Increased Expression of RUNX1 in Liver Correlates with NASH Activity Score in Patients with Non-Alcoholic Steatohepatitis (NASH)

Given the important role of angiogenesis in liver pathology, the current study investigated the role of Runt-related transcription factor 1 (RUNX1), a regulator of developmental angiogenesis, in the pathogenesis of non-alcoholic steatohepatitis (NASH). Quantitative RT-PCRs and a transcription factor analysis of angiogenesis-associated differentially expressed genes in liver tissues of healthy controls, patients with steatosis and NASH, indicated a potential role of RUNX1 in NASH. The gene expression of RUNX1 was correlated with histopathological attributes of patients. The protein expression of RUNX1 in liver was studied by immunohistochemistry. To explore the underlying mechanisms, in vitro studies using RUNX1 siRNA and overexpression plasmids were performed in endothelial cells (ECs). RUNX1 expression was significantly correlated with inflammation, fibrosis and NASH activity score in NASH patients. Its expression was conspicuous in liver non-parenchymal cells. In vitro, factors from steatotic hepatocytes and/or VEGF or TGF-β significantly induced the expression of RUNX1 in ECs. RUNX1 regulated the expression of angiogenic and adhesion molecules in ECs, including CCL2, PECAM1 and VCAM1, which was shown by silencing or over-expression of RUNX1. Furthermore, RUNX1 increased the angiogenic activity of ECs. This study reports that steatosis-induced RUNX1 augmented the expression of adhesion and angiogenic molecules and properties in ECs and may be involved in enhancing inflammation and disease severity in NASH.


Taqman quantitative Real time PCRs (qRT-PCRs)
Total RNA was isolated from a lager cohort of human liver tissues (normal liver, n = 33, steatosis, n = 46, NASH, n = 43; see also Table S1) using RNeasy kit including on column genomic DNA digestion with RNase-free DNase Set (Quiagen, Hilden, Germany). For quantitative RT-PCR analysis, we used the Fluidigm's BioMark high-throughput quantitative (q) PCR chip platform (Fluidigm Corporation, San Francisco, CA, USA) with pre-designed gene expression assays from ThermoFisher (suppl . Table S3A) according to the manufacturer's instructions [4]. The data were analyzed using the ΔΔCt method [5] and the expression values were normalized to the expression levels of the housekeeping genes (HPRT, YWHAZ) and shown in Table 1 and Table S6.

SYBR Green-based qRT-PCR Assays
Briefly, total RNA from trypsinized cells (in vitro culture experiments) was isolated with Trizol reagent (Invitrogen, Carlsbad, CA, USA) following the manufacturer's instructions. RNA quantified at 260/280 nm with ThermoScientific Nanodrop 2000 Spectrophotometer. The absorption ratio A260 nm/A280 nm between 1.90 and 2 was taken into consideration for cDNA preparation. First strand cDNA was synthesized from 1µg of total RNA with reverse transcriptase (Thermo Scientific Verso cDNA synthesis kit) according to manufacturer instructions. Quantitative real time PCR was carried using SYBR green PCR master mix (Applied Biosystems, Foster City, CA, USA) carried out in real time PCR machine. Dissociation curve was generated at the end of each PCR to verify that a single DNA species was amplified. The following cycling parameters were used: start at 95 °C for 5 min, denaturing at 95 °C for 30 s, annealing at 60 °C for 30 s, elongation at 72 °C for 30 s, and a final 5 min extra extension at the end of the reaction to ensure that all amplicons were completely extended and repeated for 40 amplification cycles. The data were analyzed using the ΔΔCt method [5] and the expression values were normalized to the expression levels of the housekeeping gene, 18S RNA.

Transcription factor Analysis
The angiogenic DEGs validated in RT-PCR studies were further dissected bio-informatically to study the transcription factors controlling them in all the subject groups (normal, steatosis and steatohepatitis) using DAVID 6.8 and JASPAR.

Immunohistochemistry studies
From all tissue samples, human liver tissue sections were fixed in 10% formalin and embedded in paraffin. They were cut with a sliding microtome into 2.5 µm sections for histology. Liver sections were deparaffinized and rehydrated. Thereafter, endogenous peroxidase was blocked for 20min with 3% hydrogen peroxidase in methanol. Antigen retrieval was done in Tris-EDTA buffer (pH: 9) for 6 min at boiling temperature. After washing under running water and giving a protein block with BSA for 1 h, they were incubated overnight at room temperature with anti RUNX1 primary antibodies (list of antibodies used given in Table S4). Thereafter, the specimen was incubated with the PolyExcel Target Binder for 10 min followed by a PolyExcel HRP labeled polymer using recommended 10 min incubation (PathNSitu Biotechnologies, Hyderabad, India). Staining was completed by a 5-10 min incubation with 3, 3'-diaminobenzidine (DAB) substratechromogen which resulted in a brown-colored precipitate at the antigen site. Counterstaining was done using hematoxylin. Immunohistochemical (IHC) scoring was done on a scale of 1-4 by counting RUNX1 positive brown cells per field. Four to five different fields were randomly selected for each sample and mean was taken. Internal negative antibody controls comprised of staining the NASH samples with RUNX1 antibody but without secondary antibody. These sections were counter-stained with hematoxylin.

Isolation of mouse primary hepatocytes and liver sinusoidal endothelial cells
Primary liver cells were isolated from control C57BL6 mice for in vitro studies. Primary hepatocytes (PH) and liver sinusoidal endothelial cells (LSECs) were isolated as described elsewhere with minor modifications [6]. Briefly, mice were intraperitoneally anesthetized with a combination of 100mg/kg ketamine and 5mg/kg midazolam. A mid-abdominal incision towards the sternum was made and the intestines were displaced to expose the portal vein. Liver was perfused (27G catheter) through the portal vein with pre-warmed Buffer 1 (HEPES with EGTA) for 10 min at a flow rate of 20 mL/min. Simultaneously the cava vein was cut to allow outflow of the solution. After perfusion, the liver was digested with pre-warmed Buffer 2 (HEPES with 0.015% collagenase) for 30min at a flow rate of 5mL/min. The resultant digested liver was excised, cut up and in vitro digestion was performed with Buffer 2. Disaggregated tissue was filtered using a 100µm nylon strainer and filtrate was centrifuged at 50× g for 5 min at 4 0 C. The pellet contained hepatocytes and supernatants contained the non-parenchymal cells. For isolation of LSECs, the supernatant was centrifuged at 800g for 10min at 4 0 C, and the obtained pellet was resuspended in PBS and centrifuged at 800g for 20min through a 25-50% Percoll gradient (Sigma-Aldrich, New Delhi, India) at room temperature. The interface of the gradient containing Kupffer cells and LSEC was precultured on 37-mm tissue culture plates for 25min at 37 0 C. The non-attached cells were enriched in LSECs.

Preparation of steatotic liver cells and BODIPY staining
For the preparation of steatotic hepatocytes/hepatoma cells, primary hepatocytes (PH) or Huh7 cells were grown in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum (Himedia Laboratories, Mumbai, India), penicillin (400U/ml), streptomycin (50µg/ml), L-glutamine (300µg/ml). Free fatty acid treatment of cells was done mainly according to previously published methods, where PH or Huh7 cells at 80% confluency were exposed to Palmitic acid-BSA (Himedia Laboratories, Mumbai, India) for 48 h [7]. Briefly, Palmitic acid (51.2mg) was dissolved in 100% ethanol (1ml) to make 200mM of stock solution. About 10% FFA low-endotoxin BSA was prepared in DMEM. 4mM palmitate-BSA conjugate in 10% BSA or 10% BSA was diluted 20× in complete medium containing 10% FBS to prepare 0.2mM or 200µM palmitate working treatment or BSA control medium. The total BSA concentrations in control and palmitate containing medium were kept the same to avoid differential protein binding effect on compounds. BSA-treated cells served as controls. After 48h, conditioned media (CM) was collected from BSA-treated and palmitate-BSA (PA) treated cells, sterile filtered (0.45 µm pore size membrane filter), and stored in aliquots at −20 °C until further use.
Fat deposition in PH and Huh7 cells were studied by BODIPY staining. Cells were washed twice with PBS before being fixed with 10% formaldehyde in PBS for 15min. After two washes in PBS, cells were then stained for 15-20min in BODIPY stain (1µg/ml, Thermofisher Scientific, Waltham, MA, USA). Representative micrographs were captured at 40× magnification using a fluorescence microscope.

Maintenance of endothelial cells in culture medium from palmitic acid treated Huh7 cells
HUVECs (Thermo Fisher Scientific, Waltham, MA, USA) or mouse LSECS were plated on 0.2% gelatin or collagen coated plates, respectively. Both endothelial cells were maintained in Endothelial Cell Basal Medium (Lonza, Mumbai, India) 10% FBS, 1% antibiotics (penicillin and streptomycin, Himedia Laboratories, Mumbai, India) and growth factors (endothelial cell growth medium supplement, Lonza, Mumbai, India). Cells were cultured at 37°C in a 5% CO2 humidified environment. HUVECs were used between passages 2 and 4. LSECs or HUVECs were treated with CM from BSA and PA-treated hepatocytes or Huh7 cells, respectively, for 24 h and then assayed for RUNX1 gene expression. HUVECs were also treated directly with BSA and 200µM PA-BSA for 24h and then studied for the gene expression of RUNX1.

Matrigel assays
For angiogenesis assays, serum-starved HUVECs (50,000 cells per well) were plated on 200µL of 5mg/mL polymerized matrigel onto 24 well plates, and then treated as described. In all conditions, after 18-20h, each well of HUVECs was analyzed under a phase contrast microscope with 4x objective [6]. Tubules in each field were imaged and branch points and tube length from 3-5 random fields in each well were counted.

ELISA assays for VEGF, PDGF-BB, TGF-β and CCL2
The levels of VEGF, PDGF-BB, TGF-β and CCL2 were analyzed in culture medium of untreated, BSA-treated and PA-treated Huh7 cells by ELISA using Thermofisher Scientific ELISA kits as per manufacturer's instructions (ThermoFisher Scientific, Waltham, MA, USA; Invitrogen, Carlsbad, CA, USA). Standard curve was drawn using standards provided in the kit and each analyte concentration was calculated from the standard curve. The assays were normalized to 10 6 cells per ml.