If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
INRA UMR 1080 Dairy Production, F-35000 Rennes, FranceAgrocampus Ouest UMR 1080 Dairy Production, F-35000 Rennes, FranceBern University of Applied Science, Swiss College of Agriculture, 3052 Zollikofen, Switzerland
The aim of this study was to investigate the effects of a severe nutrient restriction on mammary tissue morphology and remodeling, mammary epithelial cell (MEC) turnover and activity, and hormonal status in lactating dairy cows. We used 16 Holstein × Normande crossbred dairy cows, divided into 2 groups submitted to different feeding levels (basal and restricted) from 2 wk before calving to wk 11 postpartum. Restricted-diet cows had lower 11-wk average daily milk yield from calving to slaughter than did basal-diet cows (20.5 vs. 33.5 kg/d). Feed restriction decreased milk fat, protein, and lactose yields. Restriction also led to lower plasma insulin-like growth factor 1 and higher growth hormone concentrations. Restricted-diet cows had lighter mammary glands than did basal-diet cows. The total amount of DNA in the mammary gland and the size of the mammary acini were smaller in the restricted-diet group. Feed restriction had no significant effect on MEC proliferation at the time of slaughter but led to a higher level of apoptosis in the mammary gland. Gelatin zymography highlighted remodeling of the mammary extracellular matrix in restricted-diet cows. Udders from restricted-diet cows showed lower transcript expression of α-lactalbumin and kappa-casein. In conclusion, nutrient restriction resulted in lower milk yield in lactating dairy cows, partly due to modulation of MEC activity and a lower number of mammary cells. An association was found between feed restriction-induced changes in the growth hormone–insulin-like growth factor-1 axis and mammary epithelial cell dynamics.
The production performance of lactating ruminants is shaped by many farm-system factors such as milking frequency and diet. The milk production potential of the mammary gland is shaped by both mammary epithelial cell (MEC) number and their secretory activity (
), thus establishing a direct link between MEC number and activity and milk yield. Mammary epithelial cell turnover is defined by 2 processes: proliferation and apoptosis (programmed cell death;
). Experiments on early lactating rats showed that increasing feeding level during gestation led to an increase in cell proliferation and a decrease in apoptosis during late lactation (
). Any energy restriction during the lactation cycle subsequently leads to a strong decrease in milk production. Experiments on lactating ewes showed that an energy restriction of 50% led to a 30% decrease in milk yield during the subsequent lactation, suggesting a permanent change in the number or activity of MEC (
). Moreover, a restriction in nutrient availability to dairy cows had effects on mammary cell turnover and on enzyme activity and could be mediated by IGF-1 plasma concentration (
showed that energy restriction during lactation negatively affected mammary cell proliferation in dairy cows. Nutrition can also affect rate of apoptosis in the mammary gland. Evidence exists that dietary antioxidants directly affect mammary levels of B-cell lymphoma-2 (Bcl-2) and Bcl-2–associated X protein (Bax)—2 proteins implicated in cell death mechanisms (
In dairy cow mammary glands, cell proliferation is more important than apoptosis during gestation, but during the lactation phase, cell proliferation decreases and apoptosis is promoted. This ratio between cell proliferation and apoptosis is also called mammary cell turnover. Many factors control this ratio, especially prolactin (PRL), growth hormone (GH), and IGF-1 of systemic or local origin (
demonstrated that administration of GH during midlactation enhanced the milk yield by increasing mammary cell proliferation (antigen Ki-67 staining) and the rate of cell renewal. The galactopoietic effect of GH is, thus, known but 2 theories exist about the action mode of IGF-1(
). The second theory postulates that an increase in metabolism within mammary glands could lead to increasing milk production. The ability of IGF-1 to induce cell proliferation has been demonstrated in several in vitro and in vivo mammary models (
). Moreover, lactation is often dependent on the mobilization of body reserves in response to strong nutrient needs—a process facilitated by high circulating GH levels (
). During this stage, the GH–IGF-1 axis becomes uncoupled due to a downregulation of liver GH receptor, leading to lower IGF-1 levels and higher GH concentrations (
Few laboratories to date have shown interest in the potential link between energy restriction or GH–IGF-1 axis and mammary epithelium dynamics, and little is known about the effects of nutrient restriction on MEC and tissue remodeling. The aim of this study was, therefore, to investigate the effects of severe nutrient restriction on mammary tissue morphology and remodeling, MEC number and activity, and hormonal status in lactating dairy cows.
Materials and Methods
All of the animal procedures used were approved by the French Ministry of Agriculture's animal ethics committee, in accordance with French regulations (Decree No. 2001–464, May 29, 2001).
Animals and Experimental Design
Sixteen Holstein × Normande crossbred multiparous dairy cows (lactation ranks 2 and 3) were used in this study. Cows were housed at the National Institute for Agricultural Research's (INRA's) experimental dairy farm in France (Le Pin-au-Haras). They were divided into 2 groups submitted to 2 feeding levels. From 2 wk before calving to wk 11 postpartum, cows were fed with a TMR composed of either 55% maize silage, 15% dehydrated alfalfa, and 30% concentrate (basal diet group as the control group, n = 8) or 60% grass silage and 40% hay (restricted diet group, n = 8).
These diets were chosen using the same experimental design as that in
: the basal diet allowed the expression of milk potential and a moderate weight loss and the restrictive diet aimed to limit milk production and to induce an intense weight loss. The nutritive value (%) of the diets analyzed by LDA22 (Ploufragan, France) was DM 48.5 versus 56.8, CP 12.6 versus 15.4, starch 0 versus 24.3, and NDF 57.7 versus 34.7, NEL 1.28 versus 1.61 Mcal, protein digestible in the small intestine when rumen nitrogen is limiting: 81 versus 104 g/kg, and protein digestible in the small intestine when rumen energy is the limiting factor: 69 versus 105 g/kg, respectively, for the restricted versus the basal diet.
After 11 wk of lactation (between 77 and 98 d after calving), cows were slaughtered 4 h (±1 h) after last milking and last feeding at the Socopa Slaughterhouse (Gacé, France).
Cows were assigned to experimental groups to achieve a similar date of future calving. Allocation was based on average performance (milk yield and fat and protein content) over the previous 305-d lactation, together with BW and BCS before calving (2.04 vs. 2.17, P = 0.42, respectively, for the basal-diet and restricted-diet groups).
Milk Production and Composition
Milk production was recorded daily and milk composition was measured 3 times/wk.
Milk protein, fat, and lactose contents were determined by an independent laboratory using infrared methods (Lilano, St-Lô, France).
Blood Samples
Blood samples to determine plasma hormone concentrations were taken 15 min after the beginning of a.m. milking at d −6, −2, and 0 before slaughter. Sampling was performed using Monovette syringes coated with EDTA and sodium heparin (Sarstedt AG & Co., Nümbrecht, Germany). An additional sample was collected weekly into heparin-coated tubes to determine NEFA concentrations. Plasma was separated by centrifugation at 3,000 × g for 15 min at 4°C and stored at −20°C until analyses.
Plasma PRL concentration was measured using an indirect competitive ELISA. Intra-assay variability was <5% and inter-assay variability was <12%. Plasma (from EDTA-coated tubes) concentrations of GH and IGF-1 were determined by RIA (
Clinical, haematological, metabolic and endocrine traits during the first three months of life of suckling Simmentaler calves held in a cow-calf operation.
Plasma NEFA concentrations were measured on a multiparameter analyzer (Kone Instruments Corp., Espoo, Finland) using a NEFA C-test kit (Wako; Oxoid, Dardilly, France).
Mammary Glands
After 11 wk of lactation, all cows were slaughtered on the same day between 10 and 11 a.m. and the mammary glands were removed at the slaughterhouse and weighed. Mammary gland samples were always taken on the rear udder from the mid-alveolar region in small aliquots (between 2 and 10 mg), directly frozen in liquid nitrogen, and stored at −80°C for RNA, DNA, and protein extraction. Other mammary tissue samples were washed in PBS (Fisher Scientific Bioblock, Illkirch-Graffenstaden, France) and prepared for immunohistology analyses.
Immunohistology: Cell Proliferation
Proliferating mammary cells in the mammary gland were determined by immunohistochemical staining for proliferating cell nuclear antigen (PCNA). Proliferating mammary cells were identified in mammary tissue as cells expressing the PCNA. Briefly, immediately after slaughter, mammary gland tissues were fixed in PBS 4% paraformaldehyde (Sigma-Aldrich Chimie, Lyon, France) for 24 h at 4°C, then cryoprotected in 20% sucrose (Sigma-Aldrich Chimie) for 48 h at 4°C, frozen in an isopentane (Sigma-Aldrich Chimie) bath, cooled on dry ice, and stored at −80°C until use. Cryosections 7 μm in thickness were mounted onto SuperFrost Plus slides (Prolabo, Bondoufle, France), then quenched in PBS 3% hydrogen peroxide and 10% methanol for 30 min. The sections were thoroughly washed in PBS, permeabilized with PBS 1% SDS for 5 min, washed 3 times and then pre-incubated in PBS 1% BSA for 1 h at room temperature. Samples were then incubated for 12 h at 4°C in the presence or absence of the primary antibody. After washing in PBS 1% BSA, samples were incubated with an appropriate secondary antibody for 1 h at room temperature. The mammary gland sections were subsequently counterstained for 3 min with 33 μg/mL 4′,6-diamidino-2-phenylindole (DAPI, D9542; Sigma-Aldrich Chimie) and then 3 min with propidium iodide at 333 μg/mL (P4864; Sigma-Aldrich Chimie).
Immunohistology: Apoptosis
Apoptotic cells in mammary gland tissue were determined by terminal deoxynucleotidyl transferase (TdT)-mediated 2′-deoxyuridine, 5′-triphosphate (dUTP) nick-end labeling (TUNEL), based on DNA fragmentation detection. Cryosections 7 μm in thickness were mounted onto 3-aminopropyltriethoxysilane (Sigma-Aldrich Chimie)-treated slides, then thawed and incubated for 30 min at 70°C in 10 mM sodium citrate (Sigma-Aldrich Chimie) 0.1% triton solution (Fisher Scientific Bioblock). The slides were washed in PBS, then incubated 30 min at 37°C in 200 ng/μL of proteinase K solution (V3021; Promega France, Charbonnières, France), followed by incubation with the DeadEnd Fluorometric TUNEL System (G3250; Promega France) reagents following the manufacturer's instructions. The mammary gland sections were DAPI-counterstained after the TUNEL reaction.
The slides for cell proliferation and apoptosis analyses were then mounted with Vectashield (Valbiotech, Paris, France) and examined by fluorescence microscopy under an Eclipse E400 Nikon microscope (Nikon France, Le Pallet, France), and pictures were captured on a DXM 1200 digital still camera (Nikon France). Microscopic height fields (magnification × 200; area 0.14 mm2 per microscopic field) were examined for each tissue sample. Cells were detected and counted using Image J software (W. Rasband, National Institutes of Health, Bethesda, MD). For each cow, 3 different samples were used and 8 pictures were analyzed per samples, for a total of 24 pictures.
Zymography
Gelatin zymography was used to quantify the presence of both activated and latent forms of matrix metalloproteinase (MMP), MMP-2, and MMP-9 in tissue extracts. Frozen tissues were ground in protein extraction buffer (50 mM Tris-HCl at pH 7.5, 5 mM CaCl2, 0.9% mM NaCl, and 0.2% Triton X-100; Sigma-Aldrich Chemie, Lyons, France) using an Ultra-Turrax homogenizer, sonicated 10 times for 20 s, and incubated on ice for 15 min. After homogenization, the mixture was centrifuged at 13,000 × g for 15 min at 4°C, and the supernatant was recovered. Protein concentration was then determined by the Bradford protein assay (Sigma-Aldrich Chimie). Proteins (5 μg per lane) were subjected to electrophoresis in 10% SDS-PAGE gels co-polymerized with 1 mg/mL of gelatin (Sigma-Aldrich Chimie). After electrophoresis, SDS was removed by washing the gels twice with 2.5% Triton X-100 in deionized water for 10 min, and twice more for 20 min in deionized water. Gels were then incubated for 24 h at 37°C in 50 mM Tris-HCl buffer, pH 8, containing 5 mM CaCl2 and 2 μM ZnCl2. Regions of proteolytic activity were visualized as clear zones against a blue background after 1 h of staining with 0.1% Coomassie blue R-250 in 25% ethanol and 10% acetic acid at room temperature, followed by destaining with the same solution without Coomassie blue. A standard BenchMarker protein (Fermentas GmbH, Courtaboeuf, France) and a human recombinant MMP-2 were used to confirm the identity of the clear bands on the gels. Gels were scanned at 8-bit/500-dpi resolution on a CanoScan D1250 U2 flatbed scanner (Canon, Courbevoie, France). The activities of proMMP-9 (92 kDa), active MMP-9 (82 kDa), proMMP-2 (66 kDa), and active MMP2 (57 kDa) were measured by quantifying the integrated optical density of the clear bands using ImageQuant software (GE Healthcare, Lyon, France).
Western Blot Analysis
Total proteins were extracted from frozen mammary gland tissue using the Tissue Protein Extraction Reagent (Pierce; Perbio Science France, Brebières, France). A mammary tissue sample (100 mg) was ground in liquid nitrogen with a mortar and a pestle, and 1 mL of Tissue Protein Extraction Reagent was added to the powder. After homogenization, the mixture was centrifuged at 10,000 × g for 5 min at 4°C and the supernatant containing total proteins was recovered. The protein concentration was determined by the Lowry method using the DC Protein Assay kit (Pierce; Perbio Science France). Lysates were then combined with sample buffer (50 mM Tris-HCl pH 6.8, 2% SDS, 0.1% bromophenol blue, 20% glycerol, and 5% β-mercaptoethanol) and boiled for 5 min at 95°C to be resolved by SDS-PAGE.
Proteins (30 μg per lane) were separated on 10% SDS-polyacrylamide gels, blotted to polyvinylidene difluoride membranes (Amersham Biosciences, Paris, France) and incubated with blocking solution [5% dry skim milk dissolved in Tris-buffered saline with Tween (TBS-T; Perbio Science France), 50 mM Tris-HCl pH 8.6, 150 mM NaCl, and 0.1% Tween] for 30 min. A set of prestained molecular mass standards was run in each gel. Membranes were incubated overnight at 4°C with the appropriate dilution of the following primary antibodies: mouse monoclonal anti-PCNA (M0879; DakoCytomation, Trappes, France), a mouse monoclonal anti-AKT (activated kinase tyrosine; Santa Cruz Biotechnology, Paris, France) and a mouse monoclonal anti-β-actin (Sigma-Aldrich Inc., Paris, France). Membranes were then washed with TBS-T before incubation with horseradish peroxidase-conjugated anti-mouse or anti-rabbit secondary antibodies (Pierce; Perbio Science France) for 1 h at room temperature. Peroxidase activity was detected using an enhanced chemiluminescence detection system (ECL System; Amersham Biosciences, Lyon, France). Membranes were exposed to Hyperfilm (Amersham Biosciences). Images were scanned at 16-bit/600-dpi resolution using a CanoScan D1250 U2 flatbed scanner (Canon), saved as TIFF files, and calibrated to an optical density scale. The integrated optical density of bands was quantified using ImageQuant software. Each sample was normalized to β-actin content.
Real-Time Quantitative PCR Procedure
Total RNA was extracted from tissue samples after grinding in liquid nitrogen with a mortar and pestle using TRIzol (Invitrogen, Paris, France). The powdered tissue was homogenized in 1 mL of TRIzol reagent, and after 5 min of incubation at room temperature (20°C), 200 μL of chloroform was added to each sample. The mixture was centrifuged at 12,000 × g for 15 min at 4°C, and the upper aqueous phase containing total RNA was recovered and mixed with 500 μL of isopropyl alcohol to precipitate the RNA. After a further 10-min incubation at room temperature, the RNA was pelleted by centrifugation at 12,000 × g for 10 min at 4°C, rinsed twice with 75% ethanol, and dissolved in sterile RNase-free water (
Total RNA concentration and the 260/280 nm and 260/230 nm absorbance ratios were measured on a NanoDrop ND-1000 spectrophotometer (NanoDrop Technologies Inc., Wilmington, DE). The RNA quality was then determined on an Agilent 2100 Bioanalyzer (Agilent Technologies, Massy, France) by RNA profile and RNA integrity number measured using Agilent 2100 Expert Software, version B.02 (Agilent Technologies).
Complementary DNA was generated from total RNA using a ThermoScript RT-PCR System (Invitrogen, Lyon, France) according to the manufacturer's protocol. Total RNA (500 ng) was incubated for 1 h at 42°C with 10 U of avian myeloblastosis virus reverse transcriptase, 0.8 μg of oligo(dT)15 primer, 0.5 μL of 50 U/μL of RNase inhibitor, 1 μL of 1 mM deoxynucleotide mix, 1 μL of reverse transcriptase 10× buffer (10 mM Tris-HCl, pH8.3, and 50 mM KCl), and 5 mM MgCl2 in a total vol of 10 μL. Finally, the mixture was heated to 95°C for 5 min to inactivate the avian myeloblastosis virus reverse transcriptase, and then cooled down to 4°C. The cDNA was diluted 1:50 before use in real-time quantitative PCR (RT-qPCR).
Following reverse transcription, RT-qPCR was run on a StepOnePlus System (Applied Biosystems, Carlsbad, CA). Triplicate 12.5-μL reactions were carried out in 96-well optical reaction plates (Applied Biosystems) using SYBR Green PCR master mix (Applied Biosystems) with 200 nM of each specific primer [Bcl2-associated death protein (BAD): forward 5′ GAGGATGAGCGACGAGTTTC 3′, reverse 5′ TCAACCAGGACTGGAGGAAG 3′; cathepsin B: forward 5′ AGACTGAAGACGGAGGCAAA 3′, reverse 5′ GCAGCTTGTCCATGAGTGAA 3′; IGF-binding protein 5 (IGFBP 5): forward 5′ GACAGAAATCTGAGCAGGGG 3′, reverse 5′ TTGTAGAACCCTTTGCGGTC 3′; calpain 2: forward 5′ GAGGACATGCACACCATTGG 3′, reverse 5′ GTTGAGCACCTCCCGCAG 3′; caspase 3: forward 5′ AGCCATGGTGAAGAAGGAATCA 3′, reverse 5′ ACCACAGTCCAGTTCTGTGCCT 3′; phosphatase and tensin homolog protein (PTEN): PTEN forward 5′ CCGCCAAATTTAATTGCAGAGTTG 3′, reverse 5′ ACACCAGTTCGTCCCTTTCCAG 3′; αLA: forward 5′ ACCAGTGGTTATGACACACAAGC 3′, reverse 5′ AGTGCTTTATGGGCCAACCAGT 3′; κCN: forward: 5′ TGCAATGATGAAGAGTTTTTTCCTAG 3′, reverse 5′ GATTGGGATATATTTGGCTATTTTGT 3′;
Cell junction disruption after 36 h milk accumulation was associated with changes in mammary secretory tissue activity and dynamics in lactating dairy goats.
The amplification program consisted of an initial denaturation step at 95°C for 10 min, followed by 40 cycles of denaturation at 95°C for 10 s and combined primer annealing-extension at 60°C for 1 min, during which fluorescence was measured. A melting curve was produced after completion of the thermal PCR program to check the presence of a gene-specific peak and the absence of a primer dimer.
Data raw cycle threshold (Ct) values obtained from raw ABI Prism 7000 Software version 1.1 (Applied Biosystems) were transformed to quantities using the comparative Ct method (
To find suitable reference genes for this study, 5 candidate genes were evaluated: β-actin, glyceraldehyde-3-P dehydrogenase, cyclophilin A, 18S ribosomal RNA, and the ribosomal protein large P0. geNorm version 3.4 (Visual Basic Application tool for Microsoft Excel: http://medgen.ugent.be/~jvdesomp/genorm) and NormFinder (a Microsoft Excel add-in: http://www.mdl.dk/publicationsnormfinder.htm) were used to calculate the stability of the candidate reference genes. geNorm helps to calculate the gene expression stability measure for the set of candidate reference genes. It is built on the principle that the expression ratio of 2 reference genes should be identical in all samples, regardless of experimental conditions. NormFinder determines the stability of the candidate genes based on estimated inter- and intra-group variation. It calculates the most stably expressed candidate genes and suggests the optimum reference genes. The cyclophilin A gene was determined by geNorm and NormFinder as the most stable reference gene for gene expression profiling.
DNA Quantification
Mammary tissue samples (50 mg) were first weighed and homogenized in 2.5 mL of sodium phosphate extraction buffer (0.05M Na2HPO4, 0.05 M NaH2PO4, 2 mM EDTA, and 2 M NaCl, pH 7.4). The mixture was ground using an Ultra Turrax homogenizer, sonicated for 1 min, and centrifuged at 4,000 × g for 1 min. The supernatant was transferred to a new tube, whereas the pellet was resuspended in 2.5 mL of the extraction buffer before being re-subjected to the 3 successive grinding, sonication, and centrifugation extraction steps described above. At the end of each centrifugation, all supernatants were recovered and pooled. The DNA mixture was then analyzed by fluorescence using Hoechst 33258 dye (Sigma-Aldrich Chimie). Aliquots of the DNA mixture were mixed with sodium phosphate buffer stained with Hoechst dye to a final dye concentration of 1 μg/mL. Triplicate reactions were performed in a 200-μL final vol in a black 96-well microplate. Serial dilutions of spectrophotometrically measured calf thymus DNA (Sigma-Aldrich Chimie) were used to generate the standard curve. Plates were read directly on a Mithras LB940 fluorescence plate reader (Berthold Technologies, Thoiry, France) at an excitation wavelength of 355 nm and an emission wavelength of 460 nm. Fluorescence values from a control sample were subtracted from each of the extract samples, and DNA concentrations were calculated by extrapolation on the standard curve corrected for aliquot volume.
Statistical Analysis
Zootechnical data and hormone concentrations were analyzed by ANOVA using the SAS MIXED procedure with the REPEATED statement (
) using week as repeated effect and cows as subject. The effects of diet, week, and the interaction diet × wk were tested. Statistical analysis of real-time PCR data was performed on the ratio ΔΔCt [(CT(target, sample) – CT(reference, sample)) – (CT(target, control) – CT(reference, control))], where CT = cycle threshold of the target gene to ΔΔCt of the reference gene, using the MIXED procedure of SAS (
). For plasma hormone concentrations and mammary gland data, the effect of diet was analyzed by Student's t-test using the general linear model procedure of SAS. Effects were considered significant at P < 0.05. All means are reported as means ± standard error of the means.
Results
Feed Restriction Decreased Milk Production and Modified Milk Composition
The milk yield was affected by diet. The milk yield was 38% lower in restricted-diet cows than in basal-diet cows (20.5 vs. 33.5 kg/d; P < 0.001; Table 1 and Figure 1). This effect was already significant 1 wk following the beginning of treatment, and remained significant through to slaughter. In addition to lower milk yield, the restricted diet decreased total milk fat yield by 39%, total milk protein yield by 45%, and total lactose yield by 40% compared with the basal diet (Table 1). Feed restriction led to lower milk lactose and milk protein contents (P < 0.001). Mean plasma NEFA concentration was higher in restricted-diet cows than in basal-diet cows (P < 0.05).
Table 1Effects of feeding level on milk yield, milk composition (fat, protein, lactose), and NEFA concentration during the first 11 wk of lactation in restrictively or normally fed dairy cows (n = 8 in each group)
Figure 1Effects of feeding level on milk yield during the first 11 wk of lactation in restrictively or normally fed dairy cows (n = 8 in each group). The basal-diet group is in dark gray and the restricted-diet group is in light gray (* indicates significant difference at P < 0.001).
Feed Restriction Decreased Mammary Gland Weight and Total DNA
Mammary gland weight was 38% lower in restricted-diet cows (P < 0.001; Table 2). The total DNA amount in mammary gland samples was 20% (P < 0.01) lower in restricted-diet cows than in basal-diet cows, whereas mammary DNA concentration was higher in restricted-diet cows (2.5 mg/g vs. 2.0 mg/g; P < 0.01).
Table 2Effects of feeding level on mammary gland weight, DNA concentration, and nuclei cell numbers after 11 wk of lactation in restrictively or normally fed dairy cows (n = 8 in each group)
Feed Restriction Did Not Affect MEC Proliferation but Activated MEC Apoptosis Pathways
Immunohistochemical staining of PCNA (Figure 2A, panels 5–6), nuclear staining with propidium iodide (Figure 2, panels 1–2), and DAPI-staining (Figure 2, panels 3–4) were used to evaluate the number of proliferative MEC. The counts per area of all nuclei and PCNA-stained nuclei showed no proliferation difference between the 2 diets (22.8 vs. 23.8% of stained nuclei for restricted-diet cows and basal-diet cows, respectively; P = 0.79). This result was confirmed by Western blot, as the PCNA levels measured in mammary gland samples from both diet groups were similar (P = 0.85; Figure 2B). These results suggest that after 11 wk of lactation, feed restriction had no effect on MEC proliferation.
Figure 2Effects of feeding level on mammary cell proliferation after 11 wk of lactation in restrictively or normally fed dairy cows (n = 8 in each group). (A) Propidium iodide (PI) staining (1 and 2), 6-diamidino-2-phenylindole (DAPI) staining (3 and 4), and proliferating cell nuclear antigen (PCNA, 5 and 6) immunohistochemistry in tissue sections of mammary glands from cows that were fed with a basal diet (left panels) or a restricted diet (right panels). Arrowheads indicate PCNA-positive nuclei. Proliferation was assessed by quantification of nuclear staining for PCNA in tissue sections. (B) Western blot analysis of total mammary gland protein extracts to assess PCNA levels. Each single band corresponds to 1 cow. Actin was used to control loading charge. MW = molecular weight. Color version available in the online PDF.
Real-time quantitative PCR is an effective method to easily quantify RNA transcript levels in tissue. We used RT-qPCR to quantify the expression levels of 6 transcripts related to proteins involved in various apoptotic signaling pathways (i.e., BAD, PTEN, caspase 3, cathepsin B, IGFBP-5, and calpain 2. All of these transcripts of interest showed higher expression in the mammary glands of restricted-diet cows (Figure 3).
Figure 3Effects of feeding level on apoptosis and milk synthesis-involved gene expression in mammary gland extracts at 11 wk postpartum, in restrictively or normally fed dairy cows (n = 8 in each group). The Livak method (2−ΔΔCt) was used to quantify gene expression using cyclophilin A gene as reference gene and basal diet as sample reference (**P < 0.01). Above (2−ΔΔCt >1) and under (2−ΔΔCt <1) the dashed line indicates gene upregulation and downregulation, in the mammary gland samples from restricted-diet cows compared with basal-diet cows. BAD = Bcl2-associated death promoter protein; PTEN = phosphatase and tensin homolog protein; Casp3 = caspase 3; CathepB = cathepsin B; IGFBP5 = IGF-binding protein 5; GT = galactosyltransferase. Color version available in the online PDF.
To confirm cell death processes in the udders of restricted-diet cows, we monitored DNA fragmentation in situ using the TUNEL method. The fluorescence images (Figure 4A) revealed the presence of MEC with fragmented DNA in mammary glands from both groups. However, counts of these TUNEL-positive cells (Figure 4A, panels 3–4) revealed more apoptotic cells in mammary tissue from restricted-diet cows compared with basal-diet cows (0.30 vs. 0.14%; P < 0.01; Figure 4B).
Figure 4Effects of feeding level on apoptosis in mammary extracts after 11 wk postpartum in restrictively or normally fed dairy cows (n = 8 in each group). (A) 6-Diamidino-2-phenylindole (DAPI) staining (1 and 2) and terminal deoxynucleotidyl transferase 2′-deoxyuridine, 5′-triphosphate (dUTP) nick end labeling (TUNEL) staining (3 and 4) in mammary gland tissue sections from cows fed with a basal diet (left panels) or a restricted diet (right panels). Arrowheads indicate TUNEL-positive nuclei. (B) Apoptosis rates were assessed by quantification of TUNEL nuclear staining in tissue sections (**P < 0.01). Color version available in the online PDF.
Thus, after 11 wk of lactation, MEC from the udders of restricted-diet cows activated the apoptosis process, and this apoptotic activation is related to the decrease in milk production.
The activity of gelatinases MMP-2 and MMP-9 in mammary glands of cows from both diet groups was studied by zymography using electrophoresis with SDS-PAGE gels containing gelatin. The pro-forms and active forms of MMP-2 and MMP-9 were visualized at 66/59 kDa and 92/82 kDa, respectively, in total mammary gland (Figure 5A). Quantitative analysis of pixel intensity showed that MMP-2 and MMP-9 activity was higher in the mammary glands of restricted-diet cows (P < 0.01; Figure 5B), suggesting an involution process. Metalloproteinase activation could involve degradation of the extracellular matrix proteins, and thus, a remodeling of the acini and the mammary gland.
Figure 5Effects of feeding level on metalloproteinase activity in mammary extracts after 11 wk of lactation in restrictively or normally fed dairy cows (n = 8 in each group). (A) Analysis of metalloprotease-9 (MMP-9) and MMP-2 protein activity using the gelatin zymography in mammary gland extract from basal-diet and restricted-diet cows. Each single band corresponds to 1 cow. Arrows represent pro- and active forms of each MMP. (B) Graphical representation of MMP-9 and MMP-2 protein activity. Activities were quantified by densitometrically scanning the photographic negative of the zymogram, and were expressed as the log10 of pixel intensity. The basal-diet group is in black and the restricted-diet group, in light gray (**P < 0.01). MW = molecular weight.
Feed Restriction Modified Both Acini Size and Structure
Acini size is generally correlated with milk production. We thus cut mammary gland sections to analyze the size and structure of the acini. Scans very clearly showed far more intensive closing of the acini in restricted-diet cows than in basal-diet cows (Figure 6A). Acini surface measurements confirmed smaller acini in mammary glands from restricted-diet cows (10,500 μm2 vs. 22,500 μm2; P < 0.01; Figure 6B). The nuclei were DAPI-stained to obtain both the number of nuclei per unit of area and per acinus. Interestingly, restricted-diet cows had a higher number of nuclei per unit of area and a similar number of nuclei per acinus compared with basal-diet cows (Table 2). These results suggest that the restricted diet caused mammary gland remodeling without strongly affecting the number of mammary cells per acinus after 11 wk of lactation.
Figure 6Effects of feeding level on the morphology of mammary gland acini after 11 wk of lactation in restrictively or normally fed dairy cows (n = 8 in each group). (A) Hematoxylin–eosin staining of mammary gland tissue sections from cows fed with a basal diet (left panel) or a restricted diet (right panel). (B) Acini areas in mammary gland from cows fed with a basal diet (black) or a restricted diet (gray; **P < 0.01). Color version available in the online PDF.
Feed Restriction Appeared to Negatively Affect MEC Secretory Activity
The number of nuclei per acinus did not vary with diet, but milk production largely differed. We thus studied the secretory activity of MEC via RT-qPCR quantification of α-LA, κ-CN, and galactosyltransferase transcripts. Expression of α-LA (38%) and κ-CN (46%) was less stimulated in udders from restricted-diet cows (P < 0.01; Figure 3), whereas galactosyltransferase levels were similar between groups.
Feed Restriction Was Closely Associated With Changes in the GH–IGF-1 Signaling Pathway
Studies have suggested that local and systemic production of IGF-1 mediates the mammary gland response to GH and stimulates mammary development. Furthermore, IGF-1 is energy balance sensitive. We thus measured the plasma concentrations of IGF-1 and GH. Results showed that feed restriction negatively affected IGF-1 concentration and positively affected GH concentration (Table 3). Because PRL has been recently shown to have a galactopoietic effect in ruminants (
), we also investigated plasma PRL concentration in both groups. Feeding level had no effect on plasma PRL concentration (42 ng/mL vs. 52 ng/mL; P = 0.12).
Table 3Effects of feeding level on plasma IGF-1 and growth hormone (GH) concentrations and on pro-survival activated kinase tyrosine (AKT) protein in mammary extracts after 11 wk of lactation in restrictively or normally fed dairy cows (n = 8 in each group)
We tested whether the feed restriction-related decrease in IGF-1 concentration had an effect on MEC survival. Western blot analysis on the amount of AKT in mammary gland extracts showed less AKT in the restricted-diet animals (40%; P < 0.001; Table 3).
Discussion
This experiment was designed to obtain a high level of overall nutrient supply and milk yield from the basal diet, whereas the restricted diet would hamper the cows’ milk potential. The feed restriction led to a 39% decrease in average milk production over the first 11 wk of lactation. α-Lactalbumin expression was lower in feed restriction in agreement with the lower lactose production. Diet restriction also affected milk protein content, leading to lower protein content and weaker κ-CN expression in restricted-diet cows. Studies have shown that a severe decrease in dietary AA is associated with a decrease in AA availability for protein synthesis (
). Diet restriction also triggered a greater decrease in BCS and BW (data not shown), as confirmed by higher plasma NEFA concentrations. Our data thus suggest that the effects of the different dietary supplies on milk synthesis were partly compensated by a greater mobilization of body reserves. However, in restrictively fed cows, which were under severe negative energy balance, the MEC visibly lacked the precursors necessary to maintain milk protein levels.
Udder weights and total DNA contents were lower in restricted-diet cows than in basal-diet cows. As these results suggested that feed restriction caused a decrease in cell numbers, we investigated the consequences of feed restriction on mammary cell turnover and the mechanisms involved in mammary gland remodeling.
The 3 phases of lactation (early, established, and post) are directly tied to key cellular processes modulating MEC number. This modulation is related to cell proliferation, cell death, and metalloproteinase activities (
). Immunohistochemistry and Western blot analyses of mammary tissue did not show any differences in MEC proliferation between the 2 diets. However, in mammary glands sampled at 11 wk, following the peak of lactation, the epithelial cells no longer showed intense proliferation. The effect of diet restriction on mammary cell proliferation is probably dependent on lactation stage. Indeed, a low-energy-density diet has been reported to inhibit mammary cell proliferation at 8 wk but not at 16 wk postpartum (
). Thus, the difference in mammary cell proliferation could have been involved earlier in the change of mammary cell number due to feed restriction. In this study, the proliferation rate measured is relatively high. In fact, PCNA staining overestimates proliferation, as PCNA is also involved in DNA repair in G0 phase and has a high half-life (
A comparison of proliferating cell nuclear antigen (PCNA) immunostaining, nucleolar organizer region (AgNOR) staining, and histological grading in gastrointestinal stromal tumours.
Characterization of the proliferation state in canine mammary tumors by the standardized AgNOR method with postfixation and immunohistologic detection of Ki-67 and PCNA.
). The TUNEL results, together with analyses of the transcripts involved in various apoptotic pathways, showed that their expression was increased by feed restriction. This study clearly showed that different pathways implicated in cell death were activated. Our findings suggest that feed restriction activated similar mechanisms to those activated during mammary involution.
A protein- or energy-restricted food regimen is classically associated with an increase in plasma GH concentration in many species (
Effect of restricted feeding on the concentrations of growth hormone (GH), gonadotropins, and prolactin (PRL) in plasma, and on the amounts of messenger ribonucleic acid for GH, gonadotropin subunits, and PRL in the pituitary glands of adult ovariectomized ewes.
). Growth hormone mediates lipolysis of adipose tissue by potentiating the effects of catecholamines on monosensitive lipase. One marker of this GH activity on fat tissue is the release of NEFA. In our study, the high GH and high NEFA concentrations in the restricted-diet group indicate that cows drew on their body reserves to maintain basal metabolism and milk production. Due to the uncoupling commonly observed between GH and IGF-1 in the beginning of lactation in dairy cows (
Functional differences in the growth hormone and insulin-like growth factor axis in cattle and pigs: Implications for post-partum nutrition and reproduction.
), the high GH levels in restricted-diet cows did not lead to high IGF-1. The feed restriction probably even reinforced the uncoupling of the GH-IGF-1 axis (
), as restricted-diet cows had higher plasma GH concentrations and lower plasma IGF-1 concentrations than did basal-diet cows. Growth hormone resistance was, therefore, severe in restricted-diet cows. These animals, under strong negative energy balance, increase their plasma GH levels to maintain homeostasis rather than using GH to stimulate growth and cellular proliferation. This is in accordance with results of
who demonstrated that the GH injection enhanced secretion of milk protein. Insulin-like growth factor-1 levels are also related to a differential regulation of IGF-1 transcripts. In diet-restricted rats, GH increased IGF-1 transcript levels without increasing plasma IGF-1 production (
Evidence that pretranslational and translational defects decrease serum insulin-like growth factor-I concentrations during dietary protein restriction.
), suggesting a process that blocks the action of GH on the transcription of the gene coding for IGF-1.
Numerous studies have demonstrated that the signaling pathways involved in the IGF system play a pivotal role in cell survival and in the inhibition of apoptosis (particularly in epithelial cells). The best-characterized anti-apoptotic pathway is mediated by phosphoinositol 3-kinase (PI3-K) by phosphorylation of AKT protein (
). Once activated, AKT is able to phosphorylate and inactivate many cell death effectors’ proteins (BAD, caspase 9) and increase the levels of anti-apoptotic proteins (Bcl-2 and Bcl-x). Here, we showed that feed restriction led to a 42% decrease in AKT levels (P < 0.05). Our assumption is that feed restriction led to a blocking of the anti-apoptotic pathway exerted by the AKT–PI3-K pathway induced by the decrease in plasma IGF-1 concentration. Among the transcripts tested, we found increases in BAD, PTEN, caspase 3, cathepsin B, and calpain 2 in mammary gland extracts from restricted-diet cows. The in situ detection of apoptotic cells confirmed this result.
Real-time quantitative PCR analysis of transcripts coding for α-LA and κ-CN, which are indicators of mammary gland activity, showed that their expression decreased with feed restriction. The infusion of free IGF-1 into the mammary arterial supply enhances milk secretion and mammary blood flow in intact goats (
). Thus, we suggest that the decrease in plasma IGF-1 concentration may also be partly responsible for the decrease in the expression of transcripts involved in milk secretion.
A nonlactating period is necessary between lactations for optimal milk production during the subsequent lactation. After milking stops, the mammary alveolar structures are maintained and massive apoptosis of epithelial tissue occurs. This phase of mammary involution is accompanied by strong degradation of various extracellular matrix (ECM) components (
). The degradation of the ECM is due to the action of metalloproteinases with gelatinase activity. Matrix metalloproteinases produced by MEC degrade collagen IV in the basal membrane and collagen I in the stroma (
). Here, histology analysis showed that the mammary glands from restricted-diet cows had closed acini surfaces with a disorganized structure. Complementary zymography analysis of the mammary tissue also revealed higher MMP activity (MMP-2 and MMP-9) in restricted-diet cows. Taken together, these results suggest that the feed restriction caused mammary gland involution with a decrease in milk production, an increase in cell death, an activation of MMP, and a regression of the mammary acini. Growth hormone is considered to play a key role on lactation persistency (
). In our study, GH and PRL concentrations were measured close to milking time. These could eventually explain high hormonal concentrations compared with those in the literature. As both groups were milked and blood collected similarly, GH and PRL level changes were due to the severe feed restriction. Prolactin appears to directly affect the mammary gland (
). The action of IGF-1 is itself controlled by a specific protein family, the IGFBP. Studies showed that during involution, IGFBP-5 production increased to inhibit the anti-apoptotic action of IGF-1 and triggered the initiation of MEC apoptosis. It was shown that the loss of MEC by apoptosis during involution was correlated with a decrease in PRL and IGF-1 concentrations (
, who showed that the decrease in circulating PRL involved a 20 to 25% loss in secretory cell population together with a decrease in milk production. Our results argue that PRL is closely associated with the control of mammary cell involution, but further studies are needed to investigate the link between feed restriction and PRL concentrations.
In conclusion, nutrient restriction resulted in a lower milk yield in lactating dairy cows. Our hypothesis is that the decrease in milk yield partly could be due to modulation of MEC activity and a lower number of mammary cells as a result of higher levels of apoptosis, but also to mammary remodeling by the extracellular matrix. Furthermore, a link was suggested between the GH–IGF-1 axis and MEC dynamics.
Acknowledgments
The authors thank the staff at the INRA's experimental farm in Le Pin-au-Haras (France), especially Yves Gallard, Jean-Luc Deust, Sébastien Blandamour, Samuel Boulant, Dominique Lainé, Bernard Camp, Laurent Vandenbrouck, Julien Gentil, Maxime Jacquet, and Denis Pichonat for their time spent sampling, observing the animals, and recording and organizing the data. We also thank Pierre-Yves Treguer (INRA) for his participation in the molecular analysis, and Luc Delaby (INRA) for the experimental design. The authors thank GALA association for financial support. The authors thank the translation company A.T.T (Clermont, France) for English corrections.
References
Accorsi P.A.
Pacioni B.
Pezzi C.
Forni M.
Flint D.J.
Seren E.
Role of prolactin, growth hormone and insulin-like growth factor 1 in mammary gland involution in the dairy cow.
Cell junction disruption after 36 h milk accumulation was associated with changes in mammary secretory tissue activity and dynamics in lactating dairy goats.
Clinical, haematological, metabolic and endocrine traits during the first three months of life of suckling Simmentaler calves held in a cow-calf operation.
Characterization of the proliferation state in canine mammary tumors by the standardized AgNOR method with postfixation and immunohistologic detection of Ki-67 and PCNA.
Functional differences in the growth hormone and insulin-like growth factor axis in cattle and pigs: Implications for post-partum nutrition and reproduction.
Evidence that pretranslational and translational defects decrease serum insulin-like growth factor-I concentrations during dietary protein restriction.
Effect of restricted feeding on the concentrations of growth hormone (GH), gonadotropins, and prolactin (PRL) in plasma, and on the amounts of messenger ribonucleic acid for GH, gonadotropin subunits, and PRL in the pituitary glands of adult ovariectomized ewes.
A comparison of proliferating cell nuclear antigen (PCNA) immunostaining, nucleolar organizer region (AgNOR) staining, and histological grading in gastrointestinal stromal tumours.
00474-7&id=gr1.jpg){kind=link}
00474-7&id=gr2.jpg){kind=link}
00474-7&id=gr3.jpg){kind=link}
00474-7&id=gr4.jpg){kind=link}
00474-7&id=gr5.jpg){kind=link}
00474-7&id=gr6.jpg){kind=link}