Fig 1.
Experimental procedures and measurement of blood pressure.
The experimental procedures (A). The 8-week-old male SHRs were randomly divided into 4 groups (n = 6) and fed control diet with or without hydralazine treatment (7.36 mg/kg body weight/day) for 2 weeks. Then they were fed HFC or control diet in the presence or absence of hydralazine for 8 weeks before the sacrifice. The effect of hydralazine on systolic blood pressure of SHRs (B). *P < 0.05 between control diet and HFC diet groups (two-way ANOVA); †P < 0.05 between with and without hydralazine groups (two-way ANOVA); ‡P < 0.05 vs control diet group (one-way ANOVA); §P < 0.05 vs control diet with hydralazine treatment (one-way ANOVA).
Fig 2.
Images of liver sections in the first experiment.
Representative images of liver sections subjected to immunostaining with CD68 antibody. Liver sections from normotensive WKY rats fed control (A) or HFC diet (B) and hypertensive SHRs fed control (C) or HFC diet (D) [magnification: ×100]. Scale bar, 100 μm. CV, central vein; CD68, cluster of differentiation 68.
Fig 3.
Effects of HFC diet on the expression of liver fibrosis-related genes.
Western blots of the protein expression of α-SMA (A, B) and PDGFR-β (A, C) in the livers of WKY and SHRs fed the control or HFC diet. RT–qPCR measurements of the hepatic expression of COL1A1 mRNA (D). Serum TBARS levels representing levels of oxidative stress (E). n = 6/group. The values in parentheses represent the fold changes compared with the respective control. Significant interaction between the strain and HFC diet was observed in the hepatic levels of α-SMA and serum TBARS. *P < 0.05 vs respective control diet group with two-way analysis of variance (ANOVA); †P < 0.05 vs WKY control diet group (one-way ANOVA); ‡P < 0.05 vs SHR control diet group (one-way ANOVA). α-SMA, alpha-smooth muscle actin; PDGFR-β, platelet-derived growth factor receptor-β; COL1A1, collagen, type I, alpha-1 chain; TBARS, 2-thiobarbituric acid reactive substances.
Fig 4.
Effects of the HFC diet on serum MMPs/TIMP1.
ELISA of serum TIMP1 (A) and MMPs (MMP2, MMP8, and MMP9) (B–D) in WKY and SHRs fed the control or HFC diet. The ratios of MMPs to TIMP1 were calculated as a proxy of MMP activity, n = 6/group. The values in parentheses represent the fold changes compared with the respective control. Significant interaction between stain and HFC diet was observed in serum TIMP1 and ratios of all MMPs to TIMP1. †P < 0.05 vs WKY control diet group (one-way ANOVA); ‡P < 0.05 vs SHR control diet group (one-way ANOVA). TIMP1, tissue inhibitor of metalloproteinase; MMP, matrix metalloproteinases.
Fig 5.
Images of liver sections in the second experiment.
Representative images of liver sections subjected to H&E staining (A, B) [magnification: ×200], EVG staining (C, D) [magnification: ×100], and immunostaining with CD68 antibody (E, F) [magnification: ×100]. Liver sections from SHRs fed HFC diet in the absence (A, C, E) or the presence (B, D, F) of hydralazine. Scale bar, 100 μm. CV, central vein.
Fig 6.
Fibrotic area and inhibitory effects of hydralazine on HFC-induced changes in the expression of fibrosis-related genes in the liver of SHRs.
The quantification of fibrotic areas (%) in the EVG-stained liver sections (A). Hepatic expression of COL1A1 mRNA (B). Western blots of the hepatic expression of α-SMA (C, D) and PDGFR-β (C, E). n = 6/group. Significant interaction between hydralazine treatment and HFC diet was observed in the levels of COL1A1 mRNA, α-SMA, and PDGFR-β proteins. ‡P < 0.05 vs control diet group (one-way ANOVA); §P < 0.05 vs control diet with hydralazine treatment (one-way ANOVA); ‖P < 0.05 vs HFC diet group (one-way ANOVA). COL1A1, collagen, type I, alpha-1 chain; α-SMA, alpha-smooth muscle actin; PDGFR-β, platelet-derived growth factor receptor-β.
Table 1.
The effects of hydralazine on body and liver weights as well as the levels of various biochemical indices in serum and liver.
Fig 7.
Suppressing effects of hydralazine on HFC-induced changes in the serum levels of pro-inflammatory and pro-fibrotic cytokines and ratios of MMPs to TIMP1 in SHRs.
Serum levels of TNF-α (A), TGF-β1 (B), TIMP1 (C), MMP2 (D, MMP8 (E) and MMP9 (F). The ratio of MMP2 to TIMP1 as a proxy for MMP2 activity. n = 6/group. Significant interaction between hydralazine treatment and HFC diet was observed in the levels of serum TNF-α. *P < 0.05 between control diet and HFC diet groups (two-way ANOVA); †P < 0.05 between with and without hydralazine groups (two-way ANOVA); ‡P < 0.05 vs control diet group (one-way ANOVA); ‖P < 0.05 vs HFC diet group (one-way ANOVA). Abbreviations: TNF-α, tumor necrosis factor-alpha; TGF-β1, transforming growth factor-beta 1; TIMP1, tissue inhibitor of metalloproteinase-1; MMP, matrix metalloproteinases.
Fig 8.
The possible mechanism underlying hepatic fibrogenesis during the progression of hypertension-associated NASH [9,10,14] and inhibition by hydralazine.
The combined action of hypertension and HFC diet induces increased aggregation of CD68-positive Kupffer cells and results in the elevation of both serum TGF-β1 and TNF-α, the cytokines involved in liver inflammation and fibrosis. These cytokines, derived from Kupffer cells or other types of liver cells, induce increased activation of HSCs, which then differentiate into fibrogenic myofibroblasts and produce the major components of ECM (such as collagen), indicated by elevated upregulation of α-SMA and PDGFR-β. Meanwhile, myofibroblasts also participate in TIMP1 expression, the inhibitor of MMPs. HFC diet induced greater increase in serum TIMP1 as well as a greater decrease in MMP activities in hypertensive context compared with the ones under normotensive conditions. Since increased collagen synthesis was not noted in the hypertensive context, hypertension mainly enhanced the effects of HFC diet on ECM degradation and further resulted in more severe liver fibrosis. On the other hand, hydralazine, the antihypertensive agent, significantly attenuates the progression of HFC-induced liver fibrosis under hypertensive conditions by suppressing the aggregation of Kupffer cells and the elevation of serum TNF-α. It also reduces HFC-induced increases in the hepatic expression of PDGFR-β protein and COL1A1 mRNA, suggesting that hydralazine suppresses HFC-induced ECM synthesis. Furthermore, hydralazine significantly suppresses HFC-induced elevation of serum TIMP1, whereas its effects on the levels of MMPs are not prominent. Therefore, the effect of hydralazine on ECM degradation is still unclear. In conclusion, hypertension enhances HFC-induced hepatic fibrogenesis through increasing the suppression of MMP-mediated ECM degradation, whereas hydralazine attenuates liver fibrosis development mainly by suppressing HFC-induced ECM synthesis under hypertensive conditions.