The Role of Bone Morphogenetic Protein Signaling in Non-Alcoholic Fatty Liver Disease

Figures

• 

  Figure 1. Pharmacologic inhibition of BMP signaling with LDN-193189 reduces hepatic steatosis. Representative photomicrographs of liver sections from leptin receptor-deficient (db/db) mice treated with vehicle or the BMP inhibitor LDN-193189 for 14 days beginning at 4 weeks of age. Sections were stained with hematoxylin and eosin (A). LDN-193189 treatment reduced hepatic tissue triglyceride (TG) toward wild-type (WT) levels (B). Treatment of db/db mice with LDN-193189 reduced hepatic BMP signaling as measured by inhibitor of DNA binding 1 (Id1) mRNA levels and was associated with reduced diacylglycerol O-acyltransferase 2 (Dgat2) mRNA levels (C,D respectively). There were seven replicates per treatment group. Comparisons were performed using 1-way ANOVA with Sidak’s multiple comparison testing. Scale bars represent 250 µm and apply to all six photomicrographs.

• 

  Figure 2. Pharmacologic inhibition of BMP signaling with ALK3-Fc reduces hepatic steatosis. Representative photomicrographs of liver sections from db/db mice treated with vehicle or the BMP inhibitor ALK3-Fc for 21 days beginning at 4 weeks of age. Sections were stained with hematoxylin and eosin (A). ALK3-Fc treatment reduced hepatic triglyceride levels (B). Treatment of db/db mice with ALK3-Fc reduced hepatic BMP signaling as measured by Id1 (inhibitor of DNA binding 1) mRNA levels and was associated with reduced Dgat2 (diacylglycerol O-acyltransferase 2) mRNA levels (C,D respectively). There were six replicates in the ALK3-Fc treatment group and eight replicates in the vehicle treatment group. Comparisons were performed using the two-tailed Student’s t test. Scale bars represent 250 µm.

• 

  Figure 3. In vitro hepatocyte lipid accumulation is induced by BMP signaling and is prevented by BMP inhibition. Treatment of HepG2 cells (grown in media containing 1.2 mM oleic acid (OA)) for 24 h with increasing concentrations of LDN-193189 had reduced intracellular lipid content compared to vehicle-treated cells as assessed by measurement of Oil-Red-O staining normalized to total protein content (A, 10 replicates per group). Similarly, HepG2 cells (grown in media containing OA) treated with BMP2 (20 ng/ml) had increased intracellular lipid content and this induction was reversed by co-treatment with LDN-193189 (B, six replicates per group). Treatment of HepG2 cells with OA for 4 h or 24 h, compared to control cells treated for the same durations with bovine serum albumin (BSA) induced mRNA expression of ID1 (inhibitor of DNA binding 1) in HepG2 cells (C, six replicates per group). Similarly, incubation of HepG2 cells with OA also induced DGAT2 (diacylglycerol O-acyltransferase 2) expression at 4 h and 24 h (D, six replicates per group). ID1 and DGAT2 expression were both inhibited by treatment with LDN-193189 in HepG2 cells (E, six replicates per group). Comparisons were performed using…

• 

  Figure 4. BMP2 increases DGAT2 expression and activity. Treatment of HepG2 cells with BMP2 resulted in increased expression of DGAT2 (diacylglycerol O-acyltransferase 2) mRNA (A, six replicates per group) as well as increased DGAT2 protein levels at 24hrs (B). In parallel experiments, DGAT activity was assessed by quantification of fluorescently-labeled triglyceride production in HepG2 cells treated with BMP2 or control at 24hrs with DGAT2 overexpression and control plasmid as additional controls (C, three replicates per group). Comparisons were made using the two-tailed Student’s t test.

• 

  Figure 5. BMP-induced DGAT2 (diacylglycerol O-acyltransferase 2) expression is mediated through SMAD signaling. HepG2 cells treated with siRNA directed against SMAD 1/5 had reduced SMAD 1/5 protein levels (A) and decreased canonical BMP signaling as evidenced by decreased mRNA expression of ID1 (inhibitor of DNA binding 1) when stimulated with the BMP ligand BMP2 (B). BMP2-induced DGAT2 expression is SMAD1/5 dependent (C). Using chromatin immunoprecipitation combined with qPCR (ChIP-qPCR), BMP2 stimulation resulted in increased binding of SMAD 1/5 to a putative SMAD promoter region (1000 base pairs upstream of the DGAT2 start site) at 10-fold higher levels compared to control levels (D). Experiments had five replicates per group for panels (B,C) and three replicates per group for panel (D). SMAD western blots were run on 2 separate gels given similar protein weights. Comparisons were performed using 1-way ANOVA with Sidak’s multiple comparison testing.

• 

  Figure 6. The ALK6 (Activin Receptor-Like Kinase 6) mutant R371Q is a constitutively active BMP type 1 receptor that is hyper-responsive to BMP ligands. C2C12BRA cells were stably transfected encoding a luciferase downstream of the BMP response element promoter (BRE-LUC), as well as either wild-type ALK6 (WT), a constitutively active ALK6 (CA), or R371Q ALK6. The ALK6 mutant R371Q (which results from the rs34970181_A variant) had increased baseline BMP signaling activity compared to WT ALK6 (A, eight replicates per group). Compared to WT ALK6, the R371Q ALK6 variant had increased BMP signaling in response to BMP2 (B, five replicates per group). Compared to WT ALK6, expression of the R371Q ALK6 variant in HepG2 cells resulted in increased BMP-induced DGAT2 (diacylglycerol O-acyltransferase 2) mRNA levels (C, six replicates per group). Equal levels of expression of WT and R371Q ALK6 was confirmed by Western blot using antibodies detecting the HA epitope tag (D). GAPDH was used as a loading control. Immunofluorescence staining of HepG2 cells for pSMAD1/5/8 and DGAT2 transfected with wild-type ALK6 or R371Q ALK6 (E) and quantified using quantitative microscopy (F, % intensity represents signal…