TGF-β signaling in hepatocytes is implicated in negative regulation of the growth response. TGF-β type I receptor (TβRI) differentially phosphorylates COOH-tail serine residues of receptor-activated Smad (R-Smad, which include Smad2 and Smad3 to create three phosphorylated forms (phosphoisoforms) [³⁰]: COOH-terminally phosphorylated R-Smad (pSmad2C and pSmad3C), linker-phosphorylated R-Smad (pSmad2L and pSmad3L), and dually phosphorylated R-Smad (pSmad2L/C and pSmad3L/C) [⁹, ⁸³, ⁸⁴]. While pSmad2L resides in the cytoplasm, the other phosphoisoforms are localized to cell nuclei [³¹]. In homeostasis, TGF-β-mediated pSmad3C signaling in hepatocytes opposes proliferative responses induced by mitogenic signals, causing arrest of cell cycle progression in the G1 phase by downregulation of c-Myc and induction of p21^WAF1 [²⁸, ⁸⁵]. Acute liver injury induces secretion of proinflammatory cytokines, as well as TGF-β and activin A [⁹]. In turn, the loss of liver parenchyma triggers proliferation of resting hepatocytes. Even though TGF-β and activin concentrations are elevated, mitogenic proinflammatory cytokines regulate liver regeneration by shifting from cytostatic pSmad3C signaling to mitogenic pSmad3L signaling. This phenomenon allows hepatocytes to acquire “temporary resistance” to TGF-β (and activin A) and proliferate during liver regeneration [⁸⁶, ⁸⁷]. Inflammatory cytokine-induced pSmad3L stimulates c-Myc transcription [⁸⁸, ⁸⁹], which increases the proliferation of regeneration of hepatocytes and suppresses the cytostatic action of the pSmad3C/p21^WAF1 pathway [⁹].

TGF-β-Smad2/3 signaling in hepatocytes has also been implicated in epithelia-to-mesenchymal transition (EMT), a process in which fully differentiated epithelial cells undergo phenotypic transition to fully differentiated mesenchymal cells (fibroblasts or myofibroblasts) [⁹⁰]. During EMT, epithelial cells detach from the epithelial layer, lose their polarity, downregulate epithelial markers (e.g., hepatocyte marker albumin, cytokeratins, cadherins) and tight junction proteins (zonula occludens-1, ZO-1), increase their motility, and obtain a myofibroblast phenotype [⁹¹], with upregulated expression of α-smooth muscle actin (SMA) and vimentin in EMT-originated myofibroblasts. Epithelial cells transitioning into myofibroblasts are also reported to express fibroblast specific protein-1 (FSP1, S100A4), which is used as a marker of EMT in fibrogenesis and cancer [⁹¹–⁹⁴]. Hepatocytes have been implicated in EMT in response to liver injury [⁹⁵], suggesting that mature hepatic epithelial cells can contribute to fibrogenic myofibroblasts and collagen production in response to injury. EMT has been originally described during embryonic development [⁹¹] and plays a critical role in TGF-β-induced organogenesis. However, the role of EMT in fibrogenic disease has been recently questioned [⁹⁶]. Studies based on the genetic labeling of hepatocytes, using Albumin-Cre mice, have demonstrated that hepatocytes are capable of differentiating into myofibroblasts in vitro, but not in vivo [⁹⁷]. Moreover, FSP1 is not a robust marker for EMT, since its expression is not restricted to EMT-transitioning cells, but is expressed by myelomonocyic lineage cells [⁹⁸–¹⁰⁰].

TGF-β induces expression of growth factors and cytokines by hepatocytes. Moreover, hepatocytes serve as a significant source of BMP-7, a natural inhibitor of the TGF-β1-signaling pathway belonging to the TGF-β superfamily. Administration of BMP-7 in pharmacological doses attenuates the development of kidney fibrosis and liver fibrosis [¹⁰¹–¹⁰³]. Hepatocyte-specific deletion of Smad7 results in spontaneous liver dysfunction and aggravation of alcohol-induced liver injury [¹⁰⁴].

Consistent with their filtering/detoxification function, hepatocytes express TLRs which are constitutively engaged by bacterial products in the liver [¹⁰⁵]. Primary cultured hepatocytes express mRNA for all TLRs, but are capable of responding only to TLR2 and TLR4 ligands in vitro. However, TLR2 and TLR4 signaling in hepatocytes is fairly weak in vivo [¹⁰⁶–¹⁰⁸]. Under inflammatory conditions, hepatocytes upregulate TLR2 and become more sensible to TLR2-mediated signals. At the same time, TLR4 expression in hepatocytes is not strongly upregulated [¹⁰⁹]. Although hepatocytes express TLR4 and are capable of transmitting TLR4 signals in vitro, the contribution of TLR4 signaling in hepatocytes is limited. Meanwhile, the TLR/MyD88-mediated pathway is required for the initiation of liver regeneration after partial hepatectomy (PH) [¹⁰⁸, ¹¹⁰].

Chronic injury causes an imbalance between the production of protective and damaging cytokines, resulting in the activation of apoptotic signals in hepatocytes. TNF-α and related cytokines play a key role in mediating hepatocyte homeostasis by regulating both anti- and proapoptotic pathways. TNF-α signals through TNF-R1 and TNF-R2, of which TNF-R1 plays a critical role in TNF-α-mediated activity in the liver. The proapoptotic effects of TNF-α result from a cascade of caspase activation. This pathway is initiated by TNF-α-induced TNF-R1 receptor crosslinking, recruitment of TRADD and FADD (adaptor protein TNF receptor TRADD and Fas-associated death domain FADD) and cleavage of caspase 8, which activates the downstream proapoptotic caspases (caspases 3, 6, 7). In turn, the activation of TNF-α-dependent prosurvival signals is mediated by NFκB activation and involves transcriptional expression of suppressors of apoptosis, including Bcl-2, Bcl-xL, and Bfl-1 [¹¹¹]. However, TNFα also activates NFκB, rendering hepatocytes resistant to apoptosis unless also treated with cycloheximide, actinomycin D or the super-repressor of IκB [¹¹², ¹¹³].

In hepatocytes, IL-6 plays a crucial role in liver regeneration and transmits its mitotic signals mainly through Stat3. IL-6 stimulates hepatocytes to produce acute-phase response proteins, including serum amyloid A, complement C3 and C-reactive protein. In IL-6-deficient mice, Stat3 activation is dramatically suppressed in hepatocytes [⁸⁰]. Although Stat3 signaling can be induced by other cytokines, such as G-CSF [¹¹⁴] and leptin [¹¹⁵], current data suggests that Stat3 in hepatocytes is almost exclusively activated by IL-6. Thus, hepatocyte regeneration in response to partial hepatectomy triggers activation of the IL-6/Stat3 signaling pathway, composed of IL-6 receptor, gp130, receptor-associated Janus kinase (Jak), and Stat3. The IL-6 receptor forms a complex with two molecules of gp130 [⁶²].

Stat3 promotes liver regeneration by promoting cell cycle progression from G1 to S phase [¹¹⁶]. Stat3 regulates the expression of cyclin D1 [¹¹⁷], which is required for hepatocyte proliferation [¹¹⁸]. As expected, hepatocyte-specific Stat3-deficient mice exhibit impaired DNA synthesis and mitotic activity of hepatocytes after partial hepatectomy.

Other target genes of Stat3 include antiapoptotic genes FLIP, Bcl-2, and Bcl-xL [¹¹⁸, ¹¹⁹]. Therefore, it is believed that Stat3 prevents liver damage by its antiapoptotic and promitogenic effects [⁷⁸]. Deletion of the gp130/Stat3 pathway in hepatocytes leads to increased hepatotoxicity and accelerates liver injury and inflammation [⁷¹]. This effect is likely mediated via the Stat3 induction of EGFR and IGF-1 signaling pathways [¹²⁰]. Interestingly, Stat3 also possesses antioxidative capacity. Hypoxia and reperfusion injures hepatocytes via the generation of reactive oxygen species (ROS) and activation of redox-sensitive caspases such as caspase-3/-9. Stat3 upregulates Ref-1 [¹²¹] and Mn-SOD [¹²²], which protects hepatocytes from ROS-mediated apoptotic cell death [¹¹⁸]. Thus, activation of Stat3 in hepatocytes has hepatoprotective and anti-fibrotic effects [¹²³].

Kupffer cells are well established targets for the TLR4 ligand LPS and produce inflammatory and fibrogenic cytokines, which may activate HSCs [¹²⁶]. However, TLR4 signaling in Kupffer cells is not critical for the pathogenesis of experimental liver fibrosis. Deletion of TLR4 signaling in BM-derived inflammatory cells and Kupffer cells was achieved in BM-chimeric mice, pretreated with clodronate (to reconstitute long-lived Kupffer cells). Interestingly, only a modest inhibition of liver fibrosis was observed in these mice in response to liver injury [³, ⁵¹]. However, LPS, but not TGF-β, is a strong activator of Kupffer cells in vitro and in vivo. LPS-stimulated Kupffer cells secrete TNF-α and TGF-β. Furthermore, experimental alcoholic liver disease requires TLR4 on BM derived macrophage and Kupffer cells [⁵¹, ¹²⁷]. Interestingly, only a modest inhibition of liver fibrosis was observed in these mice in response to liver injury. These results indicate that LPS-induced fibrosis does not need Kupffer cell-mediated activation of HSC.

IL-6 and IL-10 induce opposing effects on macrophages. IL-6 signals via gp130 and IL-6R and promotes inflammatory responses in Kupffer cells/macrophages. In turn, IL-10 secreted by Th1 and T cells stimulate IL-10R1 and IL-10R2 on Kupffer cells/macrophages, causing their prolonged activation. Activation of IL-10-Stat3 signaling inhibits inflammatory responses. Stat3 upregulates expression of the suppressor of cytokine signaling 3 (SOCS3) [¹²⁸], which binds to gp130, limiting IL-6-induced inflammatory responses [⁷⁸]. Thus, while IL-6 triggers proinflammatory responses in macrophages, IL-10 mediates anti-inflammatory responses that are associated with decreased liver fibrosis [⁷⁸, ¹²⁸].

VEGF is an important regulator of angiogenesis and vascular tone. VEGF controls LSEC survival, proliferation, migration, and angiogenesis. VEGF binds to the receptor VEGFR2 and mediates its biological responses through reactive oxygen species (ROS) [¹³⁶]. Another strong angiogenic factor, Angiopoietin 1 (Ang 1), regulates maturation and stability of blood vessels. Moreover, neovascularization induced during the development of livers fibrosis is mediated by Angiopoietin 1, expressed mostly by activated HSCs [¹³⁷]. In turn, Ang 1 signals through endothelial receptor tyrosine kinase Tie2 and synergistically enhances VEGF's effects [¹³⁸]. VEGF binds to its receptor and activates an Akt signaling cascade to increase vascular tone and the formation of collateral circulation [¹³⁹, ¹⁴⁰].

Furthermore, proliferation and migration of endothelial cells depends on pericyte coverage of vascular sprouts for vessel stabilization. This process is regulated by VEGF and platelet-derived growth factor (PDGF) through their cognate receptors [¹⁴¹]. VEGFR is expressed on endothelial cells, and PDGFR is expressed on HSCs and vascular smooth muscle cells (VSMCs). Moreover, it is believed that PDGF-Rβ is exclusively expressed by HSCs in the liver and strongly upregulated in HSCs in response to fibrogenic liver injury [¹]. PDGF induces neovascularization by priming VSMCs/pericytes to release pro-angiogenic mediators. VEGF acts as a negative regulator of neovascularisation. Specifically, while pericyte-derived PDGF mediates angiogenesis, VEGF ablates pericyte coverage of nascent vascular sprouts, leading to vessel destabilization. VEGF-mediated activation of VEGF-R2 suppresses PDGF-Rβ signaling in HSCs/pericytes/VSMCs through the assembly of a previously undescribed receptor complex consisting of PDGF-Rβ and VEGF-R2 [¹⁴¹]. Thus, VEGF ameliorates development of liver fibrosis and can serve as a novel target for anti-fibrotic therapy [¹⁴²].

TGF-β signaling in endothelial cells plays a critical role in vascular development and maintenance of vascular homeostasis. Mice deficient for various TGF-β signaling components have an embryonic lethal phenotype due to vascular defects, abnormal yolk sac vasculogenesis and/or angiogenesis [¹⁴³, ¹⁴⁴]. TGF-β is also essential for vascular integrity in the adults due to its role in regulation of anti-inflammatory characteristics of endothelial cells, growth and migration [¹⁴⁵]. Similar to other cell types, TGF-β signaling in endothelial cells results in activation of TGF-β receptors with Ser/Thr kinase activity. The effect of TGF-β on endothelial cells is dose-dependent. Low levels of TGF-β promote angiogenesis, while higher doses inhibit angiogenesis [¹⁴⁶]. TGF-β regulates the activation of the endothelium via two opposing type I receptor/Smad pathways, activin receptor-like kinase (ALK)1 and ALK5 [¹⁴⁵]. The classical TGF-β type I receptor that activates Smad2/3 signaling is ALK5 (TGFβ-RI). ALK2 (ActRI) is typically used by BMPs to activate Smads1/5/8. Use of ALK2 by TGF-β is rather an exceptional nonhepatic event [¹⁴⁷]. Typically the Smad2/3 pathway is activated by the type I receptors ALK4, 5 or 7 [²⁸]. Furthermore, a coreceptor of TGF-β, endoglin (CD105), is upregulated on proliferating endothelial cells and facilitates effective TGF-β-ALK1 signaling [¹⁴⁸, ¹⁴⁹]. Another molecule which regulates TGF-β signaling is VE-cadherin. VE-cadherin-deficient endothelial cells demonstrate a loss of TGF-β-induced inhibition of endothelial cell proliferation and motility [¹⁴⁵, ¹⁵⁰].

TGF-β signaling in endothelial cells may contribute to fibrosis via transition to mesenchymal cells (EndMT), giving rise to myofibroblasts in response to fibrogenic injury. EndMT has been reported to contribute to cardiac [¹⁵¹] and renal [¹⁵²] fibrosis. The generation of mesenchymal profibrotic cells from endothelial cells by this process appears to recapitulate the transdifferentiation of endothelial cells that leads to the formation of the cardiac valves in embryonic development [¹⁵³]. EndMT is identified by expression of myofibroblasts-like genes [⁹¹] in endothelial cells that are expressing or have a “history” of expressing PECAM-1/CD31, Tie-1 [¹⁵¹], Tie-2 and CD34 [¹⁵², ¹⁵⁴]. A difficulty in interpreting these studies is that it is now recognized that Tie-2 is not a specific marker for endothelial cells in that it is also expressed in BM derived hematopoietic cells. Although endothelial cell injury [¹⁰] and neovascularization play a critical role in liver fibrosis, the role of EndMT in liver fibrosis is unknown.

LSEC are exposed to endogenous LPS liver injury. LPS induces upregulation of TLR4 expression in LSECs to facilitate angiogenesis [¹³¹]. In vitro, this effect is dependent on Myd88 activation and is associated with secretion of MMP2 by LSEC. In vivo studies have supported this data, demonstrating that TLR4-deficient mice exhibit attenuated angiogenesis and fibrosis [¹⁵⁵].

Low, physiological concentrations of endotoxin are continuously present in portal venous blood, and the liver mediates intrinsic signals to develop tolerance [¹⁵⁵]. LPS induces the release of IL-10 from LSECs and Kupffer cells and also downregulates CD4⁺ T cell activation by LSECs through down-modulation of the expression of MHC class II, CD80 and CD86. In contrast, TLR4 activation of professional APC by endotoxin increases T cell activation. These observations explain why the tolerogenic effect in the liver seems to be related to the continuous exposure of sinusoidal cells to bacterial products from the gut (reviewed in [¹⁵⁵]). Following initial activation of LSECs, Kupffer cells are a tolerogenic cellular population contributing to the tolerogenic properties within the liver [¹⁵⁵].

In response to liver injury, release of endogenous LPS mediates release of TNF-α, which in turn triggers expression of target genes in LSECs. However, a specific role of NFκB in LSECs in the pathogenesis of liver fibrosis has not been evaluated [¹⁵⁶]. Experiments in transgenic mice overexpressing the IkB-α super-repressor in endothelial cells, have demonstrated that inhibition of the NFκB signaling pathway in LPS-stimulated mice causes a defect in expression of endothelial tight junction proteins, and as a result, a loss of integrity of the endothelium and increased vascular permeability [¹⁵⁷], suggesting that NFκB is responsible for the stress-induced responses of the endothelium to septicemia or TNF-α.

The role of Stat3 in endothelial cells has not been carefully studied. It has been suggested that Stat3 facilitates protection of endothelial cells (LSEC) exposed to endogenous LPS liver injury and inflammation. IL-6 has a protective effect on hepatic LSECs by activation of Stat3 signaling [¹⁵⁸–¹⁶⁰]. Consistent with this, endothelial-specific Stat3-deficient mice are more susceptible to alcohol-induced injury, demonstrating a critical function of the endothelium and LSECs in chronic liver injury [¹⁶⁰]. Recent study has implicated Stat3 signaling in endothelial cells in mediating dual anti-inflammatory and antiapoptotic functions, of attenuating hepatic inflammation and SEC death during alcoholic liver injury [¹⁶¹].

Difficulties associated with the isolation and culturing of a pure population of cholangiocytes is a limiting factor in investigating the role of cholangiocytes in fibrogenic liver injury. It has been suggested that similar to hepatocytes, cholangiocytes are capable of differentiation into fibrogenic myofibroblasts via epithelial-to-mesenchymal transition (EMT) in response to TGF-β-induced liver injury [¹⁶⁷, ¹⁶⁸]. Although EMT in hepatocytes has been documented in vitro, in vivo studies in adult mice using Cre-lox-based cell fate mapping have not confirmed this finding [⁹⁷]. Similarly, genetic labeling of K19⁺ cholangiocytes has demonstrated that cholangiocytes do not contribute to fibrogenic myofibroblasts in experimental cholestatic liver injury [¹⁶³, ¹⁶⁹]. Moreover, hepatic epithelial cells and their precursors, genetically labeled in alpha-fetoprotein-Cre mice, do not differentiate into fibrogenic myofibroblasts in adult mice [⁹⁶, ¹⁷⁰].

A few studies have mostly linked TLR signaling in cholangiocytes to anti-microbial immunity [¹⁷¹, ¹⁷²]. Cholangiocytes may participate in microbe-associated, hepatic proinflammatory responses. In vitro studies of cultured human cholangiocytes suggest that LPS-TLR-signaling pathway activate the small GTPase Ras that mediates cholangiocyte proinflammatory cytokine production and proliferation [¹⁷²].

NFκB and/or Stat3 signaling pathways in cholangiocytes have not been specifically evaluated. Meanwhile, conditional inactivation of Stat3 in hepatocytes and cholangiocytes (stat3(Deltahc) of multidrug resistance gene 2 mdr2^(−/−) mice strongly aggravated bile acid-induced liver injury and fibrosis [¹²⁰].

TGF-β signaling plays a critical role in initiating and promoting the activation of qHSCs into myofibroblasts. Nuclear localization of pSmad2 and pSmad3 is seen in the activated HSC [⁹]. Transgenic mice have demonstrated that overexpression of TGF-β produces liver fibrosis [¹⁸¹], and conditional induction of TGF-β has demonstrated that the severity of fibrosis is proportional to the level of produced TGF-β [²⁴]. Smad3 is a direct mediator of matrix production in aHSCs. Mice lacking Smad3 are protected from fibrosis [²⁵, ¹⁸²]. Activation of TGF-β signaling causes transient expression of Smad7, regulated by a feed-back mechanism. Smad7, in turn, inhibits HSC differentiation into fibrogenic myofibroblasts and attenuates experimental fibrosis in vivo [¹⁸³, ¹⁸⁴]. BMP-7, another member of the TGF-β superfamily, also acts as a TGF-β antagonist and administration of BMP-7 in pharmacological doses attenuates development of kidney and liver fibrosis [¹⁰¹–¹⁰³].

Although activation/phosphorylation of Smad2/3 is considered to be the main fibrogenic pathway in HSCs, TGF-β1 has been also found to mediate its profibrogenic action via an alternative ALK1/Smad1 pathway in HSCs by upregulation of Id1 (the inhibitor of differentiation (1) that facilitates HSC activation [¹⁸⁵]. TGF-β1 controls the transdifferentiation process in HSCs. The recent study, aimed to elucidate TGF-β1 target genes responsible for fibrogenesis, has analyzed the Smad7-dependent mRNA expression profiles in HSCs, and identified that Id1 protein was strongly reduced by ectopic Smad7 expression in HSCs. In concordance, Id1 overexpression in HSCs enhanced cell activation, while knock-down of Id1 in HSCs inhibited HSC differentiation into myofibroblasts. Treatment of HSCs with TGF-β1 resulted in Id1 upregulation implicating Id1 to be an alternative but critical mediator of HSC activation into myofibroblasts signaling via TGF-β1/ALK1/Smad1 pathway [¹⁸⁵].

Other factors can facilitate TGF-β signaling in HSCs. In particular, stimulation of aHSCs with platelet-derived growth factor (PDGF) and TGF-β produces a synergistic effect on migration and expression of MMPs [¹⁸⁶, ¹⁸⁷]. Moreover, PDGF promotes the activation of HSCs via activation of the PI3K-Akt signaling pathway. PI3K (phosphatidylinositol-3-kinase) activity phosphorylates PIP₂ to generate PIP₃ (3,4,5-trisphosphate) [¹⁸⁸]. PIP₃ binds to the pleckstrin homology domain of Akt, directing it to the cell membrane where it becomes activated by phosphorylation events to initiate cell survival mechanisms. Consistently, inhibition of PI3K activity suppresses cell proliferation and type I collagen gene expression in activated HSCs [¹⁸⁹, ¹⁹⁰]. PDGF also activates ERK in HSCs by sequential activation of Ras-Raf-MEK signaling [¹⁹¹] and further facilitates proliferation of aHSCs [¹⁹²].

The tumor suppressor protein phosphatase and tensin homolog deleted on chromosome ten (PTEN) is a dual specificity protein and lipid phosphatase that dephosphorylates PIP₃ [¹⁸⁸, ¹⁹³]. PTEN is a negative regulator of PI3K and ERK signaling [¹⁹⁰]. Overexpression of PTEN attenuates collagen Type I production when aHSCs induces HSC apoptosis. Deletions of PTEN occur during malignant transformation in various tissues [¹⁹³]. Decreased PTEN expression is also associated with dysregulation of tissue remodeling, such as pulmonary fibrosis, bronchial asthma, and rheumatoid arthritis [¹⁹⁴–¹⁹⁶]. Since the PI3K/Akt pathway stimulates proliferation and activation of HSCs, inhibiting PTEN promotes liver fibrosis [¹⁹⁷].

Platelet-derived growth factor (PDGF) is a powerful mitogen for HSCs. In fibrotic liver, PDGF induces HSC proliferation, synergistically facilitating TGF-β1-mediated HSC activation [¹⁹²]. PDGF signals through the transmembrane receptor tyrosine kinases initiating multiple signaling pathways [¹⁹⁸, ¹⁹⁹], including activation of the mitogen-activated protein kinase (MAPK) family implicated in cellular proliferation and transmigration. This includes the extracellular signal-regulated protein kinase (ERK) pathway and two stress-activated protein kinase (SAPK) pathways: the c-Jun N-terminal kinase (JNK) and the p38 pathway [³¹].

LPS activates the NFκB and JNK/AP-1 pathways in aHSCs. LPS enhances expression of the adhesion molecules ICAM-1 and VCAM-1 and TLR2 and the secretion of IL-8, MCP-1, MIP-1α, MIP-1β, RANTES, KC, MIP-2, and IP-10 in aHSCs. In turn, LPS downregulates the expression of bone morphogenetic protein (BMP) and activin membrane bound inhibitor (Bambi), a transmembrane suppressor of TGF-β signaling (Figure 5). Bambi is a TGF-β pseudoreceptor that lacks an intracellular kinase domain, and similar to activin, prevents TGF-β signaling. Signaling via TLR4 downregulates Bambi and facilitates TGF-β signaling in HSCs. Overexpression of Bambi inhibits HSC activation, while overproduction of a dominant negative form of Bambi enhances TGF-β signaling, and induces activation of HSCs [⁵¹].

LPS signaling is blocked by inactivation of NFκB and JNK, demonstrating the role of NFκB and JNK in TLR4 signaling in HSCs. TLR4 signaling in HSCs is critical for development of liver fibrosis. Bone-marrow chimeric mice with a TLR4 deficiency in recipient liver cells, including HSCs, were resistant to liver fibrosis. Since hepatocytes exhibited no response to LPS in vivo, HSCs were proposed to be the major cell population in the injured liver transmitting TLR4-induced fibrogenic signals [³].

TNF-α has a dual role in HSC biology. TNF-α can produce antiapoptotic (via NFκB activation), or proapoptotic (via caspase activation) and antiproliferative responses in HSC [¹¹³, ²⁰⁰]. The later effect is mainly attributed to the ability of TNF-α to regulate CD95L expression.

Activation of HSCs in response to fibrogenic liver injury is associated with increase of the basal NFκB activity in comparison with qHSCs [²⁰¹, ²⁰²]. NFκB promotes antiapoptotic signals in aHSCs predominantly via the classic p65 : p50 complex and low levels of a p65 homodimer [²⁰⁰]. Increased basal activity of NFκB in activated HSCs is linked to downregulation of IκB-α [²⁰¹]. Interestingly, as a result of liver injury, elevated levels of TNF-α further stimulate NFκB activity. In turn, NFκB mediates antiapoptotic functions and protects aHSCs from TNF-α-induced apoptosis. TNF-α-induced apoptosis of aHSCs can be achieved in the presence of cycloheximide, or pharmacological inhibition of IkB [²⁰⁰].

Some Jak2-Stat3-signaling cytokines may have a direct effect on aHSCs by facilitating ECM deposition [²⁰³, ²⁰⁴]. Leptin increases collagen production in aHSCs/myofibroblasts in fibrotic liver [²⁰⁵–²⁰⁷] and promotes HSC survival [²⁰⁵]. Treatment with leptin increases the numbers of HSCs in S and G2/M phases of the cell cycle as well as increases cyclin D1 expression. Leptin mediates its function via activation of the Stat3-Jak2 signaling pathway with downstream activation of ERK, AKT and PI3K [¹¹³]. Moreover, other agonists, such as PDGF, EGF and HGF, also activate Stat3 and produce a direct profibrogenic effect on HSCs. As expected, deletion of their corresponding receptors in mice inhibits liver fibrosis [⁶², ⁶³]. Taken together, there is emerging evidence that supports a significant role of Stat3 in stimulating liver fibrosis. However, the specific role of Stat3 in HSC activation using conditional ablation of Stat3 has not been investigated.

It is anticipated that in response to TGF-β signaling, PFs upregulate collagen Type I and activate the Smad2/3 signaling pathway, similar to HSCs and other myofibroblasts [²¹⁶]. Our general understanding of TGF-β signaling suggests that mitogenic signaling synergistically promote the growth and invasion of mesenchymal cells [⁸⁴, ²²³]. Blocking of phosphorylation of Smad2 abrogates the synergistic responses of fibroblasts to TGF-β and mitogens [⁹, ⁸⁴].