It has been established that congenital hepatic fibrosis, and indeed Caroli’s disease closely resemble each other pathophysiologically, in that both occur as a result of ductal plate malformation. The ductal plate is a cylindrical layer of cells that surround a branch of the portal vein, and is the embryonic precursor of the intrahepatic bile ducts, as both interlobular and intralobular bile ductules develop from the ductal plate. Progressive remodeling starts at 12 wk of gestation, and full maturation is usually complete by 20 wk. Arrest of maturation and the lack of remodeling of the ductal plate that occurs as a result leads to the persistence of an excess number of immature embryonic duct structures. This abnormality has been termed the ductal plate malformation. The persistence of these immature duct elements stimulates the formation of portal fibrous tissue, and it is this periportal fibrosis that contributes to the clinical picture of recurrent cholangitis or portal hypertension and associated symptoms (Figure ). Although long standing portal hypertension is known to result in secondary portal vein thrombosis, and eventually portal vein cavernous transformation (PVCT), it is firmly believed that PVCT is actually a component of the disorder, present at the onset rather than developing at a later stage. Embryologically speaking, the development of bile ducts and hepatic vasculature are closely related. The ductal plate malformation has been shown to be associated with a “pollard willow” malformation of the portal vein, which results in too many small and closely branched portal veins, which supports the idea that PVCT may be congenital. Histologically, enlarged portal tracts containing immature ductal plates surrounding several hypoplastic or even obliterated portal vein branches are observed[7
]. In one report, PVCT was observed in almost 50% of patients with congenital hepatic fibrosis, and such patients had relatively larger splenomegaly than those without PVCT, as well as suffering from more frequent bleeding episodes from esophageal varices[8
Pathogenesis of congenital hepatic fibrosis. Embryological and molecular perspective. ET-1: Endothelin 1;
Furthermore, depending on the stage of arrest of maturation, either the small interlobular bile ducts (congenital hepatic fibrosis), or the medium intrahepatic bile ducts (Caroli’s disease) may be involved. Involvement of both simultaneously results in what is known as Caroli’s syndrome. In this context, the clinical picture of Caroli’s disease (recurrent cholangitis) may be so predominant that co-existing congenital hepatic fibrosis may easily be overlooked. A liver biopsy is therefore warranted in all patients with suspected Caroli’s disease to confirm the presence or absence of Caroli’s syndrome[9,10
The hepatic stellate cell (HSC) is at the center of the hepatic fibrotic process associated with liver disease, and has also been shown to play a role in the progression of the disease in congenital hepatic fibrosis. It is widely accepted that transforming growth factor (TGF)-β is a potent growth inhibitory and profibrotic cytokine which plays a pivotal role in the physiological process of wound healing as well as in the pathogenesis of organ fibrosis[11
]. TGF-β expression has been shown to be increased in a wide range of fibrotic diseases. Initiation of HSC activation is primarily induced by TGF-β1 derived from Kupffer cells. TGF-β1 mediates its profibrotic actions by stimulating fibroblasts and related cell types, including the HSC in the liver, to secrete a wide range of extracellular matrix proteins. In pathological conditions this leads to accumulation of fibrotic matrix or in a more physiological context to the efficient healing of wounds[12-14
]. Latent TGF-β is also activated by MMP-9, another product of Kuppfer cells. TGF-β has other important actions, namely its immunomodulatory properties and its antiproliferative effects on epithelial cells, including hepatocytes.
Several studies have attempted to establish the pathophysiological mechanism behind the abnormal and excessive fibrotic response associated with CHF. Degradation of the basement membrane and extracellular matrix (ECM) constituents, and the remodeling of the ECM are important processes of embryonic development. Basal laminar components such as laminin and type IV collagen along with the coordinated expression of proteolytic enzymes are thought to be essential for the normal development of intrahepatic bile ducts[15-18
]. Most of the proteolytic enzymes involved in these processes belong to the matrix metalloproteinases (MMPs) and the serine proteinases, in particular the plasminogen activator (PA)/plasmin system[19,20
]. Both tissue PA (tPA) and urokinase type PA have been shown to contribute to the plasminogen-dependent lysis of basement membrane laminin in human carcinoma cell lines. Furthermore, plasmin contributes to the activation of MMP-9 and MMP-13 which also play an important role in the degradation of basement membrane components including type IV collagen. In a recent study by Yasoshima et al[21
], it was postulated that biliary overexpression of plasminogen and tPA leads to the generation of excessive amounts of plasmin, and subsequent plasmin dependent lysis of the ECM molecules which may contribute to biliary dysgenesis in CHF.
Overexpression of the osteopontin gene has also been implicated in the pathophysiology of biliary atresia, as well as congenital cholestatic syndromes such as CHF and Caroli’s disease. Osteopontin is a stimulant of fibro-inflammation, and its overexpression has been shown to be regulated by the presence of excessive amounts of regulatory factors such as NF-κB and TGF-β1[22
In an effort to establish how the presence of excessive immature bile ducts contributes to the process of fibrosis, Sato et al[23
] managed to demonstrate in a rat model that in the presence of TGF-β1, cholangiocytes acquire mesenchymal features, thus resembling fibroblasts. They speculated that excess production of extracellular matrix molecules by these transformed cells may contribute to the progressive periportal/hepatic fibrosis.
On a different note, a possible role of microRNA has been postulated in the pathogenesis of fibropolycystic disorders involving both the liver and the kidneys. Chu et al[24
] demonstrated decreases in the levels of the microRNA miR15a in the livers of patients with ARPKD, ADPKD and CHF. They reported that this resulted in an increase in the expression of a cell-cycle regulator known as cell division cycle 25A gene product (Cdc25A), which is directly responsible for cellular proliferation and cystogenesis in vitro