Study subjects
Whole-exome sequencing on lymphocytes and further blood analyses were performed in the index patient and his healthy parents. In addition, fibroblast analysis of the index patient was performed. All analyses were carried out with written informed consent and the study was approved by the local medical ethics committee. For lipidome analysis, plasma of both the father (dietary control) or nine gender-matched, healthy controls have been used. Fatty acid and gene expression data from liver biopsies from patients of different BMI were included in a biobank study conducted in Ulm, which was already described in detail 17. FASCINATE-1 is a randomized 12-week placebo-controlled study of the FASN inhibitor TVB-2640 in NASH at 10 US sites (ClinicalTrials.gov number NCT03938246)10. Adults with ≥8% liver fat, and evidence of liver fibrosis by MR-elastography (MRE) ≥2·5kPa or liver biopsy were randomized to receive placebo or TVB-2640 orally, once-daily for 12 weeks. The primary endpoints were safety and relative change in liver fat, and lipidomics was an exploratory endpoint.
Plasma Lipidomics
Plasma lipidomic analysis was performed using the Lipidyzer™ Platform from SCIEX. Briefly, plasma samples were spiked with Lipidyzer™ Internal Standards (SCIEX) and lipid extraction was performed employing an adjusted MTBE/methanol extraction protocol47. Extracted lipids were concentrated and reconstituted in a mixture of dichloromethane (50): methanol (50) containing 10mM ammonium actetate. Separation and targeted profiling of lipid species was performed combining differential mobility spectrometry and a QTRAP® system (QTRAP® 5500; SCIEX). Quantification of lipids was conducted by the Lipidyzer™ software (Lipidomics Workflow Manager software; SCIEX) employing specific multiple reaction monitoring transitions.
For the FASCINATE-1 study lipidomics of triacylglycerols, patient plasma was extracted with chloroform/methanol. The organic phase was dried, reconstituted in acetonitrile/isopropanol (1:1), centrifuged, and transferred to vials for UHPLC-MS analysis as previously described48 . An appropriate test mixture of standard compounds was analyzed before and after the entire set of randomized sample injections in order to examine the retention time stability, mass accuracy, and sensitivity of the system. Metabolomics data were pre-processed using the TargetLynx application manager for MassLynx 4.1 (Waters Corp., Milford, MA).
GC/FiD fatty acid profiling
Fatty acid composition in total lipid extracts of liver was determined by gas chromatography as described previously 17 . Briefly, 250 µl of butylhydroxytoluene (0.1 mol l:1 in methanol) and 6ml of chloroform/methanol (2/1, v/v) were added to ~100 mg of tissue. After homogenization using an Ultraturrax homogenizer, the samples were heated to 50 °C for 30 min and centrifuged (1,800 g, 5 min). Fatty acid methyl esters were prepared by heating 100 ml of tissue extract, 2ml methanol/toluene (4/1, v/v), 50 µl heptadecanoic acid (200 µg/ml in methanol/toluene, 4/1) and 200 µl acetylchloride in a capped tube for 1 h at 100 °C. After cooling to room temperature, 5ml of 6% sodium carbonate and 1.6 ml of toluene were added. The mixture was centrifuged (1,800 g, 5 min) and 150–200 µl of the upper layer was transferred to auto sampler vials. Gas chromatography analyses were performed using an HP 5890 gas chromatograph (Hewlett Packard) equipped with flame ionization detectors (Stationary phase: DB-225 30m x 0.25mm i.d., film thickness 0.25 µm; Agilent). Peak identification and quantification was performed by comparing retention times and peak areas, respectively, to standard chromatograms. All calculations are based on fatty acid methyl ester values. Concentration of fatty acid species was calculated as mg% (mg fatty acid species/100mg total fatty acids).
Animal models
All animal studies were performed with permission of the Animal Welfare Officers at University Medical Center Hamburg-Eppendorf. Mouse lines (C57BL6/J and Mlxipl-deficient mice (ChREBP-/-) were purchased from Jackson Laboratories or Janvier. FASNR1812W mice were generated as described before 39 . Studies at Washington University were approved by the institutional Animal Studies Committee. Unless indicated otherwise, age matched male mice (8-20 weeks) were used. Routinely, mice were kept in single cages in Memmert climate chambers with ad libitum access to food and water. Mice were housed at indicated environmental temperatures. DHA supplementation was performed after two weeks of high fat diet feeding via DHA supplemented high fat diet (2.2%) at room temperature. Radioactive DHA incorporation into VLDL was performed by conjugation of radioactive DHA to albumin and intravenous injection into wild type or ChREBP-KO mice, which received a dose of tyloxapol to inhibit lipases as described before 49. Blood for EDTA plasma measurements was collected by cardiac puncture of anaesthetized mice.
Gene expression
RNA was isolated from tissues using peqGOLD TriFast (Peqlab), by homogenizing tissues with TissueLyser (Qiagen) and purifying RNA by NucleoSpin RNAII Kit (Macherey-Nagel). Afterwards, cDNA was prepared with High-Capacity cDNA Archive Kit (Applied Biosystems). Gene expression was assessed using Taqman assays (liver samples) supplied as assays-on-demands or SYBR green (Applied Biosystems) (fibroblast samples), and data were normalized to the gene expression levels of the housekeeper Tbp for mice or TAF1, RPLP0 or GAPDH for human samples as described before 17. Validation of the sample and RNA and cDNA quality was assessed by 260/280 nM absorbance ratio measured with a Thermo Scientific NanoDrop™ and robust expression of housekeeping genes. Samples, which showed non-detectable housekeeping or target gene amplification were excluded. SYBR green assays were designed to avoid single nucleotide polymorphism sites and PCR products were designed to span exons, which preferably have long introns to reduce unspecific amplification of possible genomic DNA contamination.
Cell culture
Fibroblasts of the index patient or age and/or healthy controls were seeded into 6 cm dishes or 6-/12-well plates and were proliferated to confluence. HepG2-CIRSPR/Cas9 cells were generated by using commercially available HepG2-Cas9 overexpressing cells and gRNA was designed using the IDT and GenScript tools using the guide sequence 5’-GAGTTGCCGCACCTCGCGCACGG-3’ within an pGS-gRNA-Neo plasmid and a HDR repair template was designed using the IDT design tool with modifications of the Arg2177Cys mutation, a silent mutated PAM sequence and a silent Csp6I restriction enzyme site for genotyping. pGS-FASN-gRNA-Neo and HDR template was transfected with jetOPTIMUS® transfection reagent (Polyplus) into Cas-9 HepG2 cells, HDR enhancer™ was used and single cell clones were sorted by FACS sorting. DNA of cell clones was extracted via standard methods and was sequenced by Sanger Sequencing using the primer 5’-GCATCCGCGACTTGGCTGCTG-3’. Protein turnover studies were performed with 40-80 µmol Cycloheximide (Cat.-No. C4859, Sigma-Aldrich) for the indicated time frame. Overexpression of wild type or FASNArg2177Cys was performed via jetOPTIMUS® transfection of HEK293T cells with a FASN-pcDNA3.1 vector purchased from Genscript or HDAC3-pcDNA3.1 vector purchased from Addgene. Immunoprecipitation studies were performed using Pierce™ Anti-HA Magnetic Beads , Cat.-No. 88836, Thermo Scientific, incubated for 2h or overnight at 4°C. In vitro acetyl-CoA incubation was performed using HA-immunoprecipitated FASN incubated with 10mM acetyl-CoA (Cayman Chemicals, Cat-no. 16160) as described in 37. Wild type and DGAT2-KO mouse embryonic fibroblasts were cultured at 10%FCS, 4,5g Glucose/l DMEM supplemented with 1% Penicillin/Streptomycin. Fatty acid incubations were performed by complexing the respective fatty acid to fatty-acid-free albumin in molar excess and supplementing the cells in lipoprotein-deficient FCS serum (LPDS). C75 FASN inhibitor was purchased at Cayman chemicals and was added at 20 µM to the cells to pharmacologically inhibit FASN. InSolution™ MG132 was purchased from Calbiochem (Cat.-No. 474791) and was added for 6h to the cell culture supernatant at a dose of 20 µM.
Lipogenesis activity assays
14C-acetate was used to study lipogenesis activity in fibroblasts. To this extend, fibroblasts were incubated with 10%FCS/4,5g Glucose/l DMEM supplemented with 0.074 Bq 14C-acetate. After 24h, cells were washed with ice cold PBS, harvested in PBS and spun down at 100g for 5 min. 1/5th of the pellet was used for BSA protein concentration measurement (PierceTM BCA Protein Assay Kit). 4/5 of the pellet was used for lipid extraction as described by Mc Donald et al. 50 and the triacylglycerol fraction was separated via subsequent thin layer chromatography. Triacylglycerol fraction was collected, and tracer amounts were counted by scintillation counting.
FASN activity was performed using HepG2 lysates from FASNWT/WT or FASNWT/R2177C HepG2 cells using a modification of previously described assays 51-53. Cells were rinsed and harvested in ice-cold PBS and pelletized by centrifugation for 5min at 650 x g. Pellets were homogenized in extraction buffer (0,1M potassium phosphate (pH 7), 0,1mM EDTA, 2mM Glutathion) and lysed by sonication. Lysates were centrifuged at 13.000 x g for 10 min at 4°C and supernatants collected. Protein concentration was obtained by using Bradford reagent. 20µl sample containing 70µg protein was added to 70µl of assay buffer (0,1M potassium phosphate (pH 7), 2,3mM EDTA, 4,6mM Glutathion, 57 µM Acetyl-CoA, 0,4 mg/ml NADPH). At last, 10µl of 1mM malonyl-CoA (Sigma) or water (for blanks) was added. NADPH-decay was measured at 340 nm with a photometer (BioTek SYNERGY H1) for 30min at 37°C. Slope was calculated at the linear range of the experiment and decay was calculated to NADPH (Roche) standard series.
Western blot
Cells were cultured as specified, washed with PBS and scraped off in ice-cold cell lysis buffer (50 mM Tris–HCl, pH 8.0; 150 mM NaCl; 1% Nonidet P-40; 0,5% Na-Deoxycholat; 5mM EDTA, 0,1% SDS), organs were harvested and homogenized in RIPA-buffer supplemented with complete Mini Protease Inhibitors and PhosStop (Roche)). Cell lysates or organ lysates were clarified by centrifugation (14,000 rpm, 10 min, 4°C) and supernatants were supplemented with sample buffer. Proteins were separated on SDS-polyacrylamide gels and transferred to PVDF membranes using the Transblot Turbo Transfer System (Bio-Rad laboratories). Following blocking (20 mM Tris–HCl, pH 7.4; 150 mM NaCl; 0.1% Tween-20; 5% non-fat dry milk) and washing (20 mM Tris–HCl, pH 7.4; 150 mM NaCl; 0.1% Tween-20), membranes were incubated in primary antibody solution (20 mM Tris–HCl, pH 7.4; 150 mM NaCl; 0.1% Tween-20; 5% BSA or 5% non-fat dry milk) containing the appropriate antibodies. Membranes were washed and incubated with donkey anti-rabbit IgG Horseradish Peroxidase secondary antibody (GE Healthcare; no. NA934V; 1:7,500 dilution) or with sheep anti-mouse IgG Horseradish Peroxidase secondary antibody (GE Healthcare; no. NA931V; 1:7,500 dilution). After final washing, proteins were visualized using the ChemiDoc MP Imaging System (Bio-Rad laboratories).
Modeling of a mutant FASN dimer
The hyperacetylated sites identified by our studies were mapped on the surface of the available 3.22Å mammalian full length FASN structure, PDB 2VZ8 54. FASN lysines displaying hyperacetylation are shown as red spheres in one of the chain (residues corresponding to UNP FAS_HUMAN). ACP, acyl carrier protein is shown as a cartoon in the figure. In porcine FAS (UNP A5YV76_PIG), Lys776 is a glutamate, and Lys1927 is an arginine. In the structure, the location of Lys673, Lys776, Lys1582, Lys1704, Lys1927 in the mammalian FASN is shown. Location of hyperacetylated Lys2206, Lys2436, and Lys2449 could not be indicated, because of the unavailability of the thioesterase domain (TE), in the PDB 2VZ8 structure. The figure was prepared using ChimeraX 55.
Whole Exome Sequencing and variant validation
Genomic DNA was extracted from blood samples and trio whole-exome sequencing was performed with DNA samples of the index patient and both healthy parents For this purpose, coding DNA fragments were enriched with a SureSelect Human All Exon 50Mb V5 Kit (Agilent) and libraries were sequenced on a HiSeq2500 platform (Illumina). Reads were aligned to the human reference genome (UCSC GRCh37/hg19) using the Burrows-Wheeler Aligner (BWA, v.0.5.87.5). Detection of genetic variation was performed with SAMtools (v.0.1.18), PINDEL (v. 0.2.4t), and ExomeDepth (v.1.0.0). The impact of predicted amino acid substitutions for protein function was assessed by the pathogenicity prediction tools CADD, M-CAP and ClinPred. The FASN variant was validated by Sanger-sequencing. Primer pairs were designed to amplify selected coding exons of the candidate gene. Amplicons were directly sequenced using the ABI BigDye Terminator Sequencing kit (Applied Biosystems) and a capillary sequencer (ABI 3500, Applied Biosystems). Sequence electropherograms were analyzed using the Sequence Pilot software (JSI Medical Systems).
Acetyl-Proteomics
Identification of the proteins captured by the IP were performed by the following steps: The protein band of the SDS-PAGE of the IP eluate fraction was cut and proteins in the band digested with trypsin according to Shevchenko et al. 56. Briefly, the gel was shrinked and swelled with 100% acetonitrile (MeCN) and 100 mm NH4HCO3, respectively. Proteins in the gel were reduced with 10 mm dithiothreitol, dissolved in 100 mM NH4HCO3 and alkylated with 55 mM iodacetamide (dissolved in 100 mm NH4HCO3). Tryptic digestion was performed with 8 ng /µL sequencing-grade trypsin, dissolved in 50 mM NH4HCO3 and 10% MeCN at 37°C for 12 h. Peptide extraction was done with 5% formic acid and 50% MeCN in water and thereafter the peptides were dried. Next, the peptides were dissolved in 0.1% formic acid (20 µL) and injected with a flow rate of 5 µL/min into a nano-LC system (Dionex UltiMate 3000 RSLCnano, Thermo Scientific) containing a trapping column (Acclaim PepMap µ-precolumn, C18, 300 µm x 5 mm, 5 µm particle size, 100 Å pore size, Thermo Scientific. Buffer A: 0.1% formic acid in H2O; buffer B: 0.1% formic acid in MeCN) coupled to an electrospray ionization (ESI) source, part of a tribrid mass spectrometer equipped with a quadrupole, a linear ion-trap, and an orbitrap (Orbitrap Fusion, Thermo Scientific). Salts and other hydrophilic compounds were washed from the trapping column with 2% buffer 5 min B using a flow rate of 5 µL/min. The desalted tryptic peptides were fractionated with a reversed-phase capillary column (Acclaim PepMap 100, C18, 75 µm x 250 mm, 2 µm particle size, 100 Å pore size, Thermo Scientific). The ESI spray was formed by a fused-silica emitter (I.D. :10 µm, New Objective, Woburn, USA) using a capillary voltage of 1650 V. The positive ion mode was used. The mass spectrometer was operated in the data-dependent acquisition (DDA) / top speed mode. Further parameters were 28% HCD collision energy, intensity threshold of 2 x 105, and an isolation width of m/z = 1.6. For the MS scan a m/z 400–1500 range was chosen, performed every second, with resolution of 120000 full width at half maximum height (FWHM) at m/z 200, a transient length of 256 ms, a maximum injection time of 50 ms and an AGC target of 2 x 105. Fragment spectra were measured in the ion-trap with a scan rate: 66 kDa/s, a maximum injection time of 200 ms and a AGC target of 1 x 104. With Proteome Discoverer 2.0 (Thermo Scientific) LC-MSMS data were processed. Proteins were identified using the search engine Sequest HT and the human Swiss-Prot protein database (www.uniprot.org). As parameters for the searches a precursor mass tolerance of 10 ppm, a fragment mass tolerance of 0.2 Da, and tryptic/semitryptic digestion were chosen. Two missed cleavages were allowed. As a fixed modification carbamidomethylation of cysteine residues, as a variable modification oxidation of methionine residues and in addition acetylation were applied for the search. A false discovery rate of 1% by using Percolator was applied.
Statistical methods
Data were transformed to natural logarithm when necessary to achieve normal distribution and/or homoscedasticity. Associations of lipids to gene expression, other lipid species or clinical data have been performed by linear regression analysis and Pearson’s correlation coefficient. Comparisons of two groups were examined using Students T-Test with adjustment for multiple testing as indicated. Comparisons of three or more groups were analyzed using ANOVA with adjustment for multiple testing as indicated. GraphPad Prism was used for all statistical analyses.