Met stimulates ARID1A degradation and activation of the PI3K-SREBP1 signaling to promote milk fat synthesis in bovine mammary epithelial cells

Abstract Methionine (Met) can promote milk fat synthesis in bovine mammary epithelial cells (BMECs), but the potential molecular mechanism is largely unknown. In this report, we aim to explore the role and molecular mechanism of AT-rich interaction domain 1A (ARID1A) in milk fat synthesis stimulated by Met. ARID1A knockdown and activation indicated that ARID1A negatively regulated the synthesis of triglycerides, cholesterol and free fatty acids and the formation of lipid droplets in BMECs. ARID1A also negatively regulated the phosphorylation of PI3K and AKT proteins, as well as the expression and maturation of SREBP1. Met stimulated the phosphorylation of PI3K and AKT proteins, as well as the expression and maturation of SREBP1, while ARID1A gene activation blocked the stimulatory effects of Met. We further found that ARID1A was located in the nucleus of BMECs, and Met reduced the nuclear localization and expression of ARID1A. ARID1A gene activation blocked the stimulation of PI3K and SREBP1 mRNA expression by Met. In summary, our data suggests that ARID1A negatively regulates milk fat synthesis stimulated by Met in BMECs through inhibiting the PI3K-SREBP1 signaling pathway, which may provide some new perspectives for improving milk fat synthesis.


Introduction
Milk is mainly composed of water, fat, lactose, whey protein and minerals, which can provide essential nutrients for growth and development. 1,2The health and nutritional benefits of edible cow milk include promoting growth, enhancing immune function, lowering blood pressure, preventing gastrointestinal infections, regulating the intestinal microbiota and improving body function. 3,4Milk fat is one of the most important and energy-rich substances in milk and is also an important indicator to evaluate the milk quality. 5As a raw material for processing butter, cheese and whole milk powder, it has a high economic value. 6Triglycerides account for 98-99% of milk fat, while the remaining 1-2% are cholesterol, free fatty acids, monoglyceride and phospholipids. 7terol regulatory element binding proteins (SREBPs) are a family of transcription factors that regulate lipid homeostasis by controlling the expression of a series of enzymes required for cholesterol, fatty acids, triglycerides and phospholipid synthesis. 8,9The SREBPs family includes three subtypes: SREBP-1a, SREBP-1c and SREBP-2.SREBP-1a and SREBP-1c are formed by transcription of different promoters of the SREBF1 gene.SREBP1 has been proven to be a key factor in regulating fat synthesis. 10,11SREBP1 is synthesized as an inactive precursor (full length of SREBP1, FL-SREBP1).SREBP cleavage activating protein (SCAP) transfers it from the endoplasmic reticulum to the Golgi apparatus, where FL-SREBP1 is cleaved and activated to form nuclear SREBP1 (N-SREBP1) to regulate gene expression. 12n addition to being the substrate for protein synthesis, amino acids are also crucial maternal nutrients that provide potential energy for the growth and development of newborns. 13Some functional amino acids such as methionine (Met) can regulate milk fat synthesis through different cellular pathways. 13,14REBP1 is a key factor in amino acid regulation of milk fat synthesis. 15It has been reported that CRTC2 mediates Met and Leu to regulate the SREBP1 signaling pathway to promote milk fat synthesis. 16Lys stimulates SREBP1 expression and maturation through the GPRC6A-FABP5 signaling pathway to regulate milk fat synthesis. 17Furthermore, recent reports have shown that amino acids and hormones can regulate SREBP1 expression and maturation through the PI3K signaling pathway. 18,19However, the underlying mechanisms and signaling pathways by which amino acids regulate milk fat synthesis are largely unknown.
2][23][24][25] The protein stability of ARID1A is an important factor affecting its function of chromatin remodeling and transcriptional regulation.Our previous research found that Met degraded ARID1A protein through the E3 ubiquitin ligase TRIM21 to promote mTOR expression and milk synthesis in mammary epithelial cells (MECs). 26In addition, increasing reports have indicated that ARID1A regulated the PI3K-AKT signaling pathway to perform its biological functions. 21,27It is still unknown whether ARID1A can play a role in Met stimulation on the SREBP1 signaling pathway and milk fat synthesis in MECs.
In this study, we investigated the role and molecular mechanism of ARID1A in Met-stimulated milk fat synthesis in bovine MECs (BMECs), building on previous studies.

Cell culture, identification and treatment
According to our previous research methods, BMECs were isolated and cultured from the tissues of dairy cows at lactation. 19All animal experiments involving cows were approved and conducted according to the regulations and guidelines of the Animal Experiment Committee of Foshan University.In short, 1 mm 3 small pieces of mammary gland tissue were inoculated in cell culture bottles (CLS430639, Corning, New York, USA) pretreated with rat tail collagen, and cells were grown for passage in basal DMEM/F12 (11320082, Gibco, California, USA) medium contained 10% fetal bovine serum, penicillin (100 U/mL) and streptomycin (0.1 mg/mL).Cells were used for subsequent experiments.Before the experiment, CK18 and b-casein subcellular localization were detected by immunofluorescence observation to determine the purity of BMECs.To detect the effects of Met, the growth medium was replaced with serum-free OPTI-MEMI (31985070, Gibco) medium for 12 h when BMECs were grown to 50% confluence.OPTI-MEMI medium is an improved minimum necessary medium for cell culture, which is a unique medium containing transferrin, hypoxanthine, thymidine and trace elements and very suitable for cationic liposome transfection and detection of nutrient effects.Then 0.6 mM Met was added into the culture medium (OPTI-MEMI medium contained 0.1 mM of Met).Cells were cultured for 24 h at 37 � C in a humidified atmosphere of 95% air and 5% CO 2 after treatments and collected for detection.

Construction and transfection of ARID1A gene activation vector
CRISPR/dCas9 system was used to construct ARID1A gene activation vector, to analyze the effects of ARID1A gene activation.We have previously successfully constructed the recombinant vectors targeting ARID1A, and verified that transfection with recombinant vector pSPgRNA-85 and SP-dCas9-VPR (hereinafter referred to as VPR) can significantly increase ARID1A expression. 26Recombinant vector pSPgRNA-85 and pSPgRNA-970 (as a negative control) were selected for the experiment in this study.The DNA sequences of these sgRNAs were: ARID1A-85: 5 0 -GGCAGCGTGAACTTTACTCG-3 0 , ARID1A-970: 5 0 -GCGAGCGCAGCGCAAAAGCCG-3 0 .Lipofectamine 3000 was used to transfect these plasmids into BMECs for 48 h to increase ARID1A expression, according to the manufacturer's instructions.
The efficiency of the primers utilized in this study was assessed through standard curve experiments, and all primers exhibited an efficiency ranging from 90% to 100%.The b-actin mRNA was used as an internal control, and qRT-PCR results were obtained using the 2 -DDCt calculation method.

Detection of triglycerides content
The content of triglycerides (TGs) in BMECs was detected using a TGs assay kit (E1013, Applygen, Beijing, China).The content of TGs was corrected in units of cell number.Cell number in each experimental group was counted by an automated cell analyzer, as previously reported. 28The final result showed the concentration of TGs in 10 5 cells.

Observation of lipid droplets
BMECs were grown on coverslips in a six-well plate (CLS3414, Corning) to 30 − 50% confluence.Cells were fixed with 4% paraformaldehyde in PBS and incubated with 1 mg/mL BODIPY 493/503 (D3922, Invitrogen, California, USA) for 20 min, then washed with PBS three times, cell nuclei were dyed with DAPI (C1002, Beyotime) for 10 min.Lipid droplets labeled with a green fluorescent signal on the coverslips were observed with a laser confocal microscopee (TCS-SP2 AOBS, Leica, Heidelberg, Germany).The area-integrated optical density (AIOD) of lipid droplets was analyzed by ImageJ.

Cholesterol content detection
The content of total cholesterol (TCs) in BMECs was detected using a TCs assay kit (E1015, Applygen), according to the manufacturer's instructions.Similar to TGs content detection, the content of total cholesterol was corrected based on cell counting.

Detection of secretion of free fatty acids
The amount of free fatty acids (non-esterified fatty acids, NEFA) secreted by BMECs into the medium was detected using a NEFA assay kit (A042-2-1, Jiancheng, Nanjing, China), according to the manufacturer's instructions.

Separation of nucleus and cytoplasm
Nucleus and cytoplasm of BMECs were separated by the Nuclear and Cytoplasmic Protein Extraction Kit (P0027, Beyotime).b-tubulin was used as a reference marker in cytoplasm, and lamin B1 was used as an internal reference marker in nucleus, to test the purity of cytoplasmic and nuclear proteins of BMECs.

Statistical analysis
All experimental data are expressed as mean ± SEM from three independent experiments.Statistical analyses were conducted by ANOVA with the software of SPSS 19.0 (IBM).The Tukey post hoc test was used as a post hoc test following ANOVA.Statistical significance was declared at p < .05.

ARID1A negatively regulates milk fat synthesis in BMECs
The expression of CK18 and b-casein in cells were identified by immunofluorescence observation to detect the purity of BMECs.The results showed that CK18 and b-casein were almost expressed in all of the observed cells, indicating that the purity of BMECs is high and they can be used for subsequent experiments (Supporting Information Fig. S1A,B).To investigate whether ARID1A is a key influencing factor in milk fat synthesis, a siRNA targeting ARID1A was transfected into BMECs and incubated for 24 h to knockdown the expression of ARID1A (Supporting Information Fig. S1C,D).The results showed that the silencing efficiency of siRNA sequences was approximately 75% (Supporting Information Fig. S1D) and approximately 80% (Supporting Information Fig. S1E), respectively.
The content of TGs (Supporting Information Fig. S1F), lipid droplet formation (Supporting Information Fig. S1G,H), TCs content (Supporting Information Fig. S1I) in cells and NEFA content in the culture medium (Supporting Information Fig. S1J) were significantly increased in ARID1A knockdown cells.The effects of ARID1A gene activation on milk fat synthesis were also observed.Transfection of the pSPgRNA-85 vector and VPR significantly increased the expression of ARID1A (Supporting Information Fig. S2A,B).The content of TGs (Supporting Information Fig. S2C), lipid droplet formation (Supporting Information Fig. S2D,E), TCs content in cells (Supporting Information Fig. S2F) and NEFA content in the culture medium (Supporting Information Fig. S2G) were significantly reduced in ARID1A gene activation cells.Taken together, these results suggest that ARID1A negatively regulates milk fat synthesis in BMECs.
The effects of ARID1A gene activation and gene knockdown on the PI3K signaling pathway were also observed.AKT is a main downstream effector of PI3K, and ARID1A knockdown significantly increased the ratios of p-PI3K/PI3K (Fig. 1A,C) and p-AKT/ AKT (Fig. 1A,D).Conversely, ARID1A gene activation significantly reduced the ratios of p-PI3K/PI3K (Fig. 2A,C) and p-AKT/AKT (Fig. 2A,D).Previous reports have shown that PI3K is a key factor in regulating SREBP1 expression and maturation. 19,28,29In summary, these results suggest that ARID1A negatively regulates SREBP1 expression and maturation through inhibiting PI3K activation.

ARID1A is a key negative regulator of Met stimulation on the PI3K-SREBP1 signaling
Previous reports have demonstrated that Met plays a crucial regulatory role in milk production, and the addition of Met (0.6 mM) for 24 h can significantly increase milk protein and fat synthesis in BMECs and cell proliferation. 30To further investigate the effects of ARID1A on Met-stimulated SREBP1 expression and maturation, the ARID1A gene activation vector was transfected into BMECs, and Met (0.6 mM) was added to the Opti-MEM medium for 24 h (37 � C).The results showed that Met significantly decreased the protein levels of ARID1A (Fig. 4A,B), but significantly increased the ratios of p-PI3K/PI3K (Fig. 4A,C) and p-AKT/AKT (Fig. 4A,D) and the protein levels of FL-SREBP1 (Fig. 4A,E) and N-SREBP1 (Fig. 4A,F).ARID1A gene activation almost totally blocked Met stimulated protein levels of p-PI3K, p-AKT, FL-SREBP1 and N-SREBP1 (Fig. 4A-F).These results suggest that ARID1A is a key negative regulator of Met stimulation on the PI3K-SREBP1 signaling pathway.

Met stimulate ARID1A protein degradation to promote SREBP1 mRNA expression and milk fat synthesis
Our previous research results have shown that Met can induce the degradation of ARID1A protein through the ubiquitin proteasome pathway. 26In this study, we found that ARID1A was located in the nucleus of BMECs, and adding Met significantly reduced the nuclear localization and protein levels of ARID1A (Fig. 5A,B).Met stimulated the synthesis of TGs, lipid droplet formation, increased TCs content and NEFA content in BMECs, but these effects were blocked by ARID1A gene activation (Fig. 5C-G).The stimulation of Met on the mRNA levels of PI3K and SREBP1 were also blocked by ARID1A gene activation (Fig. 6A-C).These results demonstrate that ARID1A inhibits Met-stimulated milk fat synthesis through  decreasing PI3K and SREBP1 mRNA expression, Met can function through decreasing ARID1A protein level (Figure 7).

Discussion
ARID1A plays an important role in regulating gene transcription as a chromatin remodeling protein.However, it is unclear whether ARID1A can regulate the SREBP1 pathway and milk fat synthesis.In this study, we demonstrate that ARID1A is a negative regulator of milk fat synthesis and negatively regulates SREBP1 expression and maturation through inhibiting PI3K activation.ARID1A gene activation blocks Met-stimulated milk fat synthesis, PI3K activation and SREBP1 expression and maturation, and PI3K and SREBP1 mRNA expression.ARID1A is located in the nucleus of BMECs, and the addition of Met decreases the expression of ARID1A protein.
Research on the function of ARID1A gene has found that ARID1A negatively regulated the synthesis of TGs, TCs and NEFA and the formation of lipid droplets in BMECs, and ARID1A also negatively regulated the expression and maturation of SREBP1.SREBP1 is a key transcription factor that regulates milk fat synthesis.Recent reports have shown that ARID3A of the ARID family is a positive regulator of SREBP1 mRNA expression and protein maturation. 28t is speculated that the ARID family may play a unique role in regulating lipid homeostasis.ARID1A is a chromatin remodeling factor, previous reports have shown that ARID1A can affect fatty acid oxidation by regulating the epigenetics of PPARa and some metabolic genes or directly combining with the promoter of genes that regulate fatty acid synthesis. 24,25RID1A shares a binding site on the mTOR promoter with the catalytic subunit of the chromatin remodeling SWI/SNF complex BRG1, and BRG1 has been reported to bind to the SREBP1 promoter at −3546 to −3940 bp. 26,31Therefore, we speculate that ARID1A might bind to the SREBP1 promoter to promote SREBP1 gene transcription.PPARa is a key mediator in lipid metabolism, 32 and whether ARID1A affects milk fat synthesis by regulating PPARa is worth considering in future research.The specific binding site of ARID1A on the SREBP1 promoter also needs to be identified in future studies.Our data demonstrate that ARID1A is an important negative regulator of milk  fat synthesis.To the best of our knowledge, this is the first report that ARID1A can regulate SREBP1 expression and maturation.
We found that ARID1A knockdown significantly increased the levels of p-PI3K and p-AKT proteins.The PI3K signaling pathway has been proven to be an upstream pathway for regulating SREBP1 signaling. 19,28-Box and WD-40 domain protein 7 (FBXW7) was involved in high glucose-induced SREBP1 expression in renal tubular cells of diabetic nephropathy under the regulation of the PI3K signaling pathway.33 Several recent reports have shown that ARID1A loss can occur simultaneously with PTEN mutations and PIK3CA activation mutations or lead to activation of the PI3K/ AKT signaling by upregulating the PI3K regulatory subunit PIK3R.34,35 These reports provide favorable evidence for our experimental results. Taen together, we consider that ARID1A can negatively regulate SREBP1 expression and maturation through inhibiting the PI3K signaling.
The experimental data showed that ARID1A negatively regulated Met-stimulated milk fat synthesis and PI3K and SREBP1 mRNA expression.Met is a limiting amino acid in milk synthesis, 36 increasing the supply of Met leads to greater milk protein percentage and milk yield. 37,38Previous reports have shown that Met can transmit signals to PI3K through amino acid transporters or G-protein-coupled receptors to stimulate the expression of SREBP1. 30,39Our experimental data also suggest that ARID1A might also affect PI3K gene transcription.ARID1A is located in the nucleus of BMECs, and the addition of Met significantly reduces the nuclear localization and protein levels of ARID1A.Our previous report showed that Met can induce the ubiquitination degradation of ARID1A protein. 26From these previous reports and our experimental data, we consider that Met can promote milk fat synthesis through relieving the inhibitory effects of ARID1A on PI3K and SREBP1 gene expression.

Conclusion
In summary, our research shows that ARID1A is a negative regulator factor of milk fat synthesis in BMECs.Met stimulates ARID1A degradation and mRNA expression and activation of PI3K to stimulate SREBP1 mRNA expression and protein maturation.Our findings provide additional fundamental knowledge to the role of Met in BMECs that can be used to explore feeding strategies to improve milk fat synthesis.

Figure 1 .
Figure 1.Effects of ARID1A knockdown on the PI3K-SREBP1 signaling pathway.(A) Cells were treated as shown in Supporting Information Fig. S1C.Relative protein levels were detected by Western blotting.B, Blank; NC, negative control; KD, Knockdown.(B-F) The protein levels of ARID1A (B), p-PI3K/PI3K (C), p-AKT/AKT (D), FL-SREBP1 (E) and N-SREBP1 (F) in (A) were quantified by ImageJ.Phosphorylated proteins were normalized to their nonphosphorylated counterparts, and others were normalized to b-actin.Data were expressed as mean ± SEM (n ¼ 3).In a bar chart, different lowercase letters in the superscript indicate statistically significant differences between various groups (p < .05).

Figure 2 .
Figure 2. Effects of ARID1A gene activation on the PI3K-SREBP1 signaling pathway.(A) Cells were treated as shown in Supporting Information Fig. S2A.Indicated protein levels were detected by Western blotting analysis.(B-F) The protein levels of ARID1A (B), p-PI3K/PI3K (C), p-AKT/AKT (D), FL-SREBP1 (E) and N-SREBP1 (F) in (A) were quantified by ImageJ.Phosphorylated proteins were normalized to their non-phosphorylated counterparts, and others were normalized to b-actin.Data were expressed as mean ± SEM (n ¼ 3).In a bar chart, different lowercase letters in the superscript indicate statistically significant differences between various groups (p < .05).

Figure 3 .
Figure 3. Effects of ARID1A knockdown and gene activation on the mRNA levels of PI3K and SREBP1.(A-C) Cells were treated as shown in Supporting Information Fig. S1C, the expression of ARID1A (A), PI3K (B) and SREBP1 (C) mRNA were analyzed by qRT-PCR.(D-F) Cells were co-transfected with VPR and pSPgRNA-85, the expression of ARID1A (D), PI3K (E) and SREBP1 (F) mRNA were analyzed by qRT-PCR.All qRT-PCR data were normalized to b-actin.Data were expressed as mean ± SEM (n ¼ 3).In a bar chart, different lowercase letters in the superscript indicate statistically significant differences between various groups (p < .05).

Figure 4 .
Figure 4. Effects of ARID1A gene activation on the PI3K-SREBP1 signaling pathway stimulated by Met.(A) Cells were treated with 0.6 mM Met for 24 h and co-transfected with VPR and pSPgRNA-85, indicated protein levels were detected by Western blotting.(B-F) The protein levels of ARID1A (B), p-PI3K/PI3K (C), p-AKT/AKT (D), FL-SREBP1 (E) and N-SREBP1 (F) in (A) were quantified by ImageJ.Phosphorylated proteins were normalized to their non-phosphorylated counterparts, and others were normalized to b-actin.Data were expressed as mean ± SEM (n ¼ 3).In a bar chart, different lowercase letters in the superscript indicate statistically significant differences between various groups (p < .05).

Figure 5 .
Figure 5. Subcellular localization of ARID1A and the effects of ARID1A gene activation on milk synthesis stimulated by Met.(A) The subcellular location of ARID1A affected by 0.6 mM Met for 24 h was detected by immunofluorescence observation.Nuclei labeled with DAPI, blue; ARID1A, green.Scale bar ¼ 20 lm.(B) Nuclear and cytoplasmic proteins were extracted from BMECs.Western blotting was used to detect the subcellular location of ARID1A.Lamin B1 and b-tubulin were used to display the purities of nucleus and cytoplasm.Nuc, nucleus; Cyt, cytoplasm.(C) Cells were treated as show in Fig. 4A, the content of TGs in culture medium was detected by a TG detection kit.(D) Lipid droplets in cells were observed by BODIPY staining.Nuclei were labeled with DAPI (blue) and lipid droplets were labeled with BODIPY (green).Scale bar ¼ 20 lm.(E) ImageJ was used to quantitatively analyze the fluorescence intensity of lipid droplets in (D), 15 cells per sample were analyzed.(F) The content of TCs in BMECs was detected by a TCs detection kit.(G) The amount of NEFA secreted by BMECs into the medium was detected using a NEFA assay kit.Data were expressed as mean ± SEM (n ¼ 3).In a bar chart, different lowercase letters in the superscript indicate statistically significant differences between various groups (p < .05).

Figure 6 .
Figure 6.Effects of ARID1A gene activation on PI3K and SREBP1 mRNA expression stimulated by Met.(A-C) Cells were treated as show in Fig. 4A, the expression of ARID1A (A), PI3K (B) and SREBP1 (C) mRNAs were analyzed by qRT-PCR.All qRT-PCR data were normalized to b-actin.Data were expressed as mean ± SEM (n ¼ 3).In a bar chart, different lowercase letters in the superscript indicate statistically significant differences between various groups (p < .05).

Figure 7 .
Figure 7. Schematic diagram of the role and molecular mechanism of ARID1A in Met -stimulated milk fat synthesis in BMECs.Met promotes milk fat synthesis through inducing ARID1A degradation, which leads to increased PI3K and SREBP1 mRNA expression and subsequent protein activation.