KRAS Affects the Lipid Composition by Regulating Mitochondrial Functions and MAPK Activation in Bovine Mammary Epithelial Cells

Simple Summary Milk fat is mainly composed of triglycerides, which are synthesized in the endoplasmic reticulum of mammary epithelial cells. Many uncertain regulatory factors and mechanisms are involved in this process. In this study, we found that inhibition of the Kirsten rat sarcoma viral oncogene homolog changed the lipid composition by regulating mitochondrial functions, autophagy, endoplasmic reticulum stress, and mitogen-activated protein kinase in a dairy cow mammary epithelial cell line. These results are helpful for exploring the synthesis and secretion of milk fat at the molecular level and providing a theoretical basis for improving the flavor and nutrition of milk. Abstract Kirsten rat sarcoma viral oncogene homolog (KRAS), or guanosine triphosphatase KRAS, is a proto-oncogene that encodes the small guanosine triphosphatase transductor protein. Previous studies have found that KRAS can promote cytokine secretion, cell chemotaxis, and survival. However, its effects on milk fat synthesis in bovine mammary epithelial cells are unclear. In this study, the effects of KRAS inhibition on cell metabolism, autophagy, oxidative stress, endoplasmic reticulum stress, mitochondrial function, and lipid composition as well as the potential mechanisms were detected in an immortalized dairy cow mammary epithelial cell line (MAC-T). The results showed that inhibition of KRAS changed the lipid composition (especially the triglyceride level), mitochondrial functions, autophagy, and endoplasmic reticulum stress in cells. Moreover, KRAS inhibition regulated the levels of the mammalian target of rapamycin and mitogen-activated protein kinase (extracellular regulated protein kinases, c-Jun N-terminal kinases, p38) activation. These results indicated that regulation of KRAS would affect the synthesis and composition of milk fat. These results are also helpful for exploring the synthesis and secretion of milk fat at the molecular level and provide a theoretical basis for improving the percentage of fat in milk and the yield of milk from cows.


Introduction
Among the livestock species that produce milk around the world, cows have always been the dominant species. Milk is rich in nutrition, and in addition to providing the essential nutrients for the growth and development of newborns, milk contains all the amino acids required by humans and an abundance of minerals, vitamins, milk protein, affect adipogenesis by regulating mitogen-activated protein kinase (MAPK) in various cell types [24,25]. The MAPK signaling pathway is a key pathway in milk fat synthesis and the entire lactation process [26,27]. Therefore, we predicted that KRAS also affects milk fat synthesis in cows.
MAC-T, the immortalized bovine mammary epithelial cell line, is a cell model for researching mammary developmental biology, mammary pathology, mammary bioreactors, and lactation engineering [28][29][30]. It can also be used to explore the effects of functional genes [31,32] and compounds [33] on the development of mammary components and milk fat synthesis. In this study, we investigated the effects of KRAS inhibition on autophagy, mitochondrial functions, oxidative stress, ER stress, cell metabolites, lipid composition, and MAPK activation in MAC-T cells. This study is helpful for understanding the synthesis and secretion of milk fat and provides a theoretical basis for improving the percentage of fat in milk and the yield of milk from cows. At the same time, the results may also provide new directions for an in-depth understanding of the prevention, occurrence, and therapy of ketosis in dairy cows.

Materials and Methods
All reagents and chemicals were purchased from Sigma-Aldrich (St. Louis, MO, USA), unless indicated otherwise.

Cell Transfection
In brief, MAC-T cells were plated in 12-well plates (NEST Biotechnology, Wuxi, China). Then, the culture medium was replaced with DMEM/F12. Lipofectamine 2000 Transfection Reagent (Invitrogen, Rochester, NY, USA)/siRNA (GenePharma, Suzhou, China) complexes were added at 1:2 ratios to the plates. The final concentration of siRNAs was 80 nM in DMEM/F12 medium. After 6 h, the medium was changed to DMEM/F12 with 10% fetal bovine serum, and the culture plates were incubated for 18 h. The siRNA sequences are shown in Supplementary Table S1.

Autophagy Inhibition by Chloroquine
In brief, MAC-T cells were cultured in 12-well plates. Then, the cells were treated with 50 µM chloroquine (Selleck, dissolved in DMSO) for 24 h when they reached 70% confluence. Next, the autophagy levels were detected by Western blot and immunofluorescence assays.

Apoptosis Detection
Cell apoptosis was detected with flow cytometry (Beckman Coulter, Brea, CA, USA) according to the instructions provided in Annexin V-FITC apoptosis analysis kit (Tianjin Sungene Biotech Co., Ltd., Tianjin, China). The details of the materials and methods are shown in the Section S1.

RNA Extraction and Quantitative Real-Time Reverse Transcription Polymerase Chain Reaction (qRT-PCR)
In brief, total RNA of MAC-T cells was extracted by using TRIzol Reagent (Life Technologies, Carlsbad, CA, USA). Gene expression was quantified by the Mastercycler ep realplex (Eppendorf, Hamburg, Germany) and 2 − CT method with β-ACTIN as the standard. The details of the methods and the primer sequences are shown in the Section S1 and Table S2.

Protein Separation and Western Blot Analysis
In brief, MAC-T cells were trypsinized and collected. Then, RIPA buffer (Solarbio, Beijing, China) containing 1% PMSF (100 mM, Solarbio) was used for lysis. The GAPDH loading controls were performed on the same membrane unless indicated otherwise. The details of the methods and antibodies used in the study are shown in the Section S1 and Table S3.

Immunofluorescence
In brief, a total of 1 × 10 5 cells were seeded on 12-well plates containing cell culture slides (NEST Biotechnology). Twenty-four hours after transfection, we performed immunofluorescence detection. The details of the methods and antibodies used in the study are shown in the Section S1 and Table S3.

TG Assay
In brief, intracellular TG was detected using a tissue/cell triacylglycerol assay kit (Applygen Technologies, Beijing, China). The details of the materials and methods are shown in the Section S1.

Reactive Oxygen Species (ROS) Assay
In brief, 24 h after cell transfection, intracellular ROS levels were detected using 10 µM 2 ,7 -dichlorofluorescein diacetate (DCFH-DA, Beyotime, Shanghai, China) and flow cytometry (Beckman Coulter, Brea, CA, USA). The details of the materials and methods are shown in the Section S1.

Total Antioxidant Capacity (T-AOC), Catalase (CAT), and Glutathione (GSH) Assays
According to the manufacturer's requirements, 24 h after cell transfection, a T-AOC assay kit (Solarbio), CAT assay kit (Solarbio), and reduced GSH assay kit (Solarbio) were used to detect the levels of T-AOC, CAT, and GSH, respectively. The absorption value was measured with a microplate reader (Tecan, Mannedorf, Switzerland). The T-AOC and activity of GSH and CAT were calculated based on the absorption value and standard curve.

Mitochondrial Morphology Assay
Briefly, mitochondrial morphology was detected using 200 nM MitoTracker Red CMXRos (Invitrogen). The details of the materials and methods are shown in the Section S1.

ATP Assay
Briefly, intracellular ATP levels were measured using an enhancing ATP detection kit (Beyotime). The details of the materials and methods are shown in the Section S1.

Mitochondrial DNA (mtDNA) Copy Number Assay
The whole genome was extracted by using the QIAamp Fast DNA Tissue Kit (Qiagen, Hilden, Germany). mtDNA was detected using qPCR. The details of the materials and methods are shown in the Section S1.

Metabolomics Analysis
In brief, 24 h after transfection, MAC-T cells were washed with PBS and digested by trypsin for collection. After sample extraction, the samples were injected into chromatography and mass spectrometry systems for detection. Then, the raw data were processed by using Progenesis QI (Waters Corporation, Milford, MA, USA). For the data extracted by the Metabolites Search, metabolites with RSD > 30% were deleted. Significantly differentially expressed metabolites between the si-NC and si-KRAS groups were identified with cutoff values of p-value < 0.05, log 2 (FC) > 1, and VIP > 1. The details of the materials and methods are shown in the Section S1.

Lipidomics Analysis
In brief, 24 h after transfection, MAC-T cells were washed with PBS and digested by trypsin for collection. After sample extraction, the samples were injected into chromatography and mass spectrometry systems for detection. Then, the raw data were processed by using LipidSearch (Thermo Fisher, Waltham, MA, USA). For data extracted by Lipid Search, lipid molecules with RSD > 30% were deleted. Significantly differentially expressed lipids between the si-NC and si-KRAS groups were identified with cutoff values of p-value < 0.05, log 2 (FC) > 1, and VIP > 1. The details of the materials and methods are shown in the Section S1.

Statistical Analysis
All experiments were performed three times unless indicated otherwise. Data are represented as the means ± standard deviation. A Bonferroni post hoc test was used to compare the data. All statistical analyses were performed using the SPSS software (version 21.0, IBM, Chicago, IL, USA). p-Values < 0.05 were considered statistically significant.

KRAS Inhibition Mainly Affects Lipid Metabolites in MAC-T Cells
The immunofluorescence results showed that KRAS was localized in the nuclei of MAC-T cells (Supplementary Figure S1). Western blot results indicated that KRAS-specific siRNA (si-KRAS) significantly inhibited the protein expression level of KRAS in MAC-T cells (Supplementary Figure S1). After determining the effectiveness of siRNA, we conducted metabolomics analysis. The results showed that the levels of 291 known metabolites were significantly changed with KRAS inhibition ( Figure 1B); only 2 metabolites were significantly upregulated, and the rest were significantly downregulated. Based on the classification of HMDB, 57% of 156 (matched in the HMDB 4.0 database) differential metabolites were lipid metabolites, followed by 14.74% organic acids and derivatives ( Figure 1C). Moreover, to further analyze feature metabolites, a variable importance in the projection (VIP) plot from the OPLS was carried out to select distinct variables as potential metabolites for exploring the effects of KRAS on cell metabolism in MAC-T cells. We found that there were 28 lipid metabolites in the top 30 metabolites with VIP scores ( Figure 1D). KEGG enrichment analysis showed that the differential metabolites were mainly enriched in the lipid metabolism pathway ( Figure 1E). The detailed data are shown in Supplementary  Tables S4 and S5.

KRAS Inhibition Affects the Lipid Composition in MAC-T Cells
The above results indicate that KRAS inhibition mainly affects lipid metabolism in MAC-T cells. To further explore the effects of KRAS on lipid composition, we carried out a lipidomic study. As shown in Figure 2A,B, after KRAS inhibition, 71 lipid metabolites  Heatmap of all the differential lipidomic metabolites after KRAS inhibition. In the figure, each column represents a sample, each row represents a metabolite, and the color indicates the relative level of metabolites expressed in the group. Red indicates that the metabolite is expressed at high levels, and blue indicates lower expression. (B) Volcano plot of the differentially lipidomic metabolites. Each dot in the figure represents a metabolite, and the size of the dot indicates the variable importance in the projection (VIP) value. The blue dots are significantly downregulated metabolites, and the red dots are significantly upregulated metabolites. (C) Identified differential lipidomic metabolites were classified. (D) Differential metabolites with VIP > 1.5 from the VIP plot based on the OPLS-DA models. (E) KEGG pathway enrichment of differential lipidomic metabolites. The size of bubbles in the figure represents the amount of differential metabolites, and the color of bubbles represents the p-values. Significant differences are represented with * (p < 0.05) and ** (p < 0.01). The size of bubbles in the figure represents the amount of differential metabolites, and the color of bubbles represents the p-values. Significant differences are represented with * (p < 0.05) and ** (p < 0.01).

KRAS Inhibition Improved Milk Fat Synthesis in MAC-T Cells
After KRAS inhibition, intracellular lipid droplet content and TG content were significantly increased in MAC-T cells ( Figure 3A-C). The protein levels of FABP4, PPARG, SREBP1, epidermal growth factor (EGF), EGF receptor (EGFR), and IGF1 receptor (IGF1R) were significantly increased with KRAS inhibition ( Figure 3D-F), while the protein level of transforming growth factor beta receptor type 1 (TGFβR1) was significantly decreased with KRAS inhibition in MAC-T cells. In addition, after KRAS inhibition, the mRNA levels of xanthine dehydrogenase (XDH), CD36, FABP3, ACSL1, ACSS2, ACC, FASN, SCD1, GPAM, LPIN1, and 1-acylglycerol-3-phosphate O-acyltransferase 1 (AGPAT1) were also significantly increased ( Figure 3G). In addition, the immunofluorescence results showed that the nuclear expression levels of PPARG ( Figure 3H) and SREBP1 ( Figure 3I) in MAC-T cells were significantly increased, and the colocalization level of lipid droplets and FABP4 with ER was significantly enhanced with KRAS inhibition ( Figure 3J). Detailed data are shown in Supplementary Table S4. , x FOR PEER REVIEW 9 of 22 After KRAS inhibition, intracellular lipid droplet content and TG content were significantly increased in MAC-T cells ( Figure 3A-C). The protein levels of FABP4, PPARG, SREBP1, epidermal growth factor (EGF), EGF receptor (EGFR), and IGF1 receptor (IGF1R) were significantly increased with KRAS inhibition ( Figure 3D-F), while the protein level of transforming growth factor beta receptor type 1 (TGFβR1) was significantly decreased with KRAS inhibition in MAC-T cells. In addition, after KRAS inhibition, the mRNA levels of xanthine dehydrogenase (XDH), CD36, FABP3, ACSL1, ACSS2, ACC, FASN, SCD1, GPAM, LPIN1, and 1-acylglycerol-3-phosphate O-acyltransferase 1 (AGPAT1) were also significantly increased ( Figure 3G). In addition, the immunofluorescence results showed that the nuclear expression levels of PPARG ( Figure 3H) and SREBP1 ( Figure 3I) in MAC-T cells were significantly increased, and the colocalization level of lipid droplets and FABP4 with ER was significantly enhanced with KRAS inhibition ( Figure 3J). Detailed data are shown in Supplementary Table S4.

KRAS Inhibition Improved Mitochondrial Function
The results showed that the levels of mtDNA ( Figure 4A) and ATP ( Figure 4B) significantly increased with KRAS inhibition in MAC-T cells. The mRNA levels of the mitochondrial-related genes cytochrome c oxidase subunit 5B (COX5B), NADH:ubiquinone oxidoreductase core subunit S8 (NDUFS8), succinate dehydrogenase complex iron sulfur subunit B (SDHB), ATP synthase F1 subunit alpha (ATP5F1A), and ubiquinol-cytochrome c reductase binding protein (UQCRB) and the mitochondrial biogenesis-related genes nuclear respiratory factor 1 (NRF1), DNA polymerase gamma catalytic subunit (POLG), mitochondrial transcription factor A (TFAM), and mitochondrial transcription factor B1 (TFB1M) in the KRAS inhibition group were also significantly increased ( Figure 4C). Interestingly, with MitoTracker staining, we found mitochondrial morphological changes in some of the MAC-T cells that produced more lipids ( Figure 4D). Moreover, the expression levels of the mitochondrial fusion-related proteins mitofusin (MFN)1, MFN2, and optic atrophy 1 (OPA1) and fission-related protein dynamin-related protein 1 (DRP1) were significantly increased with KRAS inhibition ( Figure 4E,F). Detailed data are shown in Supplementary  Table S4.
OR PEER REVIEW 10 of 22 sented by the blue, green, yellow, and red lines, respectively. Fluorescence colocalization was analyzed based on the white line in merged image. Significant differences are represented with * (p < 0.05) and ** (p < 0.01).

KRAS Inhibition Improved Mitochondrial Function
The results showed that the levels of mtDNA ( Figure 4A) and ATP ( Figure 4B) significantly increased with KRAS inhibition in MAC-T cells. The mRNA levels of the mitochondrial-related genes cytochrome c oxidase subunit 5B (COX5B), NADH:ubiquinone oxidoreductase core subunit S8 (NDUFS8), succinate dehydrogenase complex iron sulfur subunit B (SDHB), ATP synthase F1 subunit alpha (ATP5F1A), and ubiquinol-cytochrome c reductase binding protein (UQCRB) and the mitochondrial biogenesis-related genes nuclear respiratory factor 1 (NRF1), DNA polymerase gamma catalytic subunit (POLG), mitochondrial transcription factor A (TFAM), and mitochondrial transcription factor B1 (TFB1M) in the KRAS inhibition group were also significantly increased ( Figure 4C). Interestingly, with MitoTracker staining, we found mitochondrial morphological changes in some of the MAC-T cells that produced more lipids ( Figure 4D). Moreover, the expression levels of the mitochondrial fusion-related proteins mitofusin (MFN)1, MFN2, and optic atrophy 1 (OPA1) and fission-related protein dynamin-related protein 1 (DRP1) were significantly increased with KRAS inhibition ( Figure 4E,F). Detailed data are shown in Supplementary Table S4.

KRAS Inhibition Enhanced ER Stress in MAC-T Cells
The flow cytometry results indicated that the ROS level was significantly increased in the KRAS inhibition group in MAC-T cells ( Figure 5A). The protein levels of glucoseregulated protein (GRP) 78 and C/EBP homologous protein (CHOP) were also significantly increased, while the level of ATF6 was not significantly changed with KRAS inhibition ( Figure 5B,C). In addition, the relative levels of CAT, GSH, and T-AOC were significantly increased ( Figure 5D). The apoptosis levels and protein expression levels of CASPASE 3 and cleaved-CASPASE 3 were not significantly changed ( Figure 5E-H). Detailed data are shown in Supplementary Table S4.

KRAS Inhibition Enhanced ER Stress in MAC-T Cells
The flow cytometry results indicated that the ROS level was significantly increased in the KRAS inhibition group in MAC-T cells ( Figure 5A). The protein levels of glucoseregulated protein (GRP) 78 and C/EBP homologous protein (CHOP) were also significantly increased, while the level of ATF6 was not significantly changed with KRAS inhibition ( Figure 5B,C). In addition, the relative levels of CAT, GSH, and T-AOC were significantly increased ( Figure 5D). The apoptosis levels and protein expression levels of CASPASE 3 and cleaved-CASPASE 3 were not significantly changed ( Figure 5E-H

KRAS Inhibition Enhanced Autophagy in MAC-T Cells
Among MAC-T cells, the immunofluorescence results indicated that the numbers of microtubule-associated protein 1 light chain 3 beta (LC3B) puncta in the cytoplasm were significantly increased in the KRAS inhibition and chloroquine-treated groups compared with the NC group ( Figure 6A), and the protein levels of LC3B-II/LC3B-I in the KRAS inhibition and chloroquine-treated groups were also increased ( Figure 6B,C). The relative levels of phosphorylated mTOR versus total mTOR ( Figure 6B,D) were significantly decreased, while the mRNA levels of beclin 1 (BECN1) and autophagy-related 7 (ATG7) were

KRAS Inhibition Enhanced Autophagy in MAC-T Cells
Among MAC-T cells, the immunofluorescence results indicated that the numbers of microtubule-associated protein 1 light chain 3 beta (LC3B) puncta in the cytoplasm were significantly increased in the KRAS inhibition and chloroquine-treated groups compared with the NC group ( Figure 6A), and the protein levels of LC3B-II/LC3B-I in the KRAS inhibition and chloroquine-treated groups were also increased ( Figure 6B,C). The relative levels of phosphorylated mTOR versus total mTOR ( Figure 6B,D) were significantly decreased, while the mRNA levels of beclin 1 (BECN1) and autophagy-related 7 (ATG7) were significantly increased after KRAS inhibition ( Figure 6E). In addition, after KRAS inhibition, the colocalization level of CALNEXIN, LC3B, and Bodipy was enhanced, as shown by an immunofluorescence assay ( Figure 6F). Detailed data are shown in Supplementary  Tables S4 and S7.

KRAS Inhibition Regulated MAPK Activation
Western blotting was used to detect the MAPKs (extracellular regulated protein kinases, ERKs; c-Jun N-terminal kinases, JNKs; p38) phosphorylation levels in MAC-T cells with or without KRAS treatment. As shown in Figure 7, compared with the si-NC group, the si-KRAS-treated group exhibited significantly increased phosphorylation levels of ERK1/2, JNK1/2/3, and p38. Detailed data are shown in Supplementary Table S4. were represented by the blue, green, yellow, and red lines, respectively. Fluorescence colocalization was analyzed based on the white line in merged image. Significant differences are represented with * (p < 0.05) and ** (p < 0.01).

KRAS Inhibition Regulated MAPK Activation
Western blotting was used to detect the MAPKs (extracellular regulated protein kinases, ERKs; c-Jun N-terminal kinases, JNKs; p38) phosphorylation levels in MAC-T cells with or without KRAS treatment. As shown in Figure 7, compared with the si-NC group, the si-KRAS-treated group exhibited significantly increased phosphorylation levels of ERK1/2, JNK1/2/3, and p38. Detailed data are shown in Supplementary Table S4.

Discussion
Milk fat is one of the main nutrients in milk. It not only can provide energy for newborn calves but is also an important raw material for many dairy products [34]. Milk fat content is closely related to the quality and flavor of milk and even affects the development of animal husbandry. Milk fat, which is composed mainly of TG, is synthesized by the ER of BMECs [35]. During lactation, mammary mitochondria and ER have been suggested to play important roles in milk lipid biosynthesis. When excessive lipid synthesis and secretion occurs, the mitochondria and ER will continue to experience stress and damage. To maintain homeostasis, cells try to increase autophagy to remove damaged ER and excess lipids. Previous studies have shown that KRAS affects lipid accumulation. However, the specific roles of KRAS in the lactation process remain unclear.

KRAS Affects Lipid Accumulation
In mammalian mammary glands, most luminal epithelial cells express cytokeratin 18 (CK18) [36]. Through immunofluorescence studies (Supplementary Figure S1), we found that CK18 was strongly expressed in all MAC-T cells, which indicated that MAC-T cells were purified and had the characteristic of BMECs. KRAS inhibition leads to increased expression levels of PPARG and SREBP1, suggesting that inhibition of KRAS may affect the localization of key nuclear transcription factors, thus improving the ability to promote lipid droplet formation in MAC-T cells [37]. Increased levels of CD36, FABP4, ACSL1, ACC, FASN, SCD1, and AGPAT6 [38] also indicated that KRAS inhibition may increase lipogenic gene expression; the transcriptional activation of these genes is consistent with the abundant milk fat synthesis that occurs at the onset and throughout lactation. Moreover, genes related to FA metabolism, TG synthesis, and lipid droplet formation, such as ACSS2, GPAM, LPIN1, and XDH, were also increased after KRAS inhibition. Once FAs

Discussion
Milk fat is one of the main nutrients in milk. It not only can provide energy for newborn calves but is also an important raw material for many dairy products [34]. Milk fat content is closely related to the quality and flavor of milk and even affects the development of animal husbandry. Milk fat, which is composed mainly of TG, is synthesized by the ER of BMECs [35]. During lactation, mammary mitochondria and ER have been suggested to play important roles in milk lipid biosynthesis. When excessive lipid synthesis and secretion occurs, the mitochondria and ER will continue to experience stress and damage. To maintain homeostasis, cells try to increase autophagy to remove damaged ER and excess lipids. Previous studies have shown that KRAS affects lipid accumulation. However, the specific roles of KRAS in the lactation process remain unclear.

KRAS Affects Lipid Accumulation
In mammalian mammary glands, most luminal epithelial cells express cytokeratin 18 (CK18) [36]. Through immunofluorescence studies (Supplementary Figure S1), we found that CK18 was strongly expressed in all MAC-T cells, which indicated that MAC-T cells were purified and had the characteristic of BMECs. KRAS inhibition leads to increased expression levels of PPARG and SREBP1, suggesting that inhibition of KRAS may affect the localization of key nuclear transcription factors, thus improving the ability to promote lipid droplet formation in MAC-T cells [37]. Increased levels of CD36, FABP4, ACSL1, ACC, FASN, SCD1, and AGPAT6 [38] also indicated that KRAS inhibition may increase lipogenic gene expression; the transcriptional activation of these genes is consistent with the abundant milk fat synthesis that occurs at the onset and throughout lactation. Moreover, genes related to FA metabolism, TG synthesis, and lipid droplet formation, such as ACSS2, GPAM, LPIN1, and XDH, were also increased after KRAS inhibition. Once FAs with >10 carbons are formed, they are activated by ACSL1 and bound to FABP3, which allows the FAs to enter the TG synthesis pathway [39]. Short-chain FAs are inserted into the TG via diacylglycerol O-acyltransferase 1 [40]. Once TG is formed, it is inserted into the intraleaflet of the ER membrane to form lipid droplets. Perilipin 2 is essential for the formation of lipid droplets [41] and for the secretory pathway involving butyrophilin subfamily 1 member A1, and XDH also seems to play a role in the mechanism that involves perilipin 2 and butyrophilin subfamily 1 member A1 [42]. These results indicated that nuclear KRAS may mediate signaling that can activate or deactivate some transcription factors and affect the translocation of PPARG, SREBP1, and FABP4 to regulate TG formation in MAC-T cells. Moreover, previous studies have also shown that IGF-1 promotes adipogenesis [43], while blocking IGF-1 signaling disrupts adipogenesis [44]. However, TGFβ and TGFβR1 exhibited opposite effects on those of IGF1 on lipid formation. TGFβ inhibits adipocyte differentiation but is expressed by adipocytes [45]. TGFβ receptors strongly decrease as adipogenesis proceeds [46]. In this study, KRAS inhibition significantly increased the protein expression levels of IGF1R and decreased the protein expression levels of TGFβR1 in MAC-T cells, which suggests that KRAS has potential roles in regulating the IGF and TGF signaling pathways. The increased levels of EGF and EGFR expression after KRAS inhibition also indicate that KRAS can regulate the process of intracellular lipid formation and lipid accumulation [47].

Inhibition of KRAS Caused Cell Stress
The ER is the only place where milk fat synthesis occurs, and its activity changes during lactation [48,49]. During lactation, the ER activity of BMECs is greatly increased and even stressed because a high level of milk fat synthesis and secretion is necessary and its cellular metabolism is substantial. Changes in the ER stress pathway at different stages in the lactation cycle are normal adaptations of the tissues to the changing physiological state [50]. GRP78, or BiP, is a central regulator of ER stress due to its role as a major ER chaperone, its antiapoptotic properties, and its ability to control the activation of transmembrane ER stress sensors (inositol-requiring enzyme 1, RNA-dependent protein kinase-like ER kinase, and ATF6) through a binding-release mechanism [51]. ER stress mediates SREBP1/2 activation as well as the accumulation of TG and cholesterol [52]. Previous studies indicated that compared with the perinatal stage (day 15), the expression levels of RNA-dependent protein kinase-like ER kinase, ATF3, CHOP, and GRP78 significantly increased in the mammary gland of a cow during lactation, and the expression levels of ATF6, X-box binding protein 1, GRP94, and membrane bound transcription factor peptidase site 1 also showed the same trend [50]. These results further demonstrate that the KRAS inhibition-mediated ER stresssignaling pathway may be associated with lipogenesis. Curiously, although increased milk fat formation often leads to increased ER stress, the protein levels of GRP78 and CHOP were significantly improved, while ATF6 levels were not changed significantly after KRAS inhibition. These results showed that among MAC-T cells, the regulation of KRAS in ER stress may be different from the result that the expression of GRP78 can be activated by ATF6, a target of the ER stress response, which has been implicated in the ER stress response pathway [53]. This may mean that an increased level of milk fat synthesis does not always lead to increased levels of ER stress, in other words, ER stress from a high production level of milk fat and protein secretion may be reduced or blocked by other pathways. In addition, studies have shown that ER stress can lead to apoptosis [53]. Our results showed that the levels of CASPASE 3, cleaved-CASPASE 3 [54], and apoptosis did not change significantly with KRAS inhibition. These results indicated that KRAS-mediated ER stress did not induce cell apoptosis in MAC-T cells. However, this may be affected by the knockdown efficiency of siRNA.
Moreover, the MECs of lactating cows undergo intensive metabolism in cells and accumulate a large amount of free radicals, such as ROS [55]. MECs in cows that yield high levels of milk are more likely to produce ROS due to their increased capacity to synthesize and secrete milk [55]. However, excessive ROS accumulation incapacitates the redox system, which leads to oxidative stress and cell damage [56]. However, our study showed that although the level of ROS increased with KRAS inhibition, the total capacity to produce milk fat and mitochondrial function were not negatively affected in MAC-T cells. This indicated that after KRAS inhibition, despite the increase in ROS under the intensity metabolic influence, oxidative homeostasis was maintained, as the antioxidant defenses were sufficient. After KRAS inhibition, oxidative homeostasis may be maintained by the significantly increased levels of GSH and CAT.

Inhibition of KRAS Enhanced Mitochondrial Function
Mitochondria are recognized as essential components for meeting energy needs and maintaining metabolic homeostasis because mtDNA is a template for several polypeptides in respiratory chain complexes that are essential for ATP generation [57]. The decrease in mammary mitochondrial ATP synthesis on day 21 postpartum correlates with previously reported decreases in milk production in mice [58]. These results are consistent with the idea that ATP synthesis activity in mitochondria is temporally correlated with the production of milk during lactation. Combined with our results, KRAS may affect the synthesis of milk fat by regulating the level of mtDNA, intracellular ATP, and the related gene expression of the mitochondrial electron transport chain (NDUFS8, SDHB, UQCRB, COX5B, and ATP5F1A) and mitochondrial biogenesis (NRF1, TFAM, TFB1M, and POLG) [59]. Interestingly, after KRAS inhibition, some MAC-T cells with higher lipid production contained mitochondria with morphological changes. Through electron microscopy, previous studies found that during the early postpartum period, both the size and density of secretory cell mitochondria increase, which also indicated that the increase in milk production during early lactation is linked with a dramatic increase in mammary mitochondrial number [49]. However, it is difficult to determine whether the mitochondrial morphological changes that occur with KRAS inhibition lead to the increased level of their own biosynthesis and ATP [60] or whether the changes are caused by ROS or ER stress only based on the changes in mitochondrial morphology [61]. Therefore, we further examined fusion-and fission-related proteins. There is a close connection between mitochondrial dynamics and biosynthesis [62]. Studies have shown that the better the mitochondrial quality states are, the stronger the fusion-fission and biogenesis cycles [63]. The mitochondrial function and biosynthesis capacity of cells were significantly increased, and the highly expressed mitochondrial dynamics-related genes OPA1, MFN2, FIS1, and DRP1 were enhanced by leptin treatment [64]. This is precisely related to the inhibition of mitochondrial activity and biosynthesis by rotenone treatment, accompanied by the downregulation of DRP1, OPA1, and MFN2 protein expression levels [65]. Our results showed that the expression levels of the mitochondrial fusion-related proteins MFN1, MFN2, and OPA1 and the fission-related protein DRP1 were significantly increased after KRAS inhibition. These results suggested that after KRAS inhibition, the highly dynamic mitochondrial fusion and fission cycle was proposed to balance the compensation of damage by fusion and the elimination of damage by fission, eventually leading to an increase in mitochondrial biosynthesis.

Inhibition of KRAS Induced Cell Autophagy
After KRAS inhibition, the number of autophagosomes, protein levels of LC3B-II/LC3B-I, and mRNA levels of BECN1 [66] and ATG7 [67] were significantly increased, while the phosphorylation level of mTOR [68] was significantly decreased in MAC-T cells.
Chloroquine inhibits the degradation of autophagosomes, thereby inhibiting autophagy levels in cells. During this process, the mRNA expression levels of ATG7 and BECN 1 will not increase. However, in this study, KRAS inhibition resulted in increased mRNA levels of ATG7 and BECN 1. These results indicated that the inhibition of KRAS indeed increases the level of autophagy rather than preventing the degradation of autophagosomes and inhibiting autophagy in MAC-T cells. In addition, the colocalization level of autophagosomes with ER and lipid droplets was enhanced after KRAS inhibition, which indicates that the level of selective autophagy, that is, ER-phagy and lipophagy, was enhanced. ER-phagy and lipophagy represent the major quality control mechanisms that ensure that exhausted or harmful ERs are properly renewed and that the balance of lipid metabolism is maintained [69]. This indicated that after KRAS inhibition, TG levels were significantly increased, and lipid metabolism was disordered, so lipophagy may be initiated to maintain lipid homeostasis. When excessive lipid synthesis and secretion occurs, the ER suffers from stress and damage, and the level of intracellular ROS also increases significantly, which may together lead to an increased level of ER-phagy. Many studies have shown that the activation or inhibition of mTOR can upregulate or downregulate milk fat levels, respectively [70,71]. Interestingly, the finding that the phosphorylation level of mTOR decreased and milk fat levels increased with KRAS inhibition was the opposite of what we observed in our research. This may be because after KRAS inhibition, the increased level of stress caused by excessive TG accumulation inhibits mTOR activation, which enhances autophagy levels to maintain the homeostasis of intracellular environments [68,72].

KRAS Affects MAPK Activation
ERK, p38, and JNK MAPKs are intracellular signaling pathways that play a pivotal role in adipogenesis [73]. In this study, KRAS inhibition decreased the levels of ERK activation and PPARG expression, which suggests that after KRAS inhibition, the levels of FAS expression and lipid accumulation were increased by enhancing the expression of C/EBPα/β and PPARG that was caused by ERK inhibition [74]. Conversely, the activation of the ERK pathway inhibited lipid droplet formation, and the protein expression levels of the lipid-related genes C/EBPα, PPARG, lipoprotein lipase, and FABP4 were associated with our hypothesis [75]. Some studies have shown that the inhibition of p38 activity can promote lipid formation [76]. In this study, KRAS inhibition also decreased the phosphorylation level of p38, indicating that KRAS has the potential to regulate the activation of the nuclear factor of activated T cells 4 (NFATc4) and enhance the nuclear translocation of NFATc4 after dephosphorylation by NFAT, thus promoting the expression of PPARc and lipid formation [77]. The inhibition of p38 by microtubule affinity regulating kinase 4 also promoted adipogenesis, which together indicates that the activation of ERK and p38 can inhibit the capacity to adipogenesis. JNK activity also plays important roles in adipogenesis. Previous studies have shown that microtubule affinity regulating kinase 4 can also promote lipid formation by inhibiting p38 and activating JNK1, and JNK activation promotes mesenchymal stem cell adipogenic differentiation [78]; these results seem to suggest that JNK plays opposite roles to ERK and p38 in adipogenesis. In this study, the differential expression of related genes also demonstrated that KRAS inhibition can activate the JNK pathway, thus promoting lipid formation by regulating PPARG and FABP4 [79]. After KRAS inhibition, the phosphorylation level of JNK increased, while the levels of ERK and p38 decreased at the same time, which suggests that KRAS plays important roles in lipid accumulation in MAC-T cells.

KRAS Regulates Milk Fat Composition
The metabolomics analysis showed that after KRAS inhibition, the intracellular metabolite levels changed significantly; most of these metabolites were lipid metabolites, and the top 30 metabolites with VIP scores also contained 28 lipid metabolites. As the main component of milk fat, TG also regulates cellular homeostasis and the levels of other lipid metabolites [80]. Previous studies have shown that TG has negative regulatory effects with lysophosphatidylinositol/PI [81]. This is consistent with our results that after KRAS inhibition, various forms of TG were significantly increased, and one form of PI was significantly decreased. Moreover, the levels of SM, Hex1Cer, and Hex2Cer were negatively correlated with the Cer and SPH levels after KRAS inhibition. The Cer formed from SM turnover might be hydrolyzed by ceramidases to SPH [82,83]. Cellular accumulations of Cer and SPH, which occur in response to stress, such exposure to tumor necrosis factor alpha or oxidative stress, are associated with apoptotic responses [84]. SP not only plays important roles in cell signal transduction but also acts as an important part of MFGM together with GP [85]. After KRAS inhibition, various forms of SP and GP were differentially expressed, and VIP score analysis also showed that TG was the main contributor to the difference between groups. These results indicated that KRAS inhibition has important effects on both the milk fat globule core and MFGM and that the cellular stress caused by excessive TG accumulation promotes the transformation of Hex1Cer, Hex2Cer, and SM into Cer and SPH. This result also seems to indicate that KRAS-mediated TG level changes in the milk fat globule core may lead to the different MFGM components.
Overall, the levels of TG, mitochondrial functions, ROS, ERS, and autophagy were increased in MAC-T cells with KRAS inhibition. The phosphorylated levels of MAPKs (ERK, JNK, and p38) were also changed after KRAS inhibition (Figure 8). Moreover, after KRAS inhibition, the levels of some lipids were also regulated in addition to TG, such as PC, PE, Cer, and HexCer. These findings broaden the understanding of the effects of KRAS on milk fat metabolism and provide a new theoretical basis for further exploring the mechanism of milk fat synthesis and secretion. With KRAS inhibition, the mitochondrial function and milk fat synthesis capacity of MAC-T cells were significantly enhanced, while intracellular ROS and ER stress levels were increased. Thus, although KRAS inhibition enhanced milk fat secretion, it simultaneously stressed mammary epithelial cells and enhanced autophagy levels, which is not good for the health of the cows. Conversely, relatively high KRAS expression may lead to a decrease in milk fat secretion, which is detrimental to offspring feeding. Therefore, although KRAS is a key regulator of milk fat secretion, simply changing its expression level up-or downregulation may be detrimental to the health of either the next generation or dairy cows. This is similar to the occurrence of ketosis in mammary tissue, which suggests that abnormal expression of KRAS may cause a series of symptoms by affecting energy imbalance and play important roles in ketosis. Therefore, the regulation of KRAS may also provide new directions for exploring the specific mechanism and treatment of ketosis.

Conclusions
In MAC-T cells, KRAS regulated the phosphorylation levels of ERK, p38, and JNK. After KRAS inhibition, mitochondrial functions were improved. However, ROS and ERS were also increased. This would also lead to an increase in intracellular autophagy to maintain oxidative homeostasis and remove damaged ER. Excessive TG accumulation

Conclusions
In MAC-T cells, KRAS regulated the phosphorylation levels of ERK, p38, and JNK. After KRAS inhibition, mitochondrial functions were improved. However, ROS and ERS were also increased. This would also lead to an increase in intracellular autophagy to maintain oxidative homeostasis and remove damaged ER. Excessive TG accumulation caused by KRAS inhibition led to intensive lipid metabolism and secretion, which suggests that KRAS plays important roles in lipid accumulation.
Supplementary Materials: The following are available online at https://www.mdpi.com/article/ 10.3390/ani12223070/s1. Section S1: Supplementary Materials and Methods; Figure S1: KRAS is expressed mainly in MAC-T cell nuclei. Table S1: siRNA sequences used in this study; Table S2: Primer sequences used in this study; Table S3: Antibodies used for Western Blot and immunofluorescence (IF); Table S4: All values of folder changes (si-KRAS vs. si-NC); Table S5: Differential metabolites from metabolomics analysis with KRAS inhibition; Table S6: Differential lipid metabolites from lipidomics analysis with KRAS inhibition; Table S7: Relative folder changes of LC3B-II/LC3B-I.