DISCUSSION
Although wound-healing is a physiological process aimed at restoring normal tissue structure and function after an injury, it can be more damaging than the insult itself if becomes uncontrolled and excessive (1). In the liver, a dysregulated fibrotic response to tissue injury of various etiologies, including viral infections, toxics, biliary obstruction and nonalcoholic steatohepatitis, occurs and is associated to poor prognosis in several of the chronic hepatic disorders with elevated incidence, morbidity and mortality worldwide (2,3). Evidence indicates that a precise balance between fibrogenic and anti-fibrotic factors must exist to tune adequately the wound-healing response (43). By using two well-characterized experimental models of chronic liver fibrosis, we here recognize cortistatin as an endogenous protective factor. We found that a deficiency in cortistatin predisposes for developing exacerbated fibrotic responses in injured livers after exposition to hepatotoxic compounds, even at low doses, or after cholestatic damage, and to suffer subsequently more severe clinical signs, hepatic damage and increased mortality. Hallmarks of the exacerbated fibrogenic responses observed in injured livers of cortistatin-deficient mice included the excessive occurrence of portal-to-portal fibrous scars, ECM-deposition and activated myofibroblasts.
Our data indicate that cortistatin could acts as an endogenous negative regulator in the activation and/or differentiation of myofibroblast, a major player in the development of pathological hepatic fibrosis (44). Non-parenchymal cells isolated from livers of cortistatin-deficient mice showed excessive presence of cells compatible with an activated myofibroblast phenotype (45,46) characterized by enhanced expression of intracellular αSMA+-stress fibers and production of fibrogenic markers. In agreement with this commitment to myofibroblastic differentiation, the genetic signature of the cortistatin-deficient hepatic non-parenchymal cells displayed increased expression of a significant number genes related to collagen-containing ECM secretion, fiber formation and focal adhesion, but mainly, linked to function and development of muscle, actin cytoskeleton and contractile cellular fibers, including many components of muscular myosin complexes. Moreover, whereas genes, such as myogenin, myomarker (Mymk) and myogenic differentiation 1 (Myod1) that are involved in muscle differentiation/formation, are highly upregulated in cortistatin-deficient cells, the expression of myostatin, a gene that blocks myogenesis, was significantly decreased. This muscle-like phenotype is supported by the fact that cortistatin-deficient cells expressed genes encoding for delta and gamma subunits of cholinergic/nicotinic-receptors, which are solely expressed in muscle cells, which are almost 200-fold upregulated, but did not differentially express other cholinergic-receptor subunits that are present in other cells. Similarly, the gene encoding the muscle-specific carbonic anhydrase isoform Car3, but not other isoforms, was differentially increased (40-fold) in cortistatin-deficient cells. These findings suggest that lack of cortistatin in non-parenchymal hepatic cells favors the generation of myofibroblasts with contractile functions, and treatment with cortistatin impairs their differentiation. This entails important pathological consequences, because contractile myofibroblasts are abundant in the advanced phases of liver fibrotic disorders and are highly resistant to reversion/resolution of the fibrogenic response (47,48).
Numerous evidences support that HSCs are the major sources for activated myofibroblasts (4-6,49). However, several studies demonstrated that other fibrogenic cells could also contribute to myofibroblast generation depending of liver damage etiology (4-6). Whereas fibrosis in hepatotoxic liver injury is attributed to activated HSCs, activated PFs are implicated in liver fibrosis caused by cholestatic liver injury (4-6,49). Our and other studies showed that cortistatin and its receptors are expressed in HSCs and PFs, and therefore, it could act in both fibrogenic cells in an autocrine/paracrine manner (14,18,20). In fact, treatment with cortistatin reversed the activated myofibroblastic phenotype observed in cortistatin-deficient hepatic cells, and impaired the activation of human cell line of HSCs, pointing to these cells as major targets for the anti-fibrotic effect of cortistatin. Moreover, we observed that non-parenchymal liver cells lacking cortistatin showed significant increased expression of genes that are specifically associated to activated HSCs. However, deficiency in cortistatin also enhanced the levels of genes that are mostly expressed in activated PFs. Therefore, cortistatin could act as an endogenous break of activated HSCs and PFs and of their differentiation to activated myofibroblasts, and consequently, as a critical protective factor for developing severe liver fibrosis, independently of the hepatic injury type. Despite this direct effect on fibrogenic hepatic cells, we cannot fully discard the anti-inflammatory activity of cortistatin as an indirect additional mechanism involved in its anti-fibrotic effect in vivo, because inflammation is a major driver of fibrosis in many organs, including liver. However, the fact that cortistatin treatment efficiently reduced hepatic fibrosis when initiated once that the inflammatory response was fully established also supports the capacity of cortistatin to directly limit fibrogenic responses in injured livers.
Our findings have several clinical implications both from diagnostic and therapeutic points of view. The fact that a simple partial deficiency in cortistatin could predispose for developing exacerbated fibrotic responses could be used to anticipate the diagnostic of more severe forms of chronic hepatic disorders. Indeed, we found an inverse correlation between hepatic cortistatin levels and fibrosis/cirrhosis in patients and animals with different types of liver damage, and it will be intriguing corroborating these findings in plasma of patients to consider cortistatin as a potential biomarker of disease prognosis and susceptibility. The deficiency in cortistatin, and therefore the susceptibility to suffer exacerbated fibrosis, could be circumstantial and more or less transitory (i.e., chronic stress, sleep-deprivation) or permanent (i.e., individuals with 1p36 monosomy, the most common subtelomeric terminal-deletion syndrome, which are heterozygous for cortistatin) (50,51). In any case, our study demonstrates that a systemic cortistatin-based treatment would correct easily this deficiency and improve disease progression. Noteworthy is that cortistatin-treatment has a favorable safety profile in humans and demonstrated clinical efficacy in Cushing’s disease (52), and that the interest of pharmaceutical industry in developing cortistatin-based analogues with improved half-life in serum and clinical efficiency has increased lately (53). Previous studies described the therapeutic effect on fibrogenic responses by various agonists that signal through receptors that are recognized by cortistatin, including sstr and GHSR (13-20), suggesting that binding of cortistatin to both receptor-classes could allow a kind of synergic anti-fibrotic effect in liver, a hypothesis that is confirmed here in activated LX2 cells.
In summary, this study provides new insights into the protective function of cortistatin in liver fibrosis, acting as an endogenous break of activation/differentiation of myofibroblasts. Beside as a potential biomarker of disease susceptibility/protection, cortistatin emerges as an attractive candidate for designing anti-fibrotic therapies to treat chronic hepatic disorders of different etiologies.