This study was supported and approved by the Animal Ethics Committee of Inner Mongolia Agricultural University (No. NND2024007). All the animal experiments were performed in accordance with the National Standard Guidelines for Ethical Review of Animal Welfare (GB/T 35892–2018).

The procedure for preparing primary BMECs was detailed in our previous study (15). In summary, BMECs were isolated from three multiparous lactating Holstein cows. The purified cells were cultured in plastic flasks (Corning, cat. no. CLS430639) using a growth medium consisting of Dulbecco’s Modified Eagle’s Medium/F12 Ham (DMEM/F12, Gibco, cat. no. A4192001) supplemented with 10% fetal bovine serum (ViVacell, cat. no. C04001-500), 100 U/mL penicillin–streptomycin (Gibco, cat. no. 15140122), 10 μg/mL hydrocortisone (Sigma–Aldrich, cat. no. H0135), 5 mg/mL insulin-transferrin-selenium (Gibco, cat. no. 41400045), and 10 ng/mL epidermal growth factor (Sigma–Aldrich, cat. no. E4127). The cultures were maintained at 37°C in a humidified atmosphere with 5% CO₂.

Subcultured BMECs were cultivated in the aforementioned growth medium without epidermal growth factor but with the addition of 100 ng/mL prolactin (Sigma-Aldrich, cat. no. L6520), at 37°C in a 5% CO₂ humidified atmosphere. When BMECs reached 85% confluence, they were randomly allocated into four treatment groups, each with six replicates. The inducing medium was formulated using a custom DMEM/F12 base medium (SunnCell, cat. no. SNM-004F-DZ) devoid of nine EAA and supplemented with varying concentrations of Phe (0.07 mM as the control, 0.14, 0.28, and 0.56 mM). The concentration of 0.07 mM approximates the Phe concentration in bovine arterial plasma supplying the mammary gland (16). The other EAA were added at the following concentrations: Lys (0.37 mM), Met (0.09 mM), His (0.33 mM), Arg (0.43 mM), Leu (0.61 mM), Ile (0.43 mM), Val (0.90 mM), and Thr (0.44 mM). These concentrations were derived from our previous research identifying optimal EAA levels for BMECs culture (16), which correspond to approximately 4 times the concentrations found in mammary arterial plasma of lactating dairy cows. After 24 h of incubation, cells were harvested to assess the effects of Phe concentrations on amino acid uptake, metabolism, and casein synthesis.

BMECs viability was assessed using the CCK-8 colorimetric assay (MedChemExpress, cat. no. HY-K0301). Subcultured BMECs were seeded into 96-well plates at a density of 1 × 10⁵ cells/well. The plates were incubated in a CO₂ incubator until the cells reached 85% confluence. At this point, inducing media containing different concentrations of Phe were added, and the cells were cultured at 37°C in a 5% CO₂ incubator for an additional 24 h. Four hours before the end of the culture period, 20 μL of CCK-8 solution was added to each well under light-protected conditions. At the conclusion of the experiment, the optical density (OD) at 490 nm was measured for each well using a fully automated microplate reader. The relative growth rate (RGR) of the cells was calculated as follows: RGR = (OD₄₉₀ nm of the experimental group)/(OD₄₉₀ nm of the control group).

The expression of amino acid transporters in BMECs was quantified using quantitative real-time PCR (qRT-PCR). Total RNA was extracted from BMECs using the Trizol reagent (Takara, cat. no. 9180) according to the manufacturer’s protocol. RNA quality and integrity were assessed using a NanoDrop spectrophotometer and 1% agarose gel electrophoresis. Total RNA was reverse-transcribed into cDNA using the PrimeScript RT reagent kit (Takara, cat. no. RR047A).qRT-PCR was performed with SYBR Premix Ex Taq II (Tli RNaseH Plus; Takara, cat. no. RR820A) in a 20 μL reaction volume. Primer sequences, product lengths, and annealing temperatures are listed in Table 1. The qRT-PCR conditions included an initial denaturation at 95°C for 30 s, followed by 40 cycles of 95°C for 5 s, annealing at 59 ~ 66°C for 30 s, and extension at 72°C for 30 s. The housekeeping genes β-actin (ACTB), glyceraldehyde-3-phosphate dehydrogenase (GAPDH), and ubiquitously expressed transcript (UXT) were used as internal controls. The geometric mean of these three housekeeping genes was calculated and used for normalization to improve the reliability of the qRT-PCR data. The relative changes in mRNA expression levels were calculated using the 2^-ΔΔCT method.

Western blotting was used to analyze the protein expression of casein and phosphorylation levels of key proteins involved in signaling pathways associated with casein synthesis. Total protein was extracted from BMECs using RIPA lysis buffer (Beyotime, cat. no. P0013B) containing protease and phosphatase inhibitors. Protein concentration was determined using the bicinchoninic acid (BCA) assay kit (Thermo Scientific, cat. no. 23225) according to the manufacturer’s instructions. Subsequently, 40 μg of protein from each sample was separated by SDS-PAGE and transferred onto polyvinylidene fluoride membranes (Millipore, Billerica, MA, cat. no. IPVH00010). The membranes were blocked at room temperature for 2 h using TBS blocking buffer (Beyotime, cat. no. P0228). Subsequently, the membranes were incubated overnight at 4°C with primary antibodies at appropriate dilutions (detailed in Table 2). The next day, the membranes were washed three times with TBST (5 min per wash) and incubated at room temperature for 2 h with fluorescence-labeled anti-rabbit IgG secondary antibodies (dilution 1:10,000; DyLight™ 800 4 × PEG Conjugate, Cell Signaling Technology, cat. no. S151S). The membranes were imaged using the Odyssey infrared imaging system (Odyssey Clx, LI-COR Biosciences, Lincoln, NE). The relative expression levels of target proteins were calculated as the ratio of the target protein’s signal intensity to that of β-actin.

The amino acids content in BMECs culture media was analyzed using an automated amino acid analyzer (Hitachi, Model L-8900). Before analysis, the culture supernatant was collected and treated with 8% sulfosalicylic acid (w/v) for overnight hydrolysis to remove proteins. The samples were centrifuged at 12,000 rpm for 10 min, and the supernatant was filtered through a 0.22 μm hydrophilic membrane before amino acid analysis. The following formulas were used to calculate the net amino acid uptake and net uptake efficiency of BMECs: (1) net amino acid uptake = amino acid content before incubation - amino acid content after incubation. (2) net amino acid uptake efficiency (%) = (net amino acid uptake / amino acid supply) × 100.

The activities of key enzymes involved in the catabolism of EAA in BMECs were determined using ELISA kits (Jiangsu Enzyme Immunoassay Biotechnology) following the manufacturer’s protocols. The assessed enzymes included S-adenosylmethionine synthetase (SAMs, cat. no. MM-1044 V1), threonine dehydrogenase (TDH, cat. no. MM-1049 V1), phenylalanine hydroxylase (PAH, cat. no. MM-1053 V1), aminoadipate semialdehyde synthase (AASS, cat. no. MM-1057 V1), branched-chain amino acid transaminase (BCAT, cat. no. MM-51313O1), and arginase (ARG2, cat. no. MM-1045 V1).

Statistical analysis was conducted using SAS 9.2 software (SAS Institute Inc., Cary, NC, USA). Data were tested for normality with the Shapiro–Wilk test and for homogeneity of variance using Levene’s test, and all met the assumptions for parametric analysis. Results are presented as mean ± standard error of the mean (SEM) (n = 6). In the analysis of variance (ANOVA), replicate was treated as a random effect and treatment as a fixed effect. Differences between treatments were determined using Tukey’s multiple comparison test. Orthogonal polynomial contrasts were applied to evaluate linear and quadratic effects of treatment levels. Statistical significance was defined as p < 0.05.

As shown in Figure 1, increasing Phe concentrations from 0.07 to 0.56 mM did not significantly affect the proliferation of BMECs after 24 h of treatment (p = 0.390). No significant linear (P_linear = 0.121) or quadratic (P_quadratic = 0.709) responses were observed in relation to Phe concentration.

As shown in Figure 2, Phe concentration significantly influenced the expression of α_S1-casein, β-casein, and κ-casein in BMECs. Expression levels of α_S1-casein (p = 0.006), β-casein (p = 0.010), and κ-casein (p < 0.001) were highest at 0.14 mM Phe and lower at both 0.07 mM and higher concentrations (0.28 and 0.56 mM). α_S1-casein exhibited a significant quadratic response (P_quadratic = 0.012) to increasing Phe concentration. Similarly, β-casein showed a significant linear decrease (Pₗᵢₙₑₐᵣ = 0.011) as Phe concentration increased, with the lowest expression observed at 0.56 mM. κ-casein demonstrated both significant linear (Pₗᵢₙₑₐᵣ < 0.001) and quadratic (P_quadratic = 0.005) responses, with expression markedly decreasing at concentrations above 0.14 mM.

As shown in Figure 3, increasing the Phe concentration in the culture medium led to a linear increase in the net uptake of Phe by BMECs (P_linear < 0.001) and a linear decrease in Phe uptake efficiency (P_linear < 0.001). Phe availability also significantly affected the net uptake of other amino acids.

Among EAA, the net uptake of Met exhibited a quadratic response to increasing Phe concentrations (P_quadratic = 0.001), with significantly higher uptake in the 0.14 mM and 0.28 mM Phe groups than in the 0.07 mM and 0.56 mM Phe groups (p = 0.006). The net uptake of Thr showed both linear and quadratic increases with increasing Phe concentrations (P_linear = 0.003, P_quadratic < 0.001), with lower uptake in the 0.07 mM Phe group than in the other groups (p = 0.001). His uptake was significantly lower in the 0.28 mM and 0.56 mM Phe groups than in the 0.07 mM and 0.14 mM groups (p = 0.032). Arg uptake showed a significant linear decrease (P_linear = 0.033) as Phe concentration increased, with greatest uptake in the 0.14 mM Phe group (p = 0.032).

The branched-chain amino acids showed varying responses to Phe concentration. Val uptake demonstrated a quadratic decrease (P_quadratic = 0.012), with significantly lower uptake in the 0.28 mM Phe group compared to the 0.07 mM and 0.14 mM Phe groups (p = 0.040). Ile uptake exhibited a significant quadratic relationship (P_quadratic = 0.037), with greater uptake in the 0.14 mM Phe group than in the 0.07 mM and 0.28 mM Phe groups (p = 0.005). Leu uptake decreased linearly (P_linear = 0.003), with significantly lower uptake in the 0.56 mM Phe group compared to the other groups (p = 0.018).

Phe concentration also significantly affected the uptake of non-essential amino acids. The net efflux of Glu showed both linear and quadratic responses (P_linear = 0.019, P_quadratic = 0.004), with less efflux from the 0.07 mM Phe group than from the other groups (p = 0.004). Similarly, Ala exhibited a quadratic increase in net efflux (P_quadratic = 0.009), with greater efflux in the 0.28 mM Phe group compared to the 0.07 mM and 0.56 mM Phe groups (p = 0.033). The net uptake of Tyr decreased both linearly and quadratically (P_linear = 0.009, P_quadratic = 0.001) with increasing Phe, with significantly greater uptake in the 0.07 mM and 0.14 mM Phe groups than in the 0.28 mM and 0.56 mM Phe groups (p = 0.010). Pro uptake decreased quadratically (P_quadratic = 0.001), with significantly lower uptake in the 0.56 mM Phe group compared to the other groups (p = 0.008).

Figure 4 illustrates the effects of Phe availability on the mRNA expression of amino acid transporters in BMECs. Several transporters showed the greatest expression at moderate Phe concentrations followed by decreased expression at higher concentrations. The expression of SLC7A1 showed significant differences across treatments (p < 0.001), with peak expression at 0.14 mM Phe, significantly higher than all other groups, while the 0.56 mM group showed intermediate expression. Similarly, the expression of SLC7A5 exhibited both linear and quadratic responses (P_linear = 0.001, P_quadratic = 0.003, p < 0.001), with the greatest expression at 0.14 mM Phe. The expression of SLC38A2 showed a significant quadratic response (P_quadratic = 0.002, p = 0.018), with peak expression at 0.28 mM Phe, significantly higher than at 0.07 mM and 0.56 mM Phe concentrations. The expression of SLC7A8 and SLC3A2 displayed similar patterns with significant linear and quadratic responses. The expression of SLC7A8 showed significantly higher expression at 0.14 mM and 0.28 mM Phe compared to 0.07 mM and 0.56 mM Phe (P_linear = 0.010, P_quadratic < 0.001, p < 0.001). The expression of SLC3A2 maintained high expression at 0.07 mM, 0.14 mM, and 0.28 mM Phe, with a significant decrease at 0.56 mM Phe (P_linear = 0.004, P_quadratic = 0.039, p = 0.010). Overall, five amino acid transporter genes examined showed significant responses to Phe concentration, typically with greatest expression at moderate Phe levels (0.14 mM or 0.28 mM) and decreased expression at the greatest Phe concentration (0.56 mM).

As shown in Figure 5, increasing Phe concentrations in the culture medium affected the expression of the mTOR signaling pathway. Both total protein expression and phosphorylation levels of mTOR at Ser2448 showed a quadratic increase with increasing Phe concentrations (P_quadratic < 0.001). The total protein expression and phosphorylation levels of mTOR at Ser2448 in the 0.28 mM Phe group were higher than in the other groups (p < 0.001 and p = 0.001, respectively). The total protein expression of mTOR in the 0.07 mM Phe group was lower than in the other groups (p < 0.001). Phosphorylation levels of P70 S6 kinase (P70S6K) at Thr389 showed a quadratic increase (P_quadratic = 0.001). The phosphorylation levels of P70S6K at Thr389 in the 0.14 mM and 0.28 mM Phe groups were higher than in the 0.07 mM and 0.56 mM Phe groups (p = 0.002).

As shown in Figure 6, increasing Phe concentrations in the culture medium affected the expression of key proteins and genes in the GCN2 signaling pathway. The total protein expression of eIF2α showed a significant linear decrease with increasing Phe concentrations (P_linear = 0.002, p = 0.002). The expression of eIF2α in the 0.07 mM and 0.28 mM Phe groups was higher than in the 0.14 mM and 0.56 mM Phe groups (p = 0.002). The phosphorylation level of eIF2α and the mRNA expression of ATF4 both showed quadratic responses to increasing Phe concentrations (P_quadratic = 0.002 and 0.001, respectively). The phosphorylation level of eIF2α was higher in the 0.07 mM Phe group than in all other groups (p = 0.001). Similarly, the mRNA expression of ATF4 was higher in the 0.07 mM and 0.56 mM Phe groups compared to the 0.14 mM and 0.28 mM Phe groups (p = 0.001).

As shown in Figure 7, increasing Phe concentrations in the culture medium affected the activities of key enzymes involved in EAA catabolism. The activity of PAH was reduced in the 0.07 mM Phe group (p < 0.001), while no differences in PAH activity were observed among the other groups. The activity of ARG2 increased quadratically with increasing Phe availability (P_quadratic = 0.045). In contrast, the activities of AASS, BCAT, SAMs, and TDH showed no differences among the Phe treatment groups (p > 0.05).

As an essential amino acid for dairy cows, Phe must be supplied through the diet to sustain milk protein synthesis during lactation. Doepel et al. (8) demonstrated the importance of Phe by supplementing low-protein diet-fed cows with a mixed amino acid solution via abomasal infusion. Removing Phe from the solution reduced mammary uptake of Phe by 23.7% and decreased milk protein yield by 15.6%, highlighting the critical role of Phe availability in milk protein synthesis. In our study, increasing Phe concentration from 0.07 mM to 0.56 mM resulted in a quadratic increase in the expression of α_S1-casein and κ-casein in BMECs, with the highest expression observed at 0.14 mM Phe. Similarly, the expression of β-casein also peaked at this concentration, suggesting that 0.14 mM is the optimal Phe level for casein synthesis in BMECs under the tested conditions. Furthermore, Guo et al. (17) reported that Phe activated the mTOR signaling pathway to upregulate 4E-binding protein 1 expression and reduces the phosphorylation level of eIF2α, thereby promoting protein synthesis. Our results align with these findings, showing that the total protein expression of mTOR, its phosphorylation levels, and the phosphorylation of P70S6K are consistent with the trends in casein expression. Conversely, eIF2α phosphorylation and ATF4 mRNA expression exhibits opposite trends. These results suggest that optimal Phe concentrations may promote casein synthesis in BMECs partly by activating the mTOR and inhibiting the GCN2 signaling pathway. This highlights the potential importance of maintaining proper Phe levels to enhance milk protein production while avoiding disruptions to cellular regulatory processes.

Interestingly, our study found that increasing Phe concentrations from 0.07 to 0.56 mM did not significantly affect BMECs proliferation. This suggests that while Phe availability critically influences protein synthesis pathways and casein production, it may not directly regulate cell proliferation within this concentration range. This finding aligns with the specialized function of mammary epithelial cells during lactation, where resources are primarily directed toward milk protein synthesis rather than cell proliferation (18).