Comparison of the Effects of Nonprotein and Protein Nitrogen on Apoptosis and Autophagy of Rumen Epithelial Cells in Goats

Simple Summary Proteins provide important raw materials for the self-renewal of cells in the gastrointestinal tract. Cell self-renewal is inseparable from the coordination between apoptosis and autophagy. Previous researchers have explored whether different nitrogen sources are linked to activation or inhibition of apoptosis/autophagy. However, information on the regulatory mechanism of apoptosis and autophagy induced by different nitrogen sources among cells in the gastrointestinal tract is limited. This study attempted to compare the effects of different nitrogen sources on apoptosis and autophagy of rumen epithelial cells in goats through molecular biology techniques such as flow cytometry, transmission electron microscopy, and fluorescence microscopy. On the basis of the obtained results, we found that autophagy had a less obvious ameliorative effect on ruminal epithelial cell apoptosis after treatment with protein nitrogen than after treatment with nonprotein nitrogen. Abstract Protein nutrition is particularly important for the self-renewal processes of gastrointestinal epithelial cells. The self-renewal of cells is inseparable from the interaction between apoptosis and autophagy. However, there are few reports on the relationship between different nitrogen sources and apoptosis/autophagy. In this study, the relative protein expression of Bcl-2-associated X protein(Bax), caspase-3, and p62 was significantly higher (p < 0.05), while that of Bcl-xl, Bcl-2, Beclin1, and Microtuble-associated protein light chain 3 (LC3-II) was significantly lower (p < 0.05), in the NH4Cl group in comparison with the NH4Cl + 4-phenylbutyric acid (4PBA) group. In addition, the relative protein expression of Bax and caspase-3 was significantly higher (p < 0.05), while that of Bcl-2 and Bcl-xl was decreased significantly (p < 0.05), in the NH4Cl + 3-Methyladenine (3-MA) group and the methionine (Met) + 3-MA group in comparison with the NH4Cl group. Furthermore, the relative protein expression of Beclin1 and LC3B-II was significantly lower (p < 0.05), while that of p62 was significantly higher (p < 0.05), in the NH4Cl + Z-VAD-FMK group and the Met + Z-VAD-FMK group in comparison with the NH4Cl group. In conclusion, our results suggested that endoplasmic reticulum (ER) stress played a critical role in the crosstalk between apoptosis and autophagy induced by NH4Cl and Met. Autophagy had a more obvious ameliorative effect on ruminal epithelial cell apoptosis after treatment with nonprotein nitrogen than after treatment with protein nitrogen. These findings may reveal the molecular mechanism of apoptosis and autophagy induced by nonprotein nitrogen and protein nitrogen.


Introduction
Apoptosis, which plays a pivotal role in tissue homeostasis by eliminating damaged or unnecessary cells in response to various stimuli, is the main cause of the shedding of normal intestinal mucosal epithelial cells and is closely related to the regeneration and repair of gastrointestinal cells [1]. In addition, exfoliation of mucosal epithelial cells during the renewal of gastrointestinal tissue is the largest contributor to the metabolism of fecal nitrogen in ruminants [2]. Autophagy maintains cellular homeostasis to promote cell survival by removing long-lived proteins and damaged organelles under conditions of starvation or other stresses. In many cases, autophagy negatively regulates apoptosis, whereas apoptosis activation blocks autophagy. Apoptosis and autophagy can be induced by modulating the ROS (oxygen free radical)/MAPK (mitogen-activated protein kinase) [3], PI3K(phosphatidylinositol 3-kinase)/Akt(serine-threonine kinase)/mTOR(mammalian target of rapamyoin) [4], and JNK(Jun N-terminal kinase) and mTOR [5] pathways.
Several amino acids (AAs) are associated with apoptosis/autophagy activation or inhibition. Currently, it is recognized that these AAs participate in the mTORC1 and GCN2/eIF2(Protein kinase GCN2/Eukaryotic initiation factor 2) pathways, regulating protein translation and controlling cell AA requirements by simultaneously regulating autophagy-dependent catabolism [6]. For example, methionine (Met) can reduce apoptosis of lymphocytes in peripheral blood and increase the Bcl-2(B-cell lymphoma-2 protein)/Bax(Bcl-associated protein X) ratio [7]. Researchers have found that exogenous Met also inhibits autophagy in liver cells treated via in vitro perfusion [8]. Additionally, studies have shown that Met regulates autophagy in BMECs(bovine mammary epithelial cells) through the PI3K/mTOR signaling pathway [9]. Notably, it has been reported that nonprotein nitrogen (NPN), such as that in ammonium chloride (NH 4 Cl), induces apoptosis by increasing oxidative damage and Cyt C(Cytochrome C) release [10][11][12]. When ammonia is present in excess, it can cause alkaline chemical burns of the mucosa [13]. Furthermore, NH 4 Cl can inhibit autophagy through the p38 MAPK pathway [14,15].
AAs, the most basic units used for metabolic activity in gastrointestinal tissue, are utilized by epithelial cells for protein synthesis to meet the growth, reproduction and metabolism needs of cells. NH 4 Cl is widely used as a source of NPN in livestock production. However, there have been few reports on the differences between apoptosis and autophagy induced by different nitrogen sources. Therefore, this study compared the effects of different nitrogen sources on apoptosis and autophagy in goat rumen epithelial cells. The findings will help elucidate the mechanisms of self-renewal in gastrointestinal epithelial cells and endogenous nitrogen secretion in ruminants.

Culture and Treatment of Primary Ruminal Epithelial Cells (PRECs)
The rumen tissues of healthy black goats were repeatedly rinsed with normal saline and then placed into phosphate-buffered saline (PBS) containing streptomycin, amphotericin B, and gentamicin for storage until use. Isolation, purification, identification, and culture of the PRECs were conducted on the basis of previously described methods [16]. Briefly, the epithelial layer of rumen tissue was separated from the muscle layer, added to a digestive solution containing 2.5% trypsin, and incubated for 5 min at 37 • C. The epithelial tissues were digested 5 times; the supernatant was discarded after the first 2 digestions, while the supernatant and precipitates were saved after the final 3 digestions. Then, the cells were resuspended in complete medium, inoculated into culture dishes, and cultured in an incubator at 37 • C. The medium was changed every 2 days. The purified cells were cultured in Dulbecco's modified Eagle's medium: Nutrient Mixture F-12 (DMEM/F12; Invitrogen, Carlsbad, CA, USA) containing 5% fetal bovine serum (FBS; Gibco, Grand Island, NE, USA), penicillin (100 units/mL, Invitrogen, Carlsbad, CA, USA)-streptomycin (0.1 mg/mL, Invitrogen, Carlsbad, CA, USA), gentamicin (25 g/mL, Invitrogen, Carlsbad, CA, USA), and amphotericin B (2.5 g/mL, Invitrogen, Carlsbad, CA, USA) at 37 • C in a cell incubator containing 5% CO 2 at a constant temperature and humidity. Met-free medium was prepared on the basis of the DMEM/F12 formula. NH 4 Cl was freshly diluted in Met-free medium and added to the culture medium before treatment on the basis of the DMEM/F12 formula. When the PRECs reached 70% confluency, the cells were starved for 6 h in Met-free medium, and then cultured with the fresh medium containing 5 mM Met (control group) or 5 mM NH 4 Cl (treatment group) for 36 h (proliferation stage).

Morphological Detection
The fluorochrome Hoechst 33,342 (14,533, Sigma, MO, USA) was applied to detect chromatin condensation as previously described [17], and 4-phenylbutyric acid (4PBA) was added to confirm the role of endoplasmic reticulum (ER) stress in apoptosis. Briefly, after treatment for 36 h with 5 mM Met or NH 4 Cl, PRECs were fixed for 30 min at 4 • C with 4% paraformaldehyde. Then, the cells were treated with PBS (5 min per wash). The cells were then incubated with Hoechst 33,342 (8 µg/mL) in the dark for 30 min at room temperature. Finally, the cells were observed under a fluorescence microscope to determine the ratio of apoptotic nuclei.

Monodansylcadaverine (MDC) Staining Analysis
MDC (KGATG002, KeyGEN, Nanjing, China) was used as a tracer for autophagy [18]. After PRECs were treated for 36 h, MDC (50 µm) was added, and the cells were incubated in the dark for 20 min at a constant temperature of 37 • C in a 5% CO 2 incubator. The morphological changes related to autophagy were observed by fluorescence microscopy.

Apoptosis Detection by Flow Cytometry
PRECs were inoculated into 6-well plates at a density of 1 × 10 6 cells per well and cultured with Met or NH 4 Cl. The cells were collected, centrifuged at 1500 rpm for 5 min, and then resuspended in 500 µL of binding buffer. Finally, the cells were incubated with 5 µL of annexin V-FITC(Fluorescein Isothiocyanate, Nanjing, Jiangsu, China) and 5 µL of propidium iodide (PI) (Annexin V-FITC/PI Apoptosis Detection Kit, KGA108, KeyGEN) in the dark for 15 min at room temperature. Flow cytometry (BD FACScalibur, San Jose, CA, USA) was used to analyze the apoptosis rate.

Intracellular ROS Detection
Cells were collected and exposed to pre-warmed PBS containing DCFHDA(2,7-Dichlorodi -hydrofluorescein diacetate) (10 µM) for 1 h at 37 • C. Then, the cells were washed and resuspended in PBS. Finally, a flow cytometer was used to detect the fluorescence intensities of cells at an excitation/ emission wavelength of 488/525 nm. FlowJo software (FlowJo, Version 7.6.1; TreeStar, Ashland, OR, USA) was applied to analyze the fluorescence intensities data.

Detection of Mitochondrial Membrane Potential (MMP)
A JC-1 assay kit (MCE3520-43-2, Shanghai, China) was used to measure the MMP [17]. Cells were collected, washed with PBS, and then incubated with JC-1 (10 µg/mL) in the dark for 20 min at 37 • C. Finally, a flow cytometer was used to detect the fluorescence intensity of the cells (BD FACScalibur).

Transmission Electron Microscopy (TEM) Analysis
After treatment with 5 mM Met or NH 4 Cl for 36 h, PRECs were fixed successively with 1% osmic acid fixation solution and 2.5% glutaraldehyde for more than 2 h. Then, the cells were washed with phosphate buffer (0.1 mM) and dehydrated with a gradually increasing concentrations of acetone.
After this, they were coated with epoxy resin, the cell samples were sliced. A transmission electron microscope (JEM-2000EX, JEOL Co, Tokyo, Japan) was applied to collect the electron images [17].

RT-PCR
After treatment for 36 h with 5 mM Met or NH 4 Cl, total RNA was extracted using AccuZol Total RNA Extraction Reagent (Bioneer, Daejeon, Korea) on the basis of the manufacturer's instructions [19]. Then, DNaseI (Thermo Scientific, Waltham, MA, USA) was used to eliminate genomic DNA, and a Nano Drop 2000 (Thermo Scientific, Waltham, MA, USA) was used to evaluate the RNA quality and quantity. Then, a Prime Script RT Reagent Kit (Takara, Dalian, China) was applied to reverse-transcribe the RNA into cDNA(complementary Deoxyribonucleic acid). The 2 −∆∆CT method was used to determine the relative mRNA expression levels [20]. The details of the primers used for the various genes are shown in Table 1.

Western Blot Analysis
Protein isolation and Western blot analysis were conducted on the basis of previously described methods [21,22]. Briefly, each sample and a pre-stained standard were electrophoretically separated on 10% SDS-polyacrylamide gels (Bio-Rad Laboratories, Berkeley, CA, USA). The separated proteins were transferred to polyvinylidene fluoride (PVDF) membranes (Bio-Rad Laboratory) at a constant current 200 mA for 70 min. The PVDF membranes were incubated overnight in a solution containing primary antibodies (see Table 2) at 4 • C, washed 5 times, and then incubated with secondary antibodies (1:6000, Proteintech) at room temperature for 2 h while protected from light. An Alpha Imager 2200 digital imaging system (Digital Imaging System, Kirchheim, Germany) was used to obtain and analyze images.

Statistical Analysis
SPSS Statistical Software was used to analyze the data obtained. ROS levels, apoptosis rates, fluorescence intensity, autophagosome numbers, relative apoptotic and autophagic gene mRNA expression, LC3 spot numbers, protein abundance, and ER cavity size were analyzed using one-way ANOVA followed by Tukey-Kramer multiple comparisons tests. The data are shown as the mean ± SD.

Apoptosis and Autophagy Induction by NH 4 Cl and Met in PRECs
The PRECs in the Met group showed normal shapes with round, intact nuclei, whereas the cells treated with NH 4 Cl showed nuclear shrinkage, chromatin condensation, or fragmentation ( Figure 1A). In addition, the apoptosis rate was significantly higher (p < 0.05) in the NH 4 Cl supplementation group than in the Met group ( Figure 1B). The MMP change in the NH 4 Cl group was significantly larger (p < 0.05) than that in the Met group ( Figure 1C). Cytoplasmic ROS levels were higher (p < 0.05) in NH 4 Cl-treated cells than in Met-treated cells ( Figure 1D). Compared with Met addition, NH 4 Cl addition significantly increased (p < 0.05) the expression of Bax and caspase-3 at the mRNA and protein levels, while significantly decreasing (p < 0.05) the expression of Bcl-xl and Bcl-2 ( Figure 1E).  The cytoplasmic fluorescence intensity was higher (p < 0.05) in NH 4 Cl-treated cells than in Met-treated cells (Figure 2A). In addition, the number of autophagosomes per cell was higher (p < 0.05) in NH 4 Cl-treated cells than in Met-treated cells ( Figure 2B). After GFP-LC3 plasmid transfection, there were more LC3 spots in the NH 4 Cl group (p < 0.05) than in the Met group ( Figure 2C). The mRNA and protein expression levels of LC3-II and Beclin1 were higher (p < 0.05), while those of p62 were lower (p < 0.05) in NH 4 Cl-treated cells in comparison with Met-treated cells ( Figure 2D). The cytoplasmic fluorescence intensity was higher (p < 0.05) in NH4Cl-treated cells than in Met-treated cells (Figure 2A). In addition, the number of autophagosomes per cell was higher (p < 0.05) in NH4Cl-treated cells than in Met-treated cells ( Figure 2B). After GFP-LC3 plasmid transfection, there are more LC3 spots in the NH4Cl group (p < 0.05) than in the Met group ( Figure 2C). The mRNA and protein expression levels of LC3-II and Beclin1 were higher (p < 0.05), while those of p62 were lower (p < 0.05), in NH4Cl-treated cells than in Met-treated cells ( Figure 2D).

ER stress induction by NH4Cl and Met in PRECs
The expansion area of the ER cavity was larger (p < 0.05) in NH4Cl-treated cells than in Met-treated cells ( Figure 3A). Concurrently, the protein expression levels of DAPK1, GRP78, ATF4 and CHOP were higher (p < 0.05) in the NH4Cl group than in the Met group ( Figure 3B), while the protein expression levels of DAPK1, GRP78, ATF4 and CHOP were lower (p < 0.05) in NH4Cl + 4PBA-treated cells than in NH4Cl-treated cells ( Figure 3B). There were no differences (p > 0.05) in the protein expression levels of DAPK1, GRP78, ATF4 and CHOP between the Met group (p > 0.05) and the NH4Cl + 4PBA group.

ER Stress Induction by NH 4 Cl and Met in PRECs
The expansion area of the ER cavity was larger (p < 0.05) in NH 4 Cl-treated cells than in Met-treated cells ( Figure 3A). Concurrently, the protein expression levels of DAPK1(Death-associated Protein Kinase 1), GRP78 (glucose-regulated protein 78), ATF4 (activating transcription factor 4), and CHOP (C/EBP homoiogousprotein) were higher (p < 0.05) in the NH 4 Cl group than in the Met group ( Figure 3B), while the protein expression levels of DAPK1, GRP78, ATF4, and CHOP were lower (p < 0.05) in NH 4 Cl + 4PBA-treated cells than in NH 4 Cl-treated cells ( Figure 3B). There were no differences (p > 0.05) in the protein expression levels of DAPK1, GRP78, ATF4, and CHOP between the Met group (p > 0.05) and the NH 4 Cl + 4PBA group.

292
The cytoplasmic fluorescence intensity was higher (p < 0.05) in NH4Cl-treated cells than in 293 Met-treated cells (Figure 2A). In addition, the number of autophagosomes per cell was higher 294 (p < 0.05) in NH4Cl-treated cells than in Met-treated cells ( Figure 2B). After GFP-LC3 plasmid 295 transfection, there are more LC3 spots in the NH4Cl group (p < 0.05) than in the Met group 296 ( Figure 2C). The mRNA and protein expression levels of LC3-II and Beclin1 were higher (p < 297 0.05), while those of p62 were lower (p < 0.05), in NH4Cl-treated cells than in Met-treated cells 298 ( Figure 2D). 299

ER stress induction by NH4Cl and Met in PRECs 300
The expansion area of the ER cavity was larger (p < 0.05) in NH4Cl-treated cells than in 301 Met-treated cells ( Figure 3A). Concurrently, the protein expression levels of DAPK1, GRP78, 302 ATF4 and CHOP were higher (p < 0.05) in the NH4Cl group than in the Met group ( Figure 3B), 303 while the protein expression levels of DAPK1, GRP78, ATF4 and CHOP were lower (p < 0.05) 304 in NH4Cl + 4PBA-treated cells than in NH4Cl-treated cells ( Figure 3B). There were no 305 differences (p > 0.05) in the protein expression levels of DAPK1, GRP78, ATF4

Effects of ER stress on apoptosis induced by NH4Cl and Met in PRECs 353
The cytoplasmic fluorescence intensity was higher (p < 0.05) in NH4Cl-treated cells than in 354 Met-treated cells ( Figure 4A), and no significant differences were observed between Met-355 treated cells and NH4Cl+4BPA treated cells (p > 0.05). There were significantly more (p < 0.05) 356 apoptotic cells in the NH4Cl group than in the NH4Cl + 4PBA group, while no significant 357 difference (p > 0.05) was observed between the Met group and the Met + 4PBA group (p > 0.05). 358 The number of apoptotic cells in the Met + 4PBA group was not significantly different from 359 that in the NH4Cl + 4PBA group (p > 0.05) ( Figure 4B). As shown in Figure 4C

Effects of ER Stress on Apoptosis Induced by NH 4 Cl and Met in PRECs
The cytoplasmic fluorescence intensity was higher (p < 0.05) in NH 4 Cl-treated cells than in Met-treated cells ( Figure 4A), and no significant differences were observed between Met-treated cells and NH 4 Cl + 4BPA treated cells (p > 0.05). There were significantly more (p < 0.05) apoptotic cells in the NH 4 Cl group than in the NH 4 Cl + 4PBA group, while no significant difference (p > 0.05) was observed between the Met group and the Met + 4PBA group (p > 0.05). The number of apoptotic cells in the Met + 4PBA group was not significantly different from that in the NH 4 Cl + 4PBA group (p > 0.05) ( Figure 4B). As shown in Figure 4C, compared to cells treated with Met only, cells treated with NH 4 Cl + 4PBA exhibited significantly higher relative protein expression of Bax and caspase-3 (p < 0.05), and Bcl-xl and Bcl-2 protein expression was significantly lower (p < 0.05) in the NH 4 C group than in the NH 4 Cl + 4PBA group. There were no significant differences in the expression of apoptotic proteins in the Met + 4PBA group compared with the Met group (p > 0.05) ( Figure 4C).

Effects of ER Stress on Autophagy Induced by NH 4 Cl and Met in PRECs
There were significantly fewer (p < 0.05) autophagosomes in NH 4 Cl + 4PBA-treated cells than in NH 4 Cl-treated cells, while no significant differences (p > 0.05) were observed between Met + 4PBA-treated cells and Met-treated cells ( Figure 5A,B). After GFP-LC3 plasmid transfection, there were fewer LC3 spots in the NH 4 Cl + 4PBA group (p < 0.05) than in the NH 4 Cl group ( Figure 5C). The protein expression levels of Beclin1 and LC3-II were lower, while those of p62 were significantly higher (p < 0.05), in the NH 4 Cl + 4PBA group in comparison with the NH 4 Cl group ( Figure 5D). The protein expression levels of Beclin1, LC3-II, and p62 in the Met + 4PBA group were not significantly different from those in the Met group (p > 0.05).

Crosstalk between apoptosis and autophagy induced by NH4Cl and Met in PRECs
The relative protein expression of Bax and caspase-3 was significantly higher (p < 0.05),

Crosstalk between Apoptosis and Autophagy Induced by NH 4 Cl and Met in PRECs
The relative protein expression of Bax and caspase-3 was significantly higher (p < 0.05), while that of Bcl-2 and Bcl-xl was significantly lower (p < 0.05) in the NH 4 Cl + 3-MA(3-Methyladenine) group than in the NH 4 Cl group. In addition, the relative protein expression of Bax and caspase-3 was significantly lower (p < 0.05), while that of Bcl-2 and Bcl-xl was significantly higher (p < 0.05), in the Met + 3-MA group in comparison with the NH 4 Cl group ( Figure 6A). As shown in Figure 6B, the relative protein expression of Beclin1 and LC3B-II was significantly lower (p < 0.05), while that of p62 was significantly higher (p < 0.05), in the NH 4 Cl + Z-VAD-FMK group in comparison with the NH 4 Cl group, while no significant difference was observed between the Met + Z-VAD-FMK group and the Met group (p > 0.05). The protein expression of Beclin1 and LC3-II was significantly higher (p < 0.05), while that of p62 was significantly lower (p < 0.05), in the NH 4 Cl group in comparison with the Met + Z-VAD-FMK group.    Western blot analysis was performed in order to determine the effect of on autophagy induced for 36 h. Each bar shows mean ± SD.

Discussion
In ruminants, the self-renewal process of rumen epithelial cells is affected by more factors than that of other types of cells because the rumen is in direct contact with the diet. During self-renewal, the mucosal epithelial cells of the gastrointestinal tissue need to continuously synthesize DNA, RNA, and proteins, and thus proteins from various sources are very important for the growth, regulation, and repair of gastrointestinal mucosal epithelial cells [23,24]. In addition, the regeneration and repair of epithelial cells are inseparable from apoptosis and autophagy, which frequently occur in epithelial cells. Therefore, studying the relationships between different nitrogen sources and apoptosis/autophagy is critical.
Apoptosis is an important mechanism of NH 4 Cl toxicity, and ROS contribute substantially to NH 4 Cl-induced apoptosis [10]. High levels of NH 4 Cl promote ROS production, mitochondrial dysfunction, and cytochrome c release, and eventually lead to apoptosis [11]. In the present work, the levels of both ROS and p53 were increased in NH 4 Cl-treated cells. We hypothesized that ROS activated p53-regulated mitochondrial apoptosis pathways during NH 4 Cl-induced apoptosis. ROS concentrations are correlated with the disruption of mitochondrial membrane potential ∆ψm [25], which is consistent with the results regarding MMP obtained in the present work-the NH 4 Cl group exhibited higher ROS levels than the Met group. When apoptosis occurs, the p53-regulated Bcl-2 protein family (including Bax and Bcl2) activates the mitochondrial apoptotic pathway [26]. In the current study, changes in the relative protein expression of p53 and Bax were positively correlated with changes in the apoptosis rate. p53 can disrupt the Bax/Bcl-2 balance by interacting with Bcl-2 protein in the mitochondria, increasing the permeability of the outer mitochondrial membrane [27]; this phenomenon is consistent with our findings that p53 levels were higher, while the MMP and Bax/Bcl-2 ratio was lower, in the NH 4 Cl group in comparison with the Met group. Therefore, we conclude that NH 4 Cl induces apoptosis by causing mitochondrial dysfunction.
Autophagy is a lysosomal-dependent degradation pathway and a dynamic process, the last step of which is the fusion of autophagosomes and lysosomes to degrade cytoplasmic substances and organelles [28]. Previous studies have reported that Met deficiency leads to autophagy of glioma cells both in vitro and in vivo [29], consistent with our findings that the levels of autophagy-related proteins and the numbers of autophagosomes were higher in NH 4 Cl-treated cells than in Met-treated cells. In this study, we observed significantly increased LC3-II/LC3-I ratios and decreased p62 levels in NH 4 Cl-treated cells, which might have resulted from the fact that there is no colocalization between mitochondria and lysosomes when cells are treated with NH 4 Cl under starvation conditions, which leads to decreased autophagy [14]. Therefore, we speculate that autophagy in the NH 4 Cl group was induced by ER stress.
A previous study has revealed that NH 4 Cl induces apoptosis and autophagy through ER stress and mitochondrial dysfunction in MRP2 cells [11]. Thus, we aimed to explore the relationship between ER stress and apoptosis/autophagy induced by NH 4 Cl in this study. Pretreatment with 4PBA reduced NH 4 Cl-mediated induction of apoptosis and autophagy, implying that ER stress helps regulate apoptosis and autophagy in PRECs. In the present work, the levels of CHOP increased with upregulation of Bax and p53 expression in NH 4 Cl-treated cells, perhaps because ER stress activated CHOP; triggered Bcl-2, Bak, and Bax expression; and ultimately activated ROS-induced apoptosis [30]. Additionally, a previous study has found that PC12 cells might be injured by ER stress-mediated apoptosis [31]. Furthermore, activation of ER stress pathways has also been reported to be associated with autophagy [32,33]. In the present study, we observed significant increase in LC3-II and a decrease in p62 with increased CHOP expression in NH 4 Cl-treated cells, indicating that the increase in autophagy was associated with the activation of ER stress. In conclusion, ER stress contributes to the apoptosis and autophagy triggered by NPN in PRECs, perhaps because that NPN can promote the occurrence of oxidative stress [34].
Apoptosis and autophagy can occur simultaneously in Bel-7402 cells, and autophagy can regulate apoptosis via phosphorylation of p38 MAPK. However, autophagy promotes cell proliferation and survival in response to stress and inhibits apoptosis [35,36]. In this study, we found that downregulation of autophagy via treatment with 3-MA significantly intensified apoptosis triggered by oxidative stress in NH 4 Cl-treated cells; conversely, inhibition of apoptosis with Z-VAD-FMK alleviated autophagy in NH 4 Cl-treated cells. However, we did not find a similar phenomenon in cells treated with Met. Thus, we suggest that autophagy might have a palliative effect on apoptosis in cells treated with NPN compared with protein nitrogen, possibly because that protein nitrogen can better provide the carbon skeleton and energy needed for cellular protein synthesis and can promote cell growth.

Conclusions
Our results suggest that ER stress contributes substantially to the apoptosis and autophagy triggered by protein nitrogen and NPN. Autophagy has a more obvious ameliorative effect on apoptosis in ruminal epithelial cells treated with NPN than in ruminal epithelial cells treated with protein nitrogen, perhaps because protein nitrogen better suited than NPN for protein synthesis and promotes the resistance of cells to AA deficiency stress.

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