Tissue sources and preparation
Before the preparation of precision-cut bovine udder slices (PCBUS), udders of slaughtered cows (German Holstein Friesian) were examined for udder health by adspection and palpation immediately. The dissected udder was infused by heparinized tyrode solution right away to eliminate blood clots from the vessels  and instantly transported to the laboratory. In the laboratory, we performed the so-called California Mastitis Test (CMT) with the milk of the supposedly healthy udder. The CMT is used to determine the cell count in the milk, which is an additional indicator of udder health. A piece of approximately 10 x 10 x 5 cm was cut out with a sterile scalpel blade. A dermatome (Zimmer U.K. Ltd., Swindon, United Kingdom) was used to get tissue slices with a thickness of 250 µm. The obtained tissue slices were rapidly washed in sterile phosphate buffered saline (pH= 7.4; PBS) until a clear solution without milk contamination was achieved. Up to 200 PCBUS biopsy punches (6 mm diameter) were than taken from the slices, transferred to 24 well plates and washed with RPMI-1640 medium (Biochrom GmbH, Berlin, Germany) supplemented with 20 % fetal calf serum (FCS; Biochrom GmbH), 10 % penicillin streptomycin (10,000 international units (I.U.)/mL / 10,000 μg/mL, Biochrom GmbH), 10 % amphotericin B (PAA, Laboratories GmbH, Pasching, Austria) and 15 µg/mL gentamicin (PAA Laboratories GmbH) on an orbital shaker (130 rpm) in accordance to Magro et al. . Two additional washing steps were performed with RPMI-1640-medium supplemented with 10 % FCS, 1 % penicillin streptomycin, 1 % amphotericin B and 15 µg/mL gentamicin (maintenance medium). The slices were put into a new 24 well plate (Greiner Bio-One, Frickenhausen, Germany) and incubated in 1 mL of maintenance medium in a humidified atmosphere containing 5 % CO2 at 37 °C. Incubation was performed for 13 days totally and the medium was changed every 48 h.
Viability of PCBUS
In first experiments with four udders, viability of the PCBUS was tested from time point 0 h (6 h after slaughtering) until day 13, whereby the medium was changed every 48 h. For this purpose, the MTT assay was performed daily with 7 PCBUS of the respective udder. The MTT assay as described by Lambermont et al.  was used with some modifications. The PCBUS were transferred into a 24-well plate with 900 µL RPMI-1640-medium and 100 µL 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide solution (MTT, 7 mg/mL; Sigma-Aldrich, Steinheim, Germany). After 15 min of incubation, the supernatant was discarded and 200 µL of an intermixture (5 % formic acid + 95 % propanol; Sigma-Aldrich) was added. After a further 40 min, 100 µL supernatant were transferred to a 96-well plate (Greiner Bio-One) for extinction measurement at 570 nm (MRX-reader, Dynatech, Denkendorf, Germany). As negative control, PCBUS were digested with Triton X-100 (Sigma-Aldrich; 1 mL, 30 min, 37°C, 5 % CO2 atmosphere) 30 min before starting the MTT assay. All steps were carried out at room temperature in the dark. The main trials (stimulation trials) were performed over a period of 5 days. Therefore, 5 PCBUS of each of the 6 udders used in the trials were checked daily by the MTT assay for their viability until the end of the experiments.
Stimulation of PCBUS with LPS and PGN
One day before stimulation, PCBUS were incubated in RPMI-1640-medium, supplemented with 10 % FCS, but without any antibiotics in a humidified atmosphere containing 5 % CO2 and 37 °C. To simulate a pathophysiological reaction initiated by bacterial toxins during early phase of infection, 1 µg/mL lipopolysaccharide (LPS; Sigma-Aldrich) of E.coli O55:B5, 1 µg/mL peptidoglycan (PGN, Sigma-Aldrich) and a combination of LPS and PGN were added to the incubation medium. As a control, unstimulated PCBUS were used. These PCBUS were incubated with RPMI-1640-medium supplemented with 10 % FCS. 1 h and 2 h post stimulation, 30 µL supernatant and after 4 h, 6 h and 24 h, 130 µL supernatant were sampled. After removal of the samples the wells were refilled with fresh medium (RPMI-1640-medium + 10 % FCS; without any stimulants). The experiments were performed in triplicates on 6 different udders.
Tissue slices sampled the same time points and were used for histological examination from other wells.
Measurement of IL-1ß, TNF-α and PGE2
Concentrations of IL-1ß, TNF-α and PGE2 in supernatants were measured with ELISA kits according to the manufacture´s specification (bovine Interleukin-1ß Reagent Kit Invitrogen, Thermo Fisher Scientific; Prostaglandin E2 Express ELISA Kit Cayman Chemical Company, Ann Arbor, USA; Bovine TNF-alpha DuoSet® ELISA DuoSet® ELISA Development Systems, R&D Systems, Minneapolis, USA). Dilutions of the samples were determined in preliminary tests. For the determination of the IL-1ß- as well as PGE2-concentrations, the samples were diluted 1:10 according to the manufacturer's instructions. For the determination of the TNF-α content, the samples were applied undiluted.
For histological examination, PCBUS were fixed in 10 % buffered formalin for 48 h and embedded in paraffin wax using standard techniques. 3 µm slices were stained with hematoxylin eosin staining (HE staining). All sections were examined by light microscopy. The thickness of PCBUS were measured by using computer-aided software (AxioVision Software).
The results are presented as mean ± standard deviation (SD). The statistical evaluation was performed with the program GraphPad Prism (Version 8.0.1, GraphPad Software, Inc.). In the experiments, the results of the control groups at the respective measurement time were compared with those of the stimulated group with regard to significant differences using the Kruskal-Wallis test. The significance of error probability of 5 % was considered significant, with significances of ≤ 0.01 and ≤ 0.001 being particularly marked in the figures.