Enhanced venous thrombosis and hypercoagulability in murine and human metabolic dysfunction-associated steatohepatitis

Abstract

Background: Patients with metabolic dysfunction-associated steatohepatitis (MASH) are at an increased risk of developing venous thromboembolic events, including deep vein thrombosis (DVT). To date, the study of DVT in MASH has been hampered by the lack of reliable models that mimic the pathologic aspects of human disease.

Objectives: To evaluate DVT severity and hypercoagulability in murine and human MASH.

Methods: Transcriptional changes in the liver, plasma markers of coagulation, and DVT severity were evaluated in mice fed a standard chow diet or a high-fructose, high-fat, and high-cholesterol MASH diet for 24 weeks. Plasma analyses of coagulation markers and thrombin generation assays were performed in a well-characterized cohort of patients with or without MASH.

Results: Mice fed the MASH diet developed steatohepatitis and fibrosis, mimicking human MASH. Liver RNA sequencing revealed a significant upregulation of pathways related to inflammation and coagulation concomitant with increased levels of plasma coagulation markers including increased prothrombin fragment 1+2, thrombin-antithrombin complex, plasminogen activator inhibitor-1 levels, and endothelin 1. MASH exacerbated DVT severity in mice, as evidenced by increased thrombus weight and higher thrombosis incidence (15/15 vs 11/15 in controls, P = .0317). Higher endothelin 1 release and increased apoptosis were found in endothelial cells stimulated with supernatants of palmitate-stimulated HepG2 cells. Patients with MASH exhibited increased levels of plasma coagulation markers and delayed thrombin generation.

Conclusion: We report enhanced DVT severity and hypercoagulability, both in murine and human MASH. Our model of MASH-DVT can facilitate a better understanding of the fundamental mechanisms leading to increased venous thromboembolic events in patients with MASH.

Keywords: blood coagulation; constriction; inferior; liver diseases; pathologic; vena cava; venous thrombosis.

Figure 1:. Mice with MASH-inducing diet exhibit several features of human MASH.

(A) Schematic of experimental design. Eight-week-old male wild-type mice were fed ad libitum either a standard diet; “Chow” or an established MASH-inducing diet; “MASH”) for 24 weeks prior to plasma and liver samples collection and DVT induction. (B) Body weight (g), (C) liver weight (g), (D) liver to body weight ratio, (E) ALT (U/L), and (F) AST (U/L) at the end of 24 week. Plasma samples from mice were collected through the retro-orbital plexus in a centrifuge tube containing 0.05M EDTA, followed by centrifugation at 2,000 relative centrifugal force (RCF) for 10 minutes at 4 °C. The plasma samples were aliquoted and stored at −80°C. AST and ALT were measured in mouse plasma using the EnzyChrom Aspartate Transaminase or Alanine Transaminase Assay Kits (BioAssay Systems #EASTR-100 and #EALT-100, respectively) per manufacturer’s instructions. (G) Representative gross appearance of the abdominal cavity and liver from mice fed on chow diet and MASH diet. (H) Left, representative liver section stained with H&E from each group, right quantification for NAFLD Activity Score (NAS). (I) Left, representative liver section stained with picrosirius red for each group, right quantification for fibrosis score. For liver histology, formalin-fixed tissues were sectioned on a M355S rotary microtome (ThermoFisher Scientific) at 4 μm thickness and mounted on glass slides. Slides were stained for hematoxylin and eosin (H&E, ThermoFisher Scientific). H&E staining was used for NAFLD activity score (NAS50). Steatosis was scored from 0–3 (0: <5% steatosis; 1: 5–33%; 2: 34–66%; 3: >67%). Hepatocyte ballooning was scored from 0–2 (0: normal hepatocytes, 1: normal-sized with pale cytoplasm, 2: pale and enlarged hepatocytes, at least 2-fold). Lobular inflammation was scored from 0–3 based on foci of inflammation counted at 20X (0: none, 1: <2 foci; 2: 2–4 foci; 3: ≥4 foci). NAS was calculated as the sum of steatosis, hepatocyte ballooning and lobular inflammation scores. For Picrosirius Red staining, slides were treated with 0.2 N phosphomolybdic acid for 3 min and transferred to 0.1% Sirius Red saturated in picric acid (Rowley Biochemical Inc.) for 90 min, then transferred to 0.01 N hydrochloric acid for 3 min. Picrosirius Red staining was used to score hepatic fibrosis from 0–4 (0: no fibrosis; 1: perisinusoidal or portal fibrosis; 2: perisinusoidal and portal fibrosis; 3: bridging fibrosis; 4: cirrhosis). All data are expressed as mean ± SEM. Statistical comparisons were made using unpaired t-test (B-F, n =10), Mann Whitney U test (H-I, n = 5). All individual points and p values are shown. A p-value <0.05 was considered statistically significant. Scale bars = 100 μm.

Figure 2:. Unbiased liver transcriptomic and plasma analysis revealed upregulation of pathways related to inflammation and coagulation in with MASH diet.

(A) Principal component analysis was performed based on RNA-sequencing of liver samples of mice fed with chow diet or MASH diet for 24 weeks. (B) Volcano plots of differentially expressed genes (DEGs) based on RNA-sequencing analysis. (C) Log fold-change of coagulation pathway genes, (D) inflammation pathway genes and (E) endothelial activation pathway genes based on RNA-sequencing analysis. (F) Bubble plot and (G) GSEA based on RNA-sequencing. Plasma samples were isolated from mice fed with chow diet or MASH diet for 24 weeks and analyzed for (H) fibrinogen (mg/dL), (I) prothrombin fragments (F1+2) (fold change), (J) thrombin anti thrombin complex (TAT; pg/mL), (K) plasminogen activator inhibitor 1 (PAI-1; ng/mL), (L) endothelin 1 (pg/mL). These plasma coagulation markers were evaluated using commercially available mouse fibrinogen ELISA kit (Abcam, #ab108844), mouse prothrombin fragments (F1+2) ELISA kit (AFG bioscience, #EK730789), mouse Thrombin-Antithrombin Complexes ELISA kit (TAT) (Abcam, #ab137994), mouse serpin E1/PAI-1 ELISA kit (R&D Systems, #DY3828–05), and endothelin 1 ELISA kit (Abcam, #ab133030) as per manufacturer’s instructions. All data are expressed as mean ± SEM. The significance of the enriched pathways was determined by right-tailed Fisher’s exact test followed by Benjamini-Hochberg multiple testing adjustment (n = 5, A-G). Statistical comparisons were made using unpaired t-test (H-L, n = 10). All individual points and p values are shown. A p-value <0.05 was considered statistically significant.

Figure 3:. Mice with MASH diet exhibited exacerbated DVT severity following IVC stenosis.

Eight-week-old male wild-type mice were fed ad libitum either a standard diet; “Chow” or an established MASH-inducing diet; “MASH”) for 24 weeks prior to the IVC stenosis surgery. (A) Schematic of experimental design. (B) Left, representative IVC thrombus harvested 48-hour post-stenosis from each group. Right, thrombus weight (mg). Each dot represents a single mouse. (C) Thrombosis incidence, and (D) plasma endothelin-1 (pg/mL) 48-hour post-stenosis. Mice with surgery time more than 10 mins and mice with accidental bleeding during surgery were not included in the final analysis. For DVT surgeries, we used a block randomization method in which an equal number of mice from each group were randomly selected to perform surgery on any given day. For immunofluorescence analysis IVC thrombi were harvested 48-hour post-stenosis and processed. Sections were incubated overnight with primary antibodies for FITC-Ly6G1A8 (Thermo, #11–9668-82, 1:250), Cit-Histone H3 (Arg2, Arg8, Arg17), (Thermo, #630–180ABBOMAX, 1:300), CD41 (Thermo, #PA5–22307, 1:100), and Ly6c (Abcam, ab54223, 1:100) Fibrinogen (PA5–95397, 1:400) at 4°C overnight. Next day, sections were washed with 1X PBST thrice and incubated with Alexa Fluor 647/488 goat anti-rabbit/anti-rat (Thermo, #A21244, 1:500/Thermo, #A21208, 1:500) secondary antibody for 1 h in the dark at room temperature. Slides were then washed three times with 1X PBST. Images were acquired using a fluorescence microscope EVOS^™ M5000 Imaging System (Invitrogen) at 10 and 20 x magnification. (E) Left, representative cross-sectional immunofluorescence image of the isolated IVC thrombus (48-hour post-stenosis) from each group for Ly6G1A8 (neutrophils, green) and DAPI (blue). Magnification 20X; Scale bar 50 μm. Right, quantification. (F) Representative cross-sectional immunofluorescence image of the isolated IVC thrombus (48-hour post-stenosis) from each group for the anti-histone H3 (citrulline R2 + R8 + R17) (NETs, red) and DAPI (blue). Magnification 20X; Scale bar 50 μm. Right, quantification. (G) Representative cross-sectional immunofluorescence image of the isolated IVC thrombus (48-hour post-stenosis) from each group for the CD41 (platelets red) and DAPI (blue). Magnification 10X; Scale bar 100 μm. Right, quantification. (H) Representative cross-sectional immunofluorescence image of the isolated IVC thrombus (48-hour post-stenosis) from each group for the Ly6c (monocytes, green) and DAPI (blue). Magnification 20X; Scale bar 50 μm. Right, quantification. (I) Representative cross-sectional immunofluorescence image of the isolated IVC thrombus (48-hour post-stenosis) from each group for the fibrin(ogen) (yellow) and DAPI (blue). Magnification 20X; Scale bar 50 μm. Right, quantification. (J) Endothelin 1 levels quantification in supernatants of human saphenous vein endothelial cells (HSaVEC, 1 X 10⁵) cells incubated with cell supernatant of palmitate-stimulated hepatocytes (HepG2, 3 X 10⁵) or control hepatocytes (HepG2, 3 X 10⁵) using commercially available Endothelin 1 ELISA Kit (Abcam, #ab133030) as per manufacturer’s instructions. (K) Left, representative cross-sectional immunofluorescence image TUNEL positive HSaVEC cells incubated with cell supernatant of palmitate-stimulated hepatocytes or control hepatocytes. Right, quantification. All data are expressed as mean ± SEM. Statistical comparisons were made using Mann Whitney U test (B, n = 15), Fisher’s exact test (C, n = 15), unpaired t-test (D-I, K, n = 5–10), two-way ANOVA followed by Holm-Šídák’s multiple comparisons test JI, n = 5). All individual points and p values are shown. A p-value <0.05 was considered statistically significant.

Figure 4:. Patients with MASH exhibited significantly increased plasma markers of coagulation and altered delayed thrombin generation.

Blood samples from patients with no history of liver disease (controls) were obtained prior to a routine screening colonoscopy or mammogram (IRB protocol #2015.101.C). The biospecimen cohort was subsequently evaluated for the exclusion criteria of (a) active malignancy at the time of sampling and (b) history of chronic liver disease. History of chronic liver disease was defined as confirmation of cirrhosis by imaging, elastography, or biopsy. Blood specimens from patients with MASH confirmed by biopsy, magnetic resonance elastography or ultrasonic transient elastography were obtained during routine hepatology surveillance laboratory visits (IRB protocol #2016.131.B). The inclusion criteria was > 18 years of age with a history of chronic liver disease, as defined for control patients, or biopsy/elastography confirmation of MASLD. Exclusion criteria were active malignancy, history of surgical intervention for primary liver malignancy, or suspicion of hepatocellular carcinoma due to an abnormal ultrasound or elevated alpha fetoprotein. Specimens analyzed in the study resulted from a secondary panel of inclusion/exclusion criteria. Secondary inclusion criteria selected patients with biopsy/elastography confirmation of MASLD or MASH with secondary exclusion criteria for any secondary etiology of liver disease including history of alcoholic liver disease, hepatitis C, hepatitis B. Peripheral blood specimens were collected from patients with or without MASH in the vacutainer sodium citrate tubes (BD, catalog# 362761). Plasma was obtained following centrifugation at 1500 RCF for 20 minutes at 4°C, separation according to the manufacturer’s protocol and stored at −80°C until analysis. Demographical, lipid levels (n = 10–16), diabetes medication status (n = 19–20), and MASH parameters were obtained from electronic medical records. Child-Pugh score was not available for two patients with MASH. Plasma coagulation markers (n = 20) were analyzed by the commercially available kits, human fibrinogen ELISA kit (Abcam, #ab108842), human thrombin-antithrombin complex ELISA kit (TAT) (Abcam, #ab108907), human serpin E1/PAI-1 ELISA kit (R&D Systems, #DSE100) and human prothrombin fragment 1+2 (F1+2) ELISA kit (AFG Bioscience, #EK710680). The thrombin generation assay (n = 19–20) was performed by Technothrombin TGA kit (Diapharma, # 5006010) as per manufacturer’s instructions. Briefly, a 4-point calibration curve was generated using the Ceveron^® TGA CAL set (calibrated against the Thrombin Reference Preparation of the WHO). Before each test run, internal quality control was done using the Ceveron^® TGA CONT H and L vials (lyophilized human plasma with increased or decreased thrombin generation). The plasma samples were thawed for 15 min in a water bath at 37°C and analyzed immediately. To patient’s plasma (40 μL), recombinant soluble thrombomodulin (10 nM), Ceveron^® TGA BUF (Tris-Hepes-NaCl buffer), Ceveron^® TGA RC low trigger reagent (low concentration of phospholipid micelles containing recombinant human tissue factor in Tris-Hepes-NaCl buffer), Ceveron^® TGA SUB (fluorogenic substrate 1 mM Z-G-G-R-AMC), and 25 mM CaCl2 were added to give 150 μL final reaction mixture. The following parameters were used as readouts: (a) peak thrombin generation (nmol); (b) the lag time (min); (c) time to peak thrombin (min), (d) velocity index (nM/min), and (e) the area under the curve (AUC; nM*min). All data are expressed as mean ± SEM. Statistical comparisons were made using the Mann Whitney U test. All individual p values and n are shown. A p-value <0.05 was considered statistically significant. N.A., not available.