Milk, a key component of daily human nutrition, is an agricultural product of tremendous global economic importance. It is synthesised within bovine mammary epithelial cells (BMECs), and the milk production of dairy cows is primarily determined by the milk synthetic and proliferative capacities of BMECs (Ollier et al., Reference Ollier, Robert-Granié, Bernard, Chilliard and Leroux2007; Herve et al., Reference Herve, Quesnel, Lollivier and Boutinaud2016). Dairy cow mammary development follows a sequential physiological continuum from puberty through lactation to the dry period, a core reproductive cycle process with significant stage-dependent variation in BMEC proliferative capacity. Puberty represents a critical stage marked by rapid glandular epithelial proliferation and stromal expansion, during which mammary epithelial cells exhibit an extremely strong proliferative capacity (Lucas et al., Reference Lucas, Rudmann, Credille, Irizarry, Peter and Snyder2007). In early lactation, BMECs show strong proliferative capacity to build a functional cell pool for peak lactation, which declines progressively as lactation advances (Capuco et al., Reference Capuco, Wood, Baldwin, McLeod and Paape2001). A typical dry period includes active mammary involution post-milking cessation and subsequent pre-lactation mammary redevelopment (Hurley, Reference Hurley1989), with high early dry period apoptosis followed by increased cell proliferation (Nørgaard et al., Reference Nørgaard, Theil, Sørensen and Sejrsen2008). Therefore, it is important to maintain high cell numbers by either increasing cell survival and proliferation or reducing apoptosis during lactation.
Many nutritional factors, such as amino acids, can affect the proliferative capacity of mammary epithelial cells (Wu et al., Reference Wu, Heng, Tian, Song, Chen, Guan and Zhang2020). Methionine (Met) is one of the limiting amino acids in dairy rations, which functions not only as a core structural component of proteins but also as a key signalling molecule regulating multiple cellular processes (Huo et al., Reference Huo, Lin, Wang and Deng2025). Previous reports have demonstrated that Met regulates cell physiology through critical signalling pathways, particularly the phosphoinositide 3-kinase/protein kinase B (AKT)/mammalian target of rapamycin (mTOR) pathway (Qi et al., Reference Qi, Meng, Jin, Li, Li and Gao2018; Lin et al., Reference Lin, Qi, Guo, Wang, Zhang and Gao2023). As the central hub, mTOR has been well characterised as a master regulator of milk protein synthesis and cellular growth, while its downstream target Cyclin D1 is a core driver of BMEC proliferation (Qi et al., Reference Qi, Meng, Jin, Li, Li and Gao2018). As a key regulator of G1 phase progression in the cell cycle, Cyclin D1 forms complexes with cyclin-dependent kinases (CDKs) to facilitate the G1-to-S phase transition (Huetsch et al., Reference Huetsch, Suresh and Shimoda2019; Wang, Reference Wang2021). Additionally, the Cyclin D1-CDK complex promotes cell cycle progression into S phase by inhibiting Caspase 3, which is responsive to G1 phase expression (Woo et al., Reference Woo, Hakem, Furlonger, Hakem, Duncan, Sasaki, Bouchard, Lu, Wu, Paige and Mak2003; Satyanarayana and Kaldis, Reference Satyanarayana and Kaldis2009). However, the specific regulatory genes that mediate the effects of Met on BMEC proliferation through the mTOR signalling pathway remain poorly defined.
Activating transcription factor 4 (ATF4) is a basic leucine zipper transcription factor involved in endoplasmic reticulum (ER) stress. Classically, global translation is suppressed during ER stress, whereas protein kinase R-like endoplasmic reticulum kinase (PERK)-mediated eIF2α phosphorylation (p-eIF2α) drives selective upregulation of ATF4 (Rozpedek et al., Reference Rozpedek, Pytel, Mucha, Leszczynska, Diehl and Majsterek2016; Wek, Reference Wek2018). The upregulated ATF4 subsequently modulates stress-induced apoptosis via the well-established ATF4-C/EBP homologous protein (CHOP) axis (Zinszner et al., Reference Zinszner, Kuroda, Wang, Batchvarova, Lightfoot, Remotti, Stevens and Ron1998; Su and Kilberg, Reference Su and Kilberg2008). Beyond its canonical function, ATF4 also serves as a core regulator of cellular amino acid sensing, linking nutrient signals to cellular anabolism (Park et al., Reference Park, Reyna-Neyra, Philippe and Thoreen2017). Therefore, ATF4 can be activated in a PERK-eIF2α-independent manner. Specifically, the mTOR signalling enhances the expression of ATF4 by stabilising its mRNA in response to nutrient and growth signals (Park et al., Reference Park, Reyna-Neyra, Philippe and Thoreen2017; Torrence et al., Reference Torrence, MacArthur, Hosios, Valvezan, Asara, Mitchell and Manning2021). This process effectively coordinates amino acid availability with cellular anabolism and proliferation. The regulatory effect of ATF4 on cell proliferation exhibits pronounced cell and tissue specificity. ATF4 is an essential mediator of normal cell cycle progression in murine bone marrow stromal cells (Zhang et al., Reference Zhang, Yu, Galson, Luo, Fan, Zhang, Guan and Xiao2008). In lung cancer cells, upregulated ATF4 expression has been demonstrated to enhance cellular proliferative capacity significantly (Du et al., Reference Du, Liu, Mao, Qin and Fan2021). Conversely, forced overexpression of ATF4 exerts a marked inhibitory effect on the proliferation of murine mammary epithelial cells (Bagheri-Yarmand et al., Reference Bagheri-Yarmand, Vadlamudi and Kumar2003). Notably, ATF4 expression in the mammary tissue of dairy cows is dynamically regulated across the lactation cycle, rising at the onset of lactation and declining during late lactation (Invernizzi et al., Reference Invernizzi, Naeem and Loor2012). However, there is limited information on whether ATF4 regulates mTOR-mediated BMEC proliferation in the presence of Met.
This study aimed to elucidate the functional role and underlying mechanism of ATF4 in Met-induced BMEC proliferation. We examined ATF4 and Cyclin D1 expression in bovine mammary tissues across key developmental stages, determined the functional requirement of ATF4 in Met-induced proliferation via gene knockdown, and validated the involvement of mTOR signalling in this process. Furthermore, we explored the mechanism by which ATF4 regulates Cyclin D1 expression. These findings establish a theoretical foundation for designing nutritional interventions targeting ATF4 to enhance the proliferation of mammary cells in lactating dairy cows.
Materials and methods
Collection of mammary tissues of dairy cows
All animal experiments were approved by the Animal Experiment Committee of Northeast Agricultural University (Harbin, China; approval number: 2019-2) and conducted in accordance with institutional guidelines. The study used three pubertal heifers (12 months; 382 ± 24.9 kg), three lactating cows (parity 3.16 ± 0.19; 676 ± 48.2 kg; 90 Days in Milk; Somatic Cell Count < 50,000 cells/mL) and three dry period cows (parity 3.16 ± 0.19; 709 ± 32.6 kg; 220 DIM; SCC <50,000 cells/mL). All surgeries were performed to minimise the suffering of the experimental animals, and tissue collection was conducted as described previously (Hou et al., Reference Hou, Lin, Xing, Yang, Duan, Li, Gao and Lin2016).
Cell culture and treatment
Mammary epithelial cells were isolated from lactating bovine mammary tissues as described previously (Yang et al., Reference Yang, Lin, Duan, Lv, Xing, Li, Gao and Hou2017). Briefly, mammary tissue was minced with surgical scissors and digested with 1 mg/mL collagenase I (CG6140C, Biotopped, China) in an orbital shaker at 140 rpm for 3 h at 37°C. Subsequently, the tissue digest was filtered through a sterile nylon mesh, and the filtrate was enriched and centrifuged at 150 rcf for 10 min. The resulting dispersed cell suspension was seeded onto plastic culture dishes pre-coated with 5% rat-tail collagen (C8062, Solarbio, China) and cultured in Dulbecco’s modified Eagle’s medium/Nutrient Mixture F-12 (DMEM/F12, 12400024, Gibco, USA) supplemented with 10% foetal bovine serum (FBS) (A5256701, Gibco, USA), insulin (5 μg/mL), hydrocortisone (1 μg/mL), prolactin (1 μg/mL), amphotericin B (0.5 µg/mL), streptomycin (100 μg/mL) and penicillin (100 U/mL). Cells were maintained at 37°C in a humidified atmosphere of 5% CO2. When cells reached approximately 80–90% confluence, they were digested with trypsin-EDTA (0.25%; 11925200072, Gibco, USA) and passaged. Purified mammary epithelial cells were obtained after three passages. Cell viability was assessed post-isolation immediately using the trypan blue exclusion method. Only cell preparations with viability exceeding 85% were used for subsequent experiments. Cell purity was evaluated by morphological observation and immunofluorescence staining for cytokeratin 18 (CK18), a specific epithelial cell marker. To further confirm cell identity, we detected the expression of β-casein, the canonical milk-specific marker of mammary epithelial cells, via Western blotting. Furthermore, all cell cultures and key experimental batches were routinely confirmed to be mycoplasma-free using a commercial mycoplasma detection kit (D101-01, Vazyme, China) prior to functional assays.
The BMEC line MAC-T (ATCC, USA), initially established via SV40 large T-antigen immortalisation, was certified free of bacterial, fungal and mycoplasma contamination prior to use. Cells were cultured in DMEM (11965092, Gibco, USA) supplemented with 10% FBS, penicillin (100 U/mL), streptomycin (100 μg/mL), prolactin (1 μg/mL), insulin (5 μg/mL) and hydrocortisone (1 μg/mL, Sigma-Aldrich, USA) in a humidified atmosphere containing 5% CO2 at 37℃.
To investigate the effect of Met concentration on ATF4 expression, MAC-T cells (1 × 106 cells/well) were plated in a six-well culture plate. When cells reached 70% confluence, they were incubated in Met-free DMEM (C0891, Beyotime, China) without FBS for 12 h. Subsequently, cells were treated with fresh Met-free DMEM supplemented with Met at final concentrations of 0, 0.3, 0.6, 0.9 and 1.2 mM. Cells were incubated for 24 h prior to further experiments.
To evaluate the regulatory effect of mTOR on ATF4 expression, MAC-T cells were seeded into six-well plates (1 × 106 cells/well) and grown to 80% confluence, and then treated with 0.6 mM Met supplemented with either 100 nM rapamycin (Rap) (S1842, Beyotime, China) or 0.02% DMSO. Following 24 h of treatment, cells were collected for subsequent experiments.
To investigate the functional effect of glycogen synthase kinase-3β (GSK3β) on cell proliferation, MAC-T cells were treated with 5 μM AR-A014418 (AR) (SF2744, Beyotime, China) in the presence of 0.6 mM Met for 24 h. This concentration was selected based on preliminary dose-response assays (Supplementary Figure S1), in which it conferred maximal cell viability and effective target pathway inhibition via β-catenin stabilisation. Cells were harvested for subsequent experiments. All experiments were conducted using MAC-T cells within 15–25 passages.
Cell viability was assessed for all cell preparations immediately post-treatment using the trypan blue exclusion method. Only cell preparations with viability exceeding 85% were used for subsequent experiments. Furthermore, all cell cultures and key experimental batches were routinely confirmed to be mycoplasma-free using a commercial mycoplasma detection kit (D101-01, Vazyme, China) prior to functional assays. All experiments were performed with three independent biological replicates.
Quantitative Polymerase Chain Reaction
Total RNA was isolated from bovine mammary gland tissues or cells using the Hipure Total RNA Mini kit (R4111-02, Magen, China). The purity and concentration of RNA were quantified using a NanoDrop ND-1000 Spectrophotometer (Thermo Scientific), and cDNA was synthesised from 1 μg of total RNA using a reverse transcription kit (RR047Q, TaKaRa, China). The cDNA samples were subjected to Quantitative PCR (qPCR) using ChamQ Universal SYBR qPCR Master Mix (Q711−02, Vazyme, China) according to the manufacturer’s instructions, on the Applied Biosystems 7500 Real-Time PCR system (Applied Biosystems). The amplification program consisted of 1 cycle of 95℃ for 30 s, 40 cycles of 95℃ for 5 s and 60℃ for 34 s, followed by an additional 95°C for 15 s, 60°C for 1 min and 95°C for 15 s to generate a melting curve. The qPCR reactions on each sample were performed in triplicate. Relative gene expression levels were normalised to RPS9, UXT and β-actin levels, and calculated using the 
$2^{-\triangle\triangle}$Ct method. The primers used in qPCR are listed in Table 1.
The sequence of primers
Table 1 Long description
The table provides primer sequences for various genes, detailing both forward and reverse sequences in the 5’ to 3’ direction. It includes genes such as ATF4, CCND1, Caspase 3, β-actin, RPS9, and UXT, primarily for qPCR purposes. Additionally, it lists ATF4-siRNA and Scramble siRNA sequences for ATF4 knockdown and as a negative control, respectively. Each gene or siRNA entry specifies its purpose, highlighting the experimental context. The sequences are crucial for gene expression analysis and knockdown experiments, offering a comprehensive resource for researchers in molecular biology.
F, forward; R, reverse.
Immunofluorescence
Frozen mammary tissue sections (10 μm thick) were fixed in 4% paraformaldehyde at room temperature for 30 min, followed by blocking using 5% Bovine serum albumin for 1 h at 37°C. Sections were then incubated with a primary ATF4 antibody (1:100, BM5179, Boster, USA) overnight at 4°C. Following three rinses with Phosphate-Buffered Saline, sections were incubated with Alexa Fluor 488 goat anti-rabbit IgG secondary antibody (1:200, bs-0295, Bioss, China) for 1 h at 37°C. Nuclei were counterstained with 4′,6-diamidino-2-phenylindole (DAPI) for 10 min. Confocal images were captured using a confocal microscope (TCS SP8, Leica Microsystems GmbH). For quantitative analysis, three random fields were imaged from each of three sections per animal, with 800 mammary epithelial cells measured per section. The average fluorescence intensity of ATF4 was calculated using ImageJ software (Media Cybernetics, Warrendale, PA), with background subtraction and normalisation to the DAPI signal. All image acquisition and analyses were performed blinded to the experimental groups.
In order to identify the purity of isolated BMECs, cells were seeded onto 35 mm glass-bottom confocal culture dishes. The cells were incubated with CK18 antibody (1:100, A1022, ABclonal, China) overnight at 4°C. Following three rinses with PBS, cells were incubated with an Alexa Fluor 488 goat anti-rabbit IgG secondary antibody (1:200, bs-0295, Bioss, China) for 1 h at 37°C. Nuclei were counterstained with DAPI, and samples were observed under a confocal microscope (TCS SP8, Leica Microsystems GmbH).
Plasmids
The dairy cow-specific small interfering RNA (siRNA) targeting ATF4 and a non-targeting scrambled control siRNA were commercially obtained from GenePharma (Shanghai, China). The siRNA sequence of ATF4 is shown in Table 1.
The pSpCas9(BB)-2A-Puro (PX459) V2.0 vector (TaKaRa, China) was used for CRISPR/Cas9-mediated editing. The guide RNA (gRNA) sequence targeting the bovine ATF4 gene was designed using the Zhang Lab online tool (https://zlab.bio/guide-design-resources). The following guide RNA primer sequences were used: sense, 5′-CACCGcaaacccaactggtcgaagg-3′; antisense, 5′-AAACccttcgaccagttgggtttgC-3′. These oligonucleotides were annealed and cloned into the BbsI restriction site of the linearised PX459 vector to generate the ATF4-specific CRISPR/Cas9 construct.
Gene knockdown
BMECs within 3–10 passages were transfected with siRNA targeting ATF4 using Lipofectamine 3000 (100022234, Invitrogen, USA) according to the manufacturer’s protocol to reduce ATF4 expression. Transfections were performed in six-well plates when cells reached 65–70% confluence and lasted for 24 h. Knockdown efficiency was assessed by qPCR and Western blotting.
For CRISPR/Cas9-mediated ATF4 knockdown, the PX459-ATF4-sgRNA plasmid was transfected into MAC-T cells using Lipofectamine 3000 according to the manufacturer’s protocol. Puromycin (10 μg/mL) was added to the culture medium at 72 h post-transfection. The medium was replaced daily for 7 days to eliminate non-transfected cells. A single cell was then isolated from the surviving population by limiting dilution and cultured in 96-well plates to establish monoclonal cell lines. Gene editing efficiency was confirmed by Western blotting. All experiments were conducted three times.
Western blotting
Total protein was extracted from mammary epithelial tissues or cells using Radio Immunoprecipitation Assay buffer (P0013J, Beyotime, China) with 0.5 mM PMSF (ST506, Beyotime, China), 5 μg/mL leupeptin (L2884, Sigma-Aldrich, USA) and 5 μg/mL aprotinin (A6103, Sigma-Aldrich, USA). The total protein concentration was determined by a Bicinchoninic Acid Protein Assay Kit (T9300A, TaKaRa, China). Subsequently, equal amounts of protein lysates (20 μg) were separated by 10% SDS-PAGE. The protein was then transferred onto PVDF membranes, which were blocked with 5% skim milk or BSA. The membrane was incubated with the specified primary antibodies against ATF4 (1:750, BM5179, Boster, USA), mTOR (1:1000, bs-1992R, Bioss, China), p-mTOR (1:1000, bs-3495R, Bioss, China), Caspase 3 (1:1000, bs-00811R, Bioss, China), Cyclin D1 (1:1000, bs-0623R, Bioss, China), GSK3β (1:1000, WL01456, Wanleibio, China), p-GSK3β (1:1000, WL03683, Wanleibio, China), β-casein (1:750, A12749, ABclonal, China), β-catenin (1:750, WL0962a, Wanleibio, China) and β-actin (1:5000, bs-10966R, Bioss, China), followed by incubation with horseradish peroxidase-conjugated secondary antibodies (1:5000, bs-80295, Bioss, China). Finally, the protein bands on the Polyvinylidene Fluoride membrane were visualised by an ECL reagent (180-5001, Tanon, China). For normalisation purposes, β-actin was employed as the loading control. Three samples were assessed per treatment. The target proteins were analysed using ImageJ software (Media Cybernetics Inc., Warrendale, PA).
Cell cycle assays
BMECs were harvested by trypsin digestion and centrifugation. The cell pellet was treated with a Cell Cycle and Apoptosis Analysis Kit (C1052, Beyotime, China) according to the manufacturer’s instructions. The proportions (%) of cells in the G1, S and G2/M phases were analysed using a flow cytometry system (Guava EasyCyte HT system, Merck Millipore, Billerica, MA, USA).
Cell counting
According to the provided guidelines, BMECs were counted using an automated cell counter (DT CASY Cell Counter, Scharfe System GmbH, Reutlingen, Germany).
Cell counting kit-8 assay
The MAC-T cell suspension (3 × 104 cells/mL) was collected from each group and seeded into 96-well plates containing 100 μL per well for 48 h. Then, 10 μL of cell counting kit-8 (CCK-8) solution (C0037, Beyotime, China) was added to each well, and the plates were subsequently incubated in a cell incubator for 4 h. Finally, the optical density at 450 nm was measured for each well.
5-Ethynyl-2’-deoxyuridine assays
MAC-T cells (5 × 105 cells/well) were seeded in six-well plates. After transfection or exogenous stimulant treatment for the indicated durations, proliferation was assessed in each group using the EdU Imaging Kit (K1175, APExBIO, USA) according to the manufacturer’s instructions. EdU incorporation was visualised using a fluorescence microscope (EVOS™ M7000, Invitrogen, USA).
Co-immunoprecipitation
MAC-T cells (1 × 10⁷) were treated with 0.6 mM Met for 48 h. Cells were then harvested in ice-cold RIPA buffer containing PMSF, gently mixed at 4℃ for 10 min, and centrifuged at 14,000 × g for 15 min at 4℃. The resulting supernatant was aliquoted and incubated overnight at 4℃ with either ATF4 antibody (1:20; BM5179, Boster, USA) or IgG antibody (bs-0295P, Bioss, China). Protein A or G agarose beads (P2055, Beyotime, China) were added to each sample and incubated for an additional three hours at 4℃. Immunoprecipitated beads were washed at least three times with lysis buffer, boiled in 1 × SDS loading buffer and analysed by Western blotting.
Statistical analysis
All data were presented as mean ± standard error of the mean (SEM) determined from 3 independent experiments, each performed with triplicate technical replicates. Data were analysed with GraphPad Prism 10 (GraphPad Software Inc.). Statistical significance of differences was determined using a two-tailed unpaired t-test when comparing two groups or a one-way Analysis of Variance with post-Bonferroni’s multiple comparisons when comparing more than three groups. Statistical significance was assigned at P < 0.05.
Results
ATF4 expression in mammary glands of dairy cows
To explore the stage-dependent expression of ATF4 in mammary tissues of dairy cows, we analysed tissues from puberty, lactation and dry period. Immunofluorescence staining of sections showed that strong staining of ATF4 was mainly observed in the acinar epithelial cells of mammary tissues (Fig. 1A). Immunofluorescence quantification further revealed that ATF4-positive staining in lactating dairy cows was significantly lower than that in pubertal dairy cows, but higher than in dry-period dairy cows (Fig. 1A, B). qPCR results showed that the mRNA levels of ATF4 in lactating dairy cows were significantly lower than those in pubertal dairy cows and higher than those in dry-period dairy cows (Fig. 1C). The mRNA expression trend of CCND1 was consistent with that of ATF4 in the bovine mammary glands. The protein levels of ATF4 and Cyclin D1 detected by Western blotting showed similar results (Fig. 1D, E).
The expression of ATF4 in the mammary gland of dairy cows. (A) ATF4 protein expression in dairy cow mammary tissues by immunofluorescence. ATF4 was labelled by Alexa Fluor 488 (green). Nuclei were labelled with 4,6-diamidino-2-phenylindole (DAPI, blue). Scale bar, 75 μm. (B) Statistical analysis of ImageJ-positive signals. (C) The mRNA levels of ATF4 and CCND1 in mammary tissues of dairy cows. (D) The protein levels of ATF4 and Cyclin D1 were measured by Western blotting. (E) The grey scale scan of protein expression detection. Data were presented as the mean ± SEM from three independent experiments, each with an n = 3. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001.
Figure 1 Long description
The image A shows three rows of immunofluorescence staining of mammary tissues from dairy cows during puberty, lactation and dry period. Each row contains three columns labeled DAPI, ATF4 and Merge, showing different staining results. The image B presents a bar graph labeled 'Average intensity' comparing the intensity of ATF4 staining across the three stages: puberty, lactation and dry period. The image C displays a bar graph labeled 'Relative mRNA levels' comparing the mRNA levels of ATF4 and CCND1 during puberty, lactation and dry period. The image D shows Western blot results for ATF4, Cyclin D1 and beta-actin, with molecular weights labeled as 50 kDa, 26 kDa and 42 kDa respectively. The image E presents a bar graph comparing protein levels of ATF4 and Cyclin D1 across the three stages: puberty, lactation and dry period.
ATF4 knockdown inhibits BMEC proliferation
BMECs were isolated from bovine mammary tissues. The purity of the cells was confirmed by immunofluorescence staining for CK18, an epithelial cell marker (Fig. 2A). Western blotting analysis demonstrated that β-casein, another functional marker of mammary epithelial cells, was expressed in BMECs (Fig. 2B). To investigate the role of ATF4 in BMEC proliferation, we initially evaluated the effect of ATF4 knockdown on the cell cycle. Flow cytometry results revealed an increase in the percentage of cells in the G1 phase in the siATF4 group, while the percentages of cells in the S and G2/M phases were significantly decreased (Fig. 2C, D). Correspondingly, cell numbers were reduced in ATF4-knockdown BMECs (Fig. 2E). qPCR analysis indicated that the mRNA levels of ATF4 and CCND1 were significantly lower in BMECs transfected with siATF4, while Caspase 3 expression was upregulated (Fig. 2F). These findings were consistent with Western blotting results (Fig. 2G, H).
ATF4 knockdown inhibits cell proliferation in BMECs. (A) The expression of CK18 in BMECs. CK18 was labelled by Alexa Fluor 488 (green). Nuclei were labelled with 4,6-diamidino-2-phenylindole (DAPI, blue). Scale bar, 30 μm. (B) The β-casein protein level was measured by Western blotting. (C) Cell cycle transitions were measured by flow cytometry, and the proportions (%) of cells in G1, S, and G2/M phases were calculated. (D) The percentage of cells in different cell cycle phases. (E) Cell numbers were analysed using a cell counter. (F) Relative gene mRNA levels in BMECs transfected with negative control (NC) or siATF4. (G) Indicated protein levels were measured by Western blotting analysis. (H) The grey scale scan of protein expression detection. Data were presented as the mean ± SEM from three independent experiments, each with an n = 3. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001; ns, P > 0.05.
Figure 2 Long description
The image A shows BMECs stained with CK18 and DAPI, confirming cell purity. The image B displays Western blotting results for β-casein and β-actin in BMECs. The image C presents flow cytometry histograms for cell cycle phases: G1, S and G2/M, comparing control, NC and siATF4 groups. The image D shows a bar graph of cell cycle phase percentages across the three groups. The image E illustrates a bar graph of cell numbers in control, NC and siATF4 groups. The image F presents qPCR analysis results for ATF4, CCND1 and Caspase 3 mRNA levels across the groups. The image G shows Western blotting results for ATF4, Cyclin D1, Caspase 3 and β-actin. The image H displays a bar graph of relative protein expression levels for ATF4, Cyclin D1 and Caspase 3 across the groups.
Met induces cell proliferation and ATF4/Cyclin D1 expression in MAC-T cells
To determine the effect of Met on cell proliferation, MAC-T cells were treated with increasing concentrations of Met (0, 0.3, 0.6, 0.9 and 1.2 mM) for 24 h. As revealed by CCK-8 assays, cell viability initially increased and peaked at 0.6 mM Met treatment (Fig. 3A). qPCR analysis revealed that mRNA levels of ATF4 and CCND1 also increased initially, peaked at 0.6 mM Met and then decreased (Fig. 3B, C). The results of Western blotting were consistent with the qPCR results (Fig. 3D–F).
Effects of Met on cell viability and expression of ATF4 and Cyclin D1 in MAC-T cells. (A) Cell viability was detected by CCK-8. (B, C) The mRNA expression level of ATF4 and CCND1 in MAC-T cells. (D, G) Indicated protein levels were measured by Western blotting. (E, F, H) The grey scale scan of protein expression detection. Data were presented as the mean ± SEM from three independent experiments, each with an n ≥ 3. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001; ns, P > 0.05.
Figure 3 Long description
The image contains multiple panels illustrating the effects of Met on MAC-T cells. Panel A shows a bar graph of cell viability (OD 450 nm) across Met concentrations (0, 0.3, 0.6, 0.9, 1.2 mM), with viability peaking at 0.6 mM. Panel B and C display bar graphs of relative ATF4 and CCND1 mRNA levels, respectively, also peaking at 0.6 mM. Panel D presents Western blot results for ATF4, Cyclin D1 and beta-actin across Met concentrations. Panel E and F show bar graphs of relative ATF4 and Cyclin D1 protein levels, peaking at 0.6 mM. Panel G displays Western blots for ATF4, GRP78, p-eif2alpha, eif2alpha, CHOP and beta-actin comparing control and Met-treated cells. Panel H shows a bar graph of relative protein levels for ATF4, GRP78, p-eif2alpha/eif2alpha and CHOP, with significant differences noted for ATF4. Statistical significance is indicated by asterisks, with ns for not significant and varying levels of significance denoted by one to four asterisks.
To investigate whether 0.6 mM Met induced ER stress, we assessed levels of key ER stress markers. Western blotting analysis showed a significant increase in ATF4 expression following 0.6 mM Met treatment (Fig. 3G, H). Additionally, no significant effect of Met was observed on the ratio of p-eIF2α/eIF2α and the protein expression levels of glucose-regulated protein 78 (GRP78) and CHOP. These findings indicate that Met regulated cell proliferation and ATF4/Cyclin D1 expression independently of ER stress.
Met-driven cell proliferation relies on the mTOR-ATF4 axis
To investigate whether the regulatory effect of Met on cell proliferation depends on ATF4, MAC-T cells were stimulated with Met and transfected with sgATF4. EdU assay results demonstrated that ATF4 knockdown significantly abrogated the Met-induced increase in the proportion of EdU-positive cells (Fig. 4A, B). ATF4 knockdown also significantly abrogated the Met-induced upregulation of ATF4 and Cyclin D1, while concomitantly increasing Caspase 3 expression (Fig. 4C, D). Additionally, no significant effect of ATF4 knockdown was observed on the Met-induced increase in p-mTOR protein levels.
The effect of ATF4 knockdown on Met regulation of cell proliferation. (A) A fluorescence microscope detected the EdU-positive staining cells. The positively staining cell was labelled with 6-Carboxyfluorescein Azide (6-FAM Azide, green). Nuclei were labelled with Hoechst 33342 (blue). Scale bar, 75 μm. (B) The proportion of EdU-positive cells. (C) Indicated protein levels were measured by Western blotting. (D) The grey scale scan of protein expression detection. Data were presented as the mean ± SEM from three independent experiments, each with an n = 3. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001; ns, P > 0.05.
Figure 4 Long description
The composite image consists of four panels. Panel A shows fluorescence microscopy images of cells under three conditions: Control, Met and sgATF4 plus Met. Each condition is represented in three rows: Hoechst 33342 staining, 6-FAM Azide staining and a merged image. Scale bars indicate 75 micrometers. Panel B is a bar graph showing the ratio of EdU-positive cells in percentage for Control, Met and sgATF4 plus Met conditions, with statistical significance indicated by asterisks. Panel C displays protein expression analysis with Western blot results for p-mTOR, mTOR, ATF4, Cyclin D1, Caspase 3 and beta-actin under different conditions: Met, sgATF4 and sgATF4 plus Met. Molecular weights are labeled next to each protein band. Panel D is a bar graph showing relative protein levels for p-mTOR/mTOR, ATF4, Cyclin D1 and Caspase 3 across Control, Met and sgATF4 plus Met conditions, with statistical significance indicated by asterisks.
We also investigated whether mTOR is involved in ATF4 expression upon Met stimulation. MAC-T cells were treated with rapamycin, an mTOR inhibitor, upon Met stimulation. Rapamycin treatment significantly reduced the Met-induced increase in the proportion of EdU-positive cells (Fig. 5A, B). Meanwhile, rapamycin treatment did not alter total mTOR protein levels, but significantly decreased the expression of p-mTOR, ATF4 and Cyclin D1, while increasing the Caspase 3 levels (Fig. 5C, D).
The effect of rapamycin on Met regulation of ATF4 expression and cell proliferation. (A) A fluorescence microscope detected the EdU-positive staining cells. The positively staining cell was labelled with 6-Carboxyfluorescein Azide (6-FAM Azide, green). Nuclei were labelled with Hoechst 33342 (blue). Scale bar, 75 μm. (B) The proportion of EdU-positive cells. (C) Indicated protein levels were measured by Western blotting. (D) The grey scale scan of protein expression detection. Data were presented as the mean ± SEM from 3 independent experiments, each with an n = 3. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001; ns, P > 0.05.
Figure 5 Long description
The image A shows two rows of cell staining images labeled 'Met' and 'Met-Rapamycin'. Each row contains three images: Hoechst 33342, 6-FAM Azide and Merge, with scale bars indicating 75 micrometers. The image B shows a bar graph comparing the ratio of EdU-positive cells between Met and Met-Rapamycin treatments, with Met showing a higher ratio. The image C displays protein expression bands for p-mTOR, mTOR, ATF4, Cyclin D1, Caspase 3 and β-actin under conditions of Met and Met-Rapamycin treatment, with molecular weights labeled. The image D presents a bar chart comparing relative protein levels of p-mTOR/mTOR, ATF4, Cyclin D1 and Caspase 3 between Met and Met-Rapamycin treatments, showing significant differences in expression levels.
The ATF4-GSK3Β interaction regulates Met-induced Cyclin D1 expression and cell proliferation
We next explored the molecular mechanism by which ATF4 promotes Cyclin D1 expression upon Met stimulation. Co-immunoprecipitation (Co-IP) analysis showed that ATF4 bound to GSK3β (Fig. 6A). Western blotting showed that Met treatment elevated p-GSK3β (Ser9), whereas this effect was reversed in ATF4-knockdown cells treated with Met (Fig. 6B, C). These results suggest that ATF4 is involved in modulating GSK3β activity.
Met regulates cell proliferation through the ATF4-GSK3β/Cyclin D1 axis. (A) Co-IP was performed with the ATF4 antibody, and the Western blotting was used to detect indicated protein levels. (B, F) Indicated protein levels were measured by Western blotting. (C, G) The grey scale scan of protein expression detection. (D) A fluorescence microscope detected the EdU-positive staining cells. The positively staining cell was labelled with 6-FAM Azide (green). Nuclei were labelled with Hoechst 33342 (blue). Scale bar, 75 μm. (E) The proportion of EdU-positive cells. Data were presented as the mean ± SEM from three independent experiments, each with an n = 3. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001.
Figure 6 Long description
The image contains multiple panels. Panel A shows a protein analysis with bands labeled ATF4 and GSK3B, indicating Met stimulation. Panel B displays Western blot results with bands for ATF4, p-GSK3B, GSK3B and beta-actin, comparing conditions with and without sgATF4. Panel D includes cell staining images with Hoechst 33342, 6-FAM Azide and merged views under control and AR-A014418 conditions. Panel C presents a bar graph comparing relative expression levels of ATF4 and p-GSK3B under different conditions, with statistical significance indicated by asterisks. Panel E shows a bar graph of the relative fluorescence ratio of 6-FAM Azide under Met and Met plus AR-A014418 conditions. Panel F displays Western blot results for ATF4, p-GSK3B, GSK3B, Cyclin D1, Caspase 3 and beta-actin under Met and Met plus AR-A014418 conditions. Panel G includes a bar graph comparing relative expression levels of ATF4, p-GSK3B, Cyclin D1 and Caspase 3 under Met and Met plus AR-A014418 conditions, with statistical significance indicated by asterisks.
Next, we explored whether ATF4-mediated regulation of GSK3β activity occurs via specific modulation rather than complete kinase inhibition. We treated cells with AR-A014418, a GSK3β inhibitor, in the presence of Met. AR-A014418 treatment significantly decreased the Met-induced increase in the proportion of EdU-positive cells (Fig. 6D, E). Moreover, AR-A014418 treatment decreased Cyclin D1 expression in AR-A014418-treated cells, but markedly increased Caspase 3 expression (Fig. 6F, G). Meanwhile, we observed that AR-A014418 treatment did not change the Met-induced increases in ATF4 and p-GSK3β protein levels. Taken together, these data demonstrate that ATF4 specifically modulates GSK3β activity rather than completely inhibiting GSK3β activity, an effect that is critical for MAC-T cell proliferation.
Discussion
Met serves as one of the limiting amino acids for BMEC proliferation and milk production (Qi et al., Reference Qi, Meng, Jin, Li, Li and Gao2018). Our study demonstrates that ATF4, which is highly expressed in pubertal and lactating mammary tissues of dairy cows, is essential for BMEC proliferation. Met stimulates ATF4 and Cyclin D1 expression in MAC-T cells in a dose-dependent manner. Met-induced ATF4 and Cyclin D1 upregulation depends on mTOR activation. ATF4 knockdown experiments demonstrate that ATF4 drives proliferation by binding to GSK3β to upregulate Cyclin D1 upon Met stimulation.
ATF4 is a multifunctional protein widely expressed in mammalian tissues, and is a key regulator of normal metabolic homoeostasis and stress response processes (Sovolyova et al., Reference Sovolyova, Healy, Samali and Logue2014). A recent study reported high ATF4 expression in pubertal mammary glands of mice (Du et al., Reference Du, Liu, Mao, Qin and Fan2021). In this study, we found that abundant ATF4 protein was mainly localised in the mammary epithelium of pubertal dairy cows. The mammary gland is an organ that undergoes extensive cell proliferation during puberty. Studies in mouse endothelial cells have demonstrated that ATF4 is crucial for cell proliferation and functional maintenance (Bishop et al., Reference Bishop, Nien, Dauphinee and Too2006). Our in vitro experiments showed that ATF4 knockdown arrested the cell cycle in G1 phase and downregulated Cyclin D1 expression, indicating the involvement of ATF4 in cell proliferation.
Met, an essential amino acid for diverse physiological processes in dairy cows, regulates gene expression to promote cell proliferation (Martínez et al., Reference Martínez, Li, Liu, Bin, Yan, Más, Valdivié, Hu, Ren and Yin2017; Zhang, Reference Zhang2018). Circulating Met levels in dairy cows undergo physiological fluctuation, with a basal plasma concentration of 16.8–28.6 μM that can be elevated by 30.7–46.5% via rumen-protected Met supplementation (Vailati-Riboni et al., Reference Vailati-Riboni, Zhou, Jacometo, Minuti, Trevisi, Luchini and Loor2017; Toledo et al., Reference Toledo, Stangaferro, Gennari, Barletta, Perez, Wijma, Sitko, Granados, Masello, Van Amburgh, Luchini, Giordano, Shaver and Wiltbank2021). In the present study, we observed that Met addition between 0 and 0.6 mM linearly upregulated ATF4 and Cyclin D1 expression in cultured MAC-T cells. Although 0.6 mM Met represents a supraphysiological dose, this concentration is a well-established in vitro paradigm for amplifying signalling responses to characterise regulatory pathways.
Previous studies have reported that excessive Met leads to the accumulation of homocysteine, a toxic metabolite, via metabolic conversion in animals (Benevenga, Reference Benevenga1974). Homocysteine activates ER stress to promote ATF4 expression through the PERK-eIF2α pathway (Zhang et al., Reference Zhang, Lv, Luo, Weng, Qi, Bai, Zhao, Zeng, Bao, Dai, Zhang, Chen, Liu, Hu, Li and Jia2023). GRP78, a key ER chaperone, serves as a hallmark indicator of the ER stress response (Luo et al., Reference Luo, Fan, Zhang, Ngo, Zhao, Zhao, Huang, Li, Wang and Rong2022). Consistent with these prior reports, our experimental data demonstrate that 0.6 mM Met dose does not induce the expression of ER stress-related markers (GRP78, p-eIF2α and CHOP). These findings suggest that Met regulates ATF4 expression in an ER stress-independent manner. A study in BMECs showed that mTOR activation is involved in the response to 0.6 mM Met stimulation (Duan et al., Reference Duan, Lin, Lv, Yang, Jiao and Hou2017). In this study, we confirmed that mTOR inhibition attenuated Met-induced proliferation and Cyclin D1 expression, and abolished ATF4 upregulation. Consistent with previous studies, the mTOR pathway regulates the ATF4 protein abundance to execute its biological functions in an ER stress-independent manner (Park et al., Reference Park, Reyna-Neyra, Philippe and Thoreen2017; Yang et al., Reference Yang, Xie, Pan, Guan, Tu, Ye, Huang, Fu, Li, Huang, Li, Shi, Li and Zhang2024). These data demonstrate that ATF4 is the key downstream mediator through which Met drives BMEC proliferation via mTOR activation.
Moreover, we next explored the potential regulatory mechanism by which ATF4 modulates Cyclin D1 expression upon Met stimulation. In this study, we confirmed that ATF4 is required for Met-induced Cyclin D1 upregulation. Furthermore, Co-IP analysis revealed a physical interaction between ATF4 and GSK3β upon Met stimulation. GSK3β is a core regulatory kinase involved in multiple cellular processes, including cell differentiation, proliferation and apoptosis (Diehl et al., Reference Diehl, Cheng, Roussel and Sherr1998). GSK3β activity is primarily regulated by phosphorylation. Phosphorylation at Tyr216 activates the kinase, while phosphorylation at Ser9 inhibits its activity (Beurel et al., Reference Beurel, Grieco and Jope2015; Chen et al., Reference Chen, Fang, Ge, Qiu, Dworkin and Gong2024).
Notably, ATF4 and GSK3β physically interact upon Met stimulation, and this interaction is accompanied by elevated GSK3β Ser9 phosphorylation levels. However, it remains unclear whether ATF4 directly modulates GSK3β phosphorylation via the protein-protein interaction or indirectly participates in maintaining GSK3β phosphorylation mediated by upstream kinases such as AKT. AKT-mediated GSK3β Ser9 phosphorylation is a well-established mechanism driving cell proliferation via β-catenin signalling in multiple bovine cell models (Dong et al., Reference Dong, Wang, Fei, Jiang, Jiang, Guo, Liu, Cui, Meng, Li and Wang2024; Fang et al., Reference Fang, Wang, Liu, Qu, Chunyu, Xu, Xiang and Li2024). Therefore, although ATF4 modulates GSK3β Ser9 phosphorylation upon Met stimulation, further mechanistic studies are required to elucidate whether this regulation is dependent on AKT signalling.
Previous studies have demonstrated that GSK3β phosphorylation at Ser9 stabilises Cyclin D1 expression (Lan et al., Reference Lan, Qi and Du2014; Lai et al., Reference Lai, Zhao, Cao, Lu, Luo, Chen, Liu, Jiang, Yu, Lee, Li, Yu, Xu, Zhu, Gao, Li and Xue2016), whereas phosphorylation at Tyr216 facilitates its degradation (Diehl et al., Reference Diehl, Zindy and Sherr1997, Reference Diehl, Cheng, Roussel and Sherr1998). Consistent with these prior reports, our experimental data suggest that ATF4 may upregulate Cyclin D1 by inhibiting GSK3β activity upon Met stimulation. These findings indicate that ATF4 may serve as a crucial signalling intermediary that transduces Met-mediated signals to the core cell cycle regulator Cyclin D1. We also observed that ATF4 only partially inhibited GSK3β activity, rather than exerting a complete inhibitory effect on the kinase. In contrast, complete inhibition of GSK3β by AR-A014418 not only abolished Met-induced proliferation but also impaired cellular survival, without altering the levels of ATF4 and p-GSK3β (Ser9). This observation is consistent with previous studies demonstrating that GSK3β activity is essential for maintaining cell survival (Santini et al Reference Santini, Tassinari, Altomare, Loi, Ciani, Trazzi, Piccarducci, Daniele, Martini, Pagliarani, Tarozzi, Bersani, Spyrakis, Danková, Poeta, Raimondi, Davani, Aldini, Andrisano, De Simone and Milelli2025; Kunnimalaiyaan et al., Reference Kunnimalaiyaan, Gamblin and Kunnimalaiyaan2015). Therefore, ATF4-mediated regulation of GSK3β is a complex biological process, rather than the simple inhibition of its enzymatic activity. This precise regulatory mechanism was abrogated by global GSK3β inhibition, which resulted in proliferation arrest and impaired cellular survival. Collectively, in vitro data confirm that ATF4 physically interacts with GSK3β to suppress its enzymatic activity, thereby attenuating Cyclin D1 degradation and mediating mammary epithelial cell proliferation under Met-induced stimulation. Future studies will focus on elucidating the precise mechanism underlying the ATF4-GSK3β interaction and validating the physiological relevance of this mechanism in vivo.
In conclusion, ATF4 is expressed in the mammary glands of dairy cows and regulates mammary epithelial cell proliferation. Met induces ATF4 expression via mTOR signalling, independent of ER stress pathways. Mammary epithelial cell proliferation is mediated by physical interaction between ATF4 and GSK3β, which stabilises Cyclin D1 expression upon Met stimulation (Fig. 7). These findings establish a nutritional regulatory basis by which Met sustains mammary epithelial cell proliferation and provide an experimental foundation for targeted modulation of mammary cell proliferation in lactating dairy cows.
The schematic model of Met-induced cell proliferation via the ATF4-GSK3β axis. The figure was created in BioGDP (https://biogdp.com/).
Figure 7 Long description
The diagram illustrates the process of Met-induced cell proliferation via the ATF4-GSK3β axis. Methionine interacts with LATs, leading to the activation of p-mTOR. This activation influences ATF4, which interacts with GSK3β and phosphorylated GSK3β. The interaction stabilizes Cyclin D1 expression, promoting cell proliferation. The diagram also shows nonactivated ERs within the cytoplasm, indicating their role in the process. The membrane is depicted as the boundary of the cell, with the cytoplasm inside. The flow of interactions is marked with arrows, showing the sequence from methionine to cell proliferation.
Supplementary material
The supplementary material for this article can be found at 10.1017/S0022029926102416.
Funding statement
This work was supported by the Natural Science Foundation of Heilongjiang Province of China (LH2023C019 to Na Zhang), the National Natural Science Foundation of China (32402707 to Yang Yang) and the Young Talents Project of the Northeast Agricultural University (54970212 to Yang Yang).
Competing interests
The authors have declared that no competing financial interests exist.
