Extraction of total RNA
The total RNA extracted was analyzed by using 1% agarose gel electrophoresis, and all three RNA bands (28S, 18S, and 5S) were clearly visible (Supplementary Fig. 1). Spectrophotometric results revealed an OD 260/OD 280 ratio of 2.0-2.1, indicating high purity of the extracted RNA. The total RNA concentration was approximately 800 ng/μL, making it suitable for subsequent experiments.
Tissue expression analysis of candidate differential genes in milk fat metabolism
The qRT-PCR results demonstrated that the relative expression of the ENPP2 gene (2.17±0.76) in mammary gland tissue was significantly higher than that in other tissues (P < 0.05). The expression level of the PI4K2A gene in mammary gland tissue (1.40±0.26) was slightly higher than in the uterus (1.31±0.14) and significantly higher than in other tissues. The CTSH gene exhibited high expression in mammary gland and small intestine tissues, with expression levels of 1.18 ± 0.23 and 1.00 ± 0.03, respectively, which were significantly higher than in other tissues (P < 0.05). Although the relative expression of PTPRR in mammary gland tissue was the highest (3.35±0.67), it was not significantly different from that in the uterus, heart, ovary, but it was significantly higher than in the small intestine, liver, and kidney (P < 0.05). The expression levels of ID1 and PDGFD in mammary gland tissue were lower than in the uterus, but significantly higher than in other tissues (P < 0.05). The ATP8A2 gene was highly expressed in the ovary and ranked second in mammary gland tissue (10.09 ±2.15), which was significantly higher than in other tissues (P < 0.05). In comparison with other tissues, the expression of VEGFD, ZFYVE28, DKK1, APOL3, SLC16A1, and CES4A genes in mammary gland tissue was moderate or low, while the expression of KCNMA1 and BCAT1 was lower in mammary gland tissue (Fig. 1).
Expression of candidate differential genes related to milk fat metabolism in mammary gland tissue and BMECs
As demonstrated in Fig. 2A-B, the expression levels of most candidate differentially genes in mammary gland tissue correspond with those in BMECs. SLC16A1, PI4K2A, CTSH, APOL3, BCAT1, and DKK1 exhibit high expression in mammary gland tissue and BMECs, while ATP8A2, ID1, and ZFYVE28 show low expression in both. Our findings indicate that SLC16A1 has the highest expression levels in both mammary gland tissue and BMECs compared to other genes, suggesting it may play a significant role in regulating milk secretion in cows. However, the expression of the SLC16A1 gene in mammary gland tissue is not dominant across different tissues, which leads us to discard it as a key candidate gene for milk fat metabolism in cows. The expression of the CTSH gene ranks second in both mammary gland tissue and BMECs, and its expression in mammary gland tissue was significantly higher than in other tissues, except for the small intestine (P < 0.05). However, our previous research discovered that the expression of the CTSH gene in the high milk fat group was significantly lower than in the low milk fat group (P < 0.05), and identifying it as a down-regulated differentially expressed gene. Consequently, we will not consider further functional verification.
It is worth noting that PI4K2A gene has the highest expression level in mammary gland tissue compared to other tissue. Among all the differentially expressed candidate genes involved in milk fat metabolism, the expression of PI4K2A is relatively high in both mammary gland tissue and BMECs (ranked third), indicating that PI4K2A gene expression is relatively stable in these tissues. Previous transcriptome sequencing studies have also demonstrated differential expression of the PI4K2A gene in high and low milk fat groups, which is an upregulated differential gene (Supplementary Table. 1). Moreover, prior research has utilized GWAS and single nucleotide polymorphism arrays at the genomic level to genotype Italian Simmental and Holstein cows, determining that PI4K2A is an important candidate gene for milk fat metabolism in cows[14]. Therefore, this study will focus on the functional mechanism of the PI4K2A gene in cow milk fat metabolism.
Subcellular localization of PI4K2A
Subcellular localization is crucial for understanding protein function, as proteins can move within cells and possess one or more subcellular localization signals. In this study, the subcellular localization of the PI4K2A protein was predicted by using three online software tools: Euk-mPLOC 2.0, YLoc, and MultiLoc2. The results indicated that the protein was primarily located in the cytoplasm (86.0% to 87.1%), with a minor presence in the nucleus (12.8% to 13.0%, Table. 1). Although the PSORT II software predicted a lower content of PI4K2A protein in the cytoplasm (34.8%), the overall trend of the results from the four software tools was consistent.
To further confirm the authenticity of the prediction results, we extracted both nuclear and cytoplasmic RNA from BMECs. Fig. 2C shows that the concentration of cytoplasmic RNA was 1004.4 ng/μL, with an OD260/OD280 of 2.12, and the 28S and 18S bands were very clear and bright. In contrast, the nuclear RNA had a lower concentration (88.8 ng/μL) and brightness of bands, with an OD260/OD280 of 1.87, but subsequent experiments could still be carried out normally. Before reverse transcription, we diluted the concentration of cytoplasmic RNA to match that of nuclear RNA. qRT-PCR analysis revealed that U6 accounted for nearly 60% of the nuclear RNA content, while the reference genes GAPDH and β-actin accounted for more than 80% of the cytoplasmic RNA content. This indicates that both nuclear and cytoplasmic RNA were successfully isolated (Fig. 2D). Fig. 2 also demonstrates that the PI4K2A gene is expressed in both nuclear and cytoplasmic fractions, the expression level in the cytoplasm is as high as approximately 80%, while it is only about 20% in the nucleus. This suggests that the PI4K2A gene primarily plays an important role in the cytoplasm.
The effect of overexpression of PI4K2A on the expression of milk fat metabolism-related genes
The infection efficiency of both the overexpression group (Ad-PI4K2A) and the control group (Ad-EGFP) exceeded 70% (Supplementary Fig. 2). Compared to the Ad-EGFP group, the mRNA expression level of the PI4K2A gene in the Ad-PI4K2A group was significantly upregulated (P < 0.01). We use western blot (WB) to detect the expression level of the 3*Flag-tagged protein fused with PI4K2A, the results indicated that the expression level of the tagged protein 3*flag was also significantly higher in the Ad-PI4K2A group (P < 0.01). These findings suggest that the recombinant adenovirus Ad-PI4K2A was successfully constructed.
To investigate the effect of the PI4K2A gene on milk fat metabolism, BMECs were infected with Ad-PI4K2A adenovirus for 48 hours, and the mRNA expression levels of milk fat synthesis-related marker genes were detected by qRT-PCR. The results showed that overexpression of PI4K2A had no significant effect on the mRNA expression of de novo FA synthesis, LCFA uptake from blood, and desaturation marker genes (Fig. 3A-C). However, it significantly upregulated the mRNA expression of the FA transport marker gene FABP3 (Fig. 3C, P < 0.01). In addition, we detected other genes and transcriptional regulatory factors that regulate lactation. We discovered that PI4K2A overexpression upregulated ELOVL6 gene expression (Fig. 3D) and significantly increased mRNA expression levels of PPARG and ACOX1 (P < 0.01/P < 0.05, Fig. 3D), while PPARD displayed the opposite trend (P < 0.05, Fig. 3D). Subsequently, we randomly selected three significantly different genes for WB detection. The results revealed that protein expression levels of PPARG, PPARD, and FABP3 were consistent with their mRNA levels (Fig. 3E-F). These findings suggest that PI4K2A does not significantly affect de novo FA synthesis, LCFA uptake from blood, or desaturation in BMECs. However, it can promote FA transport and LCFA elongation. We hypothesize that PI4K2A can enhance milk fat synthesis in dairy cows because most lipogenic gene expression levels were upregulated after PI4K2A overexpression.
The effect of overexpression of PI4K2A on TAG synthesis and lipid droplet secretion in BMECs
To determine whether the PI4K2A gene affects TAG synthesis in BMECs, we examined the expression levels of TAG synthesis-related marker genes by using qRT-PCR. Our results revealed that the overexpression of PI4K2A significantly increased both mRNA and protein expression levels of DGAT1 (Fig. 3G, I-J, P < 0.05). Furthermore, we investigated marker genes associated with TAG degradation and discovered that mRNA expression of the ATGL gene was significantly reduced following PI4K2A overexpression (Fig. 3H, P < 0.01). In comparison to the Ad-EGFP group, PI4K2A overexpression ultimately resulted in a significant elevation of TAG content within BMECs (Fig. 3K, P < 0.05), while cholesterol content unchanged (Fig. 3L). The expression of lipid droplet secretion-related genes in BMECs was detected by using qRT-PCR, and it was found that the mRNA expression levels of XDH and PLIN2 genes tended to be upregulated after the overexpression of PI4K2A (Fig. 3M). Additionally, Oil Red O staining revealed a significant increase in the lipid droplets secreted by BMECs (Fig. 3N-O). Furthermore, by dissolving lipid droplets bound with Oil Red O in 100% isopropanol and measuring their absorbance values using an enzyme-linked instrument, the results were found to be consistent with the Oil Red O staining results (Fig. 3P). In summary, the overexpression of PI4K2A gene promotes the synthesis of milk fat in cows, more significant effects on fatty acid transport, TAG synthesis, and lipid droplet secretion. Additionally, the overexpression of PI4K2A gene has a notable inhibitory effect on TAG degradation.
The effect of interfering with PI4K2A on the expression of genes related to milk fat metabolism
The infection efficiency for both the control group (NC) and the interference group (siRNA-PI4K2A) exceeded 85% after BMECs were infected for 48 hours (Supplementary Fig. 3). Compared with NC, the mRNA expression level of the PI4K2A gene was significantly downregulated following interference (P < 0.01), achieving an interference efficiency of 59.75%. This demonstrates that the recombinant lentivirus siRNA-PI4K2A was successfully constructed.
The qRT-PCR quantification results demonstrated that PI4K2A could completely inhibit the marker genes related to bovine milk fat synthesis. Among these genes, the mRNA expression of de novo FA synthesis key genes ACSS2 and ACACA, key genes for LCFA uptake in blood (ACSL1 and CD36), and key genes for FA desaturation (SCD) were significantly reduced following PI4K2A interference (P < 0.01/P < 0.05, Fig. 4A-C). After PI4K2A interference, the key genes FASN and FABP3 involved in FA synthesis and transport showed a decreasing trend, but the difference was not significant (P > 0.05). Additionally, we examined the mRNA expression of genes and transcription factors involved in milk fat synthesis regulation and observed a downward trend in the expression of all genes (Fig. 4D). The downregulation of ELOVL6, EGFR, PPARA, and PPARD genes reached a significant level (P < 0.01/P < 0.05). We then randomly selected four genes with reduced expression for WB analysis, and the results revealed that the protein expression levels of PPARG, PPARD, SCD, and FABP3 were consistent with their mRNA levels (Fig. 4E-F). These findings suggest that interfering with the PI4K2A gene can inhibit de novo FA synthesis, LCFA uptake from blood, FA transport, and desaturation processes in BMECs. This interference may also indirectly impact some key transcription factors involved in milk fat synthesis, and leading to a reduction in bovine milk fat synthesis.
The effect of interfering with PI4K2A on TAG synthesis and lipid droplet secretion in BMECs
The detection results of TAG synthesis and decomposition-related marker genes revealed that interference with PI4K2A significantly reduced the mRNA expression of the key gene AGPAT6 for TAG synthesis (Fig. 4G, P < 0.01) and had no effect on TAG decomposition (Fig. 4H). When compared to the NC group, interference with the PI4K2A gene resulted in significant decrease of TAG content in BMECs (Fig. 4I, P < 0.01), and the total cholesterol content also showed a decreasing trend (Fig. 4J, P = 0.058). The detection results of lipid droplet secretion-related marker genes indicated that the mRNA expression levels of XDH, BTN1A1, and PLIN2 genes were significantly downregulated after interfering with PI4K2A (Fig. 4K, P < 0.05). It was observed that the lipid droplet secretion in the siRNA-PI4K2A group was significantly lower than that in the NC group (Fig. 4L-N, P < 0.05) by using Oil Red O to detect lipid droplet secretion in BMECs. In conclusion, interference with the PI4K2A gene can inhibit the synthesis of cow milk fat. Simultaneously, combined with the overexpression results of the PI4K2A gene, we concluded that the PI4K2A gene positively regulates the synthesis of cow milk fat.
The effect of PI4K2A on the proliferation and apoptosis of BMECs
After overexpressing the PI4K2A gene for 48 hours, the number of proliferating cells in the Ad-PI4K2A group was significantly higher than in the Ad-EGFP group. Using ImageJ software for cell counting and calculating the EdU labeling index, it was found that overexpression of the PI4K2A gene could significantly increase the proliferation rate of BMECs (P < 0.05, Fig. 5A-G). Further analysis of the protein expression levels of cell proliferation marker genes revealed that the expression of CCND1 in BMECs treated with Ad-PI4K2A adenovirus was also significantly higher than in the Ad-EGFP group (P < 0.05), which was consistent with the EdU detection results (Fig. 5H-I). In contrast, the experimental results of interfering with the PI4K2A gene showed the opposite effect, indicating that PI4K2A positively regulates the proliferation of BMECs (Fig. 5J-R).
Adenoviruses Ad-EGFP and Ad-PI4K2A were infected respectively in BMECs, and the apoptosis of BMECs was detected after 48 hours by using the Annexin V-mCherry/SYTOX Green kit. The results showed that overexpression of PI4K2A significantly reduced BMEC apoptosis, and the apoptotic index of the Ad-PI4K2A group was also significantly lower than that of the Ad-EGFP group (P < 0.05, Fig. 6A-I). Furthermore, the detection of protein expression levels of apoptotic marker genes showed that the expression level of caspase-3 in the Ad-PI4K2A group were also significantly lower than that in the Ad-EGFP group (P < 0.05), and consistent with the results of the apoptosis kit detection (Fig. 6J-K). In addition, the results after interfering with the PI4K2A gene were opposite (Fig. 7), indicating that PI4K2A negatively regulates the apoptosis of BMEC.