Polycystic liver diseases are genetic disorders

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

Polycystic liver diseases are genetic disorders that affect mainly the bile duct and the renal tubule epithelia. Autosomal dominant polycystic kidney

disease (ADPKD) is one of the most common inherited diseases,

occurring in 1:400 to 1:1,000 individuals; it is characterized by the formation

of multiple cysts in the kidney, liver, and pancreas. Although

synthetic liver function is usually preserved in ADPKD, severe cyst

complications (mass effect, hemorrhage, infection, or rupture) may

develop and thus require urgent liver transplantation. Autosomal recessive

polycystic kidney disease (ARPKD) and its liver-related phenotypes

Caroli disease (CD) and congenital hepatic fibrosis (CHF) are, by

contrast, rare disorders with an estimated prevalence of 1:20,000 live

births. CD and CHF are characterized by recurrent acute cholangitis and

severe portal hypertension due to an excessive peribiliary fibrosis that

can be further complicated by the development of biliary malignancies.

The even rarer "isolated" polycystic liver disease (PCLD) is phenotypically

similar to ADPKD, except that the kidney is not affected. In all

cases of cystic liver disease, the pathologic condition targets the biliary

epithelium, justifying the inclusion of these forms among the genetic

cholangiopathies [1].

ADPKD is caused by mutations in one of two genes, PKD1 (polycystic

kidney disease 1) (85-90% of the cases) or PKD2 (10-15%)

encoding for polycystin-1 (PC1) and polycystin-2 (PC2), respectively.

AQ1 Polycystins act as mechanoreceptors, chemoreceptors, and calcium

(Ca2+) channels, able to sense changes in apical flow. ARPKD/CD

and CHF are caused by mutations in the PKHD1 (polycystic kidney

and hepatic disease 1) gene, encoding for fibrocystin, a protein

whose functions remain largely unknown. Polycystins and fibrocystin

are expressed in the primary cilia of cholangiocytes. In secretory epithelia,

primary nonmotile cilia are involved in the regulation of multiple

epithelial functions including secretion, proliferation, differentiation,

and interactions with cell matrix. In the liver, cilia are preferentially

expressed by cholangiocytes. Although the impact of ciliary dysfunction

on cholangiocyte physiology is unknown, animal models with

defects in ciliary proteins, such as polycystins, fibrocystins, and polaris,

show different degrees of biliary dysgenesis. In the liver, both ADPKD

and ARPKD/CHF/CD are morphologically characterized by aberrant

development of the biliary epithelium that retains an immature, ductal

plate-like architecture with the formation of multiple biliary microhamartomas

that progressively dilate to macroscopic cysts, scattered

throughout the liver parenchyma.

AQ2 The isolated polycystic liver disease (PCLD) on the other hand is

caused by mutations in PRKCSH, a gene coding for protein kinase

C substrate 80 K-H, also called hepatocystin, or in the SEC63 gene.

SEC63 encodes for a component of the molecular machinery regulating

translocation and folding of newly synthesized membrane

glycoproteins. Hepatocystin and SEC63 are not expressed in cilia,

but in the endoplasmic reticulum, thus cystic diseases of the liver

can also be caused by defects in proteins that are not expressed in


Diseases of the biliary epithelium caused by single-gene defects that

alter a critical physiologic process provide an invaluable clue for understanding

epithelial function and pathophysiology. As a consequence, in

the last few years, interest for polycystic liver diseases has consistently

grown. In this chapter we will review the aspects of cholangiocyte biology

that more closely relate to the pathogenesis and treatment of cystic

diseases of the liver.



The biliary epithelium forms a branching system of conduits within

the liver where bile flows from the hepatocytes to the gallbladder and

intestine. The biliary tree is organized in a complex tridimensional

network that starts at the canals of Hering, located at the limiting

plate with hepatocytes, and forms tubules (<15 μm in diameter) which

gradually converge to create ducts of progressively larger size (up to

300-800 μm): interlobular, septal, major ducts, and hepatic ducts

embedded into the portal space [2]. The biliary tree is lined by

cholangiocytes. These are epithelial cells with absorptive and secretory

properties that actively contribute to bile formation, regulating

its volume, pH, and composition according to physiological needs.

Ductular secretion may account for about 40% of bile flow in humans,

a percentage that can be rapidly increased during the digestive phase.

The morphology of cholangiocytes, as well as their function,

varies along the biliary tree: cholangiocytes in the small interlobular

bile ducts are cuboidal epithelial cells, but become columnar

and mucus-secreting in larger ducts approaching the extrahepatic

portion [3]. This morphological heterogeneity also corresponds to

a functional regional specialization: cholangiocytes lining the large

interlobular and major ducts are mostly involved in secretory functions.

Conversely, cholangiocytes in the smaller bile duct branches,

cholangioles and ducts of Hering, perform other important biological

properties such as the ability to proliferate in response to liver

damage, participate in the inflammatory response, and undergo limited

phenotypic changes. Furthermore, liver progenitor cells are believed

to arise from subpopulations of cholangiocytes residing in the canal

of Hering [4]. This functional specificity is substantiated by the fact

that most cholangiopathies show a site-restricted bile duct injury.

For instance, primary biliary cirrhosis (PBC) targets specifically the

interlobular bile ducts, whereas primary sclerosing cholangitis (PSC)

affects the larger intrahepatic and extrahepatic ducts. Interestingly,

the "small duct" variant of PSC, where damage is restricted to the

finest branches of the biliary tree, has distinct clinical manifestations.

The biliary tree runs along the portal spaces between the hepatic lobules,

in close vicinity to a branch of the portal vein and to one or two

branches of the hepatic artery. While portal blood perfuses hepatocytes

in the hepatic lobules, the cholangiocyte blood supply is provided by

the hepatic artery. Branches of the hepatic artery, at the periphery of

the liver lobule, create a peribiliary vascular plexus (PBP), a network of

capillaries which nourishes the cholangiocytes and eventually merges

into the hepatic sinusoids.

Bile formation starts at the hepatocyte canalicular membrane, with

the secretion of bile acids, other organic and inorganic solutes, electrolytes,

and water. As the primary bile flows through the bile ducts on

its route toward the duodenum, its composition is regulated by the intrahepatic

bile duct epithelium that reabsorbs fluids, amino acids, glucose,

and bile acids, while secreting water, electrolytes, and immunoglobulin

A (IgA) [5, 6]. Fifteen years of investigation have partly unveiled the

complexity of the transport function of cholangiocytes and of its regulation

(see Fig. 2.1). Here we will limit our discussion to the aspects

that may be relevant for understanding cystogenesis in the liver.

Ultimately, secretion and alkalinization in the bile ducts is mainly

associated with a net flux of chloride (Cl-) and bicarbonate (HCO3−)

into the lumen which induces the secretion of water and regulates

bile pH. In contrast with hepatocytes, where the major driving force

for bile production is the active secretion of bile acids by adenosine

triphosphate (ATP)-driven transporters, cholangiocytes secrete fluid

and electrolytes in response to paracrine or endocrine stimuli. A number

of different ion channels and transporters have been identified and

shown to be specifically located at the basolateral or apical membrane.

As in all mammalian cells, the driving force for facilitated membrane

transport in cholangiocytes is provided by the Na+/K+ ATPase,

which actively extrudes sodium (Na+) from the cell and, together with

potassium (K+) channels, maintains the transmembrane potential. At

the basolateral side, the Na+ gradient regulates the Na+/H+ exchanger

isoform 1 (NHE1) and the Na+:HCO−3 symporter (or Na+-dependent

Cl-/HCO−3 exchanger in humans, NCHE) which mediate the reabsorption

of HCO−3 necessary for acid extrusion, while the Na+/K+/2Cl-

cotransporter (NKCC1), a major determinant of fluid secretion, mediates

the chloride uptake into the cell. On the apical side of the cell,

Cl- efflux is mainly mediated by a cyclic adenosine monophosphate

(cAMP) activated, slow conductance, Cl- channel encoded by the cys-

AQ3 tic fibrosis transmembrane conductance regulator (CFTR). The opening

of chloride channels (CFTR) in the apical membrane leads to an efflux

of Cl- and the generation of a lumen-negative potential which induces

the release of water into the lumen through aquaporines (AQP-1 and

AQP-4). The Cl- gradient regulates the Na+-independent Cl-/HCO−3

exchanger (AE2) which extrudes bicarbonate into the bile providing biliary

alkalinization, in accordance with intracellular pH. Other carriers

such as the Na+-dependent glucose transporter (SGLT1), the glutamate

transporter, and the ileal bile acid transporter (iBAT) expressed

on the apical membrane of cholangiocytes mediate the reabsorption of

Fig. 2.1. Secretory function of cholangiocytes and its regulation. Left: secretion

and alkalinization in bile ducts is ultimately associated with a net flux of

Cl- (which induces fluidification) and HCO−3 (alkalinization) into the lumen

mediated by specific transporters localized to the apical or basal membrane of

cholangiocytes. The Na+/K+ pump creates the membrane potential necessary

for cell homeostasis and maintains the Na+ gradient across the membrane

necessary for facilitated transports. At the basolateral side, the Na+/H+

exchanger NHE1 and the Na:HCO−3 symporter NCHE1 mediate the reabsorption

of HCO−3 into the cell and the acid extrusion, while chloride uptake occurs

through the Na+/K+/Cl- cotransporter NKCC1. On the apical side, the Cl- is

released into the bile by cystic fibrosis transmembrane conductance regulator

(CFTR) inducing a parallel osmotic movement of H2O through aquaporines.

The Cl-/HCO−3 exchanger AE2 (located at both the basolateral and the apical

membrane) mediates carbonate release into the bile. Specific Na+-dependent

apical carriers, GT and iBAT, mediate the reabsorption of glutamate and taurocholate,

respectively. Biliary acids are then secreted in the peribiliary plexus

via t-ASBT. Right: Choleretic hormone secretin stimulates cAMP production

by adenylate cyclase with consequent activation of CFTR via PKA mediated

phosphorylation. Cl- released by CFTR promotes HCO−3 secretion by AE2

and bile alkalinization. Similarly luminal purinergic nucleotides can activate a

Ca++-dependent Cl- channel and stimulate secretion. In contrast, somatostatin

decreases bile secretion and alkalinization by adenylate cyclases inhibition.

biliary constituents, such as glucose and glutathione breakdown products

and conjugated bile acids. This is particularly important because

bile acids can stimulate proliferation of biliary epithelial cells. Biliary

bile acids are then secreted in the peribiliary plexus via t-ASBT, a

truncated isoform of the apical sodium-dependent bile acid transporter

(ASBT), or via MRP3 (multidrug-resistant protein 3), a p-glycoprotein.

This cholehepatic circulation of bile acids is also important in the

overall regulation of bile secretion.

The secretory function of the bile ducts is finely regulated by rapid

hormone-mediated signaling. The net amount of fluid and secreted

HCO−3 is determined by the integration of different pro-secretory

(secretin [5], glucagon [7], VIP [8], acetylcholine [9], bombesin [10])

and anti-secretory (somatostatin [11], endothelin-1 [12]) stimuli. All

these hormone signals ultimately act on the adenylyl cyclases (ACs),

the transmembrane enzymes that regulate the intracellular level of the

second messenger cAMP, converting ATP to cAMP. Secretin, the main

choleretic hormone, increases cAMP/PKA (protein kinase A). This

activates CFTR, and consequently stimulates Cl- and HCO−3 efflux and

inhibits the Na+/H+ exchanger (NHE)-dependent Na+ absorption [13,

14]. Cholinergic agonists, β-adrenergic agonists, and HCO−3 -mediated

signals also regulate bile secretion through the cAMP and PKA pathway.

ACs may thus represent an important means of integration of

multiple secretory signals. So far nine different isoforms of AC have

been identified (AC1-9), each displaying tissue-specific expression and

regulation. Interestingly the AC6 isoform was found to be located in

cholangiocyte cilia, thus further suggesting a correlation between ductal

bile secretion and ciliary function [15].

The secretory functions of the biliary epithelium are also regulated by

molecules (such as bile salts, glutathione, and purinergic nucleotides)

secreted by hepatocytes into the canalicular bile and delivered to receptors

and transporters located in the apical membrane of cholangiocytes

[5]. For instance, ATP, which is released into the bile by hepatocytes

or by cholangiocytes themselves, can bind to apical P2Y2 purinergic

receptors and stimulate apical Ca2+-activated Cl- channels and basolateral

Na+/H+ exchanger (NHE-1), thus promoting Cl- efflux into the bile

and basolateral HCO−3 influx [16]. Certain bile acids may also stimulate

cholangiocyte secretion of HCO−3 by inducing ATP secretion by CFTR

and purinergic activation of apical Ca++-activated or volume-activated

Cl- channels [17].


Cholangiocytes possess receptors for a number of cytokines,

chemokines, and growth factors and angiogenic factors that enable

an extensive cross talk with other liver cell types, including hepatocytes,

stellate cells, and endothelial cells [6]. This property becomes

particularly relevant when the liver or the biliary tree is damaged.

In fact, the cholangiocyte compartment can significantly expand in

response to liver injury. Cholangiocyte proliferation occurs in most

pathologic conditions, including cholestasis, viral hepatitis, and hepatic

necrosis, and represents a key mechanism of regeneration and repair,

which ensures the integrity of the biliary tree following liver damage.

These "reactive" or "activated" cholangiocytes are believed to

arise from a progenitor cell compartment located in close contact

with the smallest radicals of the biliary tree, the terminal cholangioles

at the canals of Hering. Reactive cholangiocytes show a less differentiated

secretory phenotype, but acquire the capability to secrete a

number of proinflammatory and chemotactic cytokines and growth factors.

They can recruit inflammatory and mesenchymal cells and induce

them to proliferate and to produce extracellular matrix (ECM) components

[18]. There are, in fact, intimate contacts and exchange of

signals between mesenchymal cells and reactive cholangiocytes. While

mesenchymal cells are considered the effectors of fibrosis, reactive

cholangiocytes are considered the "pacemaker of liver fibrosis" [19].

The list of cytokines, chemokines, inflammatory factors and growth factors,

and receptors that mediate the epithelial/mesenchymal cross talk

in the liver is continuously increasing. It includes interleukin-6 (IL-6),

IL-8, tumor necrosis factor-α (TNFα), interferon-γ (IFNγ), monocyte

chemotactic protein-1 (MCP-1), cytokine-induced neutrophil chemoattractant

(CINC), and nitric oxide, which regulate the immune activity

of lymphocytes and polymorphonuclear cells. Reactive cholangiocytes

also produce growth factors such as vascular endothelial growth factor

(VEGF), endothelin-1 (ET-1), platelet-derived growth factor-BB

(PDGF-BB), transforming growth factor-β2 (TGF-β2), and connective

tissue growth factor (CTGF).

In addition to establishing paracrine communications with mesenchymal

cells, cholangiocytes may also participate in the generation of

liver fibrosis through a process of epithelial to mesenchymal transition

(EMT). EMT is a process of cellular reprogramming whereby epithelial

cells acquire some of the phenotypic and functional characteristics of

mesenchymal cells, such as the expression of fibroblast-specific markers

(FSP-1, vimentin), the ability to migrate by locally dismantling the

basement membrane upon which the epithelial sheet resides, and the

ability to generate different connective tissue components (fibronectin,

collagen, elastin, tenascin). EMT may thus contribute to the accumulation

of activated fibroblasts in association with the loss of bile ducts.

This biological process has also been described in the pathogenesis of

organ fibrosis in the kidney [20] and lung [21]. Recent studies suggest

that EMT may also be involved in liver fibrosis [22, 23].

In response to liver injury, reactive cholangiocytes also acquire a

neuroendocrine-like phenotype and express receptors that enable them

to respond to regulation by neural terminations. In fact, reactive cholangiocytes

express β1 and β2 adrenergic receptors, the M3 acetylcholine

receptor [24], serotonin 1A and 1B receptors [25]. During cholestasis,

2 Cholangiocyte Biology as Relevant to Cystic Liver Diseases

cholangiocytes can also directly secrete serotonin, thus further limiting

the growth of bile ducts by an additional autocrine inhibitory

loop. Furthermore, in experimental cholestasis, cholangiocytes secrete

neuropeptides, such as nerve growth factor (NGF), that can stimulate

cholangiocyte proliferation [26]. A large number of other regulatory

neurotransmitters and neuropeptides are expressed by reactive cholangiocytes,

but will not be mentioned here. The role of neuroendocrine

signaling in cyst growth in the polycystic liver diseases has not been

experimentally addressed yet.



A novel concept in biliary physiology is that primary cilia are involved

in the regulation of fundamental biological activities including cell

differentiation, proliferation, and secretion [27, 28]. The presence of

primary cilia at the apical domain of cholangiocytes has long been

known; however, their role in biliary physiology remained undefined.

The discovery that mutations in several proteins relevant to ciliary function

are associated with cystic diseases in several organs, including

kidney, liver, and pancreas, strongly revived interest in the function of

this organelle [29].

Primary cilia of cholangiocytes, as opposed to cilia of lung epithelial

cells, are nonmotile, but can be bent in response to changes in luminal

fluid flow, thereby transducing a mechanical force into an intracellular

calcium signal [30]. Masuyk et al. investigated ciliary function in the

biliary epithelium using microperfused rat intrahepatic bile duct units

(IBDU) and showed that changes in luminal flow increased [Ca2+]i and

inhibited forskolin-stimulated cAMP production [15]. These changes

were significantly inhibited by removal of cilia with chloral hydrate

or by silencing of ciliary protein (such as PC1 or PC2) or of adenylyl

cyclases 6 (AC6) [15, 31]. PC2 is believed to function as a nonselective

Ca2+ channel, activated by PC1 through its C-terminus, while

AC6 is a Ca2+-inhibitable AC expressed in cilia, which interacts with

PC2. Ciliary dysfunction in ADPKD would thus reduce intracellular

calcium levels, thereby increasing AC6 activity and the levels of

cAMP. It is well known that cAMP stimulates cholangiocyte secretion

through the Src/Ras/MEK/ERK1-2 pathway and thus can promote

cyst formation. Additionally, PC1 may have a direct transcriptional

effect mediated by the proteolytic cleavage and nuclear translocation

of its carboxy-terminal tail to the nucleus [32]. In kidney cells, PC1

was also shown to regulate the signal transducer mTOR [33, 34]. The

constitutive activation of both these pathways results in progressive cyst

growth [33].

Much of the attention has been focused on ciliary function; however,

it should be noted that morphologic alterations of cilia have

not been consistently reported in cystic liver diseases. Recent electron

microscopy studies on human ADPKD liver described heterogeneous

abnormalities on the apical surface of cyst epithelium, depending on

cyst size [35]. The epithelial cells lining small cysts (1 cm) showed a

relatively normal apical surface with cilia and microvilli represented in

the expected number and size. On the other hand, the apical surface

of medium-sized cysts (2-3 cm) showed areas free of microvilli with

rare and shortened cilia, while large hepatic cysts (3 cm) totally lack

microvilli and primary cilia. These progressive morphological abnormalities

of primary cilia and microvilli may represent a mechanic effect

of enhanced endoluminal pressure during cyst growth, rather than a

primary consequence of defective ciliary proteins.

Most "ciliary proteins" are not exclusively expressed in cilia. For

example, PC2 is also strongly expressed in the endoplasmic reticulum

(ER), where it interacts with ryanodine or InsP3 receptors to regulate

ER and cytoplasmic calcium levels. Furthermore, the proteins encoded

by the PRKCSH and SEC63 genes are not expressed in cilia, but in the

ER. Defects of other enzymatic activities associated with the ER, such

as xylosyltransferase 2, an initiator of heparin sulfate and chondroitin

sulfate biosynthesis, have been linked to the development of renal and

liver cysts [36].

It is interesting that cyst formation is not a common reaction of the

biliary epithelium to liver damage. In obstructive cholestasis the biliary

epithelium reacts by forming multiple branching tubules, while in

inflammatory biliary disease the epithelium forms a ramified mesh of

reactive cells that for the most part lack a lumen. On the other hand,

conditional PC1 or PC2 mice in which the genetic deletion is induced

after birth develop multiple cysts in the liver, indicating that polycystins

remain key determinants of biliary architecture during adult life.

An important property of epithelial sheets is planar cell polarity -

the capacity to orient the axis of cell division in such a way that the

growth of the epithelial sheet is "polarized" within the plane of the cell

sheet. This means, for example, that the epithelial cells of the kidney

align their mitotic spindle along the tubule axis so that the daughter

cell will be inserted in a way that elongates the tubule rather than

increases the size of its lumen. Interestingly, in rodent models with

low Pkhd1 expression, the orientation of the mitotic spindle is distorted

[37]. Pkhd1 is also localized to the basal body [29, 38-40], a

sub-cellular organelle that originates from the mother centriole in the

centrosome and is responsible for the assembly of the cilium. Centrioles

organize the mitotic spindle and serve as microtubule organizing center

(MTOC). An anomalous cell division would result in tubule enlargement

rather than tubule elongation. Direct experimental evidence for

this model in cholangiocytes has not yet been produced.



Mechanisms leading to the progressive growth of liver and kidney

cysts are being actively investigated. In both autosomal dominant and

recessive forms, liver cysts arise from an aberrant development of intrahepatic

bile duct epithelium. During cyst expansion different factors,

including excessive fluid secretion, extracellular matrix remodeling,

increased proliferation of the epithelial cells lining the cyst, and the

pericystic vasculature, variably take part to promoting the progressive

cyst growth.

Altered Biliary Developmental Program (Ductal Plate


The developmental role of polycystins is evident from studies in genetically

modified mice and was clear to early pathologists that recognized morphology suggestive of a blockage in ductal plate maturation. They

classified cystic liver disease as a malformative condition.

The ontogenesis of the intrahepatic biliary tree begins around the

8th week of gestation and proceeds centrifugally from the ileum to the

periphery of the liver. Still immature at birth, the biliary tree completes

its development during the first year of life. Its formation starts when the

periportal hepatoblasts surrounding branches of the portal vein undergo

a phenotypic switch and assemble into a sheath of small flat epithelial

cells, called "primordial ductal plate." Over the following weeks,

some segments of the ductal plate perimeter are duplicated by a second

layer of cells (double layered ductal plate), while the remaining single

layer portions are deleted by apoptosis. The double layered ductal

plate then dilates and starts to form a tubular structure which is incorporated

into the mesenchyme of the developing portal space (migratory

stage) and later undergoes a branching process to form the biliary tree

[41]. Ductal plate remodeling during fetal and postnatal development

is thus a fine balance between proliferative and apoptotic processes.

A failure in ductal plate remodeling causes a number of developmental

cholangiopathies, hence classified as ductal plate malformations (DPM)

of which PKD is one [42].

However, in humans, cysts appear to develop throughout adult life.

Mice with conditional knockout of Pkd1 or Pkd2 show a progressive

formation of liver and renal cysts reminiscent of human diseases even

when the induction is performed weeks after birth [43-45]. This indicates

that there is also a role for polycystin in maintaining a normal

biliary architecture during adult life. It is interesting to note that there

are fundamental structural differences between ADPKD and ARPKD

liver cysts. In ADPKD, the nascent cysts detach from the original duct

and form autonomous structures that no longer communicate with the

duct; in ARPKD, cysts mostly remain open. This fundamental difference

helps to explain the different clinical manifestations between


Altered Epithelial Fluid Secretion

Studies performed by Everson et al. on ADPKD patients have shown

that hepatic cysts are able to generate ion secretion under basal

conditions and when stimulated with secretin [46]. The increased intraluminal

pressure may contribute to cyst expansion as secretion into

the closed cyst would stretch the lining epithelial cells and induce

proliferation. In cell culture models of epithelial cysts, increasing intraluminal

pressure increased the rate of cell proliferation [47, 48]. Stretch

may activate apical secretion of purinergic agonists [49], major players

in the regulation of cholangiocyte secretion and proliferation [16,

50]. Studies in kidney cyst cells from both ARPKD and ADPKD have

shown that the cyst epithelium releases substantial amounts of ATP in

culture [51] and expresses P2X and P2Y purinergic receptors, along

with Ca2+-stimulated Cl- channel activation [52], leading to cystic

fluid accumulation. It is presently unclear if these findings apply to

cholangiocytes as well and if purinergic signaling is actually different

from the normal epithelium. Cystic cholangiocytes appear to have

an altered intracellular Ca2+ homeostasis, so without direct experimental

evidence, it is difficult to predict the overall role of purinergic


On the other hand, there is important cross talk between the cAMP

and the Ca2+-dependent pathways. Increased cellular cAMP content

due to the reduced Ca2+-dependent inhibition of AC6 would favor

CFTR-dependent secretory events. It is interesting to note that the

severity of ADPKD was milder in two cases in which the diseases coexisted

with cystic fibrosis [53], a disease that impairs CFTR-dependent

Cl- secretion [54]. Consistent with these reports and the role of CFTR-dependent

secretion, small molecule CFTR inhibitors slowed cyst

growth in experimental polycystic kidney disease [55, 56].

2 Cholangiocyte Biology as Relevant to Cystic Liver Diseases

In spite of the evidence of active fluid secretion in cystic kidney and

liver disease, there is no definitive proof that unregulated fluid secretion

is actually the major pathophysiologic mechanism leading to cyst

growth in the liver. Furthermore to account for the very slow growth

rates of the cysts, the net difference between absorption and secretion

should be very subtle and constant in the face of increasing intraluminal

pressure [57].

Cholangiocyte Proliferation

Most observations indicate that increased proliferative activity of the

cystic epithelium may be the major determinant of cyst growth. As

discussed earlier, cholangiocytes lining liver cysts present functional

similarities with the reactive ductules, and express a vastly similar array

of growth factors, growth factor mediators, cytokines, and chemokines.

While in reactive ducts this property facilitates progenitor cell-mediated

liver repair, their expression in cystic cholangiocytes likely represents

the phenotypic and functional signature of a relative loss of differentiation.

For example, we have shown that the pattern of angiogenic factor

expression by cystic cholangiocytes in ADPKD is similar to that of the

ductal plates.

Increased epithelial levels of cAMP is one of the factors determining

the increased proliferative activity of cystic cholangiocytes and fluid

secretion [58-60]. The increased cAMP level in cystic epithelia may be

related to changes in the intracellular Ca2+ homeostasis and the cross

talk with the Ca2+-inhibitable adenylate cyclase 6. The relevance of

cAMP in promoting cholangiocyte growth is demonstrated by the fact

that in vivo treatment of normal rats with the adenylate cyclase (AC)

stimulator forskolin induces cholangiocyte proliferation in association

with increased activity of protein kinase A (PKA) and the activation

of the cAMP/PKA/Src/ERK1/2 cascade [61]. Inhibition of cAMP production

has been exploited as a therapeutic strategy in ARPKD. In

particular somatostatin and its analogs, such as octreotide, were used

to inhibit the secretin-induced increases in cAMP levels observed in

cholangiocytes from bile duct-ligated (BDL) rats [62]. Somatostatin

represses AC function through its receptor SSTR2, which is expressed

in the liver only by cholangiocytes [63]. Masyuk et al. showed that

octreotide, given in vivo to PCK rats, reduced liver and kidney weight,

hepatic and renal cyst volume and fibrosis, and diminished the rate of

cell proliferation in hepatic and renal epithelia [60].

The cystic fluid from patients with ADPKD also contains elevated

cell proliferation and cyst expansion [64-66]. These include high levels

of IL-8, IL-6, EGF, and VEGF [65, 67, 68]. Histological studies

have shown a marked over-expression of estrogen receptors, insulin-like

growth factor (IGF), IGF-receptor, growth hormone receptor, and

pAKT [35, 68, 69]. Estrogens and IGF-1 are major factors able to

induce proliferation of cyst epithelium, given their capability to activate

specific proliferative and/or survival pathways. Like cAMP, estrogens

[70], IGF-1 [71] and VEGF [71] may promote cholangiocyte growth via

the ERK pathway, which is the main pathway of regulation of cholangiocytes

proliferation. However, the strong immunoreactivity for IGF1,

IGFR-1, and pAKT in liver cysts from ADPKD patients also indicate

activation of the PI3-kinase pathway. Through this pathway IGF1 can

activate mTOR (mammalian target of rapamycin) and thus promote cell

proliferation via cyclins. In fact, phospho-mTOR is over-expressed in

the liver cystic epithelium of ADPKD patients and mouse models. It

is interesting to note that mTOR can also stimulate HIF1α-dependent

VEGF secretion.

Cystic cholangiocytes over-express VEGF, angiopoietin-1, and their

cognate receptors VEGF receptors 1, 2 and Tie-2 [68]. Thus, VEGF and

angiopoietin-1 may exert an important autocrine proliferative effect and

the growth of liver cysts.

Autocrine and Paracrine VEGF Signaling

Ductal plate malformations are frequently associated with an abnormal

vasculature ramification: a morphologic feature known as "pollard willow

pattern" which derives from an alteration in the normal relationship

between bile ducts and portal vascular structures. The close anatomic

relationship between intrahepatic bile ducts and hepatic arterial vascularization

is already evident during the developmental stages and it

appears to be crucial for the maintenance of the integrity and function

of the biliary epithelium [72]. In immunohistochemical studies, Fabris

et al. investigated the expression of angiogenic growth factors (VEGF

and angiopoietins) and their cognate receptors during biliary and arterial

development in humans and showed that, during development,

VEGF released by cholangiocytes promotes the angiogenesis of the

PBP in close vicinity to the maturing bile ducts [73]. Likewise, the PBP

support is also fundamental during ductular reaction in response to liver

damage or disease. Indeed, in many forms of liver injury, cholangiocyte

proliferation is accompanied by an increase in number of the surrounding

vascular structures [74]. In rat models of cholangiocyte proliferation

2 Cholangiocyte Biology as Relevant to Cystic Liver Diseases

(common bile duct ligation), Gaudio et al. observed that a marked proliferation

of the PBP became apparent only after the extension of the

bile duct system occurred, underscoring the role of proliferating cholangiocytes

in directly promoting angiogenesis [75]. Indeed they showed

a significantly higher expression of VEGF in cholangiocytes isolated

from BDL rats compared to normal rats [71].

VEGF is one of the most potent angiogenic factors and its role in

vascular proliferation associated with tumor growth or wound healing

has been widely documented in different organs. Also in human diseases

related to developmental ductal plate malformation (i.e., PKD)

the dysmorphic bile ducts are surrounded by hyperplasic vascular structures.

Fabris et al. showed that in the cystic biliary epithelium of

fibropolycystic liver diseases (ADPKD and Caroli disease), VEGF and

angiopoietin-1 are markedly up-regulated, together with their receptors

VEGFR2 and Tie2, and their enhanced expression is closely related

to the microvascular density around biliary cysts [68]. In polycystic

diseases, the biliary epithelium retains thus an immature phenotype

characterized by up-regulation of VEGF and angiopoietins. The growing

cysts stimulate angiogenesis to meet their need of vascular supply

for metabolic support.

VEGF production is controlled in most tissues by hypoxia-inducible

factor 1 (HIF-1), one of the key regulators of oxygen homeostasis.

HIF-1 is a transcription factor active in hypoxic conditions, and the loss

of microvillar structure and decreased microvillar density in ADPKD

liver cyst epithelium are also features consistent with an ischemic

damage [35]. However, isolated cystic cholangiocytes also overproduce

VEGF, indicating that this is the direct result of the loss of

PC1 and PC2 function, rather than the consequence of the cystic

epithelium becoming hypoxic (C. Spirlì et al., submitted). Mice deficient

in PC2, in particular, show a severe liver phenotype with higher

proliferation rate of the cystic epithelium and higher expression of

VEGF and its receptor VEGFR-2, pERK1/2 and HIF1α, suggesting

that PC2 acts as repressors of the Raf/MEK/ERK cascade in physiological

conditions and the lack of its function leads to activation of this

pathway and the consequent increase in proliferation (C. Spirlì et al.,


The significance of this effect is even more relevant in light of

the fact that many of the growth factors that stimulate cholangiocytes

appear to act through this pathway. The list of HIF-1-regulated genes

is ample and includes genes coding for proteins involved in angiogenesis,

energy metabolism, erythropoiesis, cell proliferation and viability,

vascular remodeling, and vasomotor responses. Interestingly, HIF-1α

transcription can also be stimulated in normoxic conditions by a number

of growth factors, cytokines, and extracellular mediators (IL-1, IL-6,

EGF, HGF, TGFβ, 17-β-estradiol, IGF-1) which can stabilize or phosphorylate

HIF-1α via PI3K/AKT/tuberin/mTOR or Raf/MEK/ERK or


Extracellular Matrix Remodeling

In the normal liver ECM is scarce, and it is essentially composed of

elastin, fibronectin, collagen type I, and collagen types III, IV, V, and

VI in low quantity [76]. However, progressive accumulation of ECM

components with an altered composition in the portal tract is a common

feature in ADPKD, where a remodeling of the ECM is a prerequisite to

allow the expansion of the cyst wall [76, 77]. These ECM abnormalities

have also been found in the renal interstitium. The progressive establishment

of fibrosis within the renal interstitium as well as within the

hepatic portal space is particularly abundant in ARPKD, where dense

fibrotic tissue is formed in close vicinity to aberrant bile ducts. In congenital

hepatic fibrosis, Ozaki et al. recently showed that connective

tissue growth factor (CTGF) is diffusely retained in the heparan sulfate

proteoglycan web in the portal tract where it can be responsible for

non-resolving fibrosis due to persistent activation of many mast cells

and portal myofibroblasts [78]. Not surprisingly, the growth of liver

cysts requires the enhanced degradation of ECM induced by an altered

interplay between matrix metalloproteinases (MMPs) and their specific

tissue inhibitors (TIMP). MMPs are typically involved in the breakdown

of extracellular matrix in embryonic development as well as in

tissue repair and remodeling. IL-8, a cytokine that has been shown to be

up-regulated in the liver cyst epithelium of animal models of ARPKD,

stimulates MMP2 and MMP9 production by endothelial cells and portal

myofibroblasts [79, 80].

Studies in PKD rodent models [81, 82] and in cultured mouse renal

tubular cells [83] showed that MMP2 was consistently increased in

cystic epithelial cells: in the PKD mouse model (C57BL/6 J-cpk),

Rankin et al. observed an increased kidney cystic cell content of

MMP-2 both in vivo [81] and in vitro [83]. Once isolated, cultured

cells were able to migrate through collagen gels, further indicating

they possess proteolytic activity. In addition, high MMP2, MMP9,

MMP3, TIMP1, and TIMP2 levels were found in the culture medium

[83]. Thus, increased expression of MMPs contributes to the overall

reorganization and restructuring of the ECM necessary for cyst




In this chapter, we reviewed the aspects of cholangiocyte physiology

and pathophysiology that are relevant for polycystic liver diseases. In

particular, we have reviewed the anatomy and the function of the biliary

tree and the mechanisms of cholangiocyte secretion and its regulation.

We have discussed the role of some of the chemokines, cytokines,

and growth factors expressed by cholangiocytes in pathologic conditions

and in polycystic liver diseases. Many of these factors play a

role in liver cysts growth. We have also discussed the signaling pathways

that mediate cholangiocyte function. Each pathway represents a

potential target for therapies aimed at reducing the growth of liver cysts.

For example, the VEGF and HIF-1α signaling pathways could be targeted

by inhibiting VEGFR2 or using blocking antibodies to VEGF

receptors. We have shown that the treatment of ADPKD mice models

with VEGFR-2 inhibitors significantly reduced liver cysts area and

decreased the VEGF-induced pERK-1/2 activation providing a proof of

concept for the potential use of anti-angiogenic therapy in polycystic

liver diseases (C. Spirlì et al., submitted).