The Knockout of the ASIP Gene Altered the Lipid Composition in Bovine Mammary Epithelial Cells via the Expression of Genes in the Lipid Metabolism Pathway

Simple Summary According to the FDA, a litre of milk contains about 33 g of total lipids, 95% of which are triacylglycerols made up of varying lengths and saturation of fatty acids. The increase in free fatty acids in milk may lead to changes in the flavour of milk and milk products. Bovine mammary epithelial cells (bMECs) function to synthesize and secrete milk and are good in vitro models for cells in milk quality research. In this research, we report for the first time the effect of the ASIP gene knockout in bovine mammary epithelial cells using CRISPR/Cas9 technology. We found that ASIP knockout could down-regulate the expression of PPARγ, FASN, and SCD, thus affecting the saturation of fatty acids in milk and upregulating the expression of FABP4, ELOVL6, and ACSL1, increasing the synthesis of medium-long chain fatty acids. These results may allow the selection of potential targets in future molecular breeding efforts for dairy cows. Abstract Agouti signalling protein (ASIP) is a coat colour-related protein and also is a protein-related to lipid metabolism, which had first been found in agoutis. According to our previous study, ASIP is a candidate gene that affects the lipid metabolism in bovine adipocytes. However, its effect on milk lipid has not been reported yet. This study focused on the effect of the ASIP gene on the lipid metabolism of mammary epithelial cells in cattle. The ASIP gene was knocked out in bMECs by using CRISPR/Cas9 technology. The result of transcriptome sequencing showed that the differentially expressed genes associated with lipid metabolism were mainly enriched in the fatty acids metabolism pathways. Furthermore, the contents of intracellular triglycerides were significantly increased (p < 0.05), and cholesterol tended to rise (p > 0.05) in bMECs with the knockout of the ASIP gene. Fatty acid assays showed a significant alteration in medium and long-chain fatty acid content. Saturated and polyunsaturated fatty acids were significantly up-regulated (p < 0.05), and monounsaturated fatty acids were significantly decreased in the ASIP knockout bMECs (p < 0.05). The Q-PCR analysis showed that knockout of ASIP resulted in a significant reduction of gene expressions like PPARγ, FASN, SCD, and a significant up-regulation of genes like FABP4, ELOVL6, ACSL1, HACD4 prompted increased mid-to long-chain fatty acid synthesis. Overall, ASIP plays a pivotal role in regulating lipid metabolism in bMECs, which could further influence the component of lipid in milk.


Introduction
As one of the essential sources of nutrition for humans, milk contains several different components, such as protein, fats, and amino acids. Milk's major energy component is fat. It contains several lipid fractions including triglycerides (over 95%), diacylglycerol (2%) free fatty acids (0.5%), cholesterol (0.5%), and phospholipids (1%) [1]. Research has found that high-fat milk may cause heart disease, weight gain, and obesity [2,3]. Tucker and colleagues found that Americans who habitually consumed high-at milk had shorter telomeres than those who drank low-fat milk [4]. The abundant fatty acids in milk could take responsibility for this adverse effect on humans, especially saturated and partially unsaturated fatty acids. Some studies have suggested that people need to reduce the intake of saturated fatty acids. Myristic acid (14:0) and palmitic acid (16:0) contribute to an increase in low-density lipoprotein (LDL) cholesterol in the blood, which increases the risk of cardiovascular disease [5]. Furthermore, linoleate (18:2N6) and linolenic (18:3N3) are the main polyunsaturated fatty acid in milk [5]. The 18:2N6 is a precursor for arachidonate (20:4N6), the main precursor for the synthesis of arachidate (20:0). Arachidate may enhance platelet aggregation, but it may also increase the risk of coronary artery disease [6]. On the other hand, oleic acid (C18:1N9), about 25% of milk lipid, has health benefits for humans. Studies showed that the abound of monounsaturated fatty acids (MUFA) could reduce the concentration of cholesterol, LDL-cholesterol and triacylglycerol in plasma [7]. Thereby, MUFA contributes to decreasing coronary artery disease risk [5]. The molecule mechanism of gene regulation on milk lipids has been paid more attention in the research area of milk quality. By interference of stearoyl-CoA desaturase 1 (SCD1) in bMECs, a positive correlation was found between SCD1 expression and the amount of unsaturated fatty acid in bMECs [8]. Kadegowda et al. identified the peroxisome proliferator-activated receptor γ (PPARγ) as a key lipid synthesis pathway in bMECs and a target gene for fatty acid regulation of lipid metabolism [9]. Li et al. reported that single nucleotide polymorphisms (SNPs) within fatty acid synthase (FASN), peroxisome proliferator activated receptor gamma coactivator 1 alpha (PPARGC1A), ATP binding cassette subfamily G member 2 (ABCG2) and insulin like growth factor 1 (IGF1) were significantly associated with milk fatty acid composition in dairy cattle [10]. However, these regulatory mechanisms for the content and composition of milk lipid are still unclear. Therefore, more candidate genes involved in the regulation of milk lipid metabolism need to be explored.
Agouti signalling protein (ASIP) was first discovered and identified as a protein associated with hair colour in agoutis [11]. Another study on the ASIP gene reported that the mutation of ASIP in the promoter region is associated with the overexpression of ASIP in various tissues [12]. The ASIP mutant mice showed yellow coat colour, dramatically enhancing their weight, bulimic behaviour and diabetes mellitus [12]. Jones et al. found significant up-regulation of FASN and SCD1 expression in a study of yellow agouti mutants [13]. They are involved in de novo fatty acid synthesis and desaturation in the liver and adipose tissue. This may lead to a significant increase in unsaturated fatty acids in medium-long chain fatty acids. Despite the ability of ASIP for the regulation fat deposition and lipid metabolism in adipose, no studies have focused on its effect on the synthesis and metabolism of milk lipid in cattle. Previous studies on ASIP in cattle have mainly focused on coat colour [11,12]. Recently, its function on bovine fat deposition was explored. Albrecht et al. found that Japanese black cattle with higher intramuscular fat content had higher ASIP mRNA levels in the longissimus dorsi muscle than Holstein cattle [14]. Liu et al. reported that ASIP mRNA expression levels in subcutaneous fat of 229 Bulls of the F2-generation of a Charolais × Holstein cross were significantly correlated with bovine fat deposition traits carcass fat weight [15]. Furthermore, Liu et al. screened three SNPs in ASIP genes in a Chinese Simmental cattle population. The results indicated that two SNPs correlated with carcass and fat-related traits significantly, such as live weight and backfat thickness [16]. These further suggest a significant association between ASIP and fat accumulation in cattle. Previous studies in our laboratory found that the recombinant ASIP protein could regulate the expression of adipogenic and fatty acid metabolism-related genes in adipocytes, such as leptin (LEP), adiponectin (ADIPOQ), solute carrier family Animals 2022, 12, 1389 3 of 17 27 member 6 (SLC27A6), etc. Hence, ASIP could be a potential candidate gene involved in milk lipid metabolism.
To explore the effect of ASIP on bovine milk lipid, ASIP knock-out bMECs were constructed by CRISPR/Cas9 technology in this study. The triglycerides, cholesterol, and fatty acid contents in the knock-out cell lines were detected. Moreover, the transcriptome of ASIP knock-out bMECs was obtained and analysed by RNA-seq. The results of this study can reveal the potential effects of ASIP in bovine mammary epithelial cells on milk lipid metabolism. It could provide useful information about candidate genes for the future breeding of high-quality dairy cows.

Cell Culture
Lactating Chinese Holstein dairy cows were selected to isolate bovine mammary epithelial cells (bMECs). Fresh mammary tissue was taken and placed in a solution of PBS (Tiangen, Beijing, China) containing 5% (v/v) penicillin and streptomycin (Tiangen, Beijing, China). It was stored in a 37 • C thermostat and transported immediately to the laboratory. The tissue was washed several times with PBS containing antibiotics until the solution was no longer turbid. The bulk breast tissue was divided into small pieces, cut into 1 mm 3 , and washed with 1 PBS. Subsequently, the bMECs were isolated by the tissue block culture method [17]. These bMECs were preserved in the Laboratory of Molecular Genetics, College of Animal Sciences, Jilin University. According to our previous work, bMECs were cultured in DMEM/F12 (HyClone, Logan, UT, USA) supplemented with 10% foetal bovine serum (FBS, 11011-6123, Tian Hang, Zhejiang, China) at 37 • C and 5% CO 2 in an incubator (Thermo Fisher Scientific, Waltham, MA, USA). All the experiments were performed according to the guidelines for the care and use of laboratory animals by Jilin University (Animal Care and Use Committee permit number: SY201901007) [18].

Construction of KO-ASIP bMECs
The target sequences of sgRNAs in the bovine ASIP exon 1 were predicted using Zhang Lab (https://zlab.bio/guide-design-resources (accessed on 3 June 2017)), and the details of the sgRNAs sequence were showed in Table 1. All the DNA Oligos were synthesized by Sangon (Shanghai, China). Each sgRNA was inserted into the PX459 vector (BioVector NTCC, Beijing, China) and named gASIP/PX459. Knockout of the ASIP gene fragment using transient transfecting technology in bMEC. After 24 h, the bMECs with ASIP gene knockout were screened using 2 µg/mL puromycin dihydrochloride (Beyotime, Jiangsu, China). Viable cells were cultured in 20% FBS and 1% penicillin-streptomycin solution. These cells were trypsinized and collected by centrifugation, resuspended in medium and diluted well. Cells were cultured in 96-well plates at one cell per well.

Identification of KO-ASIP bMECs Lines
The genomic DNA of bMECs was extracted following the Phenol-Chloroform Extraction protocol [19]. The PCR amplification was performed using primers to identify the knockout ASIP gene in bMECs. The forward primer and reverse primer sequences were 5 -CAGGTCCATCCTATCCTTCTG-3 and 5 -CAGGTCCATCCTATCCTTCTG-3 , respectively. The PCR products were sent to GENEWIZ (Suzhou, China) for Sanger sequencing. The single-cell cloning of bMEC with ASIP gene-edited was screened, and the bMEC was named KO-ASIP.

Determination of Triglycerides, Cholesterol, and Intracellular Fatty Acid Content in bMECs of ASIP Gene
Cells were seeded at 6 × 10 6 cells/well in 6-well culture plates with triplicates for KO-ASIP bMECs and WT bMECs. These bMECs were digested with 0.25% trypsin, the cell suspension was collected with a 1.5 mL clean centrifuge tube, and cells were collected by centrifugation at 700 r. It was washed again with PBS. Add Lysis Buffer for Triglyceride or Cholesterol Extraction. Triglycerides and CHOL were measured using a TG detection kit and a tissue/cell total CHOL assay kit (E1015-105 and E1013, Applygen Technologies, Beijing, China), respectively. The cellular contents of TG and CHOL were normalized by protein content. The concentrations were measured using a microplate reader (SM600, Shanghai YongChuang Medical Instrument Co., Ltd., Shanghai, China). Cells were collected at a cell volume of 50 mg per tube, three tubes each from KO-ASIP bMECs and WT bMECs and sent to Shanghai Applied Protein Technology Inc to determine the intracellular fatty acid content.

RNA Extraction, Library Preparation and RNA-Seq Analysis
Total RNA was extracted from the bMECs by TRIzol reagent (Invitrogen, Carlsbad, CA, USA). The RNA quality detection was carried out using the Agilent bioanalyzer 2100 (Agilent Technologies, Palo Alto, CA, USA). A total of 1 µg RNA with RIN value above 7 was used following library preparation. Next-generation sequencing library preparations were constructed according to the manufacturer's protocol (NEBNext ® Ultra™ RNA Library Prep Kit for Illumina ® ). The poly(A) mRNA was captured with NEBNext ® Poly(A) mRNA Magnetic Isolation Module (NEB, Beijing, China). The mRNA fragmentation and priming were performed using NEBNext First Strand Synthesis Reaction Buffer and NEB-Next Random Primers. First-strand cDNA was synthesized using ProtoScript II Reverse Transcriptase, and the second-strand cDNA was synthesized using Second Strand Synthesis Enzyme Mix. The RNA libraries were structured using cDNA NEBNext Ultra RNA Library Prep Kit for Illumina and purified by Beckman Agencourt AMPure XP beads (Beckman Kurt, Shanghai, China). Library detection and quantification were carried out using Agilent bioanalyzer 2100 and Qubit. TruSeq PE Cluster Kit V4 was used for cBOT automatic mounting. The HiSeq sequencing was implemented using TruSeq SBS Kit v4-HS. DEGs were analysed between KO-ASIP and WT bMECs using the DESeq Bioconductor package [20]. This analysis was based on the negative binomial distribution and was adjusted using Benjamini and Hochberg's approach for controlling the false discovery rate. The fluorescent quantitative PCR validation assay was performed using |log2FC| ≥ 1.5 and p < 0.05 as screening conditions for differential gene expression, and 10 genes related to lipid metabolism and related marker genes were selected for quantitative validation. The gene ontology (GO) analysis of DEGs was conducted by the Cluster-Profiler R package [21], in which the gene length bias was corrected. The GO terms with corrected p < 0.05 were significantly enriched by DEGs. The KEGG is a database resource for understanding high-level functions and utility of the biological systems from molecular-level information (http://www.genome.jp/kegg/ (accessed on 6 February 2018)). We used the cluster Profiler R package software to test the statistical genes.

Determination of Gene Expression Related to Lipid Metabolism
To analyse gene expression of fatty acid metabolism in KO-ASIP cells, SYBR greenbased Q-PCR was used. Initially, total RNA was extracted from 4 × 10 6 cells using FastPure ® Cell/Tissue Total RNA Isolation kit (Vazyme, Nanjing, China) following the manufacturer's instructions. The RNA was quantified by NANODROP 2000 spectrophotometry (Thermo, Shanghai, China). cDNA was prepared using the EasyScript ® All-in-One First-Strand cDNA Synthesis SuperMix for the qPCR kit (Trans, Beijing, China) following the manufacturer's instructions. The cDNA obtained by reverse transcription of 1 µg of RNA was used as an RT-qPCR template using the Essential DNA Green Master Kit (Roche, Beijing, China) and PCR-Max-EC048 Real-Time PCR system (Bibby scientific, Staffordshire, UK). The RT-qPCR reactions were performed at 95 • C for 5 min, followed by 45 cycles of 95 • C for the 30 s, 60 • C for 30 s, and 72 for 20 s. A melting curve was run for each assay. All RT-qPCR reactions were performed in three biological replicates and three technical replicates. mRNA level of each gene was normalized to β-actin. The information on primers pairs was listed in Table 2. The expression fold change was calculated through a data analysis web portal using the ∆∆Ct method, in which ∆Ct is calculated between the gene of interest and an average of reference genes, followed by ∆∆Ct calculations [∆Ct (test group) −∆Ct (control group)]. Fold change was then calculated using the 2 −∆∆Ct formula [22].

Statistical Analysis
Data from experiments are shown as mean ± standard error of the mean. Differential significance was determined by p < 0.05. Gene expression data from qRT-PCR experiments were analysed with the comparative Ct method (2 −∆∆Ct ). A two-tailed t-test was conducted on the data using GraphPad Prism 7.0 (GraphPad Software Inc., San Diego, CA, USA).

Construction of KO-ASIP bMECs
In this study, the gASIP/pX459 re-combinant plasmid was successfully constructed by designing gRNA in the CDS promoter part of the ASIP gene (Figure 1a,b). After screening by stylomycin, the KO-ASIP bMECs were identified with PCR. Electrophoresis on an agarose gel (Figure 1c), identified that the PCR product had a length of 980 bp; after that, we sequenced this product. Comparing the KO-ASIP cell line sequence with the WT sequence revealed that the gRNA sequence had a deletion of 47 bases near the PAM site when the cell succeeded in transfecting the gRNA1 sequence carrier, resulting in shift mutation (Figure 1d). This result proved that the ASIP gene had been knock-outed successfully inside the bMEC. The result of qPCR found that ASIP expression was significantly decreased in the KO-ASIP bMEC line compared to the WT bMEC line (Figure 1e).

Detection of Triglycerides, Cholesterol and Fatty Acid in KO-ASIP bMECs
The results showed that the intracellular triglyceride content was increased (p < 0.05), and the cholesterol content was not significantly changed in the ASIP knockout bMECs (Figure 2a,b). Fatty acids were extracted from the wild-type and the ASIP gene knockout cells, respectively. The composition and content of the fatty acids were analysed. There were significant reductions in monounsaturated fatty acids (MUFA) and saturated fatty

Functional Enrichment Analysis of the DEGs in KO-ASIP bMECs
The RNA-seq analysis of the KO-ASIP bMECs was performed to elucidate how ASIP genes regulate lipid metabolism. A total of 485 differentially expressed genes (DEGs), including 174 up-regulated and 311 down-regulated genes, were identified in KO-ASIP bMECs compared with wild type (Figure 3a,b), which were also presented in the form of a heat map (Figure 3c). Examples: RARRES2, FABP4, ELOVL6, HACD4 and NR4A1. All differential genes in the comparison group were selected and set for clustering analysis. The FPKM values of genes were clustered using hierarchical clustering, and genes or samples with similar expression patterns in the heat map were clustered together (Figure 3d,e). Ten differentially expressed genes (seven up-regulated differentially expressed genes and three Animals 2022, 12, 1389 9 of 17 down-regulated differentially expressed genes) were selected for qPCR analysis to verify the results of RNA-seq. The qPCR results showed that the relative expression trends of the 10 differentially expressed genes detected in WT and KO-ASIP bMECs were consistent with the RNA-seq results, confirming the accuracy of the RNA-seq data ( Figure 4 and Table 4).
bMECs compared with wild type (Figure 3a,b), which were also presented in the form of a heat map (Figure 3c). Examples: RARRES2, FABP4, ELOVL6, HACD4 and NR4A1. All differential genes in the comparison group were selected and set for clustering analysis. The FPKM values of genes were clustered using hierarchical clustering, and genes or samples with similar expression patterns in the heat map were clustered together ( Figure  3d,e). Ten differentially expressed genes (seven up-regulated differentially expressed genes and three down-regulated differentially expressed genes) were selected for qPCR analysis to verify the results of RNA-seq. The qPCR results showed that the relative expression trends of the 10 differentially expressed genes detected in WT and KO-ASIP bMECs were consistent with the RNA-seq results, confirming the accuracy of the RNAseq data (Figure 4 and Table 4). (c) (d)

Prediction and Analysis of Signal Pathway Interaction after ASIP Knockout
They were significantly enriched in 28 GO terms (p < 0.05), including 22 biological processes, 3 cellular components, and 3 molecular function-related GO terms. The GO terms most significantly different in biological processes, cellular components, and molecular function categories are shown in Figure 3d and Table 5. The GO terms of the most important biological process include the immune system, single-organism, cellular, biological regulation, response to stimulus, metabolic, and multicellular organismal processes. The most significant cell components of the GO terms, include the cell part, membrane part, organelle, extracellular region part, and membrane. The most significant molecular function GO terms include binding, catalytic activity, structural molecule activity, signal transducer activity and transporter activity.

Prediction and Analysis of Signal Pathway Interaction after ASIP Knockout
They were significantly enriched in 28 GO terms (p < 0.05), including 22 biological processes, 3 cellular components, and 3 molecular function-related GO terms. The GO terms most significantly different in biological processes, cellular components, and molecular function categories are shown in Figure 3d and Table 5. The GO terms of the most important biological process include the immune system, single-organism, cellular, biological regulation, response to stimulus, metabolic, and multicellular organismal processes. The most significant cell components of the GO terms, include the cell part, membrane part, organelle, extracellular region part, and membrane. The most significant molecular function GO terms include binding, catalytic activity, structural molecule activity, signal transducer activity and transporter activity.

Verification of Differential Gene Expression Level
A KEGG enrichment analysis of differential expression genes showed that differential expression genes were significantly enriched into 46 KEGG pathways, with the first 30 most significant ones shown in Figure 3e, including the PPAR signalling pathway, Adipocytokine signalling pathway and AGE-RAGE signalling pathway in diabetic complications. Differentially expressed genes related to lipid metabolism are mainly enriched in fatty acid metabolism pathways, such as long-chain fatty-acyl-CoA biosynthetic process, long-chain fatty acid metabolic process, unsaturated fatty acid biosynthetic process et al. (Table 6). We validated the SLC26A2, BGN, ACSL1, PGM2L1, SCD, RARRES2, FABP4, ELOVL6, and HACD4 and NR4A1 genes in the differentially expressed genes by RT−qPCR, The quantitative results showed a significant upregulation of SLC26A2, BGN, ACSL1, FABP4, ELOVL6, and NR4A1 (p < 0.05, Figure 4) when compared with WT bMECs, but not significantly for HACD4. PGM2L1, SCD, and RARRES2 were significantly down-regulated (p < 0.05, Figure 4).

Effect of ASIP Knockout on the Milk Lipid Metabolism Pathways
A quantitative analysis of genes associated with the fatty acid metabolism pathway was performed using qPCR technology to resolve the regulatory effect of ASIP knockout on genes involved in the fatty acid metabolism pathway in bovine mammary epithelial cells. The results showed that the knockout of ASIP affected genes related to fat and fatty acid metabolism in bovine mammary epithelial cells. Knockout of ASIP resulted in a significant down-regulation of mRNA levels in key fatty acid metabolism genes such as PPARγ, FASN, and SCD1 (p < 0.05, Figure 4). However, key genes like FABP4, ELOVL6, and HACD4 associated with fatty acid transport and elongation were significantly up-regulated (p < 0.05, Figure 4).

Discussion
Previous work in our laboratory has demonstrated that the addition of recombinant protein ASIP altered mRNA expression of genes related to lipid metabolism and significant increases triglycerides and cholesterol content in bMECs (unpublished data). These studies could imply the involvement of ASIP genes in the synthesis and catabolism of lipids in cattle. To further explore the effect of the ASIP gene on lipid metabolism in bMECs, we successfully constructed the ASIP knockout bMECs using CRISPR/Cas9 technology. The results showed that the expression of ASIP was significantly reduced at the mRNA level in the knockout cell lines.
The content of triglycerides, cholesterol and fatty acids was firstly measured to investigate further the effect of ASIP knockout on lipid metabolism in bMECs. The results presented that ASIP knockout affected the composition of medium and long-chain fatty acids, increasing saturated and polyunsaturated fatty acids content and a decreased level of monounsaturated fatty acids in bMECs. The content of saturated fatty acids myristic acid (C14:0), pentadecanoic acid (C15:0), palmitic acid (C16:0), heptadecanoic acid (C17:0), stearic acid (C18:0), arachidonic acid (C20:0), 21-carbonic acid (C21:0), and ditetradecanoic acid (C24:0) were significantly up-regulated. The significant increase in saturated fatty acids of 14-24 C-atoms, which are substrates for triglycerides production [23], could provide the fundamental element for the elevation of triglycerides in bMECs after knockout of the ASIP gene. It has been widely accepted that saturated fatty acids and TG in dairy fat, which may lead to obesity, and high blood pressure, are detrimental to health. Moreover, saturated fatty acids such as myristic (14:0) and palmitic (16:0) acids could enhance the level of low-density lipoprotein (LDL-and high-density lipoprotein (HDL-) cholesterol in the blood [24]. Despite their inherent ability to inhibit bacteria and viruses, a high intake of these fatty acids could raise blood cholesterol content which could cause heart disease, weight gain, and obesity [5]. For some polyunsaturated fatty acids, there were significant increases in linoleic acid (C18:2N6), gamma-linoleic acid (C18:3N6), and alpha-linolenic acid (C18:3N3) in bMECs with ASIP knockout. The elevation of these fatty acids could lower blood lipids and improve vascular health status [25,26]. Ni Dan et al. investigated the effect of the complete absence of certain long-chain fatty acids (LCFA) on milk lipid metabolism in bMECs using the univariate principle. It was found that triglycerides synthesis was significantly down-regulated in bMECs with the absence of C18:0, C18:2N6 or C18:3N3 [27]. This implies that elevated C18:2N6 and C18:3N3 may also have a facilitating effect on triglycerides synthesis. Furthermore, the monounsaturated fatty acid oleic acid (C18:1N9), which accounted for the largest proportion of milk lipid [28], and other monounsaturated fatty acids such as palmitoleic acid (C16:1N7) and C20:1N9 were significantly down-regulated under ASIP knockout conditions. It has been indicated that consuming diets containing high amounts of monounsaturated fatty acids effectively reduces plasma lipids, such as cholesterol and triacylglycerol concentrations [29]. Furthermore, monounsaturated fatty acids are more protective against atherosclerosis than polyunsaturated fatty acids [30]. Thus, the reduction of ASIP can alter fatty acid composition in the intracellular bMECs and may further in milk. However, this alteration in milk may be detrimental to human health. It suggested that the lower expression level of ASIP in mammary epithelial cells was not recommended to breed dairy cows with high milk quality in the future.
To preliminarily investigate the molecular mechanism of the effect of ASIP on the lipid metabolism in bMECs, the transcriptome sequencing was analysed. The results showed that 485 differentially expressed genes were detected, including 311 up-regulated genes and 174 down-regulated genes. The GO terms analysis of differentially expressed genes indicated 112 differentially expressed genes were significantly enriched in 28 GO entries (p < 0.05), including 22 biological processes, three cellular components, and three molecular functionally relevant GO entries. The KEGG analysis indicated significant enrichment of differential expressed genes into 46 KEGG pathways, including PPAR signalling pathway, Adipocytokine signalling pathway and AGE-RAGE signalling pathway in diabetic complications. We screened 10 differentially expressed genes to satisfy the conditions of |log2FC| ≥ 1.5 and p < 0.05 as significant differential expression. The quantitative results showed that although our transcriptome was not biologically replicated, the sequencing results and relative gene quantification results were relatively consistent. Differentially expressed genes related to lipid metabolism are mainly enriched in fatty acid metabolism pathways, such as long-chain fatty-cyl-CoA biosynthetic process, long-chain fatty acid metabolic process, unsaturated fatty acid biosynthetic process, etc. The qPCR results of differentially expressed genes indicated that expression of ACSL1, FABP4 and ELOVL6 genes was elevated and the SCD FASN expression was decreased in the knockout cell lines, consistent with the transcriptome sequencing results.
Key genes of the fatty acid metabolic pathway were detected to explore the molecular mechanisms of fatty acid alterations in bMECs after ASIP knockout. As an enzyme directly related to lipid synthesis and fatty acid transport and degradation, ACSL1 can significantly influence fatty acid metabolism. By overexpressing ACSL1 in bovine adipocytes, Zhao et al. found a significant increase in saturated and polyunsaturated fatty acid (PUFA) content, particularly C16:0 and C18:0. Overexpression of ACSL1 further increased the proportion of eicosapentaenoic acid (EPA) [31]. In the present study, ACSL1 expression was significantly up-regulated in ASIP knockout bMEC, which may lead to SFA accumulation such as palmitic acid (C16:0) and C18:0 content as well as an increase in PUFA and C20:5N3. Moreover, the primary function of FABP4 is to transport fatty acids across the membrane, and overexpression increases FA transport to enhance energy and lipid metabolism [32]. The deletion of FABP4 results in an impaired fatty acid efflux, leading to an increase in fatty acids in FABP4-deficient cells. It has been shown that FABP4 maintains eicosanoid homeostasis in macrophages in mice [33]. The FASN gene is a multifunctional peptidase for producing saturated fatty acids. It is responsible for all steps in the ab initio synthesis of palmitic acid (C16:0) from acetyl-coenzyme an (acetyl coenzyme a) and malonyl coenzyme a (malonyl coenzyme a) [34]. The specific knockout of FASN in mouse mammary glands significantly reduces the content of medium and total long-chain fatty acids in milk [35]. This may indicate the down-regulation of FASN may lead to a decreased synthesis rate of C16:0 and finally lower the level of C16:0 in bMEC with ASIP knockout. Furthermore, SCD is the rate-limiting enzyme that catalyses the synthesis of monounsaturated fatty acids (MUFAs) from saturated fatty acids (SFAs). The major substrates of SCD are C16:0 and C18:0 FA, which can be converted to C16:1 cis9 and C18:1 N9 [36]. Association analysis of SNP loci of SCD with milk lipid ty acid traits in 297 Holstein cows was conducted by Mele et al. They found that cows with the AA genotype had higher levels of cis-9C18:1, total monounsaturated fatty acid levels and C14:1/C14:0 ratios in milk compared to cows with the VV genotype [37]. In this study, there was a significant down-regulation of SCD expression in ASIP knockout bMEC, leading to SFA accumulation, such as palmitic acid (C16:0) and C18:0 content and the reduction of MUFA, like C16:1N7 and C18:1N9. For C16:0, both FASN and SCD could influence its level in bMEC. As C16:0 content was lower in the control group, it may imply that the consumption rate of C16:0 by SCD could be lower than the synthesis rate by FASN in ASIP knockout cells. Furthermore, Junjvlieke et al. used adenovirus to overexpress the ELOVL6 gene, a rate-limiting enzyme in the long-chain fatty acid elongation reaction, in bovine adipocytes and found a significant increase in the proportion of C18:0 and C20:4N6 fatty acids [38]. Our results suggested that the mRNA expression level of ELOVL6 was significantly increased in bMEC with ASIP knockout which could contribute to the upregulation of C18:0. Additionally, the expression of HACD4 mRNA was detected in this study, as it is an important enzyme at the third step of fatty acid extension. Even though there was a significant elevation in 18:2N6 and 18:3N3 fatty acids after ASIP knockout in bMECs, the higher mRNA expression of HACD4 was not detected. It may indicate that the enhancement of 18:2N6 and 18:3N3 fatty acids in ASIP knockout of bMECs was not mainly regulated by HACD4. Moreover, we found that the knockout of the ASIP gene in bMECs resulted in significant downregulation of the peroxisome proliferatoractivated receptor (PPARγ), which has a key role in lipid metabolism, promoting free fatty acid uptake and accumulation of triacylglycerols in adipose tissue and liver [39]. The down-regulation of PPARγ may lead to triglycerides down-regulation. Garin-Shkolnik et al. found that high expression of fatty acid-binding protein 4 (FABP4, also known as aP2) could down-regulate PPARγ expression [40]. Tingting Li et al. found that in human hepatocytes overexpressing ASCL1 caused a significant decrease in fatty acid synthesis pathway-related genes such as FASN and SCD1 by suppressing PPARγ expression to an increase in triglycerides levels [41]. This is consistent with the results of gene and triglycerides changes in our experiments. In another human hepatocyte line (L02 cells) overexpressing PPARγ, it was found that up-regulation of PPARγ could affect cholesterol efflux and thus lower cholesterol levels through multiple pathways [42]. Our cholesterol assay results indicate a trend of up-regulation, but one not significant in bMECs after knockout of ASIP, but its molecular mechanism needs to be further investigated. These results suggest that the knockout of ASIP can influence the expression of key genes in the fatty acid metabolism pathways, thus affecting the intracellular lipid metabolism, especially the composition of fatty acids. It has been shown that ASIP affects the cellular content of cAMP by competitive binding to MC1R [43,44]. However, whether ASIP affects the expression of key genes in lipid metabolism in bMECs by regulating cAMP content still needs further investigation.

Conclusions
In summary, the knockout of ASIP by using Cas9 technology could alter lipid composition in bMECs, such as the increased content of triglycerides, saturated fatty acids, and polyunsaturated fatty acids. Some of the differentially expressed genes that were screened by transcriptome sequencing and related to fatty acid synthesis pathway were then analysed to preliminarily elucidate the regulatory mechanism of the ASIP gene in fatty acid metabolism in bMECs. These results could provide a reference for selecting potential targets in the molecular breeding work of dairy cattle in the future.