Sirtuin 3 mitigates oxidative-stress-induced apoptosis in bovine mammary epithelial cells

Ketosis is often accompanied by a reduction in milk production in dairy cows, but the molecular mechanism has not been fully elucidated. Ketotic cows possess systemic oxidative stress (OS), which may implicate apoptosis in mammary glands. Sirtuin 3 (SIRT3) is a vital regulator of cellular redox homeostasis and is under the control of AMP-activated protein kinase (AMPK) signaling in nonruminants. Thus, we aimed to investigate (1) the AMPK-SIRT3 and apoptosis status of mammary glands from ketotic cows, (2) the effect of SIRT3 on OS-induced apoptosis in bovine mammary epithelial cells (BMEC), and (3) the role of AMPK signaling on SIRT3-mediated effects on apoptosis. Mammary gland samples were reused from a previous study, which contained healthy and ketotic cows (both n = 15). BMEC were incubated with 0, 0.3, 0.6, or 0.9 m M H 2 O 2 for 6 h with/without a 30 min incubation of an antioxidant MitoQ (1 μ M ). Then BMEC were incubated with SIRT3 overexpression adenovirus (Ad-SIRT3) for 6 h followed by a 6 h incubation with 0.6 m M H 2 O 2 . Finally, BMEC were treated with the AMPK inhibitor Compound C (Cd C,10 μ M ) for 30 min before the H 2 O 2 challenge, or cells were initially treated with the AMPK agonist MK8722 (10 μ M ) for 30 min followed by a 30-h culture with/without si-SIRT3 and eventually the H 2 O 2 exposure. Ketotic cows displayed higher levels of Bax, Caspase-3 and Bax/Bcl-2 but lower levels of Bcl-2 in mammary glands. H 2 O 2 incubation displayed similar results, exhibiting a dose-dependent manner between the H 2 O 2 concentration and the apoptosis degree. Mito Q pretreatment reduced cellular reactive oxygen species and rescued cells from apoptosis. Ketotic cows had a lower mammary protein abundance of SIRT3. Similarly, H 2 O 2 incubation downregulated both mRNA and protein levels of SIRT3 in a dose-and time-dependent manner. Ad-SIRT3 infection lowered levels of cellular reactive oxygen species, Bax, Caspase-3 and Bax/Bcl-2 but increased levels of Bcl-2. TUNEL assays confirmed that Ad-SIRT3 infection mitigated H 2 O 2 -induced apoptosis. Both ketotic cows and H 2 O 2 -induced BMEC had lower levels of p-AMPK and p-AMPK/AMPK. Additionally, Cd C pretreatment decreased SIRT3 and Bcl-2 expression but increased levels of Bax and Caspase-3. Contrary to the inhibitor, MK8722 had opposite effects and reduced the percentage of apoptotic cells. However, these effects of MK8722 were reversed upon SIRT3 silencing. In conclusion, in vivo data confirmed that ketosis is associated with greater apoptosis and restricted AMPK-SIRT3 signaling in mammary glands; in vitro data indicated that SIRT3 mitigates OS-induced apoptosis via AMPK signaling. As such, there may be potential benefits for targeting the AMPK-SIRT3 axis to help counteract the negative effects of mammary glands during ketosis.


INTRODUCTION
Ketosis is one of the most common and costly metabolic disorders in dairy cows.It is characterized by elevated levels of ketone bodies (BHB, acetone, and acetoacetate) in blood.Blood BHB concentration is usually employed to define ketosis.It is generally accepted that cows are considered to have subclinical ketosis when blood BHB concentrations are in the range of 1.2 to 2.9 mM, while cows are thought to be clinical ketosis if BHB concentrations ≥3.0 mM.The disease frequency of ketosis has been described through 2 different measures, prevalence and incidence.The reported prevalence of subclinical ketosis is ranging from 11% to 47%, and that of clinical ketosis is between 4% and 12% (Benedet et al., 2019).The incidence of ketosis is on average >40% worldwide (Overton et al., 2017).Ketosis leads to immunosuppression and poor reproductive performance, and it seriously impairs the health status and welfare of cows.Ketotic cows are more vulnerable to other diseases like fatty liver, displaced abomasum, metritis, lameness, mastitis, laminitis, and other diseases (Berge and Vertenten, 2014).The total average loss of ketosis has been estimated to be $289 per case and year in the United States (McArt et al., 2015), Can $203 per case in Canada (Gohary et al., 2016) and in the range of €130 (Mostert et al., 2018) to €257 (Raboisson et al., 2015) per case and year in Europe.Of the total loss of ketosis, milk production loss is one of the most disastrous consequences.Ketosis results in a substantial loss of milk yield in early lactation (Duffield et al., 2009).A study involving 2365 cows showed a 1.5 to 2.4 kg/d per cow reduction in daily milk production in ketotic cows compared with normal cows during the first 2 weeks after calving (Chapinal et al., 2012).Mostert et al. (2018) estimated that the cost related to reduced milk production averaged 32 €/yr, which was about a quarter of the total cost of ketosis.Gohary et al. (2016) reported that the milk production loss related to ketosis was $44 per case, representing 22% of the total cost per ketosis case.However, the molecular mechanism for lower milk production in ketotic cows is not fully elucidated yet.Therefore, investigations into these mechanisms will greatly reduce the cost of ketosis and increase the profitability of dairy farming.
Oxidative stress (OS) refers to deviation from oxidation-reduction (redox) homeostasis; it develops when oxidants overwhelm antioxidants (Sies, 2015).The common biomarkers of OS include reactive oxygen species (ROS), malondialdehyde (MDA), glutathione peroxidase (GSH-Px), catalase (CAT), superoxide dismutase (SOD) and others.Biomarkers of oxidants significantly increase in cows with ketosis, but those of antioxidants reduce in the blood of cows with ketosis (Yue et al., 2022).Indeed, there are elevated H 2 O 2 content, MDA, and ROS but declined GSH-Px, CAT, and SOD activities in mammary tissues of cows with ketosis (Song et al., 2021), indicating serious OS occurs in these milk production organs.Oxidative stress will lead to disruption of normal redox signaling and molecular damage in living cells, and if not corrected it is likely to cause cell death, for instance, apoptosis.It is well-defined that OS will result in apoptosis in bovine mammary epithelial cells (BMEC) with upregulated expressions of the proapoptotic molecules Bax, Caspase 3/9 and downregulated the antiapoptotic molecule Bcl-2 (Wang et al., 2019b).Conversely, incubation with antioxidants alleviates apoptosis in BMEC (Wang et al., 2019a, Cheng et al., 2021;Meng et al., 2022).
Sirtuin 3 (SIRT3) is a nicotinamide adenine dinucleotide (NAD + )-dependent deacetylase in mitochondria and constitutes the predominant regulator of energy metabolism and redox homeostasis in nonruminant cells (Zhang et al., 2020a).There has been accumulating evidence that SIRT3 is possibly a therapeutic target for perinatal diseases in dairy cows.The mRNA abundance of SIRT3 decreases after calving in the liver of dairy goats, and it is positively related to serum SOD activity (Liu et al., 2018).SIRT3 expression is lower in cows with fatty liver, and it turns out that SIRT3 expression is downregulated by high nonesterified fatty acids (NEFA) in calf hepatocytes; SIRT3 ameliorates fatty acid metabolism in response to high NEFA in calf hepatocytes by regulating gene expression (Liu et al., 2020).The protein level of SIRT3 in mammary glands is lower in cows with ketosis compared with healthy ones (Liu et al., 2020), suggesting a correlation between SIRT3 and local pathophysiological processes stemming from ketosis.SIRT3 inhibits NF-kappa B signaling and protects from inflammatory injury in fatty acidchallenged BMEC (Liu et al., 2021b).However, the connection between SIRT3 and apoptosis has not been evaluated directly in the mammary glands of dairy cows and BMEC.Also in nonruminants, SIRT3 activation is subject to the control of AMP-activated protein kinase (AMPK) signaling (Chen et al., 2021).We thus hypothesized that SIRT3 will mitigate apoptosis induced by OS through AMPK signaling in BMEC.Here, we compared the apoptosis and AMPK-SIRT3 axis status in mammary tissues between ketotic and healthy cows.Then the role of SIRT3 on apoptosis was investigated by adenovirus (Ad)-mediated overexpression in OSinduced BMEC, and the mechanism for SIRT3′s putative function was further studied by pharmacological activation/blocking of the AMPK signaling pathway.

Mammary Gland Tissue Samples
The protocol including animals was approved by the Institutional Animal Care and Use Committee at Hunan Agricultural University (No. 2021084, Changsha, China) and Jilin University (No. SY202012019, Changchun, China).Mammary gland tissue samples were reused from a previous study including 15 dairy cows with ketosis and 15 healthy dairy cows.All cows were similar in the number of lactations and days in milk.Cows were defined to be ketosis if serum BHB ≥ 3 mM but free of comorbidities by inspection from licensed veterinarians.Healthy cows were those free of symptoms and blood concentrations of BHB, NEFA and glucose in a normal range.Both sick and healthy cows were from the same farm and on the same diet.The details of basic descriptions of animals and their diet compositions could be found elsewhere (Sun et al., 2019;Liu et al., 2021b).
Incubation with H 2 O 2 and MitoQ.To investigate the effect of OS on apoptosis of MAC-T, we treated cells with different concentrations of H 2 O 2 (7722-84-1, Sigma-Aldrich).MAC-T were seeded at the number of 1 × 10 5 per mL to culture flasks.H 2 O 2 was dissolved in medium B (medium A without FBS supplementation) at 0, 0.3, 0.6, and 0.9 mM.Cells were incubated in these media for 6, 12, and 24 h.Cells were collected after incubation for further analysis.
An antioxidant, Mitoquinone mesylate (MitoQ), was introduced to further study the effect of OS on apoptosis.MitoQ (HY-100116A, MedChemExpress) was dissolved in DMSO (D8371, Solarbio) to prepare a 10 mM stock solution.The stock solution was further serially diluted in medium B to make the final concentration of MitoQ 1 μM.MAC-T were seeded (1 × 10 5 cells/mL) in 6-well plates and allowed to grow to a confluency of 70% to 80%.Then cells were submitted to a 30 min incubation in medium B containing 1 μM MitoQ, followed by treatment with 0.6 mM H 2 O 2 for 6 h.
Adenovirus Infection Assay.To study the role of SIRT3 in OS-induced apoptosis, MAC-T were infected with SIRT3 overexpression adenovirus (Ad-SIRT3) before H 2 O 2 treatment.Both Ad-SIRT3 and the control adenovirus vector (Ad-GFP) were products from Hanbio as in our previous study (Liu et al., 2021b).The adenovirus infection in MAC-T was guided by the vendor's instructions.The titer of adenovirus before administration was > 1 × 10 10 pfu/mL.MAC-T were seeded into cell culture flasks (25 cm 2 ) at a density of 1.5 × 10 6 cells/mL.When cell confluence reached 50%, MAC-T were infected with Ad-SIRT3/Ad-GFP at 50 multiplicity of infection for 6 h in medium B. Then cells were incubated in medium A for another 36 h.Subsequently, they were subject to 0.6 mM H 2 O 2 treatment for 6 h.Cells were collected by trypsin digestion and stored at −80°C for further use within 1 wk.
AMPK Inhibitor Assay.The AMPK inhibitors Compound C (Cd C, HY-13418A, MedChemExpress) were diluted in medium B to make its final concentration 10 μM.MAC-T were seeded (1 × 10 5 cells/mL) in 6-well plates and incubated at 37°C.When reaching 70% to 80% confluence, cells were incubated in medium B containing 10 μM Cd C for 30 min, followed by 0.6 mM H 2 O 2 exposure for 6 h.Cells were collected by trypsin digestion and stored at −80°C for further use within 1 wk.
Cell Viability Assay.Cell viability assays were performed according to the manual of a commercial cell counting kit (CCK-8, CA1210, Solarbio).After treatments, cells were washed 3 times with medium B. A 100 μL of 10% CCK-8 solution was added and incubated at 37°C for 3 h.Absorbance was measured at 450 nm.Cell viability was expressed as percent viability relative to the blank control, which was set at 100%.The intracoefficient of variation (CV) of this kit was <10%, and the inter-CV was <15%.
Quantitative Reverse Transcription PCR Assay.The Minimum Information for Publication of Quantitative Real-Time PCR Experiment (MIQE) guidelines were followed for qPCR assays (http: / / rdml .org/miqe .html).Total RNA from MAC-T was extracted using Trizol (15596026, Invitrogen) according to the manufacturer's instructions.The RNA concentration and quality were determined using K5500 MicroSpectrophotometer (Beijing Kaiao Technology) and electrophoresis (1% agarose gels).The OD260/OD280 ratio of the RNA extracts ranged from 1.8 to 2.0.RNA (1 μg) was reverse-transcribed to cDNA using an Evo M-MLV Mix Kit with gDNA Clean for qPCR (AG11728, Accu-rate Biology) by following the manufacturer's instructions.The mRNA abundance was detected using an SYBR Green Premix Pro Taq HS qPCR Kit (AG11728, Accurate Biology) with a 7500 Real-Time PCR System (Applied Biosystems).The reaction system contained 10 μL 2 × SYBR Green Pro Taq HS Premix, 0.4 μL forward primer (final concentration at 0.2 μM), 0.4 μL reverse primer (final concentration at 0.2 μM), 1 μL cDNA templates and 7 μL RNase Free dH 2 O. Conditions were as follows: initial denaturation at 94°C for 30 s, 40 cycles of amplification (denaturation at 94°C for 5 s, annealing at 60°C for 30 s, and extension at 60°C for 1 min) and extension at 72°C for 5 min.Based on the gene information provided by the GenBank database, the coding region sequence of mRNA of target genes was downloaded, and specific primers were designed according to qPCR primer design principles and Oligo 7.0 software.Primer sequences were as follows, SIRT3: F: GCTAGGTTCCTG CTGCATCT, R: GATGAG-GTCCTGGATGTCGT; β-actin: F: TCACCAACTGG GACGACA, R: GCATACAGGGACAGCACA; GAPDH: F: CACAGTCAAGGCAGAGAACG; R: TACTCAGCACCAGCATCACC.These primers were synthesized by BGI.The cycles-to-threshold values of GAPDH and ACTIN did not change upon different treatments.Thus, the relative expression of target genes was normalized to GAPDH and ACTIN.The relative expression of target genes was determined by the 2 -ΔΔCT method.The qRT-PCR experiments were performed in triplicate.
Terminal Deoxynucleotidyl Transferase-mediated Nick End Labeling Assay.To visualize the apoptosis status, we used a One-step Terminal Deoxynucleotidyl Transferase-mediated Nick End Labeling (TUNEL) Apoptosis Assay Kit (C1090, Beyotime).First, MAC-T after treatments were fixed using 4% paraformaldehyde for 30 min at room temperature and permeabilized with 0.3% Triton X-100 (BS084, Biosharp) in phosphate buffer saline (PBS) for 5 min at room temperature.Next, cells were incubated with TUNEL working solution for 1 h at 37°C.Finally, cells were washed with PBS.Images were taken and analyzed using ZEISS AxioVert.A1 (ZEISS).TUNEL assays were repeated 3 times, and each time the same area was cut for fluorescence intensity with ImageJ (Version 1.8.0-172, https: / / imagej .nih.gov/ij/ ).
Measurement of Cellular ROS and Manganese Superoxide Dismutase Enzyme Activity.Cellular ROS were measured according to the instructions of the ROS Fluorometric Assay Kit (E-BC-K138-F, Elabscience Biotechnology).The excitation wavelength was set at 488 nm, and photography was taken using Leica Microsystems (Leica).ImageJ software (https: / / imagej The intra-CV of both kits was <10%, and the inter-CV was <15%.

Statistical Analysis
Data were expressed as mean ± standard error.For in vivo studies, data from western blotting were normally distributed and analyzed with paired t-tests (n = 15).Linear and quadratic contrasts were run to determine the dose response of increasing concentrations of H 2 O 2 on SIRT3 protein abundance and apoptosis status in MAC-T (n = 3).For the AMPK inhibitor assay, the 2-way ANOVA test was used for significance analysis of effects of H 2 O 2 and Cd C and their interaction (n = 3).For other in vitro assays, the one-way ANOVA test was used for significance analysis with subsequent Bonferroni correction (n = 3).GraphPadPrism8.0.1 was used for the statistical analysis.A P < 0.05 indicated that there was a statistical difference.

Apoptosis Status in Mammary Gland Tissues of Ketotic Cows
The western blotting image showed that dairy cows with ketosis possessed higher protein abundances of Bax and Caspase-3 but a lower abundance of Bcl-2 compared with healthy controls (Figure 1A).Further semiquantitative analysis confirmed that ketotic cows had significantly elevated abundances of Bax, Caspase-3 and Bax/Bcl-2 ratio but declined abundance of Bcl-2 (Figure 1B-E, P ≤ 0.008).

Effects of H 2 O 2 on Apoptosis in MAC-T
To obtain an adequate concentration of H 2 O 2 for the induction of OS in MAC-T, we conducted a doseresponse assay.Cell viability decreased linearly and quadratically with the increasing H 2 O 2 concentrations (Figure 2A, P ≤ 0.011).Specifically, a dose of 0.6 mM H 2 O 2 started to cause a significant reduction of cell viability compared with blank controls (Figure 2A, P ≤ 0.05).The western blotting image revealed that H 2 O 2 treatments gradually increased protein levels of Bax and Caspase-3 but decreased levels of Bcl-2 compared with blanks (Figure 2B).The semiquantitative analysis demonstrated that H 2 O 2 treatments dose-dependently upregulated the abundance of Bax, Caspase-3 and Bax/Bcl-2 ratio but downregulated the abundance of Bcl-2 (Figure 2C-F, Linear P < 0.001).TUNEL assays indicated that cellular fluorescence intensity raised with the increasing concentrations of H 2 O 2 , displaying both linear and quadratic manners (Figure 2G-H, P ≤ 0.008).

Effects of MitoQ on Cellular ROS and Apoptosis in MAC-T
Compared with the H 2 O 2 group, additional MitoQ pretreatment resulted in lower green fluorescence intensity (Figure 3A), suggesting a lower content of cellular ROS.Further quantitative analysis indicated that the percentage of positive cells in the H 2 O 2 + MitoQ was significantly lower than that in the H 2 O 2 group (Figure 3B, P < 0.0001).
According to the TUNEL assays, the H 2 O 2 + MitoQ group had lower red fluorescence intensity (Figure 3C).Quantitative analysis verified the observation, suggesting additional MitoQ pretreatment significantly ameliorated the H 2 O 2 -induced apoptosis (Figure 3D, P < 0.0001).

Effects of SIRT3 on OS-induced Apoptosis in MAC-T
In vivo data revealed that ketotic cows had a lower protein abundance of SIRT3 in mammary glands (Figure 4A-B, P = 0.004).In vitro, incubation with 0-0.9 mM of H 2 O 2 for 6 h gradually decreased mRNA and protein levels of SIRT3 in both linear and quadratic manners (Figure 4C-E, P ≤ 0.002).Additionally, treatments with 0.6 mM H 2 O 2 for 6-24 h significantly re- Western blotting image displayed that the Ad-SIRT3 + H 2 O 2 group had lower protein levels of Bax and Caspase-3 but higher levels of Bcl-2 compared with the Ad-GFP + H 2 O 2 group (Figure 6A).Semiquantitative analysis of protein abundances confirmed the results (Figure 6B-E, P ≤ 0.0008).TUNEL assays further verified that SIRT3 overexpression significantly reduced the red fluorescence density (Figure 6G-H, P < 0.0001).

Effects of AMPK Signaling on OS-induced Apoptosis in MAC-T
Ketotic cows had a lower protein abundance of p-AMPKα and p-AMPKα/AMPKα in mammary glands compared with healthy cows (Figure 7A-B, P = 0.0006).Incubation with H 2 O 2 also led to lower protein abundances of p-AMPKα, p-AMPKα/AMPKα and p-ACC/ACC (Figure 7C-E, P < 0.0001).However, the AMPK inhibitor Cd C significantly reduced SIRT3 protein levels regardless of H 2 O 2 incubation (Figure 7F, P < 0.0001).Interestingly, Cd C pretreatment significantly upregulated protein levels of Bax and Caspase-3 but downregulated the level of Bcl-2 regardless of H 2 O 2 incubation (Figure 7C, G-J, P ≤ 0.0006).
Contrary to the inhibitor, AMPK agonist pretreatment decreased the abundance of Bax, Caspase-3 and Bax/Bcl-2 ratio and increased the abundance of SIRT3 and the activity of MnSOD (Figure 8 A-I, P ≤ 0.0006).However, these effects of the AMPK agonist were abolished upon SIRT3 silencing (Figure 8 A-I).Additionally, TUNEL assays confirmed that AMPK agonist decreased the relative fluorescence density compared with the H 2 O 2 group, but the MK8722 + si-SIRT3 + H 2 O 2 group displayed higher relative fluorescence density compared with its control group (Figure 9 A-B, P ≤ 0.0024).

DISCUSSION
Ketosis is often concomitant with the reduction in milk production.However, the precise mechanism for this has not been fully illustrated.This study showed that ketotic cows had a higher degree of apoptosis and hindered AMPK-SIRT3 signaling in the mammary glands.In H 2 O 2 -induced MAC-T, Ad-SIRT3 infection lowered levels of cellular ROS, Bax, Caspase-3 and Bax/Bcl-2 but increased the level of Bcl-2.Further AMPK inhibitor and agonist assays displayed opposite results in OS-induced apoptosis.Interestingly, these ef- fects of the AMPK agonist were abolished upon SIRT3 silencing.Therefore, our results supported the hypothesis that the OS in mammary glands of ketotic cows led to apoptosis and the AMPK-SIRT3 axis played a key role therein.Systemic OS occurs in ketotic cows.In particular, mammary glands from both clinical and subclinical ketotic cows contained higher contents of H 2 O 2 and MDA but lower activities of SOD, GSH-Px and CAT (Sun et al., 2021).This OS will in turn bring in deteriorating  outcomes in mammary glands.In this study, the higher levels of proapoptotic proteins (Bax and Caspase-3) and the lower level of antiapoptotic protein Bcl-2 indicated that mammary gland tissues underwent greater apoptosis in ketotic cows, consistent with the observations by Sun et al. (2021).Because H 2 O 2 contents increased in mammary glands during ketosis, we used H 2 O 2 -challenged MAC-T to simulate the in vivo situation.The in vitro assays confirmed that OS led to greater apoptosis; further addition of the mitochondria-targeted antioxidants MitoQ prevented MAC-T from apoptosis.Taken together, these data demonstrated that ketotic cows were suffering from OS in mammary glands and this OS was a major contributor to the apoptosis in the mammary parenchymal cells.As such, a greater rate of apoptosis relative to cell renewal is critical to the decreased milk yield (Capuco et al., 2001).Also, apoptosis in the mammary parenchymal cells at early stage of lactation may bring long-term impacts on milk yield in ketotic cows.A recent retrospective cohort study found that hyperketonemia during wk1 was associated with a milk yield reduction of 3.7 kg per cow per day throughout the 300-d lactation (Rodriguez et al., 2022).
SIRT3 is a major regulator of OS in cells of nonruminants (Singh et al., 2018).As a mitochondrial deacetylase, SIRT3 deacetylates key antioxidative enzymes like MnSOD.This posttranslational modification of MnSOD enhances its enzyme activity and maintains a lower cellular ROS level.However, SIRT3 protein abundances were downregulated in the mammary glands of ketotic cows.Also, it is not clear what factor attributes to the lower expression of SIRT3.In vitro assays proved that H 2 O 2 both dose-and time-dependently decreased SIRT3 expression in MAC-T, suggesting OS is a key contributor to the lower SIRT3 expression in mammary gland tissues.In addition, SIRT3 overexpression resulted in lower cellular ROS and higher enzyme activity of MnSOD, confirming that SIRT3 also exerted an antioxidative influence in ruminant cells.Moreover, Ad-SIRT3 infection caused lower levels of proapoptotic proteins and a high level of antiapoptotic protein.This implies that a SIRT3 restoration is attributed to the melioration of OS-induced apoptosis.The TUNEL assays further verified the antiapoptotic effect of SIRT3.The antiapoptotic role of SIRT3 in MAC-T is consistent with the studies in hypoxia/reoxygenation-induced Caco-2 cells (Wang et al., 2020), H9c2 cells (Zhai et al., 2017), chondrocytes (Xu et al., 2021) and others (Hu et al., 2021).
AMPK signaling plays a vital role in regulating cellular energy balance in eukaryotic cells (Carling, 2017).AMPK is activated by phosphorylation at Thr172 and then can phosphorylate ACC to p-ACC at Ser79 to partially regulate cellular energy metabolism.ROS are by-products of aerobic metabolism.Hence, there is a close cross-talk between AMPK signaling and ROS (Zhao et al., 2017).Interestingly, AMPK is in charge of the activation of SIRT3 in nonruminant cells (Yu et al., 2017;He et al., 2021).We thus speculated that the AMPK-SIRT3 axis would regulate the OS-induced apoptosis in BMEC.As expected, ketotic cows had suppressed AMPK signaling in mammary glands.In vitro assays verified that incubation with H 2 O 2 inhibited AMPK signaling because of the lower p-AMPKα/ AMPKα and p-ACC/ACC.Notably, incubation with the AMPK inhibitor and agonist displayed opposite effects on apoptosis.The observation that AMPK activation increased SIRT3 expression but AMPK inhibition did the opposite was consistent with others (Guo et al., 2022).Particularly, the antiapoptotic effect of AMPK agonist was erased upon SIRT3 silencing.These data clearly demonstrated that the activation axis would ameliorate the OS-induced apoptosis in BMEC.
In addition to elevated levels of H 2 O 2 , the increasing concentration of NEFA, the negative energy balance biomarker, also induces apoptosis in BMEC.Mechanically, high concentrations of NEFA led to unbalanced redox and uncontrolled inflammation.Intriguingly, SIRT3 emerges to be an antagonist to the lipotoxicity of high NEFA because SIRT3 has been shown to attenuate both OS and excess inflammatory responses elicited by high NEFA (Liu et al., 2021b).Similarly, SIRT3 has been confirmed to protect hepatocytes from palmitic acid and oleic acid mixture-induced lipotoxicity in mice (Zhang et al., 2020b, Liu et al., 2021a).These observations highlight a potential therapeutic role of SIRT3 on metabolic disorders implicated in the negative energy balance and OS.However, thus far there have been no in vivo studies targeting SIRT3 in dairy cows.Consequently, further investigations are needed to clarify the efficacy of SIRT3.

CONCLUSIONS
In summary, in vivo data suggested that ketosis is associated with enhanced apoptosis and blunt AMPK-SIRT3 signaling in mammary glands.In vitro data indicated that SIRT3 mitigates oxidative-stress-induced apoptosis in BMEC via AMPK signaling.There may be potential benefits for targeting the AMPK-SIRT3 axis to help prevent the negative effects of mammary glands during ketosis.
Liu et al.: SIRTUIN 3 AND APOPTOSIS IN MAMMARY CELLS .nih.gov/ij/ download .html)was used to analyze the fluorescence density, and the relative mean optical density of each treatment group was calculated based on the positive control group.The manganese superoxide dismutase (MnSOD) enzyme activity was measured with a CuZn/MnSOD activity kit (S0103, Beyotime).