Liver fibrosis is characterized by excessive scarring, caused by chronic inflammatory processes during liver diseases of different origin. In response to chronic liver injuries, various cell types get activated and transdifferentiate into myofibroblastic cells that then participate in synthesis and reorganization of connective tissue (Hautekeete and Geerts, 1997; Friedman, 1999, 2008; Knittel et al., 1999). A main source of extracellular matrix (ECM) production are hepatic stellate cells (HSC) undergoing myofibroblastic transition (Hautekeete and Geerts, 1997; Friedman, 1999, 2008). Myofibroblastic differentiation and matrix accumulation of HSC is heavily triggered by profibrogenic mediators such as transforming growth factor β (TGF-β) and the β-isoform of platelet-derived growth factor (PDGF). Hence, in response to TGF-β, HSC take center stage during fibrosis by the enhanced synthesis of ECM proteins, in particular collagen I and II. Here, we emphasize the role of dysregulated microRNA (miRNA) during liver fibrosis, their function in ECM expression and in profibrogenic signaling in HSC.

The fibrotic remodeling process and the change of gene expression during liver fibrosis are associated with an altered pattern of miRNA. miRNA are small (∼19–24 nucleotides long) non-coding RNA molecules inhibiting posttranscriptionally gene expression. It is suggested that more than one-third of all human genes are regulated by miRNA (Krek et al., 2005; Lewis et al., 2005). The primary miRNA (pri-miRNA) molecules of more than 1000 bases in length are transcribed by RNA-Polymerase II. Cleavage of the pri-miRNA then generates the precursor miRNA (pre-miRNA). This pre-miRNA is exported into the cytoplasm and further processed by RNase III and Dicer resulting in the release of a mature miRNA (Ruby et al., 2007). After integration into the RNA-induced silencing complex (RISC), interaction of miRNA sequences with the untranslated region (UTR) of transcripts causes translational repression or transcript degradation (Bartel, 2009). Several algorithms predict that one miRNA might bind to a multitude of mRNA transcripts and, in turn, that one mRNA might be targeted by a widespread panel of miRNA species (Doench and Sharp, 2004; Lewis et al., 2005).

In liver, miR-122 is the most abundant miRNA, accounting for more than 70% of the total miRNA in hepatocytes (Jopling et al., 2005). miR-122 is involved in cholesterol synthesis and hepatitis C virus (HCV) replication (Jopling et al., 2005; Esau et al., 2006). After liver injury and fibrosis it is markedly decreased depending on severity of fibrosis (Cheung et al., 2008; Morita et al., 2011). Other miRNA species, that are also highly expressed in the healthy liver, e.g., miR-125b or miR-22, were shown to be additionally reduced in fibrotic liver biopsies of chronic hepatitis C patients. Furthermore, the members of the miR-29 family, as well as miR-194 and miR-150 are reported to be downregulated during fibrogenesis (Kwiecinski et al., 2011; Roderburg et al., 2011). Additionally, miR-19b is recently found to be diminished in fibrotic liver of rat and human (Table 1) (Lakner et al., 2012). Whereas a wide range of miRNA are reduced after fibrosis, only some miRNA like the members of the miR-199, the miR-200, and the miR-34 family are known to be increased (Alisi et al., 2010; Pogribny et al., 2010; Murakami et al., 2011). Though miR-21 upregulation was only analyzed on fibrotic liver biopsies of a limited number of HCV patients (Marquez et al., 2010), in many other organs miR-21 is one of the most important miRNA prevalently expressed after fibrosis induction (Thum et al., 2008; Liu et al., 2010; Zhu and Fan, 2011).

Recent reports have shown that altered miRNA levels are also associated with the phenotypical changes of HSC during the myofibroblastic transition process including the induction of ECM proteins (Guo et al., 2009a; Ji et al., 2009) (Table 1).

The members of the miR-29 family are of particular interest because they are shown to inhibit ECM synthesis indicating an antifibrotic function. The miR-29 family consists of miR-29a, miR-29b, and miR-29c, differing only in two or three bases. miR-29a and miR-29b₁ as well as miR-29c and miR-29b₂, are encoded and transcribed in tandem by two genes located on chromosome 7 or chromosome 1, respectively (Mott et al., 2007; Wang et al., 2008). van Rooij et al. (2008) first reported the miR-29 function after myocardial infarction. The authors proved miR-29 mediated repression of elastin, collagen I and III synthesis in cardiac fibroblasts and its regulation by TGF-β. Accordingly, downregulation of miR-29 was suggested to enhance the fibrotic response after myocardial infarction, whereas overexpression of miR-29 in cardiac fibroblasts reduced collagen expression. Similarly, Ogawa et al. (2010) demonstrated that miR-29 inhibits the production of fibrillar collagen in HSC, suggesting also a function of miR-29 in liver fibrosis (Roderburg et al., 2011). Furthermore, highly conserved binding sites for the miR-29a and miR-29b are found in many 3′-UTR sequences of the various subunits of ECM proteins like collagen type V and XV subunits, laminin γ 1, and nidogen. The findings of Cushing et al. (2011) suggest that miR-29a and miR-29b are also involved in a widespread panel of other fibrosis associated genes including Adam metalloproteinases mainly type 12 and 9, ECM formatting components such as fibrillin-1 and follistatin-1, and different integrin chains. Recent work of Kwiecinski et al. (in review) showed that miR-29 not only targets gene expression of ECM associated proteins, but also the expression of profibrogenic growth factors (manuscript submitted). They found putative miR-29a and miR-29b binding sites in mRNA of IGF-I but most notably also in members of PDGF linked signaling pathways such as PDGF-β receptor, PDGF-C, and VEGF-A. Accordingly, miR-29 mediated inhibition of IGF-I and PDGF-C was proven in HSC. In addition, Sekiya et al. (2011) collect evidence that also the PDGF-β receptor is targeted by miR-29 regulation.

As a consequence, miR-29 acts as antifibrogenic miRNA by two pathways (1) by inhibition of ECM formation and (2) interfering with the profibrogenic cell communication pathways via PDGF-B and PDGF-C signaling (Figure 1B). Interestingly, miR-29 expression is in turn highly regulated by PDGF-BB (Ogawa et al., 2010). Thus, the loss of miR-29 in myofibroblastic HSC during liver fibrosis is mainly due to profibrogenic stimulation by TGF-β and PDGF-BB (Ogawa et al., 2010; Kwiecinski et al., 2011).

The decrease of miR-29 was observed after experimental fibrosis including liver intoxication in mice and cholestasis induced fibrosis after bile duct occlusion in rat. Furthermore, in liver biopsies of patients with chronic hepatitis C miR-29 levels are significantly reduced (Kwiecinski et al., 2011; Roderburg et al., 2011). These in vivo studies suggest that the loss of miR-29 during liver fibrogenesis leads to the abolishment of ECM and profibrogenic mediator repression. The findings on HSC provide additional evidence, that reduced miR-29 levels leading to profibrogenic gene expression are due to TGF-β or PDGF-BB exposure. The Smad-2/3 proteins act as the main signal transducers of TGF-β stimulation, initiating ECM synthesis and myofibroblastic transition (Dooley et al., 2000, 2001a). In renal fibrosis, silencing of the miR-29c/b₂ gene is assumed to involve Smad-3 signaling (Qin et al., 2011). However, the analysis of promoter regions of both miR-29 genes, shown in Figure 1A, has just started by defining the transcriptional NF-K_B and YY-NF-K_B control, respectively (Wang et al., 2008; Mott et al., 2010), and did not yet focus on TGF-β signaling.

Interestingly, the promoter sequence of the miR-29a/b₁ gene harbors several CAGA-boxes. CAGA-boxes function as Smad-3/4 protein binding elements (SBE) and Smad-3/4 binding is commonly followed by induction of gene transcription (Dennler et al., 1998). However, suppressive effects of Smad proteins are also possible, depending on their interaction with other regulatory factors like histone deacetylases (Liberati et al., 2001) or the activator protein 1 (Ap1; Hall et al., 2003). Thus, TGF-β induced Smad-3/4 signaling and the regulatory binding of the SBE may contribute to the transcriptional repression of the miR-29a/b₁ gene. In addition, three consensus sequences for binding of the transcription factor Ap1 are located in the upstream region of the miR-29a/b₁ gene. Ap1 is highly activated by the ras/raf pathway after TGF-β and PDGF-BB exposure (Angel and Karin, 1992; Zhang et al., 1998). Ap1 interaction with the Ap1 responsive elements (Ap1 RE) drives transcription of a broad spectrum of genes involved in the early gene response and acute inflammatory stimulation. However, Ap1 can also function as transcriptional suppressor by binding to TGF-β inhibitory elements (TIE), that are known to lead to gene silencing in the stromelysin (Kerr et al., 1990), the HGF, and the c-myc gene (Matrisian et al., 1992; Yagi et al., 2002). Thus, in future experiments it has to be proven if profibrogenic miR-29 repression is mediated by the consensus sequences of the TIE, the Ap1 RE, the SBE, or even by the interplay of different regulatory elements (Figure 1A).

miR-21 is an ubiquitously expressed miRNA, which is highly upregulated in the majority of cancer types. In cancer, miR-21 acts as an oncogenic miRNA (oncomiR) targeting PTEN and other tumor suppressor proteins (Krichevsky and Gabriely, 2009; Qi et al., 2009). Previous reports have demonstrated that miR-21 is also enhanced after initiation of myocardial, renal, or pulmonary fibrosis (Thum et al., 2008; Liu et al., 2010).

miR-21 enrichment during fibrosis is assumed to be based on profibrogenic stimulation by TGF-β, because Zhong et al. (2011) have demonstrated that Smad-3 transducing TGF-β signaling, is crucial for miR-21 transcription. Furthermore, the signal transducers of TGF-β, the receptor-regulated Smad-2 and Smad-3 (R-Smad), are able to interact with the p68 RNA helicase of the Drosha/PDCD4 microprocessor complex, promoting the restriction of pri-miR-21 into the precursor pre-miR-21 (Davis et al., 2008). Thus, the upregulation of miR-21 during fibrotic processes is based on two TGF-β stimulated mechanisms, that are mediated by the Smad proteins: (1) by transcriptional induction and (2) by enhanced miR-21 maturation (Figure 2).

A potential mechanism for the role of miR-21 in fibrosis is through regulating the protein synthesis of the inhibitory Smad-7 (Figure 2; Liu et al., 2010; Marquez et al., 2010). Smad-7 mediates a negative feed-back mechanism of TGF-β signaling. After TGF-β stimulation, Smad-7 transcription is driven by Smad-2/3 and Smad-4 (Stopa et al., 2000; Dooley et al., 2001b). Enhanced Smad-7 levels, in turn, are highly potent to antagonize TGF-β-mediated pathways (1) by blockade of the TGF-β type I receptor for further binding of the receptor Smads (Massague and Chen, 2000), and (2) by Smurf2 interaction, to form an E3 ubiquitin ligase that targets the TGF-β type I receptor for degradation (Kavsak et al., 2000).

Therefore, the inhibition of Smad-7 in response to miR-21 enhancement during fibrosis has to be considered as an important profibrogenic pathway, which might become an attractive target for future therapeutical approaches. In contrast to miR-21 that promotes TGF-β signaling, a recent study of Lakner et al. shows that miR-19b which is downregulated during fibrosis, inhibits TGF-β pathways. miR-19b represses fibrotic features after myofibroblastic activation of stellate cells by targeting the TGF-β II receptor and most notably Smad-3 (Lakner et al., 2012). Hence, upcoming knowledge of miRNA species, involved in profibrogenic signaling, will provide novel perspectives in understanding and probably also in treatment of chronic liver diseases.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.