Bovine mammary epithelial cells (MAC-T cells, preserved in the laboratory, originally from ATCC) were grown in Dulbecco's Modified Eagle Medium (DMEM) with high glucose (HyClone; Cytiva), supplemented with 10% fetal bovine serum (ExCell Biotechnology Co., Ltd.) and 1% penicillin-streptomycin at 37˚C with 5% CO₂ in an incubator. An expression vector for bovine SIRT5 was constructed using the plasmid expressing enhanced green fluorescence protein (pEGFP-N1) (Beijing Huayueyang Biotechnology Co., Ltd.) (32), and stably transfected into MAC-T cells for implementing SIRT5 overexpression. In brief, according to the CDS region of the bovine SIRT5 gene (NM_001034295), the cDNA encoding SIRT5 was inserted into the pEGFP-N1 vector to construct pEGFP-SIRT5 vector. Subsequently, according to the instructions for the Lipofectamine 3000 reagent, the pEGFP-N1 vector and the pEGFP-SIRT5 vector were respectively transfected into MAC-T cells. After 24 h, G418 was used to select SIRT5-overexpressing MAC-T cells (SIRT5^+/+). Subsequently, the fluorescence intensity in MAC-T cells was observed under a fluorescence microscope to confirm successful transfection. The cells were divided into 2 groups: Empty vector (pEGFP, transfected with pEGFP-N1 vector) and SIRT5 overexpression (pEGFP-SIRT5, transfected with pEGFP-SIRT5 vector).

The stable SIRT5 knockdown cells (SIRT5^-/-) were generated with CRISPR-CAS9 gene editing systems to target the SIRT5 gene in MAC-T cells. Three 20-bp guide sequences targeting bovine SIRT5 gene (NM_001034295) were as follows: Single guide (sg)RNA1, 5'-GTTCTACCACTACCGGCGGG-3'; sgRNA2, 5'-GGGAGTTCTACCACCGG-3'; and sgRNA3, 5'-GGAGTTCTACCACTACCGGC-3'. sgRNAs were synthesized by Sangon Biotech Co., Ltd. pX330 (Beijing Huayueyang Biotechnology Co., Ltd.) was selected as the knockdown vector. The synthesized RNA interference sequences were cloned into the pX330 vector at the NheI and BamHI sites using restriction enzymes (Thermo Fisher Scientific Inc.). After enzyme digestion identification, the vectors were named as pX330-sgRNA-SIRT5. MAC-T cells were resuscitated and supplemented with DMEM containing 10% fetal bovine serum at 37˚C in an incubator with 5% CO₂. After the cells had reached 70-80% confluency, pX330 vector or pX330-sgRNA-SIRT5 vector was transfected with Lipofectamine 3000 reagent according to the manufacturer's protocol. After 24 h, puromycin was used to screen for SIRT5^-/- cells. The cells were divided into 2 groups: Empty vector (pX330, transfected with pX330 vector) and SIRT5 knockdown (sgRNA-1, sgRNA-2 and sgRNA-3, transfected with pX330-sgRNA-SIRT5 vectors).

Overexpression and knockdown efficacy was validated by RT-qPCR and western blot. The plasmid kits and transfection reagent Lipofectamine 3000 were purchased from Thermo Fisher Scientific, Inc. RNAiso Plus, T4 DNA ligase, SYBR^® PremixExTaq^™ II (TliRNaseHPlus), LATaq enzyme and DL2000 Marker were from Takara Biotechnology Co., Ltd and were used according to the manufacturer's instructions.

NH₄Cl, LY294002 (LY; PI3K inhibitor) and Rapamycin (RA; mTOR inhibitor) were used to treat the cells and had been obtained from Sigma-Aldrich (Merck KGaA). MC3482 (SIRT5 inhibitor) was from MedChemExpress. The MAC-T cells and SIRT5^-/-, SIRT5^+/+ cell lines were washed twice with PBS and then cultured in medium supplemented with the indicated agents. The MAC-T cells were randomized into the four experimental groups: i) Control (CT); ii) NH₄Cl (NC); iii) MC3482 (MC); and iv) MC3482 + NH₄Cl (MCN). The SIRT5^+/+ cells were randomized into four experimental groups: i) SIRT5^+/+ (SO); ii) SIRT5^+/+ + NH₄Cl (SON); iii) SIRT5^+/+ + NH₄Cl + LY (SNL); and iv) SIRT5^+/+ + NH₄Cl + RA (SNR). The SIRT5^-/- cells were randomized into two experimental groups: i) SIRT5^-/- (SD); and ii) SIRT5^-/- + NH₄Cl (SDN). Each of the above groups contained three independent repeats. In the CT group, the MAC-T cells were cultured with basal medium for 12 h. In the SO and SD groups, the SIRT5^+/+ and SIRT5^-/- cells were cultured with basal medium for 12 h, respectively. In the NC, SON and SDN groups, the MAC-T, SIRT5^+/+ and SIRT5^-/- cells were exposed to 4 mM NH₄Cl for 12 h. In the MC group, MAC-T cells were incubated with 20 µM MC3482 for 12 h. In the MCN group, MAC-T cells were incubated with MC3482 (20 µM) for 30 min and then the cells were treated with 4 mM NH₄Cl for 12 h. In the SNL and SNR groups, SIRT5^+/+ cells were respectively incubated with LY294002 (20 mM) or Rapamycin (10 µM) for 30 min, and then the cells were treated with 4 mM NH₄Cl for 12 h.

Total RNA was extracted from cells with TRIzol reagent (Invitrogen; Thermo Fisher Scientific, Inc.) following the protocol provided by the manufacturer. The integrity of RNA was analyzed by agarose gel electrophoresis. cDNA was synthesized using the RT kit (cat. no. 6110A; Takara Biotechnology Co., Ltd.) following to the manufacturer's instructions. qPCR was performed using SYBR Premix Ex Taq (cat. no. RR820A; Takara Biotechnology Co., Ltd.) following the manufacturer's instructions. PCR conditions were as follows: 95˚C for 5 min, followed by 30 cycles at 95˚C for 30 sec, 58˚C for 20 sec and 72˚C for 20 sec. All experiments were performed thrice. GADPH was used as a housekeeping gene. Gene expression was quantified relative to GAPDH expression amplified in the same sample, and gene expression was analyzed using the 2^-ΔΔCq method (11). Primers used for the real-time qPCR assay are listed in Table I.

Total protein was extracted from cells in each cell group using RIPA lysis buffer (Thermo Fisher Scientific, Inc.). The protein concentration was determined using a BCA kit (cat. no. BCA01; Beijing Dingguo Changsheng Biotechnology Co., Ltd.). Next, 5X SDS loading buffer (Beijing Dingguo Changsheng Biotechnology Co., Ltd.) was added to the extracted proteins, which were then denatured at 99˚C for 10 min. The proteins (24 µg/lane) were separated by on 10% gels using SDS-PAGE and finally transferred to polyvinylidene fluoride (PVDF) membranes (EMD Millipore). The membranes were then blocked with Tris-buffered saline containing Tween-20 (TBST) with 5% skimmed milk for 1 h and incubated with the corresponding antibodies overnight at 4˚C. The PVDF membranes were washed with TBST for 30 min and incubated with horseradish peroxidase (HRP)-conjugated anti-IgG antibodies for 2 h at room temperature. Finally, the visualized protein bands were detected under a developer using enhanced chemiluminescence reagent (EMD Millipore), and ImageJ software 1.4.3.67 (National Institutes of Health) was used to analyze the grayscale values of the protein bands and calculate the relative expression of proteins in each group. Mouse anti-Bcl-2 antibody (cat. no. bsm-33047M; 1:500 dilution), rabbit anti-Bax antibody (cat. no. bs-0127R; 1:500), rabbit anti-β-actin antibody (cat. no. bs-33036M; 1:1,000), rabbit anti-Caspase-3 antibody (cat. no. bs-0081R; 1:500) were purchased from Beijing Bioss Technology Co., Ltd. Rabbit anti-light chain 3β (LC3B) antibody (cat. no. ab48394; 1:1,000 dilution), rabbit-anti-p62 antibody (cat. no. ab101266; 1:1,000) were purchased from Abcam. Rabbit anti-Beclin1 antibody (cat. no. AP0769; 1:500) and rabbit anti-SIRT5 antibody (cat. no. BS91247; 1:500) were purchased from Bioworld Technology, Inc. Goat anti-mouse IgG (H+L) (cat. no. RGAM001; 1:5,000) and goat anti-rabbit IgG (H+L) (cat. no. SA00001-2; 1:5,000) were purchased from Proteintech Group, Inc.

Regarding the analysis of autophagic vesicle formation, TEM is the gold standard for evaluating autophagic activity (11). First, the cells were digested with trypsin and isolated. Next, the cells were fixed with electron microscope fixative and then fixed with 1% ozonated acid for 2 h, at room tempreture. After dehydration, permeabilisation and embedding, ultrathin sections were observed and analyzed in representative areas by TEM (HT7700; Hitachi).

Flow cytometry was performed to analyze the apoptotic ratio. In brief, cells were cultured under the aforementioned conditions. Subsequently, they were double-stained with Annexin V FITC/PI, which had been obtained from Yisheng Biotechnology Co., Ltd. Next, cells were analyzed using a flow cytometer according to the manufacturer's instructions. All experiments were performed at least in triplicate. The apoptosis ratio was detected by flow cytometry (CytoFLEX; Beckman Coulter) and calculated with the help of the CytExpert software (11).

The cells were lysed and centrifuged. Subsequently, the supernatant was collected. The protein concentration was determined using the bicinchoninic acid protein assay kit (Abcam; cat. no. ab102536). The NH₃ content in the cell culture medium was determined using an NH₃ assay kit (Sigma-Aldrich; Merck KGaA; cat. no. AA0100) following the manufacturer's protocol. The content of glutamate, GLS activity and the ADP/ATP ratio in the cells were determined using relevant kits, including glutamate measurement kit (Nanjing Jiancheng Bioengineering Institute; cat. no. A074-1-1), GLS test kit (Nanjing Jiancheng Bioengineering Institute; cat. no. A124-1-1) and ADP/ATP ratio assay kit (Sigma-Aldrich; Merck KGaA; cat. no. MAK135). The above kits were used according to the manufacturers' specifications. All assays were performed in triplicates.

All experiments were performed at least in triplicates and quantitative data were presented as the mean ± standard deviation. GraphPad Prism software (version 6.01; GraphPad; Dotmatics) was used for statistical analysis. Differences among multiple groups were calculated using one-way ANOVA followed by post-hoc Bonferroni's correction. P<0.05 was considered to indicate a statistically significant difference.

The fluorescence intensity in MAC-T cells was observed under a fluorescence microscope to confirm successful transfection. As indicated in Fig. S1A, no fluorescence was present in non-transfected MAC-T cells. However, fluorescence was observed in cells transfected with the pEGFP-N1 vector and the pEGFP-SIRT5 vector (Fig. S1B and C). Furthermore, in Fig. 1, the expression of SIRT5 mRNA and protein in SIRT5^+/+ cells was increased compared with that in MAC-T cells (Fig. 1A and C), while the expression of SIRT5 mRNA and protein in SIRT5^-/- cells was significantly decreased (Fig. 1B and D). Specifically, the mRNA and protein expression levels of SIRT5 in the pX330-sgRNA-SIRT5-transfected cells were significantly reduced, indicating that pX330-sgRNA-SIRT5 vectors efficiently knocked down SIRT5 in MAC-T cells (Fig. 1B and D). Furthermore, the knockdown efficiency of pX330-sgRNA-2-SIRT5 vector was high and the expression of SIRT5 decreased obviously (P<0.05). Therefore, in the subsequent experiments, the pX330-sgRNA-2-SIRT5 vector was used to transfect MAC-T cells to obtain SIRT5^-/- cells. In addition, MC3482 was adopted as a specific inhibitor of SIRT5 to treat MAC-T cells. As indicated in Fig. 2, MC3482 at 20 mM inhibited the expression of SIRT5 (Fig. 2D, H and K).

There are numerous similarities between autophagy and apoptosis, e.g. both of them can cause cell death. Furthermore, a variety of autophagy-related proteins are involved in cell apoptosis (33). Substantial evidence demonstrated that the SIRT family participates in cell autophagy and apoptosis (34-36). Accumulating studies illustrated that SIRT5 regulates autophagy and apoptosis in cancer cells (20,21). Autophagy and mitophagy increased in SIRT5-silenced cells and similar results were also observed in MDA-MB-231 and C2C12 cells treated with MC3482. Of note, autophagy and mitophagy decreased in SIRT5-overexpressing cells (25). Furthermore, SIRT5 promoted autophagy and maintained the balance of autophagy and apoptosis (21). However, the effects of SIRT5 on autophagy and apoptosis in bovine mammary epithelial cells have remained elusive. Thus, in the present study, the effect of SIRT5 on MAC-T-cell autophagy and apoptosis was examined.

TEM observation was adopted to display the autophagy process initially. As shown in Fig. 3, an accumulation of autophagosomes and autolysosomes was present in SIRT5-silenced cells (SD group) and in MC3482-treated MAC-T cells (MC group), compared with cells in the CT and SO groups (SIRT5^+/+ cells). Next, the levels of the autophagic markers LC3 and Beclin1 and the autophagy flux marker SQSTM1/p62 were also determined. The levels of LC3B and Beclin1 increased and those of SQSTM1/p62 decreased in SIRT5^-/- cells (SD group) as compared with the CT group (Fig. 2). Again, a similar increase in LC3 and Beclin1 and decrease in SQSTM1/p62 were present in MAC-T cells treated with MC3482 (Fig. 2). However, SIRT5 overexpression decreased the levels of LC3 and Beclin1, while increasing the levels of SQSTM1/p62 in the SO group (Fig. 2). All of these results preliminarily proved that SIRT5 inhibits autophagy in MAC-T cells.

In sequence, the apoptotic role of SIRT5 in MAC-T cells was evaluated. The flow cytometry results revealed that the number of cells with positive fluorescence staining increased in SIRT5^-/- cells and MAC-T cells treated with MC3482, indicating a higher rate of apoptosis than that in the CT and SO groups (Fig. 4). At the same time, the expression levels of the proapoptotic regulatory factor Bax, the apoptosis executive factor Caspase-3, as well as the anti-apoptotic factor Bcl-2 were also determined. Compared with cells in the CT group and SO group, the mRNA and protein expression levels of Bax and caspase-3 in SIRT5^-/- cells and MAC-T cells treated with MC3482 increased significantly, while the mRNA and protein levels of Bcl-2 decreased significantly (Fig. 5). However, compared with those in the CT group, the mRNA and protein expression levels of Bax and caspase-3 in the SIRT5^+/+ cells was significantly decreased, and the mRNA and protein levels of Bcl-2 was significantly increased (Fig. 5). Hence, the aforementioned results suggested that SIRT5 inhibited the apoptotic occurrence in MAC-T cells.

Previous studies have found that the addition of exogenous NH₄Cl-induced MAC-T-cell apoptosis and autophagy (11). NH₃ in cells was mainly from glutamine catabolism. Research pointed out that SIRT5 regulates glutamine metabolism (25). However, SIRT5 was ubiquitously expressed (37). Therefore, in the present study, it was examined whether SIRT5 also participates in regulating NH₃ levels in MAC-T cells, and ultimately affects apoptosis and autophagy of MAC-T cells induced by NH₃. It was found that SIRT5 overexpression significantly reduced the NH₃ concentration in the culture medium (P<0.01; Fig. 6A). On the contrary, SIRT5 knockdown obviously promoted NH₃ accumulation compared to MAC-T cells (Fig. 6A). Furthermore, an elevation in NH₃ levels was obtained when MAC-T cells were treated with MC3482, an inhibitor of SIRT5 (Fig. 6A). NH₃ levels were further examined in SIRT5^+/+, SIRT5^-/- or MAC-T cells treated with MC3482, and combined with NH₄Cl treatment as the donor of NH₃. In Fig. 7, a decline in the content of NH₃ appeared in SIRT5^+/+ cells treated with NH₄Cl, and a significant increase was present in SIRT5^-/- cells treated with NH₄Cl and MAC-T cells co-treated with MC3482 and NH₄Cl (Fig. 7A). These results indicated that SIRT5 may reduce the generation of NH₃ in MAC-T cells.

Glutamine is a non-essential amino acid, which acts as a precursor for glutamate and NH₃, and has a critical role in various tissues (38,39). Glutamine is transformed into glutamate and NH₃ by the enzyme GLS in the mitochondria (40,41). Of note, a previous study reported that SIRT5 can desuccinylate GLS to regulate NH₃ production in MDA-MB-231 and C2C12 cell lines, and further found that SIRT5 co-immunoprecipitates with GLS (25). The enzyme GLS has a mitochondrial localization and its functionally relevant structural domain was oriented towards the mitochondrial matrix, while SIRT5 was also located in the mitochondrial matrix (40). Hence, in the present study, the activity of GLS in MAC-T cells was measured to investigate whether SIRT5 regulates the activity of GLS to control NH₃ production. To verify this point, MAC-T cells were either treated or untreated with MC3482, and the activity of GLS was then detected. Similarly, an increase in GLS levels in MC3482-treated MAC-T cells was observed (Fig. 6B). To demonstrate that SIRT5 may regulate GLS activity, the glutamate content in MAC-T cells was detected (Fig. 6B). Fig. 6C shows that glutamate levels were declined in SIRT5^+/+ cells and enhanced in SIRT5^-/- and MC3482-treated MAC-T cells. The above results indicated that SIRT5 inhibited the metabolism of glutamine in MAC-T cells, further reducing NH₃ release in cells. It was found that a decrease in the content of NH₃ and glutamate, as well as GLS activity, were present in SIRT5^+/+ treated with NH₄Cl, while a significant increase was observed in SIRT5^-/- treated with NH₄Cl (Fig. 7). Furthermore, the content of glutamate and GLS activity also increased significantly in MAC-T cells co-treated with MC3482 and NH₄Cl (Fig. 7). This indicated that SIRT5 inhibited the metabolism of glutamine in MAC-T cells. In addition, the ratio of ADP/ATP was similar in SIRT5^-/- and SIRT5 inhibitor-treated cells, and alleviation of ATP levels was more obvious in SIRT5^-/- and SIRT5 inhibitor-treated cells (Fig. 6D). By contrast, ATP levels in SIRT5^+/+ were enhanced (Fig. 6D). This indicated that the changes of cell energy may be involved in the NH₃ production regulated by SIRT5.

Previous research by our group indicated that NH₃ regulates the apoptotic and autophagic activity in bovine mammary epithelial cells via PI3K/Akt/mTOR signaling (11). As confirmed by the results of the present study, SIRT5 regulates the activity of GLS, and NH₃ originated from glutaminolysis enhanced apoptosis and autophagy (11,42). It was further explored whether NH₃ production in SIRT5^+/+, SIRT5^-/- or MC3482-treated MAC-T cells, and combined with NH₄Cl treatment as the donor of NH₃, is able to induce apoptosis and autophagy. NH₃-stimulated autophagy is characteristic of MC3482-treated or SIRT5^-/- cells with NH₄Cl co-treatment, since LC3 and Beclin1 expression were significantly elevated, while in SIRT5^+/+ cells, the opposite result was obtained, as LC3 and Beclin1 levels were significantly decreased with NH₄Cl treatment in Fig. 8 (A, B, D, E and M). The expression of SQSTM1/p62 as a marker of autophagosomal cargo lysosomal degradation was also assessed. Degradation of SQSTM1/p62 was hindered in SIRT5-overexpressing cells, whereas it was not impaired in MC3482-treated or SIRT5^-/- cells with NH₄Cl co-treatment (Fig. 8C, F and M).

In MAC-T cells, the number of cells with positive fluorescence staining in the SIRT5^+/+ and NH₄Cl co-treatment group (SON group) was reduced, which significantly reduced the number of cells with positive fluorescence staining in the MC3482 and NH₄Cl co-treatment group (MCN group), while the SIRT5^-/- and NH₄Cl co-treatment group (SDN group) exhibited a significant enhancement of the apoptosis rate compared with the NH₄Cl group (Fig. 9). Subsequently, apoptotic marker expression was also determined. Compared with the NH₄Cl group, in the cells of the SIRT5^+/+ and NH₄Cl co-treatment group (SON group), the mRNA and protein expression of Bax and Caspase-3 decreased obviously (Fig. 8G, H, J, K and M), while the levels of Bcl-2 mRNA and protein increased significantly in the SON group (Fig. 8I, L and M). Of note, the protein levels of Bax and Caspase-3 in the MC3482 and NH₄Cl co-treatment group (MCN group) and SIRT5^-/- and NH₄Cl co-treatment group (SDN group) were significantly increased, while Bcl-2 expression was significantly decreased (Fig. 8G-M). The expression of apoptosis-related factors also confirmed that SIRT5 inhibited the occurrence of MAC-T-cell apoptosis induced by NH₄Cl.

PI3K/Akt/mTOR signaling participates in controlling cell proliferation, growth, differentiation, apoptosis, autophagy and metabolism (26,27). SIRT5 reportedly inhibited cancer cell growth, invasion and migration, as well as modulated inherent and acquired immunity via the PI3K/Akt pathway (28). The mTOR signal plays a critical role in regulating autophagy (29). The mTOR signal promotes glutamine uptake through GLS and accelerates the release of NH₃ (30,31). A study demonstrated that NH₃ relies on PI3K signaling to regulate mTOR activity, further affecting apoptosis and autophagy in mammary epithelial cells (11). In the present study, LY294002 (PI3K inhibitor) or Rapamycin (mTOR inhibitor) were used to treat SIRT5^+/+ cells, and the apoptotic and autophagic markers were detected to clarify the mechanism of SIRT5 regulating apoptosis and autophagy induced by NH₃ in MAC-T cells. As indicated in Fig. 9, the number of cells with positive fluorescence staining in the SIRT5^+/+ and NH₄Cl co-treatment group (SON group) was obviously lower than that in the NH₄Cl group (NC group), showing that SIRT5 expression significantly decreased the rate of apoptosis. Of note, the addition of LY294002 or Rapamycin further decreased cell apoptosis, and the number of cells with positive fluorescence staining of apoptotic cells markedly declined and the apoptosis rate was notably reduced (Fig. 10). At the same time, LY294002 or Rapamycin treatment alleviated the increase in the expression of Bax and Caspase-3 caused by the co-treatment of SIRT5^+/+ and NH₄Cl, and enhanced the level of Bcl-2 mRNA and protein (Fig. 11). Similarly, in comparison with the SIRT5^+/+ and NH₄Cl co-treatment group, LY294002 or Rapamycin treatment led to a significant decline in the mRNA and protein expression of autophagy factor Beclin1, and p62 expression was obviously enhanced (Fig. 11). Unexpectedly, blocking the PI3K signal led to a significant increase in LC3 mRNA and protein expression, while blocking the mTOR signal led to the opposite effect on LC3 mRNA and protein expression (Fig. 11). These results indicated that PI3K/Akt/mTOR signaling has a certain role in SIRT5 regulating autophagy and apoptosis stimulated by NH₃.