Huimin project of Chengdu science and technology (2016HM01 00337SF).