AMP-activated protein kinase mediates (-)-epigallocatechin-3-gallate (EGCG) to promote lipid synthesis in mastitis cows

Abstract

Mastitis, a serious threat to the health and milk production function of dairy cows decreases milk quality. Blood from three healthy cows and three mastitis cows were collected in this study and their transcriptome was sequenced using the Illumina HiSeq platform. Differentially expressed genes (DEGs) were screened according to the |log2FoldChange| > 1 and P-value < 0.05 criteria. Pathway enrichment and functional annotation were performed through KEGG and GO analyses. Finally, the mechanism of the AMP-activated protein kinase (AMPK) mediation of (-)-epigallocatechin-3-gallate (EGCG) to promote lipid metabolism in mastitis cows was analyzed in bovine mammary epithelial cells (BMECs). Transcriptome analysis revealed a total of 825 DEGs, with 474 genes showing increased expression and 351 genes showing decreased expression. The KEGG analysis of DEGs revealed that they were mainly linked to tumour necrosis factor, nuclear factor-κB signalling pathway, and lipid metabolism-related signalling pathway, whereas GO functional annotation found that DEGs were enriched in threonine and methionine kinase activity, cellular metabolic processes, and cytoplasm. AMPK expression, which is involved in several lipid metabolism pathways, was downregulated in mastitis cows. The results of in vitro experiments showed that the inhibition of AMPK promoted the expression of lipid synthesis genes in lipopolysaccharide-induced BMECs and that EGCG could promote lipid synthesis by decreasing the expression of AMPK and downregulating the expression of inflammatory factors in inflammatory BMECs. In conclusion, our study demonstrated that AMPK mediated EGCG to inhabit of inflammatory responses and promote of lipid synthesis in inflammatory BMECs.

Introduction

Milk fat, an energy substance and important nutrient in milk, generally accounts for 4%–6% of milk. The fat in milk is in the form of droplets of milk fat, which are mainly composed of phospholipid-coated triglycerides (TGs), diglycerides, and monoglycerides, along with a small amount of cholesterol and vitamin A: this structure is also known as the milk fat globule.^1,2 Because of its small diameter, high emulsifying rate, and melting point below normal body temperature, milk fat is easily absorbed in the human body.³ The synthesis and metabolism of milk fat are affected by various factors during the growth and development of dairy cows, and its traits are closely related to genetic factors and regulated by several transcription factors.⁴ Recent research has focused on the process of synthesizing fatty acids (FAs) from scratch, including FA uptake, activation, intracellular transport, elongation, and desaturation, as well as triacylglycerol (TAG) synthesis and lipid droplet formation.⁵ FA ab initio synthesis rate-limiting enzymes such as acetyl-CoA carboxylases alpha (ACACA) and sterol regulatory element binding protein 1c (SREBP-1c) have been identified as the downstream target genes of AMP-activated protein kinase (AMPK). ACACA catalyzes the conversion of acetyl-CoA to malonyl-CoA and initiates FA ab initio synthesis.⁶ SREBP-1c is a lipid homeostasis transcription factor that regulates milk fat synthesis in dairy cows.⁷

Bovine mammary epithelial cells (BMECs) are the predominant cells in breast tissue that constitute the ‘microenvironment’ of the mammary gland. They have innate immune function and are the main site of milk secretion.⁸ Feeding cows with high concentrates of diet to meet energy requirements in early and mid-lactation can lead to a disturbance in rumen microbial metabolism, a decrease in pH, and the production of large amounts of lipopolysaccharides (LPS) and long-chain FAs into the bloodstream.⁹ Excessive LPS can enter the bloodstream and cause systemic inflammatory response, fat suppression and several metabolic diseases¹⁰ as well as inhibit the expression of lipid-synthesizing genes, thereby altering the composition of milk.¹¹ The decline in milk quality and production efficiency has severely limited the rapid development of the dairy industry.

AMPK is the central hub of mammalian metabolism that regulates gene transcription (such as lipid synthesis, oxidation and lipolysis) by directly phosphorylating proteins or via other means and plays a key role in regulating anabolism or catabolism.¹² The activation of AMPK inhibits rate-limiting enzymes and transcriptional regulators involved in the ab initio synthesis of FAs and reduces the acetylation process of peroxisome proliferator-activated receptorγcoactivator-1 (PGC-1α), which subsequently enhances the oxidation of FAs and regulates the synthesis of milk lipids in mammary epithelial cells.¹³ As a key transcription factor in lipid synthesis, SREBP-1c is involved in the transcriptional activation of genes encoding the rate-limiting enzymes in adipogenesis,¹⁴ thereby positively regulating the expression of FA synthase. In addition, the activity of SREBP-1c is negatively regulated by the negative feedback regulation of AMPK, the activation of which directly phosphorylates SREBP-1c and inhibits its nuclear translocation. This process indirectly regulates gene expression in the adipose tissue and inhibits TG synthesis, whereas the upregulation of SREBP-1c expression increases TG synthesis.¹⁵ Meanwhile, resveratrol (RES) has been shown to ameliorate lipogenesis and inflammatory responses in human SZ95 adipocytes in vitro via the AMPK signalling pathway.¹⁶ In addition, sodium butyrate inhibited anti-inflammatory effects by inhibiting the phosphorylation of AMPK, thereby regulating lipid metabolism disorders, and promoting the de novo synthesis of TGs in LPS-induced mastitis in dairy cows.¹⁷

(-)-Epigallocatechin-3-gallate (EGCG) is a bioactive substance with favourable therapeutic effects on inflammatory diseases both in vitro^18,19 and in vivo.^18,19 EGCG can directly inhibit reactive oxygen species production by scavenging free radicals; the D-ring of the gallyl group in EGCG, which has a strong free radical scavenging activity, can prevent the oxidative damage of DNA in cells by reducing the expression of cytochrome P450.²⁰ The cotreatment of BMECs with EGCG and hydroxytyrosol (HTyr) can the synthesis of inflammation-related molecules, thereby potentially protecting against the development of mastitis in periparturient dairy cows.²¹ The reduced accumulation of hepatic fat deposits in EGCG-fed mice may be due to the activation of AMPK phosphorylation, which inhibits the expression of adipogenic genes.²² However, the regulation of EGCG in lipid metabolism in mastitis dairy cows has not been reported. Based on transcriptome sequencing, this study explored differences between the blood transcriptomes of mastitis cows and healthy cows as well as the key genes of lipid regulation in mastitis cows, further revealing the regulatory molecular mechanisms of regulating lipid metabolism in dairy cows with mastitis and providing gene targets and basic theory for preventing and regulating lipid metabolism in mastitis cows.

Materials and methods

Sample collection

Holstein cows were selected from a large-scale dairy farm in Lingwu City, Ningxia Autonomous Region. Three healthy cows and three cows with California mastitis test (CMT)-positive mastitis were screened using clinical performance evaluation and CMT detection criteria, and the results revealed no other complications. Approximately 10 mL blood samples were collected from the caudal vein.

Sample preparation for sequencing

Total RNA from whole blood was extracted using the TRIzol (Takara, Japan) method. Library construction was performed using the Agilent 2100 Bioanalyzer. A total of six libraries were constructed: three inflammation groups (Y-1, Y-2 and Y-3) and three control groups (C-1, C-2 and C-3). Each sample was individually processed and individually sequenced. After library construction, the samples were sequenced using the light-based next-generation sequencing technology. Sequencing services and library construction were provided by Shanghai Baypur Biotech.

Data processing and quality control

To ensure high quality reads and accurate analyses, data quality was strictly controlled and assessed using the following measures: (1) Cutadapt was used to remove sequences with a splice at the 3′ end and (2) reads with an average quality score, i.e., lower than Q20, were removed. High quality clean data obtained after quality control were available in the FASTQ format. Transcriptomic analyses were performed using the bovine reference genome, with the following version information: Bos_taurus.ARS-UCD1.2. dna. toplevel. fa.

Transcriptome data analysis

The ‘Read Count’ value of the gene was computed and considered as the raw expression of the genes. This expression was normalized using kilobases per million fragments (FPKM), and FPKM > 1 is considered to be gene expression.

Differentially expressed gene (DEG) analysis

DEG analysis was performed using DESeq, and the screening criteria for the selected DEGs were |log2FoldChange| > 1 and P-value < 0.05. Gene ontology (GO) enrichment analysis was performed using R/topGO, and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis was performed using R/cluster analyzer.

Real-time fluorescent quantitative PCR (RT-qPCR)

RNA was extracted from the blood and cells using the TRIzol method. The 260/280 values and concentration of RNA were measured using a multifunctional full-wavelength enzyme labelling instrument (Synergy|LX, Bio-Rad, USA). cDNA was obtained via reverse transcription with a reverse transcription reagent (Takara, Japan). The RT-qPCR system is presented in Supplementary Table S1. The internal reference gene was β-actin. The information of primers synthesized by Anhui Synthesis Biological Co. Ltd (China) is shown in Supplementary Table S2.

BMEC culture

The BMEC cell lines used in this study were derived from prefrozen cultures stored in our laboratory. Approximately 8 × 10⁵ cells were inoculated into 6-well plates, and the cells were treated when the culture density reached 70–80%. BMECs in each group were treated in the following manner. For the CSN15887 + EGCG + LPS group: the AMPK inhibitor (CSN15887, CSNpharm, USA) was added to the medium and incubated for 24 h, followed by EGCG (Sigma, USA; concentration 10 µg/mL) treatment for 12 h, and 105 μL LPS (Sigma, USA; the final concentration was 50 ng/μL)²³ treatment for 12 h. For the EGCG + LPS group, EGCG was added and incubated for 12 h, followed by LPS treatment for 12 h. For the LPS + CSN15887 group, AMPK inhibitor (CSN15887) was added and incubated for 24 h, followed by LPS treatment for 12 h. For the LPS group, LPS was added and incubated for 12 h. Only the cell growth medium (10% fetal bovine serum (BI, Israel) + DME/F12 (Hyclone, USA)) was used in the CON group. All groups of cells were collected at the same time.

After incubation for 48 h, the growth medium was discarded, the cells were washed with PBS, and the induction medium (10% fetal bovine serum + DME/F12 + 5 μg/mL insulin, (Sigma, USA) + 5 μg/mL hydrocortisone (MCE, China) + 20 ng/mL prolactin (Prospec, China)) was added. This was recorded as Day 0 of the induction. The induction medium was changed every other day, and the cells were stained 4 days after the induction.

Immunofluorescence and enzyme linked immunosorbent assay (ELISA)

BMECs were inoculated into 6-well cell culture plates and incubated to a density of approximately 80%. After discarding the medium, PBS (Hyclone, USA) was used to wash the cells thrice, and 1 mL of pre-cooled 4% paraformaldehyde was added to fix the cells. After 20 min of incubation, the fixative was discarded and the cells were washed thrice with PBS for 5 min each. Approximately 1 mL of 0.5% Tritonx-100 solution (Sigma, USA) was added to cells and incubated for 10 min at room temperature, followed by three PBS washes. Then, 1 mL of 2% BSA (Roche, Switzerland) solution was added and incubated for 1 h at room temperature. After discarding the BSA solution, add 1 mL primary antibody (CK-18, Gclone, China, 1:1,000) was added and incubated at 4 °C for several hours. After incubation, the primary antibody CK-18 was discarded via three PBS washes for 10 min each. Next, 1 mL of secondary antibody (goat anti-rabbit IgG, Gclone, China) was added and incubated at room temperature for 2 h, which was subsequently washed with PBS thrice. The cell nucleus was stained with DAPI (Solabio, China) dye solution. After incubation for 10 min, the DAPI solution was discarded, and the cells were washed three times with PBS and photographed using a fluorescence microscope.

The supernatants (or blood) of each BMEC treatment group were collected and assayed for tumour necrosis factor (TNF)-α and interleukin (IL)-6 levels according to the respective ELISA kit (Ruixin, Quanzhou, China) instructions.

Cell staining

Oil red O staining

The culture medium was discarded after 4 days of BMEC induction, and the cells were fixed with 4% formaldehyde solution for 1 h. The fixed cells were washed with 60% isopropanol, left to stand for 5–6 min, and then treated with 1–2 mL of oil red O (Sorabio, China) working solution and incubated for 20 min at room temperature away from light. Finally, the oil red O working solution was discarded, and the cells were washed thrice with PBS and photographed under a microscope.

Bodipy staining

A bodipy storage solution (5 nM) was prepared by dissolving bodipy with an appropriate amount of dimethyl sulfoxide (DMSO, Amresco, USA) and stored at 4 °C in a refrigerator. The bodipy storage solution was diluted with PBS to 5 μM working solution. The cells were placed in 4% paraformaldehyde for 40 min, followed by the addition of 500 μL bodipy working solution into each well of a 6-well plate and incubation at room temperature away from light for 30 min. The bodipy working solution was discarded, and the cells were washed thrice with PBS. Next, the DAPI staining solution was added to stain the cell nucleus for 10 min. The cells were washed thrice with PBS, and photographed using a fluorescence microscope.

Data analysis

For each group, at least three independent experiments were performed. The RT-qPCR results of fluorescence quantification were analyzed using the 2^-ΔΔCt method, with the data were expressed as mean ± standard error. The data were analyzed using one-way analysis of variance (One-way ANOVA) in GraphPad Prism 9.5.1. P < 0.05, indicated with a single asterisk, was considered statistically significant, whereas two asterisks indicated P < 0.01 (highly significant). The grayscale values of the fluorescence images were analyzed using the ImageJ software.

Results

RNA-seq data

To explore transcription level changes in mastitis cows, Illumina HiSeq platform was used to analyze the transcription level of mastitis cows (Y-1, Y-2 and Y-3) and healthy cows (C-1, C-2 and C-3). The base content of Q30 in the six samples was more than 92.19%. A reference genome index was created with Bowtie2. On comparing the filtered sample reads with the cattle genome, the mapped ratio of the samples was minimum 84.14% and maximum 91.49%. The sample quality control data are shown in Supplementary Table S3. High-quality transcriptome sequencing results provided reliable raw data for subsequent assembly.

RNA-seq sample correlation and gene expression distribution

The correlation between the expression levels of RNA-seq samples was analyzed to ensure the rationality and accuracy of the results. The results of PCA showed a strong correlation between the two groups (Supplementary Figure S1A). The construction of violin maps based on the localization of FPKM fragments showed that the distribution of genes in all samples was homogeneous and symmetrical (Supplementary Figure S1B). This result indicated that the distribution of gene expression in all samples was correct.

DEG analysis

A total of 825 DEGs were identified using the DESeq analysis of gene expression differences and volcano mapping, of 351 were differentially upregulated and 474 were differentially downregulated (Figure 1A). According to the results of genomic information and RNA differential expression analysis, the differentially expressed RNA was labelled on the genome. DEGs were evenly distributed on each chromosome (Figure 1B), and the bidirectional data clustering of all differential genes showed that DEGs in the healthy cows were clustered into one category, while those in the mastitis cows were clustered into one category (Figure 1C). All DEGs were divided into nine distinct expression patterns as per K-means analysis (Supplementary Figure S2).

Figure 1.

DEG analysis. A. Volcano map of DEGs. B. Genome circle diagram. C. DEG clustering thermogram.

GO function annotation and KEGG enrichment analysis of DEGs

The results of GO enrichment analysis of DEGs were classified into GO categories according to molecular function (MF), biological process (BP), and cellular component (CC), and the 10 GO term entries with the most significant enrichment in each GO category were selected for display. MF was mainly enriched in intracellular, endosomal membrane-bound, and membrane-bound organelles; BP was mainly enriched in enzyme-bound and ion-bound organelles; and CC was mainly enriched in the regulation of immune response, regulation of cellular metabolism, and bioregulation (Figure 2A). The 20 GO entries with the lowest FDR values were selected for display, and the DEGs were mainly enriched in intracellular, organelle, and organic metabolic process entries (Figure 2B). KEGG results showed that DEGs were mainly enriched in the pathways of genetic information processing, environmental information processing and cellular processes (Figure 2C). According to KEGG enrichment results, the top 20 GO items with the lowest FDR value and the most significant enrichment were selected, and these included nuclear factor-κB (NF-κB) signalling pathway, blood lipid and atherosclerosis pathway, and TNF signalling pathway (Figure 2D).

Figure 2.

GO function annotation and KEGG enrichment analysis of DEGs. A. GO enrichment analysis of DEGs. B. FDR values and the top 20 GO entries with the most significant enrichment. C. KEGG enrichment analysis of DEGs. D. FDR values and the top 20 most significant signalling pathways.

RT-qPCR verification

Six DGEs were randomly selected, and their RT-qPCR results were compared with RNA-seq results to verify the authenticity of the transcriptome data (Figure 3). The RT-qPCR results were consistent with the RNA-Seq results, indicating that the RNA-seq data were valid and reliable.

Figure 3.

RT-qPCR verification.

Inhibition of AMPK promoted lipid synthesis in BMECs induced by LPS

The analysis of transcriptome sequencing data revealed that AMPK genes, which are located in the signalling pathway related to lipid metabolism, were downregulated in the blood of mastitis cows, suggesting that AMPK regulates lipid metabolism in mastitis cows (Supplementary Figure S3). These results were subsequently verified at the cellular level. The cells were identified via immunofluorescence, and the results showed that the positive cells containing CK-18 protein were more than 95%, indicating that the cells were BMECs (Figure 4A). ELISA for inflammatory factors in the cell culture medium revealed that the AMPK inhibitor CSN15887 inhibited the expression of TNF-α and IL-6 in LPS-induced BMECs (TNF-α, Figure 4B, P< 0.05; IL-6, Figure 4C, P< 0.05). Up-regulation of AMPK mRNA expression after LPS-induced of BMECs (Supplementary Figure S4). The expression of genes related to lipid metabolism after the inhibition of inflammatory cells with the AMPK inhibitor CSN15887 was further examined, and it was found that the expression of lipid synthesis genes was downregulated in LPS-induced cells and upregulated in cells cocultured with CSN15887 (stearoyl-CoA desaturase-1, SCD-1, Figure 4D, P< 0.05; SREBPR-1c, Figure 4E, P< 0.05; CD-36, Figure 4F, P< 0.05; fatty acid synthase, FASN, Figure 4G, P< 0.05); The lipid oxidation gene expression was promoted in LPS-induced cells and suppressed in cells cocultured with CSN15887 (acyl-coA oxidase, ACO, Figure 4H, P< 0.05; carnitine palmitoyltransferase 1, CPT1A, Figure 4I, P< 0.05; carnitine palmitoyltransferase 2, CPT2, Figure 4J, P< 0.05). LPS significantly reduced TG content, but significantly increased it after AMPK inhibition (Figure 4K, P< 0.05). The oil red O staining (Figure 4L) and bodipy staining (Figure 4M, Supplementary Figure S5A) results showed that LPS significantly reduced the lipid droplet accumulation in cells, but significantly increased the accumulation after AMPK inhibition. These results indicated that LPS can inhibit the lipid synthesis in BMECs, whereas AMPK can inhibition can effectively promote lipid synthesis in BMECs with inflammatory injury.

Figure 4.

Inhibition of AMPK promotes lipid synthesis in BMECs induced by LPS. A. Cell identification. B-C. Determination of TNF-α and IL-6 levels in cell culture media using ELISA. D-G. RT-qPCR detection of lipid synthesis gene mRNA expression. H-J. RT-qPCR detection of lipid oxidation gene mRNA expression. K. TG content. L. Oil red O staining. M. Bodipy staining. *, P < 0.05, **, P < 0.01.

AMPK mediates EGCG to promote lipid synthesis in BMECs with inflammatory injury

To explore whether AMPK can mediate EGCG to regulate the lipid synthesis in BMECs with inflammatory injury, the optimal concentration of EGCG was screened using CCK-8. It was found that the best cell viability was achieved when the concentration of EGCG was 10 μg/mL, and that cell viability gradually decreased with the increasing concentration of EGCG (Figure 5A, P< 0.05). Therefore, we chose the EGCG concentration of 10 µg/mL for subsequent experiments. RT-qPCR results revealed that cotreatment with EGCG and LPS reduced the expression of TNF-α and IL-6 compared to the LPS-treated group; however, their expression remained reduced when AMPK was inhibited (TNF-α, Figure 5B, P< 0.01; IL-6, Figure 5C, P< 0.01). Taken together, we showed that EGCG and CSN15887 had a synergistic effect in suppressing the expression of inflammatory factors. Meanwhile, the cotreatment with EGCG and LPS increased the expression of lipid synthesis genes in BMECs, but the addition of CSN15887 significantly decreased lipid synthesis gene expression (FASN, Figure 5D, P< 0.01; SREBP-1, Figure 5E, P< 0.01; SCD-1, Figure 5F, P< 0.01). The opposite was true for the expression level of lipid oxidation genes (Figure 5G, P < 0.05). The TG content in each treatment group was examined, and the results revealed that TG was significantly increased after the addition of EGCG and further increased by the inhibition of AMPK (Figure 5H, P< 0.05). The lipid droplet accumulation in each treatment group was detected using oil red O and bodipy staining assays, and the results showed that lipid droplet accumulation increased in the EGCG + LPS group compared with the LPS group, whereas the difference in lipid droplet accumulation was not significant after the inhibition of AMPK (Figure 5I-J, Supplementary Figure S5B). In summary, EGCG promoted lipid synthesis in BMECs with inflammatory injury, and the promotion effect of EGCG was weakened after AMPK was inhibited.

Figure 5.

AMPK mediates EGCG to promote lipid synthesis in BMECs with inflammatory injury. A. CCK-8-based detection of cellular activity after treatment with the different concentrations of EGCG. B-C. The concentration of TNF-α and IL-6 was detected using ELISA. D-G. The mRNA expressions of FASN, SRFBP-1, SCD-1, and ACO was detected via RT-qPCR. H. TG content. I-J. Oil red O staining and bodipy staining. *, P < 0.05, **, P < 0.01.

Discussion

Cow mastitis, considered one of the most challenging diseases in the world, is characterized by the diversity of pathogenic microorganisms and the complexity of host responses, and it is a major limiting factor for the global dairy industry.²⁴ Mastitis results in reduced milk production, lowered milk quality, and increased in treatment costs and early animal culling. Indeed, it has caused significant economic losses to the livestock industry.^25,26 To improve the mammary glands health of dairy cows, the RNA-seq technology was used to deeply analyze transcriptomic changes in the body fluids of dairy cows with mastitis to characterize changes at the gene transcription level caused by pathogenic microorganisms in mammary glands several hours before infection, help uncover the pathogenesis of mastitis, and provide technical support for exploring new targets in terms of the pathogenesis and prevention of mastitis.²⁷ Research has have found that the mastitis reaction caused by Escherichia coli is faster and more intense than that caused by Streptococcus and Staphylococcus aureus, with a more pronounced effect on the transcriptome.²⁸ At present, high-throughput methods such as transcriptomics,²⁹ metabonomics, and proteomics have been widely used in animal production³⁰ to reveal the pathogenesis of mastitis in dairy cows to the maximum extent. For example, the transcriptome analysis of milk somatic cells from cows with subclinical mastitis revealed that CIITA may play a key role in regulating the response of such animals to subclinical intramammary infections.³¹ Moreover, the serum metabolome of subclinical intramammary infected cows revealed that certain metabolites, such as 3-hydroxybutyric acid, acetone, allantoin, carnosine, citrate, and ethanol, were associated with rumen fermentation, energy metabolism, urea synthesis and metabolism, immune and inflammatory responses, and mammary permeability; that subclinical intramammary infections were associated with systemic disease; and that the metabolic profiles of animals with subclinical mastitis were altered in ways that were relevant to the pathogens of mastitis.³² Finally, tandem mass tagging-based quantitative proteomics revealed potential targets associated with episodes of subclinical mastitis in dairy cows, with the overexpression of CHI3L1, LBP, GSN, GCLC, C4 and PIGR being proteins positively correlated with events that trigger host defences, thereby stimulating the production of cytokines and inflammatory molecules.³³

BMECs, the major cell population of the mammary tissue of dairy cows, are also important immune cells.³⁴ Lactolipids are derived from de novo synthesis in blood and mammary glands and are regulated by the AMPK signalling pathway.^35,36 Menthol was shown to exert autophagy by modulating the AMPK signalling pathway, thereby reducing the inflammatory response in BMECs and restoring the synthesis of milk components in inflammation-injured BMECs.³⁷ Milk fat consists mainly of total cholesterol, diglycerides, phospholipids, cholesterol, and free FAs, which are synthesized and secreted by BMECs. SREBP-1c, a transcription factor associated with lipid metabolism, plays a crucial role in the uptake, transport, and metabolism of milk fat³⁸ and regulates FASN and SCD-1 to participate in lipid synthesis and deposition. Studies have shown that 10 μg/mL LPS can downregulate the mRNA expression of SREBP-1c and inhibit the transport of SREBP-1c to the nucleus,³⁹ which is consistent with the results obtained in this study. AMPK is a negative inflammatory mediator that can reduce the occurrence of inflammation in BMECs.⁴⁰ AMPK is readily activated under inflammatory conditions.⁴¹ The activated AMPK can upregulate the mRNA expression of lipid oxidation genes, such as CPT1A, CPT2, and ACO and downregulate the mRNA expression of lipid synthesis genes, such as SREBP-1C, FAS, and SCD-1, demonstrating that AMPK can inhibit milk fat synthesis.^42,43 However, in the present study, the inhibition of AMPK pathway in BMECs downregulated the mRNA expression of lipid oxidation genes and upregulated the mRNA expression of lipid synthesis genes in BMECs. thereby promoting lipid synthesis in BMECs. Meanwhile, the activation of AMPK can increase glucose uptake and promote the onset of glycolysis, causing lipolysis and oxidation. Different nutrients, including resveratrol, berberine, curcumin and flavonoids, activate AMPK and its downstream genes.⁴⁴ The activation of AMPK in macrophages can reduce NF-κB and TNF-α secretion,⁴⁵ and in inflammatory pain, the activation of AMPK can inhibit NF-κB, which in turn decreases the expression of IL-1β, thereby playing an analgesic role.⁴⁶ Guo et al.⁴⁷ found that niacin could attenuate LPS-induced inflammatory responses in BMECs by activating GPR2A, phosphorylating AMPK, and promoting NRF-109 nucleation and autophagy. Exogenous metformin can activate AMPK signalling transduction and help alleviate inflammatory reaction in BMECs induced by LPS.⁴⁸ These results suggest further revealed that AMPK activation is a necessary mediator to protect BMECs from LPS-induced inflammatory response and that AMPK activation not only effectively inhibits the inflammatory responses in BMECs but also inhibits lipid synthesis in BMECs. Experiments in cells, animals, and humans have shown that green tea and its main component EGCG can inhibit the gene and/or protein expression of inflammatory cytokines and inflammatory enzymes.⁴⁹ For example, EGCG inhibit the NF-κB pathway and activated of the nuclear factor 2-related factor 2(Nrf-2)/HO −1 pathway, which protected microglia from inflammation and oxidative stress caused by hypoxia.⁵⁰ In a rat model study, EGCG extended the lifespan of obese rats by significantly increasing the protein expression of forkhead box protein O1 (FoxO1), sirtuin 1, catalase, glutathione-s-transferase A1, long-chain acyl-coenzyme A synthetase, and CPT1 in rat livers; inhibiting oxidative stress; activating FA transport and oxidation; and promoting cholesterol metabolism.⁵¹ Indeed, EGCG significantly increased the expression of FA synthesis genes in mice on a high-fat diet and partially reduced the deposition of white adipose tissue in the epididymis of obese mice by inhibiting AMPK.⁵² EGCG also inhibited FA and cholesterol synthesis by activating AMPK and insulin pathways.⁵³ In the present study, EGCG promoted lactolipid synthesis in inflammation-injured BMECs by inhibiting AMPK, promoting lipid synthesis gene expression, and inhibiting lipid oxidation gene expression.

The current study also comprehensively evaluated the changes in the related genes in the blood of mastitis cows and healthy cows. The results showed that FoxO1, peroxisome proliferator-activated receptor-gamma (PPARG), protein phosphatase 2 catalytic subunit (PPP2CA), gamma-2 regulatory subunit (PRKAG2), 6-phosphofructose-2-kinase (PFKFB3), and ELAV-like RNA-binding protein 1 (ELAVL1) were significantly different in the blood of cows with mastitis and that of healthy cows. These genes are mainly involved in immune response and lipid metabolism. The FoxO1 transcription factors affect a wide range of cell types important for host responses, and their downstream gene targets, including such as proinflammatory signalling molecules (Toll-like receptor-2, Toll-like receptor-4, IL-1β, and TNF-α), wound-healing factors (TGF-β, vascular endothelial growth factor and connective tissue growth factor), and chemokine receptors (chemokine receptor 7 and C-X-C motif receptor 2), play important roles in regulating host inflammatory responses.⁵⁴ Meanwhile, FoxO1 is a key regulator of lipid metabolism.^55,56 PPARG is considered a central regulator of lipid metabolism in mammary cells and was shown to significantly promote lipid storage in goat mammary epithelial cells.⁵⁷ The KEGG pathway enrichment analysis of DEGs revealed that immune pathways such as NF-κB, IL-17, insulin and NOD receptors were highly enriched in mastitis cows, indicating that immune responses and lipid metabolism were significantly altered in mastitis cows.

Conclusion

DEGs and signalling pathways in the blood of mastitis cows were mainly enriched in immunity and lipid metabolism, and the AMPK signalling pathway was involved in lipid metabolism in mastitis cows. LPS could inhibit lipid synthesis and increase lipid oxidation in BMECs, whereas EGCG could increase lipid synthesis in inflammatory injury BMECs with inflammatory injury by downregulating AMPK.

Funding Statement

Ningxia Natural Science Foundation Program (No. 2022AAC02006, 2023AAC03042, Yinchuan, China), Ningxia Ruminant Nutrition Science and Technology Innovation Team (No. 2024CXTD008, Yinchuan, China), Ningxia Hui Autonomous Region Young Top Talent Program (No. 2023, Yinchuan, China), Yinchuan Science and Technology Innovation Team Program (No. 2023CXTD32, Yinchuan, China), Ningxia Overseas Returned Personnel Innovation Program (No. 2024CXTD32, Yinchuan, China 2023, Yinchuan, China), Yinchuan Science and Technology Innovation Team Program (No. 2023CXTD32, Yinchuan, China), and Ningxia Overseas Returned Personnel Innovation Program (No. 2024, Yinchuan, China), Ningxia Hui Autonomous Region Key R&D Program (No. 2024CXTD008, Yinchuan, China). bbf02013, 2021bef01001.

Author contributions

Writing - original draft, writing - review&editing, conceptualization, formal analysis, C.H. and W.D.; data curation, software, X.M.; methodology, resources, D.W and Y.A.; funding acquisition, project administration and supervision, Y.M.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Conflicts of interest

The authors declare they have no competing financial interest and no conflicts of interest.

Institutional review board statement

The experimental steps were approved by the Animal Experimentation Committee of Ningxia University, based on the Regulations on the Management of Laboratory Animals in China (Ethics 22–72, Ningxia University). During the execution of the experiments, we strictly followed the approved guidelines and regulations

Supplementary materials

Supplementary Table S1: RT-qPCR system; Supplementary Table S2: Primer information; Supplementary Table S3: Statistical results of the library reads and the sequencing data. Supplementary Figure S1: Distribution of sample correlation and gene expression; Supplementary Figure S2: K-means clustering analysis of differential gene; Supplementary Figure S3: AMPK relative signal pathway. Supplementary Figure S4: Bodipy staining grey value analysis.

Data availability statement

Data from the results of this study are available from the corresponding author upon reasonable request.