We read with great interest the article published in HEPATOLOGY by Guy and colleagues.1 In this study, the authors demonstrated a direct correlation between the hepatic level of Hedgehog (HH) pathway activity and the severity of liver injury and fibrosis in human nonalcoholic fatty liver disease (NAFLD). The expression of Sonic Hedgehog (SHH), a ligand of the HH pathway, was analyzed by immunohistochemistry in 84 NAFLD patients with different stages of fibrosis. In these patients, SHH expression increases concomitantly to fibrosis stage, ballooning, portal inflammation, and fibrosis. Interestingly, at the univariate analysis age, body mass index, waist circumference, homeostasis model assessment-insulin resistance, essential hypertension, and fibrosis stage strongly correlated with hepatic expression of SHH.
In HH signaling, the interaction of SHH with the cell surface receptor Patched depresses Smoothened (SMO) activity, leading to nuclear localization of glioblastoma family transcription factors (GLI1, 2, and 3) that regulate the expression of cell-specific target genes.2 Interestingly, Guy and colleagues1 observed a significant correlation between nuclear accumulation of GLI2, liver fibrosis, and other risk factors for NAFLD. Accordingly, we found that GLI2 was overexpressed in liver extracts from rats treated with high fat/high fructose (HF/HFr) diet as compared with standard diet (Fig. 1A). We previously demonstrated that rats fed a HF/HFr diet histologically resemble human NAFLD, developing a rare fibrosis with increased collagen VI.3 Here we show that this dietetic regimen also increases hyaluronic acid (HA) circulating levels (Fig. 1B). HA, as well as osteopontin, is an important ligand for CD44, a marker of cancer stem cells, whose expression is inhibited by SMO antagonists.4 We hypothesize that an interaction network may exist between HA and HH signaling. This hypothesis is strongly supported by data from Patched1 mutant mice (Ptch1+/−), in which the HH pathway is constitutively activated and displays high levels of circulating HA with respect to Ptch1+/+ mice (Fig. 1c). These results may explain why these mice are susceptible to developing fibrosis in diet-induced NAFLD.5
In conclusion, the relationships between the HH pathway and CD44 ligands, such as HA, may be critical for the comprehension of mechanisms that lead to fibrosis and hepatocellular carcinoma from NAFLD.