Liver samples derived from patients undergoing liver resection

Liver tissue from liver resections was obtained from Leipzig University Hospital, Department of Hepatobiliary Surgery and Visceral Transplantation. During liver resections, diseased tissue and the surrounding unaffected liver tissue were removed; a sample of the unaffected liver tissue was used for cell isolation. The patients provided written informed consent in accordance with the approved ethical protocols. Liver samples from 11 patients were included in this study. The donor characteristics are displayed in Table 1.

Cell isolation

PHH and hepM were isolated from the same liver tissue samples with a two-step collagenase perfusion technique described previously (Kegel et al. 2016; Damm et al. 2019). Briefly, liver tissue was perfused with a perfusion buffer containing EGTA (cat. no. 03777-10 g, Sigma-Aldrich, St. Louis, USA) followed by a digestion buffer containing collagenase P (cat. no. 11213873001, Roche, Basel, Switzerland). PHH were obtained by washing and centrifuging at 51 xg the generated cell suspension two times with phosphate-buffered saline (PBS) containing Mg2+/Ca2+ (cat. no. 14040174, Gibco, Thermo Fisher Scientific, Waltham, Massachusetts, USA). The PHH in the resulting pellet were resuspended and for PHH monoculture (see section Co-cultures and monocultures) seeded on cell culture plates coated with rat tail collagen I (produced in-house); the supernatant containing hepM and other non-parenchymal cells was centrifuged at 300 xg and then 600 xg. The resulting cell pellets containing hepM were combined. HepM were separated from the other non-parenchymal cells by adhesion separation. For this, the cells were seeded for 20 min on rat tail collagen I–coated cell culture plates. After 20 min, the hepM had attached to the cell cultures plate, and the non-adherent cells were removed resulting in hepM monoculture (see section Co-cultures and monocultures).

Cell cultureTHP-1 culture and differentiation into M0 macrophages

THP-1 cells were cultured in suspension in RPMI 1640 medium (Sigma-Aldrich, St. Louis, USA) containing fetal calf serum (FCS; cat. no. S0615-500 ml Sigma-Aldrich, St. Louis, USA), penicillin/streptomycin (cat. no. 15140122, Gibco, Thermo Fisher Scientific, Waltham, Massachusetts, USA) and L-Glutamine 200 mM (cat. no. 25030–024, Gibco, Thermo Fisher Scientific, Waltham, Massachusetts, USA) maintaining a density of 0.1–1 million cells/mL. The culture medium was refreshed every 2–3 days. Cells were collected by centrifugation at 300xg for 5 min, washed in PBS to remove debris, and resuspended in fresh medium. Cell viability and count were assessed prior to reseeding into new culture flasks.

For M0 macrophage differentiation, THP-1 cells were seeded on rat tail collagen I-coated cell culture plates in medium containing 10 ng/mL phorbol 12-myristate 13-acetate (PMA, P8139-1MG, Sigma-Aldrich, St. Louis, USA). After 24 h of incubation, adherent cells were washed twice with 400 µL PBS resulting in M0 monocultures.

Co-cultures and monocultures

For co-cultures PHH were seeded in a ratio of 4:1 on the respective macrophage monoculture and for the PHH monoculture the respective share of PHH were seeded in Williams E medium with GlutaMAX™ (cat. no. 32551087, Gibco, Thermo Fisher Scientific, Waltham, Massachusetts, USA) with fetal calf serum (FCS; cat. no. S0615-500 ml Sigma-Aldrich, St. Louis, USA), insulin (cat. no. PZN: 02526396, Lilly, Germany), dexamethasone (cat. no. PZN: 08704404, Jenapharm/MIBE GmbH, Germany), HEPES (cat. no. 15630–056, Gibco, Thermo Fisher Scientific, Waltham, Massachusetts, USA), sodium pyruvate (cat. no. 11360–039, Gibco, Thermo Fisher Scientific, Waltham, Massachusetts, USA), minimum essential medium non-essential amino acids (MEM-NEAA; cat. no. 11140–035, Gibco, Thermo Fisher Scientific, Waltham, Massachusetts, USA), and penicillin/streptomycin (cat. no. 15140122, Gibco, Thermo Fisher Scientific, Waltham, Massachusetts, USA). In parallel the medium of the macrophage monocultures were changed to the same medium mentioned before. The PHH were allowed to adhere overnight (12–16 h). Some analyses used supernatant directly after the isolation. The later was obtained 2 h after seeding (shortest sampling time point after cell adherence = 0 h of cultivation). The next day, the cells were washed two times with PBS containing Mg2+/Ca2+ to remove dead cells, FCS, insulin, and dexamethasone (sample = initial 12 h of cultivation). Thereafter, the cells were cultured in Williams E medium with GlutaMAX™ with HEPES, sodium pyruvate, MEM-NEAA, and penicillin/streptomycin, but without FCS, insulin, and dexamethasone (starvation medium) for another 24 h. After 24 h culturing in starving medium (sample = initial 36 h of cultivation), the initial characterization was performed (see Fig. 1). Then, the cultures were treated with a low or high dose of MEN (1 and 10 µM; cat. no. M5625-25G, Sigma-Aldrich, St. Louis, USA), DIC (0.5 and 5 mM; cat. no. D6899-10G, Sigma-Aldrich, St. Louis, USA), or APAP (0.5 and 5 Mm; cat. no. A5000-100G, Sigma-Aldrich, St. Louis, USA). MEN, DIC and APAP were dissolved in dimethyl sulfoxide (DMSO). The doses and incubation time were selected according to pretests (Fig. S1). After the treatment (3 h for MEN, and 6 h DIC and APAP), the DILI characterization was performed with the assays outlined in Fig. 1.

XTT assay

The Cell Proliferation Kit II (XTT) Kit (cat. no.11465015001, Roche, Switzerland) was used to measure metabolic activity according to the manufacturer’s instructions. Briefly, the culture medium was aspirated from the cells and the cells were washed once with PBS. XTT labeling solution was added to the cells and incubated for 2 h at 37°C and 5% CO2. The optical density (OD) was determined at 492 and 690 nm with a microplate reader (Synergy H1, BioTek Instruments, Inc., Winooski, VT, USA). For analysis, the OD reading at 690 nm was subtracted from the OD reading at 492 nm.

2′,7′-dichlorofluorescein (DCF) assay

To quantify intracellular ROS levels in the cultures 2′,7′-dichlorofluorescein (DCF) assay was performed. The assay uses 2′,7′-dichlorofluorescein diacetate (DCF-DA), a non-fluorescent compound that diffuses into cells. Intracellular esterases convert it into 2′,7′-dichlorodihydrofluorescein, which is then oxidized by ROS to form fluorescent 2′,7′-dichlorofluorescein (DCF). The fluorescence intensity, measured with a fluorometric plate reader, reflects the intracellular ROS levels.

Cells were washed with PBS and incubated with 20 µM DCF-DA (sc-209391, Santa Cruz Biotechnology, Dallas, USA) diluted in serum-free media for 30 min at 37°C, 5% CO2. DCF-DA was removed from the cells and replaced by serum-free media. Cells were incubated for 1 h at 37°C, 5% CO2. The supernatant was transferred to a 96-well plate and fluorescence intensity was measured with excitation at 480 nm and emission at 530 nm using a microplate reader.

Bicinchoninic acid (BCA) protein assay

The protein content of cultures was assessed for normalization purposes using the BCA protein assay. Briefly, cultures were washed twice with PBS containing Mg2+/Ca2+ and then lysed using a BCA lysis buffer containing 0.1% sodium dodecyl sulfate (SDS), 0.5% Triton X-100, and 50 mM Trizma base, followed by ultrasonication. Samples were incubated with a BCA reaction mixture (cat. no. B9643-1L (BCA reagent A) and cat. no. C2284-25ML-D (CuSO4 Solution 4%), Sigma-Aldrich, St. Louis, USA) and bovine serum albumin (BSA, cat. no. A4503, Sigma-Aldrich, St. Louis, USA) used to prepare a standard curve according to the manufacturer’s instructions. The samples were incubated for 30 min at 37°C in the dark and the absorbance at 550 nm was measured using a microplate reader.

ALAT, ASAT, GGT, and LDH activity assay

ALAT, ASAT, GGT and LDH are enzymes released upon cell membrane damage indicating cell death. The activity of these liver enzymes was assessed in the culture supernatants to determine occurrence of cell death and hepatotoxicity of substances. Culture supernatants were analyzed with commercially available reagents and kits (Diacon N [multi-color serum, normal], cat. no. D14481; Diacal Auto [multi calibration serum], cat. no. D98485; GGT, cat. no. D95604; LDH, cat.no. D95604; GOT [ASAT], cat. no. D95604; GPT [ALAT], cat. no. D95604, all from DIALAB, Wiener Neudorf, Austria) according to the manufacturer’s instructions. Absorbance was measured at 340 nm for ALAT, ASAT, and LDH, and at 405 nm for GGT with a microplate reader.

Cytokine array

Conditioned media from PHH, hepM, M0, CoC1, and CoC2 (derived from donors 10, 11, and 12) were pooled to determine the cytokine profile using the RayBio® Human Cytokine Array C5 (BioCat, Heidelberg, Germany) in triplicate. The arrays were performed according to the manufacturer’s instructions, with overnight incubation of the samples at 4 °C. Chemiluminescent signals were detected with a CCD camera (INTAS, Göttingen, Germany) and quantified with ImageJ v1.47 (National Institutes of Health, Bethesda, MD, USA). After background correction (blank values), all spots from one membrane were normalized to the respective positive controls. Targets with signal intensities below the standard error of the mean of the six positive controls (detection limit) were excluded from further analyses. Relative changes in the cytokine levels are presented as a heat map.

ELISA

ELISA were performed on cell culture supernatants of PHH, CoC1 and CoC2 cultures with and without DIC and APAP low dose treatment. The following kits were used: human anti-TNFα ELISA (cat. no. 900-TM2, PeproTech, Cranbury, USA), human anti-IL1b ELISA (cat. no. 900-M95, PeproTech, Cranbury, USA), human anti-IL6 ELISA (cat. no. 900-M16, PeproTech, Cranbury, USA), human anti-IL10 ELISA (cat. no. 900-M21, PeproTech, Cranbury, USA), human anti-CCL22 DuoSet ELISA (cat. no. DY336, R&D Systems, Minneapolis, USA), human anti-CCL23 DuoSet ELISA (cat. no. DY131, R&D Systems, Minneapolis, USA), human TGF-beta 1 DuoSet ELISA (cat. no. DY240, R&D Systems, Minneapolis, USA), and human anti IL-16 ELISA (cat. no. D1600, R&D Systems, Minneapolis, USA). The ELISAs were performed according to the manufacturers’ instructions. A second-order polynomial (quadratic) standard curve was used for analysis.

Immunofluorescence

CD68, CD86, and CD206 immunofluorescence was performed according to a standard protocol. Briefly, the cultures were fixed for 15 min at room temperature, followed by permeabilization with 0.2% Triton X-100 for 10 min. The cells were incubated with 2% BSA for 20 min, and FcR was blocked by an FcR blocking reagent (cat. no. 130–059–901, Miltenyi Biotec, Bergisch Gladbach, Germany) for 10 min at 4°C. The cells were incubated with the appropriate primary antibody overnight at 4°C: mouse anti-CD68 (cat. no. ab955, Abcam, Cambridge, United Kingdom), rabbit anti-CD86 (cat. no. ab239075, Abcam, Cambridge, United Kingdom), and goat anti-CD206 (cat.no. sc34577, Santa Cruz Biotechnology, Dallas, Texas, USA). Subsequently, the cells were incubated with the appropriate secondary antibody for 1 h at room temperature: donkey anti-mouse Dylight 550 (cat.no. ab98767, Abcam, Cambridge, United Kingdom), donkey anti-rabbit Alexa Fluor 488 (cat.no. ab181346, Abcam, Cambridge, United Kingdom), and donkey anti-goat Alexa Flour 647 (cat. no. 705-605-147, Jackson ImmunoResearch, Ely, United Kingdom). Nuclear counterstaining was performed by Hoechst 33,342 for 10 min at room temperature. Mowiol (cat.no. 0713, Carl Roth, Karlsruhe, Germany) generated according to manufacturer’s instructions was used for mounting. Images were taken with a Keyence microscope BZ-9000 (Keyence GmbH, Neu-Isenburg, Germany).

Western blotting

The protein concentrations of the samples were determined by using the BCA Protein Assay (Thermo Fisher Scientific, Waltham, Massachusetts, USA). 25 µg of each sample were separated by SDS–polyacrylamide gel electrophoresis (PAGE). The separated protein was transferred onto a 0.2 µm nitrocellulose membrane by electroblotting. The membrane was incubated with 5% milk in Tris-buffered saline containing 0.1% Tween 20 (TBS-T) for 60 min, prior to incubating with the appropriate primary antibody diluted in milk overnight at 4 °C: PARP1 (1:2000; cat. no. 9542, Cell Signaling, Danvers, Massachusetts, USA) and beta-actin (1:5000; cat. no. 4970, Cell Signaling, Danvers, Massachusetts, USA). The next day, the blots were washed three times in TBS-T, incubated with the appropriate horseradish peroxidase-conjugated secondary antibody (1:5000) for 60 min, and then washed three times with TBS-T. Finally, protein bands were visualized by chemiluminescence using the SuperSignal™ West Femto Maximum Sensitivity Substrate (Thermo Fisher Scientific, Waltham, Massachusetts, USA) and scanned with the ImageQuant LAS-4000 chemiluminescence detection system (GE Healthcare, Chicago, Illinois, USA). A second way to normalize the Western blot data was the Revert™ 700 Total Protein Stain (LI-COR Biosciences, Bad Homburg, Germany). It is a non-permanent staining procedure to measure the total amount of protein in the different samples. Briefly, after transfer of the proteins, the membrane was rinsed with water and incubated in Revert 700 Total Protein Stain. After one wash step with the Wash solution, the total protein stain was acquired by using the Odyssey Imaging System. The staining was removed by using the Revert Destaining Solution and the membrane was used for further Western blot analysis, as described above.

Graphical design and statistical analysis

Experiments were performed with three biological replicates (three donors per experiment) except for evaluating the drug treatment times and concentrations and Western blotting, which was only performed for two donors. The data in Figs. 2 and 7 are presented as the means of at least three biological replicates + standard deviations (SD). The data in Figs. 3 and 5 are presented as the mean of three technical replicates ± SD. Figure 4 presents data of three pooled donors. GraphPad Prism 7 software (GraphPad Software, San Diego, CA, USA) was used for statistical analyses and to generate the graphs. To compare multiple groups, one-way and two-way analysis of variance (ANOVA) were performed followed by the post hoc Tukey test. Statistical significance is denoted in the figures with asterisks: * p < 0.05, ** p < 0.01, *** p < 0.001, and **** p < 0.0001. Servier Medical Art was used to generate the experimental design scheme in Fig. 1.