Isolation and purification of bSCs and bADSCs
bSCs were isolated from 160 g of fresh masseter muscle samples (within 6 h of euthanasia) of 27-month-old Japanese black cattle obtained at the slaughterhouse (Tokyo Shibaura Zouki, Tokyo, Japan, and JA ZEN-NOH Kanagawa, Kanagawa, Japan). The freshly harvested bovine muscle was kept on ice, transferred to a clean bench, and washed with cold 70% ethanol for 1 min, followed by cold PBS 1 x 2 times. Then, the fat tissue part was disposed of, the remaining muscle was cut into small pieces with a knife, and minced with a food processor mechanically. The bovine minced muscle was washed with cold PBS 1x with 1% penicillin–streptomycin (PS) (Lonza, 17-745E) for 1 min. The washed muscle was transferred to a bottle and mixed with 160 ml of 0.2% collagenase II (Worthington, CLS-2) in DMEM (Invitrogen, 41966-29) supplemented with 1% PS. The bottle was incubated and shaken every 10 min for 1.5 h at 37 °C. After digestion, 160 mL of 20% FBS in DMEM supplemented with 1% PS was added and mixed well. The mixed solution was centrifuged for 3 min at 80 g and 4 °C. Floating tissues in the supernatant after centrifugation were removed by tweezers and then the collected supernatant was kept on ice as a mononuclear cell suspension. Precipitated debris were mixed with 80 ml of cold 1% PS in PBS 1x and centrifuged for 3 min at 80 g and 4 °C. The supernatant was collected again and mixed with the former mononuclear cell suspension. After that, the cells were filtered through a 100 μm cell strainer. After centrifugation for 5 min at 1500 g and 4 °C, the cells were suspended with 160 ml of cold DMEM with 20% FBS and 1% PS, were filtered through a 100-μm cell strainer followed by a 40-μm cell strainer. The cells were then centrifuged for 5 min at 1500 g and 4 °C. Precipitated cells were incubated with 8 ml of erythrocyte lysis buffer (ACK, 786-650) for 5 min on ice. Then the cells were washed twice with cold PBS 1x supplemented with 1% PS and the cell pellet was mixed with FBS supplement with 10% dimethyl sulfoxide and then reserved at −150 °C. The frozen cells were recovered in a 37 °C water bath and washed with cold PBS 1x twice. The cells were suspended in F10 medium (Gibco, 31550-023) containing 20% FBS, 5 ng/mL bFGF (R&D, 233-FB-025), and 1% PS supplemented with 10 μM p38i (Selleck, S1076), and then seeded at 1.1 × 105 cells/well in 6-well cell culture plates (Corning) that were coated with 0.05 % bovine collagen type I (Sigma, C4243). The cells were cultured by changing the medium every two or three days and were passaged when they reach 60 % of confluency until passage 3 for cell sorting by flow cytometry. The cells were suspended in FACS buffer (1% BSA in PBS1x) and stained with APC anti-human CD29 Antibody (BioLegend, 303008, TS2/16, dilution 1/40), PE-CyTM7 anti-human CD56 (BD, 335826, NCAM16.2, dilution 1/40), FITC anti-sheep CD31 (BIO-RAD, MCA1097F, CO.3E1D4, dilution 1/40), and FITC anti-sheep CD45 (BIO-RAD, MCA2220F, 1.11.32, dilution 1/40) for 30–45 min on ice in the dark. After antibody incubation, the cells were washed twice with cold PBS 1x and reconstituted in PBS 1x with 2% FBS. The CD31−CD45−CD56+CD29+ cells were isolated by Sony Cell Sorter SH800S as bSCs.
Subcutaneous bovine adipose tissues were isolated at the slaughter house on the day of slaughter and sent to the laboratory with an ice pack. Following the 24-h duration delivery, the tissues were first washed in PBS 1x containing 5% of penicillin-streptomycin-amphotericin B (Wako, 161-23181). Then, 8–10 g of tissue was separated into fragments to fill the six wells of a 6-well plate and was minced to get around 1mm3 in size using autoclaved scissors and tweezers, directly in 2 mL of collagenase solution containing both collagenase type I (Sigma Aldrich, C0130) and type II (Sigma Aldrich, C6885) at 2 mg/mL for each in DMEM with 1% antibiotics–antimycotic mixed solution (Nacalai, 02892-54), 0% FBS, and 5% BSA (sterilized by 0.2 µm filtration). After one hour of incubation at 37 °C with 250-rpm agitation, DMEM (Nacalai, 08458-16) with 10% FBS and 1% antibiotics was added and the lysate was filtrated using a sterilized 500 µm iron-mesh filter, before being centrifuged for 3 min at 80 g. The upper human mature adipocyte layer was then removed and the pellet containing the stromal vascular fraction (SVF) was washed two times in PBS 1x with 5% BSA and 1% antibiotics and once in complete DMEM (10% FBS + 1% antibiotics), by 3 min of centrifugation at 80 g between each wash. Finally, the pellet containing the SVF cells was resuspended in DMEM and seeded in a 10-cm dish for expansion by changing the medium every day for three days, and the cells were passaged when they reached 80% of confluency. After P1, the remaining adherent cells are considered as bADSCs and were expanded in DMEM.
2D cell culture
bSCs were cultured in high-glucose DMEM (Gibco, 10569-010) containing 1% antibiotic–antimycotic mixed solution (Nacalai, 02892-54), 10% FBS, 4 ng/mL fibroblast growth factor (Fujifilm, 067-04031), and 10 uM p38 inhibitor (Selleck, SB203580) for proliferation, and bSCs within P8 were used for all experiments. For differentiation induction of bSCs, the medium was changed to DMEM containing 2% horse serum and 1% antibiotic–antimycotic mixed solution. To investigate the cell proliferation and differentiation ratio, 50,000 cells were seeded on 24-well plates, and the differentiation medium was replaced every two days after differentiation induction.
Culture and differentiation of bADSCs to bovine endothelial cells
To monitor the endothelial differentiation of bADSCs, bADSCs at P1 previously expanded in DMEM were seeded at 5000 cells on 48-well plates and kept seven days in eight different conditions (DMEM with 1, 5, or 10% HS, DMEM with 10% FBS, DMEM with 10% human serum, DMEM with 10% calf serum, F12K with 10% HS, and F12K with 10% FBS, all containing 1% antibiotics-antimycotic mixed solution) by renewing the medium every 2-3 days.
Collagen microfiber preparation
The collagen microfibers (CMF) were first prepared from a collagen type I sponge (Nipponham) after dehydration condensation at 200 °C for 24 h crosslinking to prevent its dissolution in water-based solution. The crosslinked collagen sponge was mixed with ultra-pure water at a concentration of 10 mg/mL (pH = 7.4, 25 °C) and homogenized for 6 min at 30,000 rpm. Then, the solution was ultrasonicated (Ultrasonic processor VC50, SONICS) in an ice bath for 100 cycles (one cycle comprised 20 s of ultrasonication and 10 s of cooling), to make smaller fragments that are able to induce better vascularization and filtrated (40-µm filter, microsyringe 25-mm filter holder, Merck), before being freeze-dried for 48 h (FDU-2200, EYELA)50. The obtained CMF was kept in a desiccator at RT.
3D gel-embedded culture
To construct the adipose tissues by 3D culture, CMF were first weighted and washed in DMEM without FBS by being centrifuged for 1 min at 16,083 g to get a final concentration in the tissues of 1.2 wt%. The bADSCs were added after trypsin detachment (always used at P1–5) and centrifuged for 1 min at 1970 g to get a final cell concentration of 5 × 106 cells/mL. The pellet containing CMF and bADSCs was then mixed in a fibrinogen (Sigma Aldrich, F8630) solution at 6 mg/mL final concentration (the stock solution at 50 mg/mL prepared in DMEM with 1% antibiotics) and the thrombin solution (Sigma Aldrich, T4648) was added to get a final concentration of 3 U/mL (the stock solution at 200 U/mL prepared in DMEM with 10% FBS and 1% antibiotics). Finally, 2 µL drop tissues were seeded in a 96 well plate (Iwaki, 3860-096) and gelated for 15 min in the incubator at 37 °C. Then 300 µL of medium (DMEM with 10% FBS and 1% antibiotics) was added to the drop tissues. For adipogenic differentiation, three days of proliferation were first necessary to allow the bADSC proliferation, until reaching a suitable cell–cell interaction required for the adipogenesis51. The medium was then switched for DMEM with 10% FBS containing different adipogenic components to compare: Rosiglitazone (at 20 µM final concentration, Sigma Aldrich, R2408), Insulin (at 10 µg/mL final concentration, Sigma Aldrich, I6634), Troglitazone (at 40 µM final concentration, Sigma Aldrich, T2573), Pristanic acid (at 50 µM final concentration, Funakoshi, 11-1500), Phytanic acid (at 50 µM final concentration, Sigma Aldrich, P4060), Erucic acid (at 50 µM final concentration, Sigma Aldrich, 45629-F), Elaidic acid (at 50 µM final concentration, Sigma Aldrich, 45089), Oleic acid (at 50 µM final concentration, Sigma Aldrich, O1383), Palmitoleic acid (at 50 µM final concentration, Sigma Aldrich, 76169), Myristoleic acid (at 50 µM final concentration, Sigma Aldrich, 41788), TGF type I receptor activin-like kinase 5 inhibitor (ALK5i II, 2-[3-(6-methyl-2-pyridinyl)-1H-pyrazol-4-yl]-1,5-naphthyridine, at 1–10 µM final concentration, Cayman, 14794), or Bovine Endothelial Cell Growth Medium (Cell Applications Inc., B211K-500). The individual seven free fatty acids were compared as well as some other possible different mixtures (pristanic and phytanic acids together, or the other five remaining free fatty acids) following already published possible adipogenesis inducers for bovine ADSCs29,30. The 300 µL of differentiation medium was then renewed every 2-3 days.
Supporting bath-assisted 3D printing (SBP) and culture
G-GG was produced by preparing a 1 wt% gellan gum (Sansho) in PBS 1x, grinding it with a rotor–stator homogenizer for 6 min at 30,000 rpm, centrifuging at 2837 g for 3 min, and removing the supernatant. G-Gel was produced by preparing 4.5 wt% porcine gelatin (Sigma Aldrich, G1890) in DMEM containing 1% antibiotic–antimycotic mixed solution and 10% FBS, putting it at 4 °C overnight for gelation, adding the same volume of DMEM to the gelatin gel, grinding it with a rotor–stator homogenizer for 2 min at 30,000 rpm, centrifuging at 2837 g for 3 min, and removing the supernatant. Gellan gum and gelatin were conjugated with fluorescein to measure particle size. After the fabrication of G-GG and G-Gel with fluorescein-conjugated material, the fluorescence images were obtained. Major and minor lengths of 60 particles were measured in G-GG and G-Gel and the particle size was determined by averaging major and minor lengths. The bioink was prepared to be 5 × 107 cells/mL of bSCs in the mixture composed of 20 mg/mL fibrinogen (Sigma Aldrich, F8630) in DMEM and Matrigel (Corning, 356234) (6:4, v/v). The supporting bath was prepared by mixing G-GG or G-Gel with 10 U/mL thrombin (Sigma Aldrich, T4648) before printing. After filling the prepared supporting bath in a glass vial and loading the syringe containing the prepared bioink onto the dispenser instrument (Musashi, Shotmaster 200DS), cell printing was conducted inside the supporting bath maintaining the syringe and bed parts of the instrument at 4 °C. All parts, such as syringes, nozzles, and containers used for cell printing, were sterilized with 70% ethanol and UV treatment. The nozzle gauge, moving speed, and dispensing speed was 16 G, 1 mm/s, and 2 µL/s, respectively. The printed structures inside the supporting baths were incubated inside a sterile cabinet at RT for 1 h to ensure gelation. After gelation, the G-GG was gently removed by pipetting and printed structures were immersed in 50 mM Tris-HCl buffer (pH 7.4) at 37 °C for 30 min, and the same process was repeated one more time. G-Gel was dissolved by incubation at 37 °C for 2 h. The obtained cell fibers after removal of the supporting baths were cultured in the basic medium of bSCs for 2–3 days and then replaced with the differentiation medium. Suspension culture was simply conducted by placing the printed cell fiber on a tissue culture plate, and needle-fixed culture was conducted by fixing both ends of the printed cell fibers onto the silicone rubber with a size of 2 cm × 2 cm × 3 mm (W × D × H) placed on a six-well plate.
PDMS well fabrication
The parts of PLA molds were fabricated by the FDM 3D printer (Creality, Ender-3) after modeling by Fusion360 and slicing by Cura for the PDMS wells. PDMS (Corning, Sylgard 184) was poured into the assembled PLA mold and cured at 50 °C overnight, the PDMS wells were obtained by removal of the PLA molds.
TIP and culture
In all, 4 wt% CNFs were produced from collagen sponge (Nipponham, Type I & III mixture) based on the previous method. After cutting a small area of the PDMS wells’ side to make the media flow channel, it was sterilized by 70% ethanol and UV treatments and was then put on a slide glass. The PDMS well was filled with 4 wt% CNFs, G-Gel, and 4 wt% CNFs at the bottom, middle, and top layers, respectively. Cell printing was conducted in the same way as the SBP. After printing cells, the printed area at the top layer of PDMS well was covered one more time with 4 wt% CNFs, incubated at RT for 1 h, then it was incubated at 37 °C for 2 h to dissolve the G-Gel and induce the CNFs gelation, and finally was placed in a culture container. The bioinks were prepared in the same way as in SBP for muscle cell fiber, by mixing 5 × 106 cells/mL of bADSCs in 1.2 wt% CMF and 20 mg/mL fibrinogen solution for fat fibers and 107 cells/mL of endothelial differentiated bADSCs in 1.2 wt% CMF and 20 mg/mL fibrinogen solution for vascular fibers in DMEM.
Live/dead staining
Printed cell fibers were stained with 2 µM calcein-AM and 4 µM ethidium homodimer-1 (Invitrogen, L3224) in DMEM at 37 °C for 15 min, followed by rinsing with PBS 1x and fluorescence imaging by confocal microscopy (Olympus, FV3000) with 10x or 30x objective lens. The imaging conditions are as below:
Channel 1 (Live cell); 488 nm laser (power 0.1 ~ 2%), Ex: 500 ~ 540 nm, HV:450 ~ 550, gain 1
Channel 2 (Dead cell); 561 nm laser (power 0.1 ~ 6%), Ex: 610 ~ 710 nm, HV:450 ~ 550, gain 1
Channel 3 (Nucleus); 405 nm laser (power 2 ~ 5%), Ex: 430 ~ 470 nm, HV:450 ~ 550, gain 1
To measure the cell viability, at least six images were randomly taken. The image-based cell viability (%) was calculated by dividing the number of non-dead-stained nucleus by the total number of the nucleus in each image and averaging it. The nucleus counting was conducted by using a particle-analysis plugin in ImageJ.
Histological staining
Tissues were washed once in PBS 1x and then fixed in 4% paraformaldehyde (Wako, 163-20145) overnight at 4 °C, followed by three-time washes in PBS 1x. The samples were then maintained in PBS 1x solution before being mounted in paraffin-embedded blocks. Paraffin-embedded blocks and sections were prepared and hematoxylin/eosin (H/E) stained by the Applied Medical Research Laboratory, Inc. Some pieces of commercial (Wagyu) beef steak were immersed in 4% paraformaldehyde solution overnight at 4 °C, and then in 1/15 M phosphate buffer (pH 7.4) containing 30% sucrose. They were rapidly frozen in dry ice acetone and cut into 20 µm thick sections. The tissue sections were processed for immunostaining for sarcomeric alpha-actinin and laminin to depict myotubes of skeletal muscles and basement membranes of the muscles and adipose tissues. The immunostaining was performed by use of specific antibodies against sarcomeric alpha actinin (abcam, ab9465, EA-53, dilution 1/1000) and laminin (Sigma-Aldrich, L9393, polyclonal, dilution 1/100), and the reaction products were visualized in blue (Vector Laboratories, Vector Blue) and brown (DAKO, DAB+ chromogen), respectively.
Immunostaining
Immunostaining was conducted by a general process, 4% paraformaldehyde fixation at RT for 15 min or at 4 °C overnight, permeabilization with 0.2% Triton X-100 (Sigma Aldrich, T8787) at RT for 15 min, blocking with 1% BSA (Sigma-Aldrich, A3294) for 30–60 min, incubation with the 1st antibody in 1% BSA at RT for 2 h or at 4 °C overnight, incubation with cocktails containing fluorophore-conjugated 2nd antibody and 1 µg/mL TRITC-phalloidin (Sigma Aldrich, P-1951) for actin staining or 100 ng/mL NileRed (TCI, N0659) for lipid staining at RT for 1 h, and finally 300 nM DAPI (Invitrogen, D21490) counterstaining. Myosin 4 Monoclonal Antibody (eBioscience, 14-6503-82, MF20, dilution 1/500) for bSCs, Anti-CD31 (Wako, M0823, JC70A, dilution 1/100) for bovine endothelial cells, and PPAR gamma (Abcam, ab45036, polyclonal, dilution 1/100) and FABP4 (LSBio, LS–B4227, polyclonal, dilution 1/100) antibodies for bovine adipocytes were used as 1st antibodies. Goat anti-Mouse IgG (H+L) Cross-Adsorbed Secondary Antibody, Alexa Fluor 488 (Invitrogen, A-11001, polyclonal, dilution 1/200) and goat anti-Mouse IgG (H+L) Cross-Adsorbed Secondary Antibody, Alexa Fluor 647 (Invitrogen, A-21235, polyclonal, dilution 1/200) were used as 2nd antibodies. All fluorescence images were taken by confocal microscopy. In the case of printed cell fibers, they were treated with Rapiclear 1.52 (SUNJin Lab) at RT for 30 min for deep-tissue imaging.
Rheology measurement
Viscosity was measured by the controlled-rate mode of the rheometer (Thermo Scientific, HAAKE RheoStress 6000) to verify the thixotropy of G-Gel and G-GG. The steps were composed at 0.01/s for 30 s, at 30/s for 100 s, and at 0.01/s for 30 s.
UV–Vis
The disposable cuvette was filled with 1 mL of 4 wt% CNFs, and then transmittance was measured at the temperature-controlled steps by UV–Vis spectrometer (Jasco, V-670) at 4 °C for 2000 s, at 37 °C for 2000 s, at 4 °C for 2000 s, and then at 37 °C for 2000 s. After each temperature change, the photos of the samples were taken.
Mechanical test
The elastic modulus of these printed fibers was measured with EZ test (SHIMADZU, EZ/CE 500 N). All fibers, including printed cell fiber and fibrous tissues from commercial meat, were fixed by 4% paraformaldehyde and washed several times before measurement. After preparing the different printed fibers, several fibers were stacked on a 24-well insert for the sample surface to become larger than the geometry’s surface area at RT. Spherical mold (5 mm in diameter) was used to measure the elastic modulus at a head-moving speed of 1.0 mm/min. The compressive test protocol was employed to increase the engineering strain, until the testing stress to 200 mN. The modulus is automatically calculated by EZ test in the elastic range (10–20 mN). The total sample size was n = 3 for each fiber type.
Water-content measurement
The water content is calculated according to the mass before and after freeze-drying. Briefly, the printed fibers in PBS1x were taken out, the surface liquid was removed, and the wet weight (Wwet) of the fibers was measured by a balance. The dry weight (Wdry) of the fibers was measured after freeze-drying (24 h). The water content is given by the following formula (1):
$${V}_{rm {water}}=frac{{{{{{{rm{W}}}}}}}_{rm {wet}}-{{{{{{rm{W}}}}}}}_{rm {dry}}}{{{{{{{rm{W}}}}}}}_{rm {wet}}}times 100$$
(1)
DNA content measurement
Commercial beef was bought from the supermarket and intramuscular fat tissues as wells as muscle tissue parts were cut into small fibers of the same size as the printed fibers (Fig. S19). One fiber from the commercial fibers or printed fibers was put per microcentrifuge tube and the tissues were lysed following the DNeasy Blood & Tissue Kit (QIAGEN, 69504) to extract their DNA content, which was quantified by the NanodropTM N1000 device (Thermo Fisher Scientific). Then the DNA amount was presented normalized by the weight of the samples before lysis.
Bovine-cell fiber assembly
Each cell fiber (muscle, fat, and vascular fibers) was stacked on a plastic container one by one according to the model image obtained from the commercial meat’s histological image. After dispensing transglutaminase solution (10 U/mL, Ajinomoto) onto the stacked cell fibers, they were wrapped up and kept at 4 °C for two days. To take a cross-sectional picture, we cut it and pill off the plastic wrap.
Image processing and analysis
For the evaluation of bSCs differentiation on 2D culture, at least 4 fluorescence images of the samples stained with DAPI and MF20 were taken in size of about 6 mm × 6 mm, and the number of nuclei in all areas and MHC-positive area was measured by ImageJ. For the measurement of cell alignment degree, we randomly selected the fluorescence images (actin stained with TRITC-phalloidin) after Z-stack imaging of one printed tissue, chose three 2D images showing clear-cell morphologies, and measured the angle of individual cells to printed cell fiber’s major axis by using ImageJ. The number of measured cells in each condition was in the range of 27–56. For the calculation of lipid production from bADSCs, the 3D tissues’ Z-stack images were taken (3 slices with 50 µm step) with the same exposure time, brightness, and contrast, then the summation of Z-slice’s intensity in Nile Red and Hoechst of each 3D tissue was done by Image J. The relative intensity was calculated from the total Nile Red’s intensity divided by the Hoechst intensity in each 3D tissue.
The 4x lens (0.16 dry, WD 13.0) of the confocal quantitative image cytometer CQ1 (Yokogawa, Tokyo, Japan). was used, with the following parameters:
Channel 1 (Nile Red); 561-nm laser (power 20%), Ex: 617~673 nm, Exposure time 500 ms, Bin 1, Gain 16-bit, low noise, and high well capacity, contrast enhancement maintained constant at 100–3200. Channel 2 (Hoechst); 405-nm laser (power 100%), Ex: 447~460 nm, Exposure time 500 ms, Bin 1, Gain 16-bit, low noise, and high well capacity, contrast enhancement maintained constant at 320–1300.
For the differentiation from bADSCs to endothelial cells, the same method was applied with 2D images taken by the CQ1 confocal, using CD31’s total fluorescence intensity normalized by Hoechst total fluorescence-intensity ratios, at the same parameters for all conditions compared. The 3D-reconstructed image and the printed cell fibers’ 3D movie was obtained by Imaris software (Bitplane).
Gene expression
Gene expression was analyzed using real-time quantitative polymerase chain reaction (RT-qPCR). Adipose drop tissues at days 0, 7, and 14 of differentiation, as well as bioprinted fiber samples at days (before bioprinting) and at days 6 (myoblast fibers), 7 (endothelial fibers), or 14 (adipocyte fibers) were washed in PBS and total RNA extraction was carried out using the PureLink RNA Micro Kit (Invitrogen, Waltham, USA), with the DNAse step, following the manufacturer’s instructions. Samples’ RNA content was quantified with the NanodropTM spectrometer (N1000, Thermo Fisher Scientific, Waltham, USA). For RT-qPCR, the RNA samples were first submitted to reverse transcription into cDNA using iSCRIPT cDNA synthesis kit (Bio-Rad, Hercules, USA), before being amplified using Taqman probes and reagents (Taqman Fast Advanced Mix, Taqman gene expression assays (FAM): MYH2 (Assay ID: Bt03223147_gH), FABP4 (Assay ID: Bt03213820_m1), CD31/PECAM1 (Assay ID: Bt03215106_m1), PPARG (Assay ID: Bt03217547_m1), and PPIA (Assay ID: Bt03224615_g1), Thermo Fisher Scientific, Waltham, USA). The cDNA synthesis and RT-qPCR reactions were conducted using the StepOnePlus Real-Time PCR System (Thermo Fisher Scientific, Waltham, USA) and the gene expression was normalized by PPIA as the housekeeping gene.
Western blot
Proteins relative expression was assessed by performing western blot. Myoblast (days 0 and 6), endothelial (days 0 and 7), and adipocyte (days 0 and 14 of differentiation) fibers were washed with PBS and homogenized by pipetting in lysis buffer (RIPA Buffer R0278, Sigma-Aldrich with Protease Inhibitor Cocktail 25955-24, Nacalai-Tesque), before being quantified by a BCA Protein Assay Kit (23225, ThermoFisher Scientific). About 4–10 μg of protein was resolved on each lane of 4–15% protein gels (Mini-PROTEAN® TGX Stain-Free™ 4568084, BIORAD with Running Buffer for SDS-Tris-Glycine (10X), Cosmobio), electrotransferred onto a PVDF membrane (Immun-Blot® 1620-0176 BIORAD), and probed using specific antibodies: Myosin 4 Monoclonal antibody (eBioscience, 14-6503-82, dilution 1/1000), PPAR gamma antibody (Abcam, ab45036, polyclonal, dilution 1/1000), FABP4 antibody (LSBio, LS–B4227, polyclonal, dilution 1/1000), and β-Actin antibody (Sigma-Aldrich, A5441, AC-15, dilution 1/3000). Proteins were detected by secondary antibodies conjugated to horseradish peroxidase (Anti-Mouse IgG 170-6516, anti-Rabbit IgG 170-6515, dilution 1/5000, and Precision ProteinTM StrepTactin-HRP Conjugate 161-0380, BIORAD, dilution 1/5000). Signal detection was performed by an enhanced chemiluminescence-detection reagent (Clarity Western ECL Substrate 1705060, BIORAD) using the ChemiDoc Imaging System (BIORAD). Molecular weights were determined by comparison with the migration of prestained protein standards (Precision Plus ProteinTM KaleidoscopeTM Standards 161-0395, BIORAD). Quantitative estimation of the bands’ intensity was performed using ImageJ software.
Statistical analysis
Statistical analysis was performed using EzAnova (version 0.98, University of South Carolina, Columbia, SC, USA) and GraphPad Prism 8 (version 8.4.3 (686), San Diego, USA) software. The detail of the number of n corresponding to the number of independent experiments using isolated bovine primary cells from different bovine donors, or independent samples, is displayed on each graph in the figures and in the captions, as well as the exact p values. For paired samples when the same cells were measured at different times (Fig. 2f and h), a one-way ANOVA with a repeated measures design was used, with the Greenhouse–Geisser correction and the Tukey’s HSD post test for the multiple comparisons. When they were subjected to different treatments (Fig. 2e, I, 4 l, Supplementary Fig. 3, and Supplementary Fig. 15) a classic one-way ANOVA was used, with the Brown-Forsythe and the Bartlett’s tests as well as the Tukey’s HSD post-test for the multiple comparisons. When time and treatments were both involved (Fig. 2d), a two-way ANOVA was applied with time set as “paired or repeated measured” and the treatment as classic analysis or “unpaired”, which led to a pairwise comparison, with the Greenhouse–Geisser and Huynh–Feldt corrections as well as the Tukey’s HSD post test for the multiple comparisons. When two different treatments were applied to the cells (Fig. 2j), a classic two-ways ANOVA was performed, with the Šidák post test for the multiple comparisons. ANOVA was used when more than two conditions were compared with each other and for only two conditions, the pairwise t-test comparison was performed (Fig. 4f, i, j, m, Supplementary Fig. 17, and Supplementary Fig. 21). EzAnova software was used only for Fig. 2d analysis, for the other statistical analysis, GraphPad Prism software was used. Error bars represent SD. p values were considered significantly different at least when p < 0.05. When no marks are shown on the graphs, it means that the differences are not significant.
Reporting summary
Further information on research design is available in the Nature Research Reporting Summary linked to this article.
