SIRT5-mediated GLS and GDH desuccinylation attenuates autophagy in MAC-T cells induced by ammonia

24 Our previous research revealed that NH3 regulated autophagy dependent on 25 SIRT5 in MAC-T cells. Interestingly, SIRT5 reduced the content of NH3 and 26 glutamate by inhibiting GLS activity, ADP/ATP value also declined. In this study, 27 SIRT5 interacted with endogenous GLS and GDH, and had no effect on endogenous 28 GLS and GDH expression. SIRT5 declined significantly the succinylation levels of 29 GLS and GDH, and further reduced the enzymatic activity of GLS and GDH. SIRT5 30 declined the glutamine metabolism, which attenuated ammonia release in MAC-T 31 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted August 26, 2024. ; https://doi.org/10.1101/2024.08.26.609685doi: bioRxiv preprint cells, accompanying with cellular autophagy decline, reducing the formation of 32 autophagosome. Deletion of SIRT5 increased the content of NH3 and glutamate, as 33 well as promotes autophagy, which could be alleviated by SIRT5 overexpression. 34 SIRT5 KO was associated with increased succinylation and activity of GLS and GDH, 35 as well as autophagy response in MAC-T cells. Furthermore, SIRT5 promoted the 36 maintenance of mitochondria homeostasis. Mechanistically, SIRT5 modulated the 37 succinylation levels and enzymatic activities of GLS and GDH in mitochondria and 38 promoted the maintenance of mitochondria homeostasis, further attenuating 39 ammonia-stimulated autophagy in MAC-T cells. 40

Sirtuin5; desuccinylation; GLS; GDH; autophagy; ammonia. 41

42 Ammonia (NH 3), as one of the harmful gases in livestock houses, affect ed 43 the health and growth of livestock, and reduce d their production performance [1]. In 44 addition, NH3 could change pH, electrolyte balance and metabolic changes, inducing 45 various transform in many cell types [ 2]. In vitro , a mmonia ha d a dual role in 46 autophagy, it played an inducer at low er concentrations and an inhibitor at high 47 concentrations [3]. Ammonia at high concentrations (>20 mM) enhanced pH in 48 lysosome and increased water influx, resulting in inhibition of substrate degradation 49 and lysosome swelling, and further destroying lysosomal function [4, 5]. In contrast, 50 ammonia strongly promoted autophagy at lower concentrations (0.2~10 mM) 51 [6-12]. Ammonia derived from glutamine catabolism also strongly induce d autophagy 52 in c ancer cells [13]. Apoptosis and autophagy exert key role in maintaining the 53 quantity and ability of mammary epithelial cells, and are key factors affecting the 54 lactation performance of dairy cows. Our previous research found that NH 3 regulated 55 the autophagy in cow mammary epithelial cells through PI3K/Akt/mTOR signaling 56 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted August 26, 2024. ; https://doi.org/10.1101/2024.08.26.609685doi: bioRxiv preprint pathway [14]. 57 Sirtuin 5 (SIRT5) was an important regulator y factor in maintaining cellular 58 homeostasis, which emerged to be a n important desuccinylase enzyme , 59 desuccinylating more than half of proteins in mitochondria and controlling 60 ammonia-induced autophagy in tumor cells [15-23]. In addition, we found that SIRT5 61 inhibited the autophagy in bovine mammary epithelial cells through PI3K/Akt/mTOR 62 signal, involving in NH 3-induced autophagy dependent on SIRT5 [24]. Moreover, 63 there were many studies explaining the dual role of SIRT5 for autophagy. SIRT5 64 promoted autophagy in colorectal cancer, where its overexpression was associated 65 with low survival [20]. SIRT5 promoted autophagy in gastric cancer cells through the 66 AMPK/mTOR pathway, while SIRT5 expression was suppressed in gastric cancer 67 tissues [ 25]. In breast cancer and mouse myogenic cells, silencing SIRT5 increased 68 ammonia-induced autophagy by controlling gl utamine metabolism and mitophagy 69 [12]. The glutaminolysis in mitochondria was catalyzed by glutaminase (GLS) with 70 deamination of glutamine , by which produced ammonia and glutamate . Glutamate 71 was conver ted to α -ketoglutarate through oxidative deamination by glutamate 72 dehydrogenase (GDH) in mitochondr ia, releasing NH 4 + [26, 27]. SIRT5 ha d been 73 shown to exhibit weaker deacetylase activity, more commonly demalonylase, 74 desuccinylase, and deglutarylase activities, participating in several metabolic 75 processes of mitochondria [18, 28-31]. SIRT5 downregulated the succinylation level 76 of GLS and inhibit ed ammonia-induced autophagy in tumor cells [ 12]. Our previous 77 studies also found that SIRT5 reduced NH 3-induced autophagy in bovine mammary 78 epithelial cells, but little is known about the specific molecular mechanism. 79 Interestingly, our recent study found that SIRT5 reduced the content of NH 3 80 and glutamate in bovine mammary epithelial cells by inhibiting G LS activity, and 81 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted August 26, 2024. ; https://doi.org/10.1101/2024.08.26.609685doi: bioRxiv preprint declined the ratio of ADP/ATP [24]. Therefore, we hypothesized that SIRT5-mediated 82 GLS and GDH desuccinylation controlled ammonia production, further regulate d the 83 autophagy of mammary epithelial cells. To verify this point, we constructed SIRT5 84 overexpression or knockout cell lines. A dditionally, ammonium chloride, and some 85 inhibitors for SIRT5, GLS, and autophagy were also used to treat cells . The 86 interaction of SIRT5 with GLS and GDH were identified using immunoprecipitation 87 techniques, and the succinylation level and enzyme activity of GLS and GDH were 88 also determined. The results verified that SIRT5 interacted with endogenous GLS and 89 GDH, and had no effect on the expression of endogenous GLS and GDH. Expectedly, 90 we discovered that SIRT5 obviously decreased the succinylation levels of GLS and 91 GDH, and further reduced the enzymatic activity of GLS and GDH. Next, the content 92 of ammonia and glutamate, as well as the related autophagy markers were measured, 93 the results demonstrated that SIRT5 declined the glutamine metabolism and ammonia 94 release in MAC-T cells , accompan ying with cellular autophagy decline and less 95 autophagosome formation. Interestingly, SIRT5 enhanced the content of ATP and 96 promoted the maintenance of mitochondria homeostasis. Altogether, SIRT5 reduced 97 ammonia release by modulating the succinylation levels and enzymatic activities of 98 GLS and GDH in mitochondria , and further promoted the maintenance of 99 mitochondria homeostasis, accompanying with attenuating ammonia-stimulated 100 autophagy in cow mammary epithelial cells . In this research, we revealed that GLS 101 and GDH were two physiological substrates of SIRT5, which relied on 102 desuccinylation to reduce ammonia production by inhibiting the enzymatic activity of 103 GLS and GDH. 104

105 Generation of SIRT5 OE and KO bovine mammary epithelial cell lines 106 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted August 26, 2024. ; https://doi.org/10.1101/2024.08.26.609685doi: bioRxiv preprint Bovine mammary epithelial cells (MAC-T cells, preserved in our laboratory) 107 were cultured according to our previous method [14, 24]. To generate SIRT5 108 overexpression cells (SIRT5 OE), bovine SIRT5 gene (NM_001034295) was 109 subcloned into a pPB-EF1α eukaryotic expression vector (Haixing Biotechnology Co., 110 Ltd). The vectors with SIRT5 (pPB-EF1α-SIRT5 vector) or without SIRT5 111 (pPB-EF1α vector), which contained non-fused EGFP fluorescent tags and resisted to 112 puromycin, were electrotransfected into MAC-T cells using the Neon transfection 113 system (Neon™ transfector, Cat# MPK5000) according to the product manual 114 (Thermo Fisher Scientific Inc.). The empty vector pPB-EF1α without the insert 115 (SIRT5 cDNA) was used as the control. At 48 h post-transfection, the generation of 116 SIRT5 OE cell line was achieved by selecting with 5.0 μg/mL puromycin. 117 CRISPR-Cas9 mediated the excision of SIRT5 gene was obtained with 118 CRISPR-Cas9 RNP (supplied with Haixing Biotechnology Co., Ltd) including 119 expression element for hSpCas9 and chimeric lead RNA. To guide exon 4~exon 9 of 120 SIRT5 gene, two target RNA sequence of AGCGTGCTTTCCCGAGACAGCGG and 121 GCGGGTGACGGAGTTGTGTGTGG were selected on 122 the http://crispr.mit.edu website. Vector including the target RNA sequence was 123 electrotransfected into cells using the Neon transfection system according to the 124 product manual (Thermo Fisher Scientific Inc.). After 48 h, single colonies were 125 moved to 96-well culture plates. To measure the exist of indels in SIRT5 guided 126 colony, a Quick-DNA Miniprep kit (Zymo Research, USA, CA) was used to segregate 127 genomic DNA, and 2×Taq Master Mix (Dye Plus, Vazyme, P112) with exon flanking 128 primers was obtained for PCR amplification. Forward: 5’- AGTGGGACGGAG 129 CATTTGTT-3’; Reverse: 5’-CTGACTTAGGTAATGACAAGATGCT-3’. Vectors 130 were collected from 8-10 single clones to be sequenced by Sanger sequencing 131 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted August 26, 2024. ; https://doi.org/10.1101/2024.08.26.609685doi: bioRxiv preprint (GENEWIZ, China). Colonies with mutations in both alleles were chosen for 132 subsequent tests. SIRT5 knockout (KO) cell line was obtained according to the above 133 steps, and cultivated under the same conditions as the parental cells. Then, we used 134 PCR and western blotting method to verify SIRT5 expression in SIRT5 OE and KO 135 MAC-T cell lines. RT-PCR method was conducted according to our previous 136 literatures [14, 24]. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was 137 chosen as internal reference genes. The primers of SIRT5 and GAPDH as following. 138 GAPDH (NM_001034034.2): (upstream, 5’-3’) CCATGTTTGTGATGGGCGTG, 139 (downstream, 5’-3’) GCAGGGATGATATTCTGGGCA; SIRT5 (NM_001034295.2): 140 (upstream, 5’-3’) TTGTGGAGTTGTGGCTGAGA, (downstream, 5’-3’) 141 GTCCCCACCACTAGACACAG. The details of western blotting method were 142 described later. 143 Cell treatment 144 Ammonium chloride (NH4Cl, Sigma-Aldrich, Shanghai, A43A4899) as a 145 donor of NH3 was diluted in deionized water and adjusted to an ultimate 146 concentration of 10 mM. MC3482 (SIRT5 inhibitor) from 147 MedChemExpress (Monmouth Junction, NJ, USA) was diluted in deionized DMSO 148 and adjusted to an ultimate concentration of 20 μM. BPTES (GLS inhibitor) and 149 BafA1 (bafilomycin A1, autophagy inhibitor) were purchased from MedChemExpress 150 LLC (Shanghai). The MAC-T cell and SIRT5 KO, SIRT5 OE cell lines were cleaned 151 twice using phosphate-buffered saline and then cultivated in medium containing the 152 indicated agents. The MAC-T cells were randomized into the seven experimental 153 groups: (1) Control (CT), (2) NH4Cl (NC), (3) MC3482 (MC), (4) MC3482+NH4Cl 154 (MCN), (5) BPTES (BP), (6) MC3482+BPTES (MCBP), (7) BafA1 (BA). The SIRT5 155 OE cell lines were randomized into the three experimental groups: (1) SIRT5 OE (SO), 156 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted August 26, 2024. ; https://doi.org/10.1101/2024.08.26.609685doi: bioRxiv preprint (2) SIRT5 OE+NH4Cl (SON), and (3) SIRT5 OE+BPTES (SOBP). The SIRT5 KO 157 cell lines were randomized into the four experimental groups: (1) SIRT5 KO (SK), (2) 158 SIRT5 KO+NH4Cl (SKN), (3) SIRT5 KO+BPTES (SKBP), and (4) SIRT5 159 KO+NH4Cl+BafA1 (SKNBA). Each of the above groups contained three independent 160 repeats. In CT, SO, and SK groups, the MAC-T, SIRT5 OE, and SIRT5 KO cells were 161 cultivated for 12 h in basal medium, respectively. In the NC, SON, SKN groups, the 162 MAC-T, SIRT5 OE, and SIRT5 KO cells were treated with 10 mM NH4Cl for 12 h, 163 respectively. In the MC, BP, and BA group, MAC-T cells were incubated with 20 μM 164 MC3482, or BPTES (0.12 μM), or BafA1 (50 nM) for 12 h, respectively. In the MCN 165 and MCBP group, MAC-T cells were incubated with MC3482 (20 μM) for 1 h, and 166 then the cells were exposed to 10 mM NH4Cl or BPTES (0.12 μM) for 12 h. In SOBP, 167 SKBP groups, SIRT5 OE cells and SIRT5 KO cells were cultivated in medium 168 containing BPTES (0.12 μM) for 1 h, and then the cells were exposed to 10 mM 169 NH4Cl for 12 h. In SKNBA group, SIRT5 KO cells were treated with BafA1 (50 nM) 170 for 1 h, and then the cells were exposed to 10 mM NH4Cl for 12 h. 171 Immunoprecipitation assay 172 For detection of protein-protein interactions, cells were lysed with 500 μL of 173 specific immunoprecipitation lysis solution for 30 min with orbital vibrating on ice. 174 Then, lysates were selected and moved into cold microfuge tubes, at the centrifugal 175 force of 12, 000 g for 10 min at 4℃. For co-immunoprecipitation (Co-IP) of immune 176 complexes, cell lysate was first pre-cleared by inoculation with 50 μL of protein A/G 177 magnetic beads (Thermo Fisher Scientific Inc.) and rabbit anti-IgG (to decline most of 178 unspecific protein interactions) for 30 min at 4˚C with lightly vertical mixing (10 179 RPM). The beads were gathered with a magnet, and the lysates were moved into a 180 new, cold microfuge tube. 181 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted August 26, 2024. ; https://doi.org/10.1101/2024.08.26.609685doi: bioRxiv preprint Next, 5 μL GLS, SIRT5 or GDH antibodies were replenished into the tube, 182 respectively. The whole volume was increased to 500 μL. The samples were 183 inoculated at 4˚C for overnight with lightly vertical mixing (10 RPM) to form 184 immune complex. The magnetic beads of p rotein A/G were wetted and washed triple 185 using PBST, and then the incubated lysates were bound to the magnetic beads and 186 held in a rotator at 25˚C for 6 h. After binding, the lysates were wetted and washed 187 three times at 99°C for 10 min with PBST and 70 μL 1×SDS loading buffer. The 188 beads were collected, and the lysates were moved into a new, cold microfuge tube. 189 Samples were then electrophoresed on a 10 % SDS polyacrylamid e gel and further 190 immunoblotted with designated antibodies . Here, IgG was selected as a negative 191 control. ImageJ software was adopt ed to analyze grayscale values, calculate relative 192 protein expression levels, and perform statistical analysis. 193 Mouse anti-SIRT5 monoclonal antibody (Cat No:67257-1-Ig; 1:1000), rabbit 194 anti-GLS polyclonal antibody (Cat No: 12855-1-AP; 1:2000) , rabbit anti -β-actin 195 recombinant antibody (Cat No: 81115-1-RR; 1:5000), rabbit IgG control polyclonal 196 antibody (Cat No:30000 -0-AP; 1:2000) and goat anti -rabbit IgG (H+L) (Cat No: 197 SA00001‑2; 1:5,000) were purchased from Proteintech Group, Inc. Rabbit anti-SIRT5 198 monoclonal antibody ( Cat No: ab259967; 1:1000) and rabbit monoclonal anti -GDH 199 antibody (Cat No: ab166618; 1:2000) were purchased from Abcam (Shanghai) trading 200 Co., Ltd. Rabbit anti-succinylated lysine polyclonal antibody (Cat No: 3089; 1:2000) 201 was brought from Wuhan DIA AN biotechnology Co., Ltd. 202 Biochemical assays 203 The cells were lysed, and centrifuged. Then, the supernatant was collected. 204 The BCA protein assay kit (Abcam, Shanghai , ab102536 ), ammonia assay kit 205 (Sigma-Aldrich, Shanghai, AA0100 ), glutamate measurement kit (Nanjing Jiancheng 206 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted August 26, 2024. ; https://doi.org/10.1101/2024.08.26.609685doi: bioRxiv preprint Bioengineering Institute, China, A074-1-1), ATP assay kit (Beyotime Biotechnology, 207 China, S0026), glutaminase test kit (Nanjing Jiancheng Bioengineering Institute, 208 China, A124 -1-1), and glutamate dehydrogenase test kit (Nanjing Jiancheng 209 Bioengineering Institute, China , A125 -1-1) were used to determine the protein 210 concentration, the content of ammonia, glutamate and ATP, as well as GLS and GDH 211 enzymatic activity. The above kits were operated according to the manufacturer's 212 specifications. All assays were executed in triplicates. 213 Western blotting 214 Western blotting was used to analyze the expression of SIRT5 and 215 designated autophagy proteins. T he above treated cells were cleared twice with cold 216 PBS, the BCA method was applied to measure the concentration of protein. Next, 217 5×SDS loading buffer was increased to the extracted proteins, which were then 218 denatured at 99˚C for 10 min. The proteins (24 μg/lane) were separated with 10 % 219 SDS-PAGE gels, and then moved onto polyvinylidene fluoride (PVDF) membranes 220 (EMD Millipore) that were sealed with 5 % skim milk for 1 h. The membranes were 221 sequentially incubated with the specific first antibodies at 4˚C for overnight. The 222 membranes were cleared with TBST for 30 min and in oculated with goat anti ‑rabbit 223 IgG (H+L) antibody (Cat No: SA00001 ‑2; 1:5,000; purchased from Proteintech 224 Group, Inc.) at room temperature for 2 h. Then, the membranes were washed with 225 TBST for 30 min again. Finally, the visualized protein bands were detected under a 226 developer using enhanced chemiluminescence reagent (EMD Millipore), and Imag eJ 227 software 1.4.3.67 (National Institutes of Health) was adopted to evaluate the grayscale 228 values of the protein bands , and calculate the relative expression of proteins in each 229 group. 230 The corresponding primary antibodies used as following: rabbit anti-SIRT5 231 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted August 26, 2024. ; https://doi.org/10.1101/2024.08.26.609685doi: bioRxiv preprint monoclonal antibody (Cat No: ab259967; 1:1000) was purchased from Abcam 232 (Shanghai) trading Co., Ltd. Rabbit anti-Beclin1 polyclonal antibody (Cat No: 233 11306-1-AP; 1:2000), rabbit anti-LC3B polyclonal antibody (Cat No: 18725-1-AP; 234 1:1000), and rabbit anti-p62/SQSTM1 polyclonal antibody (Cat No: 18420-1-AP; 235 1:5000) were bought from Proteintech Group, Inc. 236 Immunofluorescence staining 237 To identify the formation of autophagosomes through LC3B puncta, 238 immunofluorescence staining was performed on the cells. Briefly, MAC-T cells were 239 dealt with or without NH 4Cl, and SIRT5 KO cells were treated with NH 4Cl or NH4Cl 240 and BafA1 together. Cells achieving a cell density of 50 % were cultured in a 24-well 241 plate. 4 % paraformaldehyde was used to fix the cells, at room temperature for 30 min 242 and rinsed with PBS. Then, 0.2 % Triton X -100 was added for 10 min at room 243 temperature, and the cells were rinsed with PBS for 30 min. Next, 10 % fetal bovine 244 serum (FBS) was added to block for 1 h at room temperature , after which the 245 anti-LC3B antibody was added to in oculation at 4°C for overnight. The secondary 246 fluorescent antibody was in oculated at room temperature with protection from light 247 for 1 h, and the cells were cleared in PBS. 4, 6-diamidino-2-phenylindole (DAPI) was 248 used to performe nuclear staining . Finally, anti -fluorescence quenching sealer was 249 added and the film was sealed and stored, and the stained LC3B positive cells was 250 observed with a laser confocal fluorescence microscope after overnight stabilization. 251 The effect of SIRT5 on the formation of autophagosomes was evaluated with t he 252 number of autophagosomes , which was calculated based on the number of green 253 fluorescent aggregated spots from three views in each sample . Rabbit polyclonal to 254 LC3B-autophagosome marker (Cat No: ab48394; 1:500; Abcam (Shanghai) trading 255 Co., Ltd ) goat anti -rabbit IgG (H+L) cross -adsorbed secondary antibody, Alexa 256 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted August 26, 2024. ; https://doi.org/10.1101/2024.08.26.609685doi: bioRxiv preprint Fluor™ 488 (Cat No: A-11008; 1:500; Cell signaling technology, Inc.) were used. 257 Mitochondrial staining 258 Mito-Tracker Red staining is a method for labeling mitochondria, which 259 binds to deoxyribonucleotides inside the mitochondria to stain the mitochondria of 260 living cells. Mito-Tracker Red CMXRos (Cat No: C1049B) was used in this study, 261 and bought from Shanghai Beyotime Biotechnology, Inc. Mito Tracker Red CMXRos 262 is a kind of cell permeable agent derivating from X -rosamine 263 (Chloroethyl-X-rosamine, abbreviated as CMXRos) that can specifically label 264 biologically active mitochondria in cells. As the aggregation of Mito Tracker Red 265 CMXRos in mitochondria depends on the membrane potential of mitochondria, it can 266 only stain live cells and detect mitochondrial membrane potential. Using 267 Mito-Tracker Red-stained MAC -T cells (CT), SIRT5 OE, and SIRT5 KO cells, 268 mitochondria was observed under confocal laser microscopy, and further analyzed the 269 effects of SIRT5 on the morphology and quantity of mitochondria, and mitochondrial 270 activity. 271 Statistical analysis 272 The quantitative data were displayed as the mean±standard deviation. 273 GraphPad Prism software (version 6.01, GraphPad, Dotmatics) was selected for 274 statistical analysis. Differences among groups were calculated using one-way ANOV A 275 followed by post -hoc sidak's correction. P<0.05 was indicate d as a statistically 276 significant difference. 277

278 Generation of SIRT5 OE and KO MAC-T cell lines 279 The fluorescence intensity in cells was evaluated under a fluorescence 280 microscope to evaluate the successful transfection. As seen in Figure 1A, no 281 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted August 26, 2024. ; https://doi.org/10.1101/2024.08.26.609685doi: bioRxiv preprint fluorescence existed in non-transfected MAC-T cells. A large amount of green 282 fluorescence was observed in cells transfected with the pPB-EF1α vector and the 283 pPB-EF1α-SIRT5 vector (Figure 1B, C). Meanwhile, the PCR and western blotting 284

verified that SIRT5 expression at the mRNA and protein level in SIRT5 KO 285 cells was almost undetectable (Fig. 2A, B), while SIRT5 expression was promoted at 286 the mRNA and protein level in SIRT5 OE cells compared to MAC-T cells, and the 287 difference significant was obvious (Fig. 2C, D). Combined with the observation of 288 fluorescence microscopy, the results showed that the SIRT5 OE and SIRT5 KO cell 289 lines were successfully generated. 290 SIRT5 catalyzed lysine desuccinylation of GLS and GDH 291 SIRT5 in mitochondria ha d a strong lysine desuccinylation capacity , 292 involving in several processes of mitochondrial metabolism [ 18, 28-31]. Interestingly, 293 our previous study found that SIRT5 reduced the content of NH 3 and glutamate in 294 bovine mammary epithelial cells by inhibiting GLS activity [ 24]. As list in Figure 3, 295 SIRT5 interacted with endogenous GLS or GDH. Strangely, SIRT5 had no effect on 296 the expression of endogenous GLS or GDH (Figure 4). Compared with MAC-T cells, 297 the protein succinylation levels were down -regulated with statistical difference in 298 SIRT5 OE cells and significant up -regulated in SIRT5 KO cells (Figure 5) . Th ese 299 demonstrated that SIRT5 declined the protein succinylation level of cells. 300 Interestingly, w e also found that the GLS or GDH succinylation level was 301 significantly down-regulated in SIRT5 OE cells , whereas the GLS or GDH 302 succinylation level w as obviously up-regulated in SIRT5 KO cells (Figure 6). Taken 303 together, SIRT5 interacted with endogenous GLS or GDH, and further catalyzed 304 lysine desuccinylation of GLS and GDH. 305 GLS and GDH enzymatic activity were enhanced by lysine succinylation 306 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted August 26, 2024. ; https://doi.org/10.1101/2024.08.26.609685doi: bioRxiv preprint To determine the functional consequence of lysine desuccinylation o f GLS 307 and GDH catalyzed by SIRT5, we detected the GLS and GDH enzymatic activity. In 308 SIRT5 OE cells, GLS and GDH enzymatic activity w ere lower than those in MAC -T 309 cells (Figure 7A, 7B). Similar changes existed in the content of NH 3 and glutamate 310 (Figure 7C, 7D ). Correspondingly, with NH 4Cl treatment, SIRT5 also reduced GLS 311 and GDH enzymatic activity, resulting in the induction of NH 3 and glutamate content, 312 alleviating glutamine metabolism (Fig ure 8). In contrast , SIRT5 KO cells led to a 313 profound increase in GLS and GDH enzymatic activity (Figure 7), which w ere 314 associated with increased succinylation level of GLS and GDH (Figure 6). Meanwhile, 315 compared with the CT group, inhibiting SIRT5 (MC3482, a specific SIRT5 inhibitor) 316 improved o bviously GLS and GDH enzymatic activity (Figure 7A, 7B), 317 accompanying with obvious reduction in the content NH 3 and glutamate (Figure 7C, 318 7D). In addition, compared to SIRT5 OE or KO cells, we observed that GLS and 319 GDH enzymatic activity in SIRT5 OE or KO cells with NH 4Cl treatment was 320 significantly enhanced by 1.27-fold and 1.33-fold or 1.20-fold and 1.08-fold (Figure 321 S1). In accord, the results in MCN group were similar to those in SKN group. These 322

again verified that SIRT5 negatively regulates GLS and GDH enzymatic 323 activity by lysine desuccinylation. 324 We further used BPTES (GLS inhibitor) to treat cells, GLS enzymatic 325 activity declined statistically significant , which suggested the significant inhibitory 326 effect of BPTES on GLS enzymatic activity (Figure 9A) . Moreover, the content of 327 glutamate and NH 3 were significantly down -regulated in MAC-T cells treated with 328 BPTES (Figure 9B, 9C) . As list in Figure 9B and 9C, c ompared with the BP group, 329 the content NH 3 and glutamate in SO BP group declined (P>0.05), while in the SKBP 330 and MCBP groups, the content NH 3 and glutamate increased (P>0.05). Altogether, 331 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted August 26, 2024. ; https://doi.org/10.1101/2024.08.26.609685doi: bioRxiv preprint SIRT5 inhibited glutamine metabolism in bovine mammary epithelial cells. 332 GLS and GDH were required for SIRT5 to regulate ammonia-induced cellular 333 autophagy 334 Our previous studies found that SIRT5 reduced NH 3-induced autophagy in 335 bovine mammary epithelial cells [ 24]. We speculated that SIRT5 control led the 336 autophagy in bovine mammary epithelial cells by mediating GLS and GDH activity. 337 As expected, compared to CT or NC group, the levels of Beclin1 and LC3II/I 338 increased significantly and p62 expression decreased obviously in SIRT5 KO cells 339 without or with NH 4Cl treatment (Figure 10 and Figure 11 ). In contrast, the levels of 340 Beclin1 and LC3II/I declined obviously and p62 expression increased significantly in 341 SIRT5 OE cells without or with NH 4Cl treatment (Figure 10 and Figure 11). 342 Nevertheless, SIRT5 KO cells had more punc tate LC3B positive cells and increased 343 significantly autophagosome counts (Figure 12). After NH 4Cl treatment, the number 344 of punctate LC3B positive cells and autophagosomes in MAC -T cells and SIRT5 KO 345 cells increased significantly, and the number of punctate LC3B positive cells and 346 autophagosomes in SIRT5 KO cells were significantly more than those in the MAC-T 347 cells (Figure 13) . Generally, SIRT5 significantly reduces the number of dot shaped 348 LC3B positive cells and autophagosomes, and inhibits the fo rmation of 349 autophagosomes. This was associated with elevated the autophagosomes number as 350 well as increased protein levels of autophagic markers. 351 Furthermore, we observed that SIRT5 KO led to NH 3 accumulation, and this 352 effect of SIRT5 KO was exacerbated by NH4Cl treatment (Figure S1 and Figure 11). 353 This was associated with elevated autophagy as well as increased protein levels of 354 autophagic markers (Figure 11). Additionally, we selected the inhibitor of autophagic 355 flux (BafA1, 50 nM) to treat cells and found that BafA1 effectively diminished SIRT5 356 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted August 26, 2024. ; https://doi.org/10.1101/2024.08.26.609685doi: bioRxiv preprint knockout-decreased p62 expression and elevated LC3II/I expression, had no effect on 357 Beclin1 expression (Figure 11). There were similar changes in the number of 358 autophagosomes (Figure 13). The above results suggested that SIRT5 interfere d 359 primarily with the autophagic flux, GLS and GDH were required as the major 360 substrate of SIRT5. To sum up, these findings provided independent and in vi tro data 361 supporting that lysine succinylation c ould enhance GLS and GDH enzymatic activity, 362 which interfered with ammonia-induced cellular autophagy. 363 SIRT5 promoted the maintenance of mitochondrial homeostasis 364 SIRT5 in mitochondria decorated proteins, which were participated in 365 metabolic activities, antioxidant pathways, energy production, apoptosis and 366 autophagy, further maintaining mitochondrial homeostasis . Mitochondrial 367 homeostasis involved in rebalancing the production and consumption of ATP during 368 periods of negative energy balance [32-34]. Interestingly, our previou s study found 369 that SIRT5 declined the ratio of ADP/ATP [ 24]. In this study, we also found that 370 SIRT5 increased the content of ATP (Fig ure 14A). Indeed, SIRT5 targeted succinate 371 dehydrogenase complex, whose activity regulated the levels of intracellular reactive 372 oxygen ( ROS). Consistently, SIRT5 KO in mice increased ROS levels [ 35]. SIRT5 373 activated homeostatic mechanisms to protect cells from stress ors [18]. Thus, we 374 speculated that S IRT5 might affect autophagy by regulating mitochondrial 375 homeostasis. As a result of SIRT5 OE cells had a greater number of mitochondria, 376 with normal shapes and linear structures (Figure 14B). Compared with MAC-T cells, 377 the number of mitochondria in SIRT5 OE cells was significantly increased, and the 378 brightness was enhanced, indicating that SIRT5 OE increases the number and activity 379 of mitochondria (Figure 14B) . SIRT5 KO decreased the number of mitochondria, 380 which were swelling and fragmentation (Figure 14B) . Hence, SIRT5 had a role of 381 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted August 26, 2024. ; https://doi.org/10.1101/2024.08.26.609685doi: bioRxiv preprint maintaining mitochondrial homeostasis in bovine mammary epithelial cells. 382

383 Ammonia affect ed the health and growth of livestock, reducing their 384 productivity [1]. NH3 increased the incidence rate and mortality of bovine respiratory 385 diseases, and also reduced the productivity, lactation performance and pregnancy 386 opportunities of cows [36, 37]. More than 98 % of ammonia in the body existed in the 387 form of NH4 +, which crosse d the cell membrane through various transport channels 388 and participated in various physiological activities [ 38]. In addition, NH 3 could 389 change pH, electrolyte balance and metabolic changes, which caused a negative 390 impact on cellular function [ 2, 39]. High concentrations of NH 3 could affect the 391 composition of microorganisms in the nasal cavity and colon of rabbits, and interfere 392 with local immune response and inflammatory processes, as well as increase the 393 incidence of respiratory diseases in rabbits [40]. 394 Autophagy exerted a significant role in maintaining the quantity and function 395 of mammary epithelial cells, and was a key factor affecting the lactation performance 396 of dairy cows. Ammonia ha d a dual role in autophagy, acting as an inducer at low 397 concentrations and an inhibitor at high concentrations [ 3]. Our previous research 398 found that NH 3 regulated autophagy of cow mammary epithelial cells through the 399 PI3K/Akt/mTOR signaling pathway [ 14]. Interestingly, SIRT5 was a key regulatory 400 factor in maintaining cellular homeostasis, which regulated protein activity in various 401 metabolic processes with post-translational modification s and control led ammonia 402 induced autophagy in tumor cells [ 15-22]. In addition, we found that SIRT5 inhibited 403 autophagy of cow mammary epithelial cells, and the PI3K/Akt/mTOR signaling was 404 involved in NH3 induced autophagy, depending on SIRT5 [ 24]. Meanwhile, SIRT5 405 promoted autophagy in colorectal cancer, and its overexpression was associated with 406 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted August 26, 2024. ; https://doi.org/10.1101/2024.08.26.609685doi: bioRxiv preprint low survival rates [20]. Moreover, the AMPK/mTOR pathway was involved in 407 SIRT5-enhanced autophagy in gastric cancer cells , while SIRT5 expression was 408 suppressed in gastric cancer tissues [ 25]. Strangely, i n breast cancer and mouse 409 myogenic cells, silencing SIRT5 increased ammonia -induced autophagy by 410 controlling glutamine metabolism and mitophagy [ 12]. However, SIRT5 ha d been 411 shown to exhibit weaker deacetylase activity, more commonly demalonylase, 412 desuccinylase, and deglutarylase activities, participating in several metabolic 413 processes of mitochondria [18, 28-31]. SIRT5 downregulated the succinylation level 414 of GLS and inhibit ed ammonia-induced autophagy in tumor cells [ 12]. Our previous 415 studies also found that SIRT5 reduced NH 3-induced autophagy in bovine mammary 416 epithelial cells, but little is known about the specific molecular mechanism. 417 Bcl-2 associated athanogene 3 (BAG3) stabilized GLS through inhibiting its 418 interaction with SIRT5, thereby preventing its desuccinylation to enhance autophagy 419 [41]. SIRT5 regulated GDH to provide glutamine with entering the tricarboxylic acid 420 (TCA) cycle in malignant colorectal cancer cells [42]. Mechanistically, the direct 421 interaction between SIRT5 and GDH led to deglutarylation and functional activation 422 of GDH, which was a key regulatory factor of glutaminolysis [42]. Interestingly, our 423 recent study found that SIRT5 reduced the content of NH 3 and glutamate in bovine 424 mammary epithelial cells by inhibiting GLS activity, and declined the ratio of 425 ADP/ATP [24]. Therefore, we hypothesized that SIRT5 -mediated GLS and GDH 426 desuccinylation controlled ammonia production, further regulate d the autophagy of 427 mammary epithelial cells. To verify this point, we constructed SIRT5 overexpression 428 or knockout cell lines. Additionally, ammonium chloride, and some inhibitors for 429 SIRT5, GLS, and autophagy were also used to treat cells. Immunoprecipitation 430 techniques verified that SIRT5 interact ed with endogenous GLS or GDH. Strangely, 431 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted August 26, 2024. ; https://doi.org/10.1101/2024.08.26.609685doi: bioRxiv preprint SIRT5 had no effect on the expression of endogenous GLS or GDH. 432 Reversible post-translational modifications are considered as key regulators 433 of mitochondrial proteins and various metabolism. The analysis revealed the potential 434 impact of lysine succinylation on enzymes of mitochondrial metaboli sm, such as 435 tricarboxylic acid cycle (TCA), fatty acid metabolism , and amino acid degradation 436 [43]. SIRT5 with efficient lysine desuccinylase inhibited the biochemical activity of 437 pyruvate dehydrogenase complex and succinate dehydrogenase [43]. SIRT5 deficient 438 in mice appeared to enhance succinylation level of carbamoyl phosphate synthase 1, 439 which was a known target of SIRT5 [29]. The absence of SIRT5 caused the 440 accumulation of medium -chain and long -chain acylcarnitines, and reduce d the 441 production of β -hydroxybutyric acid [23]. Moreover, SIRT5 mediated succinylation 442 level of ketogenic enzyme 3-hydroxy-3-methylglutaryl-CoA synthase 2 (HMGCS2) in 443 vivo and in vitro [23]. In summary, SIRT5 was a global regulator for mitochondrial 444 lysine succinylation. Interestingly, our previous study found that SIRT5 reduced the 445 content of NH 3 and glutamate in bovine mammary epithelial cells by inhibiting GLS 446 activity [ 24]. Based on SIRT5, which ha d a strong desuccinylation effect, we 447 determined the protein succinylation level and demonstrated that SIRT5 reduced the 448 protein succinylation level in cells. Interestingly, we also found that SIRT5 attenuated 449 the succinylation levels of GLS and GDH, which w ere consistent with the results of 450 total protein. 451 To determine the functional consequence of SIRT5 catalyzing the 452 desuccinylation of GLS and GDH lysine, we examined GLS and GDH enzymatic 453 activity, and demonstrated that SIRT5 reduced the enzymatic activity of GLS and 454 GDH, as well as the content of NH 3 and glutamate. Correspondingly, after treatment 455 with NH4Cl, SIRT5 also reduced GLS and GDH enzymatic activity, resulting in the 456 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted August 26, 2024. ; https://doi.org/10.1101/2024.08.26.609685doi: bioRxiv preprint induction of NH 3 and glutamate content and alleviating glutamine metabolism. 457 Meanwhile, inhibiting SIRT5 (MC3482, a specific SIRT5 inhibitor) improved 458 obviously GLS and GDH enzymatic activity, accompanying with obvious reduction in 459 the content NH 3 and glut amate. These results again verified that SIRT5 regulate d 460 negatively GLS and GDH enzymatic activity by lysine desuccinylation. We further 461 treated cells with BPTES (GLS inhibitor) and confirmed that SIRT5 inhibited 462 glutamine metabolism in bovine mammary epithelial cells. Collectively, we found that 463 SIRT5 obviously decreased the succinylation levels of GLS and GDH, and further 464 decreased the enzymatic activity of GLS and GDH. 465 Autophagy was a strictly mediated process that got rid of damaged organelles 466 or cytosolic components, the process of which started with the formation of 467 autophagosomes with double-membrane vesicles [ 44]. Then, the melt 468 autophagosomes with cargo were transported to lysosomes for degradation and 469 recovery. LC3, Beclin1 and p62 are indicated genes known as autophagy related 470 genes. Moreover, autophagy was also affected by many other genes [ 45]. There is 471 evidence that SIRT5 had a regulatory function in autophagy [25, 46]. SIRT5 promoted 472 autophagy in colorectal cancer and gastric cancer cells [ 20, 25]. In breast cancer and 473 mouse myogenic cells, silencing SIRT5 increased ammonia -induced autophagy by 474 controlling glutamine metabolism and mitophagy [ 12]. Another study showed that 475 SIRT5 increa sed autophagy and accelerated in colorectal cancer growth by 476 deacetylating lactate dehydrogenase B, thereby promoting its activity, and knockdown 477 or inhibition of SIRT5 down-regulated autophagy levels and inhibited growth [47]. 478 There was much evidence that both ammonia and SIRT5 c ould regulate the 479 level of cellular autophagy, and excess ammonia increase d the expression of the 480 classical autophag y genes LC3 and Beclin1, and decrease d the expression of 481 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted August 26, 2024. ; https://doi.org/10.1101/2024.08.26.609685doi: bioRxiv preprint p62/SQSTM1, suggesting that high ammonia induced autophagy of skeletal muscle in 482 patients with cirrhosis [ 48], and that ammonia promote d the occurrence of autophagy 483 in hepatocytes in mice with high blood ammonia [ 49]. Our previous studies also 484 found that SIRT5 reduced NH 3-induced autophagy in bovine mam mary epithelial 485 cells [24]. It was known that GLS and GDH participate in the regulation of ammonia 486 production [ 2, 50]. Therefore, we speculated that SIRT5 regulated GLS and GDH 487 activity to control the autophagy of bovine mammary epithelial cells. As expect ed, 488 SIRT5 attenuated the levels of Beclin1 and LC3II/I , and enhanced p62 expression 489 without or with NH 4Cl treatment. Nevertheless, SIRT5 significantly reduce d the 490 number of dot shaped LC3B positive cells and autophagosomes, and inhibit ed the 491 formation of autophagosomes. This was associated with elevated the autophagosomes 492 number as well as increased protein levels of autophagic markers. Furthermore, we 493 observed that SIRT5 KO led to NH 3 accumulation, and this effect of SIRT5 KO was 494 exacerbated by NH4Cl treatment. This was associated with elevated autophagy as well 495 as increased protein levels of autophagic markers (Figure 11). Additionally, we 496 selected the inhibitor of autophagic flux (BafA1) to treat cells and found that BafA1 497 effectively diminished S IRT5 knockout -decreased p62 expression and elevated 498 LC3II/I expression, had no effect on Beclin1 expression. There were similar changes 499 in the number of autophagosomes (Figure 13). Hence, these findings provided 500 independent and in vitro data supporting that lysine succinylation could enhance GLS 501 and GDH enzymatic activity, which influenced ammonia -induced cellular autophagy. 502 These results suggested that SIRT5 interfered primarily with the autophagic flux, GLS 503 and GDH were required as the major substrate of SIRT5. 504 Mitochondria are the central organelles of metabolic activities, and several 505 factors have been demonstrated to maintain these homeostatic mechanisms [51, 52]. 506 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted August 26, 2024. ; https://doi.org/10.1101/2024.08.26.609685doi: bioRxiv preprint Due to the important role of sirtuins in regulating metabolism, this protein family has 507 aroused great interest in the scientific community. Especially, SIRT3, SIRT4, and 508 SIRT5 located in the mitochondrial matrix, where they mediated proteins participating 509 in metabolic re sponses, energy metabolism, antioxidant roles, and autophagy, further 510 maintaining the homeostasis of mitochondria [32, 33]. SIRT5 had been described to 511 play a role in mediating cellular metabolism, detoxification, oxidative stress, energy 512 balance, and autophagy [53]. Pathway analysis indicated that SIRT5 targeted fatty 513 acid β-oxidation, branched chain amino acid catabolism, the citric acid cycle, ATP 514 synthesis, and ketone body synthesis , which was the most enriched target pathways 515 [23]. The activity and expression of mitochondrial sirtuins were closely related to 516 cellular metabolic status, making metabolic adaptation possible, thereby rebalancing 517 ATP production and consumption during negative energy balance periods [34]. 518 Interestingly, our previous study found that SIRT5 declined the ratio of ADP/ATP [24]. 519 In this study, we also found that SIRT5 increased the content of ATP (Figure 14A). 520 Indeed, SIRT5 target ed succinate dehydrogenase complex, whose activity c ould 521 interfere with the intracellular reactive oxygen (ROS) levels [ 35]. SIRT5 maintained 522 mitochondrial homeostasis to protect cells from stressors [ 18]. Thus, we speculated 523 that SIRT5 might affect autophagy by regulating mitochondrial homeostasis. As a 524

of SIRT5 OE cells had a greater number of mitochondria, with normal shapes 525 and linear structures (Figur e 14B). Compared with the MAC -T cells, the number of 526 mitochondria in SIRT5 OE cells was significantly increased, and the brightness was 527 enhanced, indicating that SIRT5 OE increases the number and activity of 528 mitochondria (Figure 14B). SIRT5 KO decreased th e number of mitochondria, which 529 were swelling and fragmentation (Figure 14B). Hence, SIRT5 ha d a role of 530 maintaining mitochondrial homeostasis in bovine mammary epithelial cells. 531 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted August 26, 2024. ; https://doi.org/10.1101/2024.08.26.609685doi: bioRxiv preprint Collectively, we concluded that SIRT5 reduced ammonia release by 532 modulating t he succinylation levels and enzymatic activities of GLS and GDH in 533 mitochondria and promoted the maintenance of mitochondrial homeostasis, as well as 534 further attenuated ammonia -induced autophagy in bovine mammary epithelial cells 535 (Figure 15). Consistently, we discovered that GLS and GDH were two physiological 536 substrates of SIRT5, which relied on desuccinylation to reduce ammonia production 537 by inhibiting the enzymatic activity of GLS and GDH. 538 Funding 539 This work was supported by National Natural Science Foundation of China 540 (32172809). The funding agency had no role in the study design, data collection, 541 interpretation, or the decision to submit the work for publication. 542 Availability of data and materials 543 The datasets used and/or analyzed during the present study are available from the 544 corresponding author upon request. 545 Author contributions 546 LHP and WYY conceived and designed the experiment. YHL, GSK, LGY and JJH 547 executed experiments; DJR, ZXY and LLY cultured and treated cells; ZK and GS 548 drew figures, ZGM and HLQ performed data analyses; YHL, GSK, LHP, and WYY 549 interpreted the data and wrote the manuscript. 550 Ethics approval and consent to participate 551 This article does not contain any studies with human subjects or animals performed 552 by any of the authors. All experimental protocols were approved by College of 553 Animal Veterinary Medicine, Henan Agricultural University Ethics Committee 554 (Zhengzhou, China). 555 Patient consent for publication 556 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted August 26, 2024. ; https://doi.org/10.1101/2024.08.26.609685doi: bioRxiv preprint Not applicable. 557 Competing interests 558 The authors declare that they have no competing interests. 559

560 [1] Rodrigues ARF, Maia MRG, Miranda C, Cabrita ARJ, Fonseca AJM, Pereira JLS, Trindade 561 H. Ammonia and greenhouse emissions from cow's excreta are affected by feeding system, 562 stage of lactation and sampling time. J Environ Manage. 2022, 320: 115882. 563 [2] Dasarathy S, Mookerjee RP, Rackayova V , Rangroo Thrane V , Vairappan B, Ott P, Rose CF. 564 Ammonia toxicity: from head to toe? Metab Brain Dis. 2017, 32(2): 529-538. 565 [3] Soria LR, Brunetti -Pierri N. Ammonia and autophagy: An emerging relationship with 566 implications for disorders with hyperammonemia. J Inherit Metab Dis. 2019, 42(6): 567 1097-1104. 568 [4] Reijngoud DJ, Oud PS, Kás J, Tager JM. Relationship between medium pH and that of the 569 lysosomal matrix as s tudied by two independent methods. Biochim Biophys Acta. 1976 , 570 448(2):290-302. 571 [5] Seglen PO, Reith A. Ammonia inhibition of protein degradation in isolated rat hepatocytes. 572 Quantitative ultrastructural alterations in the lysosomal system. Exp Cell Res. 1976 , 100(2): 573 276-80. 574 [6] Kumar A, Davuluri G, Silva RNE, Engelen MPKJ, Ten Have GAM, Prayson R, Deutz NEP, 575 Dasarathy S. Ammonia lowering reverses sarcopenia of cirrhosis by restoring skeletal muscle 576 proteostasis. Hepatology. 2017, 65(6): 2045-2058. 577 [7] Qiu J, Tsien C, Thapalaya S, Narayanan A, Weihl CC, Ching JK, Eghtesad B, Singh K, Fu X, 578 Dubyak G, McDonald C, Almasan A, Hazen SL, Naga Prasad SV , Dasarathy S. 579 Hyperammonemia-mediated autophagy in skeletal muscle contributes to sarcopenia of 580 cirrhosis. Am J Physiol Endocrinol Metab. 2012, 303(8): E983-93. 581 [8] Soria LR, Allegri G, Melck D, Pastore N, Annunziata P, Paris D, Polishchuk E, Nusco E, 582 Thöny B, Motta A, Häberle J, Ballabio A, Brunetti -Pierri N. Enhancement of hepatic 583 autophagy increases ureagenesis and protects ag ainst hyperammonemia. Proc Natl Acad Sci 584 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted August 26, 2024. ; https://doi.org/10.1101/2024.08.26.609685doi: bioRxiv preprint U S A. 2018, 115(2): 391-396. 585 [9] Cheong H, Lindsten T, Wu J, Lu C, Thompson CB. Ammonia -induced autophagy is 586 independent of ULK1/ULK2 kinases. Proc Natl Acad Sci U S A. 2011, 108(27): 11121-6. 587 [10] Harder LM, Bunkenborg J, Andersen JS. Inducing autophagy: a comparative 588 phosphoproteomic study of the cellular response to ammonia and rapamycin. Autophagy. 589 2014, 10(2): 339-55. 590 [11] Li Z, Ji X, Wang W, Liu J, Liang X, Wu H, Liu J, Eggert US, Liu Q, Zhang X. Ammonia 591 induces autophagy through dopamine receptor D3 and mTOR. PLoS One. 2016 , 11(4): 592 e0153526. 593 [12] Polletta L, Vernucci E, Carnevale I, Arcangeli T, Rotili D, Palmerio S, Steegborn C, Nowak T, 594 Schutkowski M, Pellegrini L, Sansone L, Villanova L, Runci A, Pucci B, Morgante E, Fini M, 595 Mai A, Russo MA, Tafani M. SIRT5 regulation of ammonia-induced autophagy and 596 mitophagy. Autophagy. 2015, 11(2): 253-70. 597 [13] Eng CH, Yu K, Lucas J, White E, Abraham RT. Ammonia derived from glutaminolysis is a 598 diffusible regulator of autophagy. Sci Signal. 2010, 3(119): ra31. 599 [14] Feng L, Liao H, Liu J, Xu C, Zhong K, Zhu H, Guo S, Guo Y , Han L, Li H, Wang Y . 600 Inhibition of PI3K/Akt/mTOR pathway by ammonium chloride induced apoptosis and 601 autophagy in MAC-T cell. Res Vet Sci. 2021, 136: 622-630. 602 [15] Fiorentino F, Castiello C, Mai A, Rotili D. Therapeutic potential and activity modulation of 603 the protein lysine deacylase sirtuin 5. J Med Chem. 2022, 65(14): 9580-9606. 604 [16] Lagunas-Rangel FA. Current role of mammalian sirtuins in DNA repair. DNA Repair (Amst). 605 2019, 80: 85-92. 606 [17] Mori M, Cazzaniga G, Meneghetti F, Villa S, Gelain A. Insights on the modulation of SIRT5 607 activity: a challenging balance. Molecules. 2022, 27(14): 4449. 608 [18] Kumar S, Lombard DB. Functions of the sirtuin deacylase SIRT5 in normal physiology and 609 pathobiology. Crit Rev Biochem Mol Biol. 2018, 53(3): 311-334. 610 [19] Wang Y , Chen H, Zha X. Overview of SIRT5 as a potential therapeutic target: structure, 611 function and inhibitors. Eur J Med Chem. 2022, 236: 114363. 612 [20] Lin ZF, Xu HB, Wang JY , Lin Q, Ruan Z, Liu FB, Jin W, Huang HH, Chen X. SIRT5 613 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted August 26, 2024. ; https://doi.org/10.1101/2024.08.26.609685doi: bioRxiv preprint desuccinylates and activates SOD1 to eliminate ROS. Biochem Biophys Res Commun. 2013, 614 441(1): 191-5. 615 [21] Zhang Y , Bharathi SS, Rardin MJ, Uppala R, Verdin E, Gibson BW, Goetzman ES. SIRT3 616 and SIRT5 regulate the enzyme activity and cardiolipin binding of very long-chain acyl-CoA 617 dehydrogenase. PLoS One. 2015, 10(3): e0122297. 618 [22] He S, Jia Q, Zhou L, Wang Z, Li M. SIRT5 is involved in the proliferation and metastasis of 619 breast cancer by promoting aerobic glycolysis. Pathol Res Pract. 2022, 235: 153943. 620 [23] Rardin MJ, He W, Nishida Y , Newman JC, Carrico C, Danielson SR, Guo A, Gut P, Sahu AK, 621 Li B, Uppala R, Fitch M, Riiff T, Zhu L, Zhou J, Mulhern D, Stevens RD, Ilkayeva OR, 622 Newgard CB, Jacobson MP, Hellerstein M, Goetzman ES, Gibson BW, Verdin E. SIRT5 623 regulates the mitochondrial lysine succinylome and metabolic networks. Cell Metab. 2013, 624 18(6): 920-33. 625 [24] He J, Feng L, Yang H, Gao S, Dong J, Lu G, Liu L, Zhang X, Zhong K, Guo S, Zha G, Han 626 L, Li H, Wang Y . Sirtuin 5 alleviates apoptosis and autophagy stimulated by ammonium 627 chloride in bovine mammary epithelial cells. Exp Ther Med. 2024, 28(1): 295. 628 [25] Gu W, Qian Q, Xu Y , Xu X, Zhang L, He S, Li D. SIRT5 regulates autophagy and apoptosis 629 in gastric cancer cells. J Int Med Res. 2021, 49(2): 300060520986355. 630 [26] Altman BJ, Stine ZE, Dang CV . From Krebs to clinic: glutamine metabolism to cancer 631 therapy. Nat Rev Cancer. 2016, 16(11): 749. 632 [27] Neu J, Shenoy V , Chakrabarti R. Glutamine nutrition and metabolism: where do we go from 633 here ? FASEB J. 1996, 10(8): 829-37. 634 [28] Hu T, Shukla SK, Vernucci E, He C, Wang D, King RJ, Jha K, Siddhanta K, Mullen NJ, Attri 635 KS, Murthy D, Chaika NV , Thakur R, Mulder SE, Pacheco CG, Fu X, High RR, Yu F, 636 Lazenby A, Steegborn C, Lan P, Mehla K, Rotili D, Chaudhary S, Valente S, Tafani M, Mai 637 A, Auwerx J, Verdin E, Tuveson D, Singh PK. Metabolic rewiring by loss of Sirt5 promotes 638 kras-induced pancreatic cancer progression. Gastroenterology. 2021, 161(5): 1584-1600. 639 [29] Du J, Zhou Y , Su X, Yu JJ, Khan S, Jiang H, Kim J, Woo J, Kim JH, Choi BH, He B, Chen W, 640 Zhang S, Cerione RA, Auwerx J, Hao Q, Lin H. Sirt5 is a NAD-dependent protein lysine 641 demalonylase and desuccinylase. Science. 2011, 334(6057): 806-9. 642 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted August 26, 2024. ; https://doi.org/10.1101/2024.08.26.609685doi: bioRxiv preprint [30] Chen XF, Tian MX, Sun RQ, Zhang ML, Zhou LS, Jin L, Chen LL, Zhou WJ, Duan KL, 643 Chen YJ, Gao C, Cheng ZL, Wang F, Zhang JY , Sun YP, Yu HX, Zhao YZ, Yang Y , Liu WR, 644 Shi YH, Xiong Y , Guan KL, Ye D. SIRT5 inhibits peroxisomal ACOX1 to prevent oxidative 645 damage and is downregulated in liver cancer. EMBO Rep. 2018, 19(5): e45124. 646 [31] Tan M, Peng C, Anderson KA, Chhoy P, Xie Z, Dai L, Park J, Chen Y , Huang H, Zhang Y , 647 Ro J, Wagner GR, Green MF, Madsen AS, Schmiesing J, Peterson BS, Xu G, Ilkayeva OR, 648 Muehlbauer MJ, Braulke T, Mühlhausen C, Backos DS, Olsen CA, McGuire PJ, Pletcher SD, 649 Lombard DB, Hirschey MD, Zhao Y . Lysine glutarylation is a protein posttranslational 650 modification regulated by SIRT5. Cell Metab. 2014, 19(4): 605-17. 651 [32] Elkhwanky MS, Hakkola J. Extranuclear sirtuins and metabolic stress. Antioxid Redox 652 Signal. 2018, 28(8): 662-676. 653 [33] Singh CK, Chhabra G, Ndiaye MA, Garcia-Peterson LM, Mack NJ, Ahmad N. The role of 654 sirtuins in antioxidant and redox signaling. Antioxid Redox Signal. 2018, 28(8): 643-661. 655 [34] Guarente L. Calorie restriction and sirtuins revisited. Genes Dev. 2013, 27(19): 2072-85. 656 [35] Zhou L, Wang F, Sun R, Chen X, Zhang M, Xu Q, Wang Y , Wang S, Xiong Y , Guan KL, 657 Yang P, Yu H, Ye D. SIRT5 promotes IDH2 desuccinylation and G6PD deglutarylation to 658 enhance cellular antioxidant defense. EMBO Rep. 2016, 17(6): 811-22. 659 [36] EFSA Panel on Animal Health and Welfare (AHAW). Scientific opinion on the welfare of 660 cattle kept for beef production and the welfare in intensive calf farming systems. EFSA J. 661 2012, 10(5): 2669. 662 [37] Zheng P, Qin X, Feng R, Li Q, Huang F, Li Y , Zhao Q, Huang H. Alleviative effect of 663 melatonin on the decrease of uterine receptivity caused by blood ammonia through 664 ROS/NF-κB pathway in dairy cow. Ecotoxicol Environ Saf. 2022, 231: 113166. 665 [38] Bosoi CR, Rose CF. Identifying the direct effects of ammonia on the brain. Metab Brain Dis. 666 2009, 24(1): 95-102. 667 [39] Cudalbu C. In vivo studies of brain metabolism in animal models of Hepatic Encephalopathy 668 using ¹H Magnetic Resonance Spectroscopy. Metab Brain Dis. 2013, 28(2): 167-74. 669 [40] Li K, Pang S, Li Z, Ding X, Gan Y , Gan Q, Fang S. House ammonia ex posure causes 670 alterations in microbiota, transcriptome, and metabolome of rabbits. Front Microbiol. 2023, 671 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted August 26, 2024. ; https://doi.org/10.1101/2024.08.26.609685doi: bioRxiv preprint 14: 1125195. 672 [41] Zhao S, Wang JM, Yan J, Zhang DL, Liu BQ, Jiang JY , Li C, Li S, Meng XN, Wang HQ. 673 BAG3 promotes autophagy and glutaminolysis via stabilizing glutaminase. Cell Death Dis. 674 2019, 10(4): 284. 675 [42] Wang YQ, Wang HL, Xu J, Tan J, Fu LN, Wang JL, Zou TH, Sun DF, Gao QY , Chen YX, 676 Fang JY . Sirtuin5 contributes to colorectal carcinogenesis by enhancing glutaminolysis in a 677 deglutarylation-dependent manner. Nat Commun. 2018, 9(1): 545. 678 [43] Park J, Chen Y , Tishkoff DX, Peng C, Tan M, Dai L, Xie Z, Zhang Y , Zwaans BM, Skinner 679 ME, Lombard DB, Zhao Y . SIRT5-mediated lysine desuccinylation impacts diverse 680 metabolic pathways. Mol Cell. 2013, 50(6): 919-30. 681 [44] Mizushima N, Levine B, Cuervo AM, Klionsky DJ. Autophagy fights disease through 682 cellular self-digestion. Nature. 2008, 451(7182): 1069-75. 683 [45] Kang R, Zeh HJ, Lotze MT, Tang D. The Beclin 1 network regulates autophagy and 684 apoptosis. Cell Death Differ. 2011, 18(4): 571-80. 685 [46] He XX, Huang CK, Xie BS. Autophagy inhibition enhanced 5-FU-induced cell death in 686 human gastric carcinoma BGC-823 cells. Mol Med Rep. 2018, 17(5): 6768-6776. 687 [47] Shi L, Yan H, An S, Shen M, Jia W, Zhang R, Zhao L, Huang G, Liu J. SIRT5-mediated 688 deacetylation of LDHB promotes autophagy and tumorigenesis in colorectal cancer. Mol 689 Oncol. 2019, 13(2): 358-375. 690 [48] Zhang YB, Li SX, Chen XP, Yang L, Zhang YG, Liu R, Tao LY . Autophagy is activated and 691 might protect neurons from degeneration after traumatic brain inju ry. Neurosci Bull. 2008, 692 24(3): 143-9. 693 [49] Ha J, Guan KL, Kim J. AMPK and autophagy in glucose/glycogen metabolism. Mol Aspects 694 Med. 2015, 46: 46-62. 695 [50] Braissant O, McLin V A, Cudalbu C. Ammonia toxicity to the brain. J Inherit Metab Dis. 696 2013, 36(4): 595-612. 697 [51] Wang CH, Wei YH. Roles of mitochondrial sirtuins in mitochondrial function, redox 698 homeostasis, insulin resistance and type 2 diabetes. Int J Mol Sci. 2020, 21(15): 5266. 699 [52] Kupis W, Pałyga J, Tomal E, Niewiadomska E. The role of sirtuins in cellular homeostasis. J 700 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted August 26, 2024. ; https://doi.org/10.1101/2024.08.26.609685doi: bioRxiv preprint Physiol Biochem. 2016, 72(3): 371-80. 701 [53] Liu B, Che W, Zheng C, Liu W, Wen J, Fu H, Tang K, Zhang J, Xu Y . SIRT5: a safeguard 702 against oxidative stress-induced apoptosis in cardiomyocytes. Cell Physiol Biochem. 2013, 703 32(4): 1050-9. 704 [54] Figures Legends 705 [55] Figure 1 Observation of green fluorescence in MAC -T cells transfected with 706 pPB-EF1α or pPB -EF1α-SIRT5 vectors . Green fluorescence was observed 707 clearly under the fluorescence microscopy in the transfected MAC -T cells (B , 708 pPB-EF1α; C, pPB-EF1α-SIRT5), but no green fluorescence in non -transfected 709 MAC-T cells (A, CT). 710 [56] Figure 2 SIRT5 expression in SIRT5 KO and OE cell lines. (A, B) The mRNA 711 and protein expression of SIRT5 in SIRT5 KO cells. (C, D) The mRNA and 712 protein expression of SIRT5 in SIRT5 OE cells. Note: ** P < 0.01 vs. MAC -T 713 cells (CT group). 714 [57] Figure 3 SIRT5 interacted with endogenous GLS or GDH. A. Endogenous 715 GLS interacts with SIRT5. SIRT5 protein in MAC -T cells was purified by IP 716 with an anti-GLS antibody, followed by western blotting to detect SIRT5 w ith an 717 anti-SIRT5 antibody. B. Endogenous GDH interacts with SIRT5. SIRT5 protein 718 in MAC -T cells was purified by IP with an anti -GDH antibody, followed by 719 western blotting to detect SIRT5 with an anti-SIRT5 antibody. 720 [58] Figure 4 SIRT5 did not change the expre ssion of GLS and GDH. A. GLS 721 expression detected by western blotting. B. GDH expression detected by western 722 blotting. β-actin is used as loading control. MAC-T cells (CT), SIRT5 OE and 723 SIRT5 KO cells were cultured in base medium, respectively. 724 [59] Figure 5 SIRT5 decreased protein succinylation in whole cell lysates. A. 725 Immunoprecipitation and western blotting were performed to assess succinylation 726 levels using succinyl-lysine-specific antibodies in whole cell lysates. B. Relative 727 intensity of SuccK/ β-actin. In stable MAC -T, SIRT5 OE, and SIRT5 KO cells, 728 the total proteins were purified by IP with beads and western blotting to detect 729 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted August 26, 2024. ; https://doi.org/10.1101/2024.08.26.609685doi: bioRxiv preprint their lysine succinylation level. Note: * P < 0.05, ** P < 0.01 vs. MAC -T cells 730 (CT group). 731 [60] Figure 6 SIRT5 decreased the GLS an d GDH succinylation. A. 732 Immunoprecipitation and western blotting were performed to assess succinylation 733 levels of GLS and GDH using succinyl-lysine-specific antibodies. B. Fold change 734 of GLS SuccK. C. Fold change of GDH SuccK. In stable MAC -T, SIRT5 OE, 735 and SIRT5 KO cells, the GLS and GDH protein were purified by IP with beads 736 and western blotting to detect their succinylation level. Note: ** P < 0.01 vs. 737 MAC-T cells (CT group). 738 [61] Figure 7 SIRT5 attenuated ammonia release. MAC-T cells (CT) were cultured 739 in base medium. Cells in SIRT5 OE and SIRT5 KO group were cultured in base 740 medium. MAC -T cells in MC group were treated by complete medium 741 containing 20 μM MC3482 for 12 h. GLS activity in cells (A), GDH activity in 742 cells (B), NH 3 content in cells (C), Glutamat e content in cells (D). Note: * P < 743 0.05, ** P < 0.01 vs .CT group. 744 [62] Figure 8 SIRT5 exacerbated ammonia release with NH 4Cl treatment. 745 MAC-T cells (CT) were cultured in base medium. MAC-T cells in NC group, and 746 SO or SK cells in SON or SKN group were treated with base medium containing 747 10 mM NH4Cl for 12 h. MAC-T cells in MCN group were pretreated with 20 μM 748 MC3482 for 30 min, and then treated with 10 mM NH 4Cl for 12 h. GLS activity 749 in cells (A), GDH activity in cells (B), NH 3 content in cells (C), Glutamate 750 content in cells (D). Note: * P < 0.05, ** P < 0.01 vs .CT group; # P <0.05, ## P < 751 0.01 vs. NC group. 752 [63] Figure 9 SIRT5 declined ammonia release with GLS inhibitor treatment. 753 MAC-T cells (CT) were cultured in base medium. MAC-T cells in BP group, and 754 SO or S K cells in SOBP or SKBP group were treated with base medium 755 containing 0.12 μM BPTES for 12 h. MAC -T cells in MCBP group were 756 pretreated with 20 μM MC3482 for 1 h, and then the cells were treated with 0.12 757 μM BPTES for 12 h. GLS activity in cells (A), NH 3 content in cells (B), 758 Glutamate content in cells (C). Note: * P < 0.05, ** P < 0.01 vs .CT group; # P 759 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted August 26, 2024. ; https://doi.org/10.1101/2024.08.26.609685doi: bioRxiv preprint <0.05, ## P < 0.01 vs. BP group. 760 [64] Figure 10. SIRT5 down-regulated the expression of Beclin1 and LC3II/I and 761 increased p62 expression. MAC-T cells (CT), SO or SK cells were cultured in 762 base medium. LC3II/I (A), Beclin1 (B), p62 (C). Note: * P < 0.05, ** P < 0.01 763 vs .CT group. 764 [65] Figure 11. SIRT5 aggravated the decrease of Beclin1 and LC3II/I expression 765 and the increase of p62 expression with NH 4Cl treatment. MAC-T cells (CT) 766 were cultured in base medium. MAC -T cells in NC group, and SO or SK cells in 767 SON or SKN group were treated with base medium containing 10 mM NH 4Cl for 768 12 h. In SKNBA group, SIRT5 KO cells were incubated with BafA1 (50 nM) for 769 1 h, and then the cells were treated with 10 mM NH 4Cl for 12 h. LC3II/I (A), 770 Beclin1 (B), p62 (C). Note: * P < 0.05, ** P < 0.01 vs .CT group; # P <0.05, ## P 771 < 0.01 vs. NC group; Δ P < 0.05 vs .SKN group. 772 [66] Figure 12. SIRT5 KO promoted the formation of autophagosomes. MAC-T 773 cells (CT), and SK cells were cultured in base medium. GFP -LC3 staining (A), 774 the number of autophagosomes (B). Note: ** P < 0.01 vs .CT group. 775 [67] Figure 13. SIRT5 KO promoted the formation of autophagosomes with 776 NH4Cl treatment. MAC-T cells (CT), and SK cells were cultured in base 777 medium. MAC-T cells in NC group and SK cells in SKN group were treated with 778 base medium containing 10 mM NH 4Cl for 12 h. In SKNBA group, SIRT5 KO 779 cells were incubated with BafA1 (50 nM) for 1 h, and then the cells were treated 780 with 10 mM NH 4Cl for 12 h. GFP-LC3 staining (A), t he number of 781 autophagosomes (B). Note: ** P < 0.01 vs .CT group; ## P < 0.01 vs. NC group; 782 Δ P < 0.05 vs .SKN group. 783 [68] Figure 14. SIRT5 enhanced the content of ATP and the activity of 784 mitochondria. MAC-T cells (CT), SO or SK cells were cultured in base medium. 785 The content of ATP (A), Mito-Tracker Red-staining of mitochondria (B). Note: 786 ** P < 0.01 vs .CT group. 787 [69] Figure 15. A model of SIRT5-mediated GLS and GDH desuccinylation 788 attenuated the autophagy of bovine mammary epithelial cells induced by 789 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted August 26, 2024. ; https://doi.org/10.1101/2024.08.26.609685doi: bioRxiv preprint ammonia. GLS and GDH were two physiological substrates of SIRT5, which 790 relied on desuccinylation to reduce ammonia production by inhibiting the 791 enzymatic activity of GLS and GDH in mitochondria. SIRT5 enhanced ATP 792 content and promoted the maintenance of mitochondrial homeostasis, as well as 793 further attenuated ammonia -induced autophagy in bovine mammary epithelial 794 cells. SIRT5, sirtuin 5; LC3B, light chain 3 β ; GLS, glutaminase; α-KG, 795 α-ketoglutarate; Gln, glutamin; GDH, glutamate dehydrogenase; Glu, glutamate; 796 BPTES, GLS inhibitor; MC3482, SIRT5 inhibitor; BafA1, bafilomycin A1, 797 autophagy inhibitor. The solid arrow represented the strengthening effect. The 798 dashed arrow represented the inhibitory effect. 799 [70] Figure S1. NH4Cl treatment exacerbated the effect of SIRT5 on 800 ammonia release. MAC-T cells (CT), SO or SK cells were cultured in base 801 medium. SO or SK cells in SON or SKN group were treated with base medium 802 containing 10 mM NH4Cl for 12 h. GLS activity in cells (A), GDH activity in 803 cells (B), NH3 content in cells (C), Glutamate content in cells (D). Note: * P < 804 0.05, ** P < 0.01 vs .CT group; ## P < 0.01 vs. SO group; Δ P < 0.05, ΔΔ P < 805 0.01 vs .SK group. 806 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted August 26, 2024. ; https://doi.org/10.1101/2024.08.26.609685doi: bioRxiv preprint