Downregulation of the MARC1 p.A165 risk allele reduces hepatocyte lipid content by increasing beta-oxidation

Abstract

Background/aims: Metabolic dysfunction-associated steatotic liver disease (MASLD) is a global epidemic. The disease has a strong genetic component, and a common missense variant (rs2642438) in the mitochondrial amidoxime-reducing component 1 (MARC1) gene confers protection against its onset and severity. However, there are contrasting results regarding the mechanisms that promote this protection.

Methods: We downregulated MARC1 in primary human hepatocytes (PHHs) using short interfering RNA (siRNA). We measured neutral lipid content by Oil-Red O staining and fatty acid oxidation by radiolabeled tracers. We also performed RNA-sequencing and proteomic analysis using LC-MS. Additionally, we analyzed data from 239,075 participants from the UK Biobank.

Results: Downregulation of MARC1 reduced neutral lipid content in PHHs homozygous for the wild type (p.A165, risk), but not for the mutant (p.T165, protective), allele. We found that this reduction was mediated by increased fatty acid utilization via β-oxidation. Consistent with these results, we found that the levels of 3-hydroxybutyrate, a by-product of β-oxidation, were higher in carriers of the rs2642438 minor allele among samples from the UK biobank, indicating higher β-oxidation in these individuals. Moreover, downregulation of the MARC1 p.A165 variant resulted in a more favorable phenotype by reducing ferroptosis and reactive oxygen species levels.

Conclusion: MARC1 downregulation in carriers of the risk allele results in lower hepatocyte neutral lipids content due to higher β-oxidation, while upregulating beneficial pathways involved in cell survival.

Keywords: Fatty acid oxidation; Ferroptosis; MASLD; MTARC1; Reactive oxygen species; Steatohepatitis.

Figure 1.

MARC1 p.A165 risk allele downregulation decreases the intracellular fat content in human primary hepatocytes (PHHs). PHHs from donor homozygous for MARC1 rs2642438 p.A165 risk (A) or p.T165 protective (B) allele were seeded (100,000 live cells/well in 8 chamber slides), transfected with siSCR or siMARC1. The cells were incubated with or without increasing concentration of FA mixture (palmitic acid+oleic acid in 1:1 ratio; conjugated in 1% [w/v] BSA in HI media; 0 µM; 30 µM FA or 300 µM FA concentration). Forty-eight hours post-transfection, cells were fixed in 4% PFA and stained with ORO staining. The ORO area in the pictures was quantified using ImageJ. Objective: 40X, DAPI: Blue, ORO: red. Scale bar=100 µm. Data are shown as mean±standard deviation of on average 12 images from 4 experimental replicates. The P-value for ORO quantification was calculated by Mann-Whitney non-parametric test. MARC1 mRNA level was measured by real-time qPCR (C). Western blot of PHH homozygous for the risk (p.A165) or protective (p.T165) protein. Cell lysates were run in 12% SDS-PAGE, blotted against MARC1 and CNX, and quantified using ImageLab software (D). Data are shown after normalization with CNX. Statistical test for Western blot was performed using unpaired Student’s t-test. RU, relative unit; R1-R3, experimental replicates; FA, fatty acid mixture of palmitic acid and oleic acid in same proportion (1:1); CNX, calnexin; MARC1,mitochondrial amidoxime-reducing component 1; ORO, Oil-Red-O; PHH, primary human hepatocytes; siSCR, scramble siRNA; siMARC1, MARC1 siRNA.

Figure 2.

MARC1 p.A165 risk allele downregulation increases β-oxidation. PHHs from donor homozygous for the MARC1 rs2642438 risk allele (p.A165) were transfected with siSCR or siMARC1. β-oxidation was assessed in the media collected from cells without or with MARC1 downregulation after 2 hours incubation with 8.3 μCi/mL [³H] palmitate and 110 µM palmitate (A). Triglyceride-rich lipoproteins were isolated from media by ultracentrifugation. Levels of the ApoB100 protein were measured by western blotting; triglycerides levels in media were assessed by MS and normalized to ApoB100 protein level (B). Data are shown as mean±standard deviation of 3 experiments. Newly synthesized triglycerides were isolated from hepatocytes, separated by one-dimensional thin layer chromatography and quantified by liquid scintillation counter without or with MARC1 downregulation after 5 hours incubation with 6 μCi/mL [³H] glycerol and 1.5 mM glycerol (C). PHH with either risk allele (p.A165) or protective allele (p.T165) were incubated with 300 µM fatty acid (oleic acid and palmitic acid 1:1 ratio) and transfected with siRNA either for scramble (siSCR) or MARC1 (siMARC1) for 48 hours, and then assessed for: β-oxidation in the media collected from PHH after incubation with 8.3 μCi/mL [³H] palmitate and 110 µM palmitate for 2 hours (D); ApoB100 levels from triglyceride-rich lipoproteins measured by Western blot, triglyceride measured by MS and normalized to ApoB100 protein levels either for the PHH with risk allele (p.A165) or protective allele (p.T165) (E). Plasma 3-hydroxybutyrate level from the UKBB was stratified based on the rs2642438 genotype in total n=239,075 individuals and adjusted for body mass index and additional clinical conditions (F). Data for radiolabeled experiments (β-oxidation and de novo triglyceride synthesis) were normalized for number of cells. The P-value was calculated by Mann-Whitney non-parametric analysis. DPM/n cells, disintegrations per min/number of cells; RU, relative unit; AU, arbitrary unit; R1-R3 (experimental replicates); CI, confidence interval; FA, fatty acid; TG, triglyceride; TAG, triacylglycerol; siSCR, scramble siRNA; siMARC1, MARC1 siRNA; PHH, primary human hepatocytes; NA, no additional clinical condition; PRS-HFC, polygenic risk score for high fat content in liver; LDL-C, low density lipoproteins containing cholesterol; HTN, hypertension; CLD, chronic liver disease; CVD, cardiovascular disease; Joint, adjusted for all the conditions mentioned above.

Figure 3.

MARC1 p.A165 knockout affects intracellular lipid composition. Lipid species with different degree of unsaturation overlaid on optical images of three representative cells per line (A); red denotes overlapping SFAs and MUFAs, purple denotes MUFAs dominated regions, and blue denotes DUFAs regions. Overall lipid content of MOCK and MARC1 KO cells, measured as the ratio of the total lipid area over the cell area (B). Calibration chart calculated as a Voronoi diagram from spectral data measured on pure FAs with different unsaturation degrees (black dots in the chart), considering Raman intensity ratio between peaks at 1,265 and 1,300 cm^-1 (horizontal axis), upshift of the t(CH₂) vibration peak at about 1,298 cm^-1 (vertical axis), and unsaturation ratio N_C=C /N_CH2 of pure FAs (color intensity) (C); the same parameters of horizontal and vertical axes are also calculated for the three lipid clusters (colored triangles), whose positions in the chart identify their unsaturation ratio. Quantification of the different FAs composition, calculated as the ratio between the area of a single type of FA (SFA-MUFA, MUFA or DUFA) over the whole area per cell (D). Basis Analysis of the cellular full datasets, using a basis composed of spectra of nucleus (blue), cytosol (gray), lipids (yellow) added with spectra for CEs of saturated (red) and unsaturated (green) FAs (E). Expression of CEs of saturated (red) and unsaturated (green) FAs computed from panel (E) as ratio of red (green) areas over the whole cell area (F). Data are shown as mean±standard deviation. The P-values were calculated by Mann-Whitney non-parametric test. MARC1, mitochondrial amidoxime-reducing component 1; SFA, saturated fatty acids; MUFA, monounsaturated fatty acid; DUFA, di-unsaturated fatty acid; KO, knockout; FA, fatty acid; CE,cholesterol ester.

Figure 4.

Proteins suppressing ferroptosis are upregulated after MARC1 p.A165 risk allele downregulation. Transcriptomic and proteomic analyses of five/three technical replicates of PHH homozygous for the MARC1 risk allele transfected with SCR siRNA or MARC1 siRNA. Top 30 differentially expressed genes after MARC1 downregulation as quantified by RNA-seq analysis of PHHs (A). Red colour in the name of the genes marks consistent result with proteomics. Volcano plot of RNA-seq data, depicting significant genes; cut off used for log2 fold change 0.5 (B). Top 30 differential protein levels after MARC1 downregulation in PHHs, as detected by LC-MS (C). Colour intensity is plotted as the fold change value from average raw protein abundance. Red colour in the name of the proteins signifies consistent result with transcriptomic data. Proteins involved in ferroptosis were overrepresented as detected in LC-MS data (D). Using FerrDB database’s overrepresentation analysis, a list of proteins for each driver, regulator or suppressor of ferroptosis was obtained and the heatmap of the proteins is plotted. Odds ratio of ferroptosis related proteins after MARC1 downregulation (E). MARC1, mitochondrial amidoxime-reducing component 1; PHH,primary human hepatocytes; siRNA, short interfering RNA.

Figure 5.

MARC1 p.A165 risk allele downregulation reduces ROS levels. Overall ROS production level in PHHs homozygous for the MARC1 risk allele (40,000 cells/well in a 96-well plate), transfected with either SCR siRNA or MARC1 siRNA (A). Forty-eight hours posttransfection, medium was replaced with 20 µM DCFDA solution for 45 minutes. Fluorescence intensity was measured on fluorescence plate reader (SpectraMax i3) at Ex/Em 485/535 nm. Lipid peroxidation level as measured by 4HNE, a lipid peroxidation marker, immunofluorescence quantification in PHH homozygous for the risk allele with or without MARC1 downregulation (B). 200,000 PHHs/well in a 24well plate were seeded and transfected with scramble or MARC1 siRNA for 48 hours. The cells were incubated with a polyunsaturated fatty acid mixture (200 µM linoleic acid and 200 µM arachidonic acid, conjugated with 1% BSA) for 48 hours. 4HNE area was quantified by ImageJ. Objective: 20X, DAPI: Blue, 4HNE: Green. Scale bar=100 µm. Data are shown as mean±SD of 4 technical replicates, each with 12 images. The P-value for 4HNE quantification was calculated by Mann-Whitney non-parametric test. Levels of the ATG7 protein measured by Western blotting and its quantification in cell lysates of PHH homozygous for the risk allele without or with MARC1 downregulation (C). CNX was used as loading control. Data were normalized to CNX and expressed as mean±SD. Heatmap of the proteins related to lipophagy and detected in the proteomics data (D). Data is shown in fold change value. MARC1, mitochondrial amidoxime-reducing component 1; ROS, reactive oxygen species; siRNA, short interfering RNA; 4HNE, 4-hydroxynonenal; SD, standard deviation; ATG7, autophagy-related gene 7; CNX, calnexin.