University of Groningen

Cystic fibrosis liver disease and the enterohepatic circulation of bile acidsBodewes, Frank

IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite fromit. Please check the document version below.

Document VersionPublisher's PDF, also known as Version of record

Publication date:2014

Link to publication in University of Groningen/UMCG research database

Citation for published version (APA):Bodewes, F. (2014). Cystic fibrosis liver disease and the enterohepatic circulation of bile acids. [S.l.]: [S.n.].

CopyrightOther than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of theauthor(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).

Take-down policyIf you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediatelyand investigate your claim.

Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons thenumber of authors shown on this cover page is limited to 10 maximum.

Download date: 05-09-2020

CYSTIC FIBROSIS LIVER DISEASE

AND THE ENTEROHEPATIC

CIRCULATION

OF BILE ACIDS

FRANCISCUS ALBERTUS JOHANNES ANDREAS BODEWES

The work described in this thesis was performed at the Department of Pediatrics, Center for Liver,

digestive and Metabolic diseases, Beatrix Children’s Hospital, University Medical Center Groningen, the

Netherlands

The author gratefully acknowledges the unrestricted financial support for printing of this thesis by:

ABBOTT B.V.

ACTELION PHARMACEUTICALS NEDERLAND B.V.

DR FALK PHARMA BENELUX B.V.

NUTRICA BABY- EN KINDERVOEDING

TRAMEDICO

RIJKSUNIVERSITEIT GRONINGEN

Cover: Tiger and Turtle Landmarke Angermund - Duisburg

Boek (Book) ISBN: 978-90-367-6828-3

Eboek : PDF zonder DRM (PDF without DRM) ISBN: 978-90-367-6827-6

Copyright © F.A.J.A. Bodewes, 2014

All rights reserved, No part of this thesis may be reproduced, distributed, stored in a retrieval system, or

transmitted in any form or by any means, without prior permission of the author ,or when appropriate,

of the publisher of the publications

Cystic fibrosis liver disease and the enterohepatic circulation of bile acids

Proefschrift

ter verkrijging van de graad van doctor aan de

Rijksuniversiteit Groningen

op gezag van de

rector magnificus, prof. dr. E. Sterken

en volgens besluit van het College voor Promoties.

De openbare verdediging zal plaatsvinden op

woensdag 19 maart 2014 om 16.15 uur

door

Franciscus Albertus Johannes Andreas Bodewes

geboren op 14 februari 1966

te Eelde-Paterswolde

Promotor:

Prof. dr. H.J. Verkade

Copromotor:

Dr. H.R. de Jonge

Beoordelingscommissie:

Prof. dr. G.H. Koppelman

Prof. dr. E.E.S. Nieuwenhuizen

Prof. dr. R.J. Porte

Paranimfen:

Bertjan Bodewes

Bas Morselt

Ik draag dit proefschrift op aan mijn lieve ouders

Jan Bodewes en Gerry Bodewes-Valens

CONTENTS OF THIS THESIS

CHAPTER 1

General introduction

CHAPTER 2

Persistent fat malabsorption in cystic fibrosis; lessons from patients and mice

Published

CHAPTER 3

Increase of serum gamma glutamyltransferase (GGT) associated with the development of cirrhotic cystic fibrosis liver disease

Submitted

CHAPTER 4

Cholic acid induces a Cftr dependent biliary secretion and liver growth response in mice.

Submitted

CHAPTER 5

Bile salt metabolism in a CF mice model with spontaneous liver disease

Submitted

CHAPTER 6

Ursodeoxycholate modulates bile flow and bile salt pool independently from Cftr in mice

Published

CHAPTER 7

General discussion and summary

CHAPTER 8CHAPTER 8

Nederlandse discussie en samenvatting

Curriculum Vitae

Dankwoord- acknowledgements

CHAPTER 1

GENERAL INTRODUCTION

Chapter 1

10

1

1) THE CLINICAL PERSPECTIVE

CYSTIC FIBROSIS DISEASE

Cystic Fibrosis (CF) is a severe, lifespan limiting, disease. Cystic fibrosis is one of the most

frequent autosomal inherited diseases in the world. The incidences differ globally according

to regional genetic variations (1). In the Netherlands around 1:1500-6000 inhabitants suffer

from CF (2, 3). The disease is usually already manifest at birth and progresses with age. To

date the median survival of CF patients is around 40 years (4). Most patients die from end

stage lung disease. Although severe lung disease dominates the clinical picture, CF is a multi-

organ disease. In particular diseases of intestine, pancreas and liver can be serious and

potentially life threatening (figure 1.).

In 1989, the CF disease causing gene was identified and named “cystic fibrosis transport

regulator” (CFTR). The CFTR gen encodes for the CFTR protein (5, 6). The CFTR gene is

localized on chromosome 7(7). Cystic fibrosis can be caused by a variety of mutations in the

CFTR gene. To date, over 1900 different mutations in the CFTR gene, are identified (8). The

CFTR gene mutation can be divided into separate groups. A mutation in one copy of the CFTR

gene that always causes CF, as long as it is paired with another CF-causing mutation in the

other copy of the CFTR gene, is a CF-causing mutation. A mutation in one copy of the CFTR

gene that does NOT cause CF, even when it is paired with a CF-causing mutation in the other

copy of the CFTR gene, is a non CF-causing mutation. A mutation that may cause CF, when

paired with CF-causing mutation in the other copy of the CFTR gene, is a mutation of varying

clinical consequence. A mutation for which we do not have enough information to determine

whether or not it falls into the other three categories is a mutation of unknown significance

(9).

The disease causing mutations can be divided into 5 different mutations classes according to

the type of malfunction of the CFTR protein (11). Based on this classification, difference in

clinical disease presentation and severity can be recognized (figure 2). The various mutation

classes offer different potential therapeutic targets for the treatment of Cystic fibrosis (12).

CF can develop if a person carries a disease causing mutations in each CFTR allele. The most

common CFTR mutation in humans is the 508del gene variation (13). In the 508del mutation

one nucleotide T, on the 508 position of the gene, is replaced by a G nucleotide. 508del is a,

so called, class 2 mutation. These class 2 mutations cause an almost complete failure (less

than 5% of normal) of the CFTR function, leading to the typical severe phenotype with

General introduction

11

1 1

progressive pulmonary disease, complete exocrine pancreatic insufficiency, diabetes and

cirrhosis (14).

CFTR is a cell membrane protein localized in various cell and in particularly in all epithelial

tissues. The protein can, for instance, be found in the alveolar cells of the airways, the

ductular cells of the pancreas, the enterocytes of the intestine, the cholangiocytes of the bile

ducts but also in the sweat glands of the skin(15). Mutations in the CFTR gene are directly

responsible for the symptomatology and disease development in these organs.

Figure 1. Cystic fibrosis is a multi-organ disease. Because so many bodily functions rely on normal water flow, a

disruption in water flow can cause a number of devastating effects, as shown in the "Manifestations of Cystic

Fibrosis" image above (illustration adapted from: Wikimedia Commons, 2011)(10). In particular diseases of intestine,

pancreas and liver can be serious and potentially life threatening.

In epithelial cells, CFTR functions as a chloride ion (Cl-), trans-membrane transporter protein

(17). CFTR actively pumps Cl- across the cellular membrane by concomitant with ATP

hydrolysis. The Cl- transport serves various roles in the different epithelia. In the lungs, the Cl

-

transport induces a Cl-ion gradient across the cell membrane (18). Based on this gradient,

Chapter 1

12

1

bicarbonate (HCO3-) is passively exchanged against Cl

- and transported out of the alveolar cell

in to alveolar lumen. Here HCO3- here serves a crucial role in maintaining viscosity and fluidity

the alveolar fluid layer. Disturbance of this HCO3- transport function causes a thick, highly

viscous fluid layer in the alveoli. The sticky secretion impairs lung mucus clearance and

thereby an increased susceptibility for bacterial pulmonary infections in CF disease. CFTR

functions as a Cl- channel in different epithelia (19). However, the contribution of CFTR

channel dysfunction, in disease development in different organs, for example, intestine and

liver, is not clear.

Figure 2. Molecular consequences of CFTR mutations. a, CFTR correctly positioned at the apical membrane of an

epithelial cell, functioning as a chloride channel. b, Class I. No CFTR messenger ribonucleic acid or no CFTR protein

formed (e.g., nonsense, frameshift, or splice site mutation). c, Class II. Trafficking defect. CFTR messenger ribonucleic

acid formed, but protein fails to traffic to cell membrane. d, Class III. Regulation defect. CFTR reaches the cell

membrane but fails to respond to cAMP stimulation. e, Class IV. Channel defect. CFTR functions as altered chloride

channel. f, Class V. Synthesis defect. Reduced synthesis or defective processing of normal CFTR. Chloride channel

properties are normal (Illustration from The Journal of Pediatrics, 127, 5, 1995)(16)

General introduction

13

1 1

GASTRO-INTESTINAL AND HEPATIC DISEASE IN CYSTIC FIBROSIS

CF presents with clinical symptomatology in various abdominal organs like pancreas, intestine

and liver (20). The clinical phenotype and contribution of different organ systems varies in CF

patients (21). In cystic fibrosis the clinical, organ specific, presentation and phenotypical

penetration depend on gene modifiers. Gene modifiers are genetic or environmental factors

that determine the actual organ specific phenotype. Different disease presentations in CF are

in greater or lesser extent dependent on the influence of gene modifiers (figure 3.). The most

important and relevant gastro-intestinal disease presentations in CF are exocrine pancreatic

insufficiency and cirrhosis (22, 23). However, also specific intestinal diseases, like neonatal

meconium ileus, the distal intestinal obstruction syndrome (DIOS) and intestinal fat- and bile

salt malabsorption, are frequently found CF patients (24-26).

Figuur 3. The relative contribution of modifier genes, CFTR, and environment on phenotype. [Adapted from Borowitz

et al.]1Borowitz D et al. Gastrointestinal outcomes and confounders in cystic fibrosis. J Pediatr Gastroenterol Nutr.

41:273-85 (2005)

EXOCRINE PANCREATIC INSUFFICIENCY AND FAT MALABSORPTION IN

CYSTIC FIBROSIS

Intestinal fat malabsorption is a serious clinical manifestation of CF. The decreased absorption

of dietary fats impairs the development and maintenance of a healthy nutritional status in

particular in growing children (27). In CF nutritional status relates to the prognosis and

Chapter 1

14

1

survival (28). Therefore, high caloric feeding and treatment of the intestinal fat malabsorption

are cornerstones in the treatment of cystic fibrosis (29).

The leading cause of the intestinal fat malabsorption in CF is exocrine pancreatic insufficiency

(PI)(22). CF causes fibrotic degeneration of the acinar tissue of the pancreas secondary to

destruction of the ductular structures due to loss of CFTR function (30). The fibrotic pancreas

is no longer able to excrete pancreatic enzymes including lipases and proteases essential for

intestinal fat and protein absorption. The pancreatic destruction, partly based on auto-

digestion, starts already in utero and, in most patients with a severe genotype, this develops

into complete PI already during infancy (31).

PI is treated with pancreas enzyme replacement therapy (PERT) (32). These products contain

pancreatic enzymes mostly of animal origin (33). However, bioengineered products are

currently developed and coming to the market (34). PERT is individually dosed based on the

dietary fat intake and its effects on intestinal fat absorption. Despite optimizing and

maximizing PERT many patients a degree of fat malabsorption persists (35, 36).

LIVER DISEASE IN CYSTIC FIBROSIS

The earliest form of liver involvement in CF is neonatal cholestasis (37). Infants can present

with prolonged jaundice and vitamin K dependent coagulopathy. Liver histology displays signs

of biliary obstruction, portal fibrosis and inflammation with bile duct proliferation. Mucous

plugs in bile ducts are described (38). The pathogenesis of CF related neonatal cholestasis is

not known, but it might be related to temporary biliary obstruction in the neonatal period.

The disease is mostly self-limiting and does not seem to be related to the development of

severe liver CF related liver disease later in life. To date CF related neonatal cholestasis is

probably earlier recognized in countries with a neonatal screenings program for CF (39).

CF patients frequently show signs of hepatic steatosis or fatty liver (40). Steatosis is often

diagnosed based during routine ultrasonography of the liver in the clinical follow up of CF

patients. An elevated fat fraction has also been observed in over 80% of subjects with cystic

fibrosis on MRI scanning (41). Histologically proven steatosis can be found in up to 35% of CF

patients (42). There is no direct correlation between steatosis and the later development of

cirrhotic liver disease (43). Steatosis in infants with CF is often suggested being related to

poor nutritional status in particular early or late recognized disease (44). It has further been

suggested that steatosis in CF has to do with to essential fatty acid deficiency due to the CF

related intestinal fat malabsorption (45).

Cholelithiasis or bile stone disease is frequent in cystic fibrosis patients (46). Asymptomatic

stones are often seen on routine ultrasonography studies (47, 48). However, CF patients do

General introduction

15

1 1

regularly present with symptomatic cholelithiasis that necessitate cholecystectomy (46).

Cystic fibrosis related cholelithiasis does not usually respond to non-lithogenic treatment like

for instance ursodeoxycholic acid (UDCA) probably because cholesterol is not the main

component of stone or sludge (49). It is hypothesized that cholelithiasis in CF is related to

CFTR dependent alterations in the biliary bile composition. However, the exact pathogenesis

of the susceptibility of CF patients for bile stone disease has remained unknown (50).

During routine laboratory checkups in CF patients often elevations of liver enzymes (AST, ALT

and GGT) are found (51). If these laboratories abnormalities persist, in repeated

measurements, they are sometime classified as signs of CF related liver disease (52).

Persistent elevation of liver enzymes above two times the upper limit of normal are suggested

as an indication to start UDCA as potential treatment option (53). However, the predictive

value and the relation of elevation of liver enzymes to the presence of relevant liver disease

has remained a subject of controversy (54).

CIRRHOSIS IN CYSTIC FIBROSIS

The most severe hepatic complication in CF is the development of hepatic fibrosis into

cirrhosis. This potentially life threatening liver disease develops in about 10% of the CF

patients (40). The clinical presentation is dominated by symptoms of portal hypertension such

as splenomegaly, hypersplenism, gastrointestinal variceal disease and bleeding (42, 53). Less

frequent are ascites and hepatopulmonary syndrome found in CF related cirrhosis (CCFLD)

(55). The liver parenchymal functions, including protein synthesis and detoxification, are

usually spared (4, 56).

The disease is not yet clinically present at birth and develops during childhood. The majority

of patients have developed clinically manifest cirrhosis before adulthood with a peak

incidence around the age of 10 years (57). Although CF related cirrhosis is rapidly progressive

during childhood, the disease tends to stabilize into adulthood. Cirrhotic decomposition is a

rare event and liver transplantation while feasible in CF patients, is relatively rarely indicated

(56).

The clinical diagnosis of CCFLD is generally made on the basis of multi nodular irregular aspect

of the liver on ultrasound in combination with the presence of splenomegaly and signs of

hypersplenism including thrombocytopenia (58). Liver biopsy can be used for histological

diagnosis of cirrhosis (59). Since CCFLD can be localized in a segmental manner, liver biopsy

poses a risk for sampling error. Therefore, it has been advised to perform multiple biopsies in

patients suspected of CCFLD (60). For the clinical situation, it is not necessary to perform liver

biopsy in the situation of established cirrhosis. However to evaluate developing fibrosis for

diagnostic of therapeutic studies biopsy is probably needed to serve as the gold standard.

Chapter 1

16

1

To date ursodeoxycholic acid (UDCA) is the only medical treatment used in CF related liver

disease. UDCA is an endogenous, relatively hydrophilic bile salt with choleretic and

antifibrotic properties (61). UDCA is given to CF patients with persistently increased liver

enzymes during routine laboratory follow up and/or hepatomegaly (53). It is proven that

UDCA is capable of reversing liver enzyme elevation (62). However, the benefit of UDCA in the

treatment or prevention of CCFLD is not known (63). Therefore, the use of UDCA in CF

remains subject of discussion (64).

THE ROLE OF CFTR IN CYSTIC FIBROSIS RELATED LIVER DISEASES.

Despite the fact that, in recent years, there has been a rapid increase in the knowledge on CF

and CFTR protein function, the pathogenesis of CCFLD remains unknown. To improve the

treatment and clinical outcome of CCFLD a more fundamental understanding of the

mechanisms causing the disease is crucial.

Proliferation and destruction of the bile ducts is are prominent histological features of CCFLD

(65, 66). In the past, these observations lead to the assumption that biliary obstruction lays at

the basis of the disease (67). In CF lung disease, due to CFTR malfunction, the mucus in the

alveoli is thick and sticky. In the liver, CFTR is exclusively expressed in the cholangiocytes or

bile duct cells. In parallel to the lung disease, it was assumed that the observed bile duct

obstruction was caused by thick and sticky bile obstructing the bile ducts (68). This hypothesis

formed one of the substantiations to try the choleretic bile acid UDCA as a treatment in CFLD

(69). It was believed that UDCA would make the bile more fluid and increase the bile flow,

thereby preventing the bile duct obstruction. However, it was never been shown that indeed

the increased viscosity of bile in CF is the primary cause of the disease.

It is not known why only up to 10% of CF patients develop CCFLD. It has been demonstrated

that only patients with a severe phenotype, including pancreatic insufficiency, can develop

CCFLD (70). However, within the group of CF patients with a severe phenotype the risk for

developing CCFLD is not “further” genotype related. It is assumed that additional genetic risk

factors or external factors have co-responsibility for the development of CCFLD (71).

Extensive genetic modifiers studies in patients with CCFLD have shown for instance that

carrier ship of the Pi allele for alpha-1 antitrypsin deficiency adds to risk for CCFLD (57).

Another striking and yet unexplained clinical phenomenon of CCFLD is the young and distinct

age of presentation of the disease. CCFLD is not present at birth. In most patients, cirrhosis

develops before the age of 18 years with a peak incidence of about 10 years (40). In

adulthood, hardly any new case of cirrhosis develops. It is known from historical autopsy

studies that liver fibrosis and biliary obstruction are relatively common in adult CF patients

(72). However not all patients did eventually develop clinical manifest cirrhosis. To date we do

General introduction

17

1 1

not know what the risks factors are that trigger the disease to start development in children.

Most CCFLD patients have a severe form of portal hypertension with splenomegaly. Gastro-

intestinal variceal disease is common and in particular young patients are at high risk for

hemorrhages (58, 73). This risk for variceal bleeding diminishes with age and is less common

in adult CF patients (74).

2) THE ROLE OF BILE SALT IN CYTOTOXICITY AND BILE FLOW

A potential mechanism for the biliary liver disease in CF involves the concept of bile salt

related cytotoxicity. Before discussing the contribution of bile salts in pathology, the

physiological roles of bile salts will be addressed.

THE ROLE OF BILE SALT IN DIETARY FAT DIGESTION

Bile salts are polarized steroids that play a vital role in intestinal fat absorption and biliary

disposal of endogenous and exogenous compounds (75). In the intestine bile salts function as

essential surfactants used to solubilize dietary fats in the hydrophilic milieu of gut (76).

However, based on the same properties, bile salts can also act as detergents to liver tissue,

for example, in the situation of bile salt accumulation (cholestasis) (77). Corresponding with

their potentially toxic effects to cells, the bile salt synthesis and their concentrations are

tightly regulated (78).

Bile salts are synthesized in the hepatocytes from cholesterol. Bile salts are excreted into the

bile and transported, to the intestine, via the intra- and extrahepatic bile ducts. In the bile and

the gut, bile salts form water-solvable aggregates, so called micelles, together with the fatty

acids originating from the dietary fats. The formation of micelles is essential to transport the

dietary fats towards the enterocytes across the aqueous intestinal lumen. Absence of bile

salts in the gut results in severe intestinal fat malabsorption. In recent years, it has become

clear that bile salts are not only involved in dietary food digestion. It has been shown that bile

salts also play vital roles a variety of systemic metabolic regulatory processes, which are,

however, outside the scoop of this thesis (79).

Chapter 1

18

1

THE ENTEROHEPATIC CIRCULATION OF BILE SALTS

Bile salts are efficiently recycled via the portal system back to the liver in the so called

enterohepatic circulation (80). Bile salts are to a large extent (>95%), absorbed in the terminal

ileum, the final section of the small intestine. The total amount of bile salts in the body is

balanced and is kept in a tight, steady state, (81). Under steady state conditions, the fecal loss

of bile salts is entirely compensated by de novo bile salt synthesis of primary bile salts in the

liver. The primary bile salts in humans are cholate and chenodeoxycholate. The primary bile

salts are excreted via de bile into the intestine. In the intestinal lumen, the bile salts can be

metabolized by the gut flora. Bacteria are capable of deconjugating bile salts and

transforming them into a variety of different secondary bile salts. Bile salt species are

amphipathic molecules with a hydrophilic and a hydrophobic domain. Bile salts differ in their

water solubility and their hydrophobic-hydrophilic balance.

Hydrophobic bile salts have a high capability for solubilizing fats and lipids (82). As a result,

hydrophobic bile salts also have the ability to solubilize the lipid structures of cell membranes.

Therefore, the more hydrophobic a bile salt, the higher their detergent cytotoxicity for cells

and tissues exposed to them.

Under physiological conditions, the hydrophobic cytotoxicity of biliary bile salts is limited by

co-secretion of phospholipids (and cholesterol), leading to the formation of mixed micelles

(83). Phospholipids are excreted into bile by the membrane transporter enzyme MDR3 (Mdr2

in mice). A genetic incapacity to excrete phospholipids into the bile results in severe bile duct

destruction and biliary liver disease. In humans, an inactivating MDR3 mutation is the basis

for the disease “progressive familiar cholestatic disease type 3 (PFIC 3)” (84). The disease

presents with severe biliary cirrhosis and cholestasis, usually at child age. A genetic mouse

model without functional Mdr 2 protein expression (Mdr2 knockout mice) spontaneously

develops biliary disease, similar to human PFIC 3 patients (85). The bile duct destruction in

Mdr2 knockout mice can be even augmented by increasing the hydrophobicity of the bile salt

pool via administration of the hydrophobic bile salt cholate to the mice (86).

BILE PRODUCTION AND BILE FLOW

The magnitude of bile flow is determined at two levels in the biliary tract. The first level is the

so called canalicular lumen, the smallest intercellular biliary domain between hepatocytes.

Canaliculi form the start of the bile ducts and are surrounded by hepatocytes. When

hepatocytes actively transport bile salt into the canalicular lumen, water passively follows as a

result of the osmotic activity of the bile salts. The amount of water transport that is generated

via bile salt osmosis is called the bile salt dependent bile flow (87).

General introduction

19

1 1

Various bile salt species differ in their choleretic capacity, i.e the capability to induce bile flow

(88). The choleretic capacity of bile salt depends on the molecular structure and their

hydrophobicity. Several hydrophilic bile salts are capable of inducing an exceptionally high

bile salt dependent bile flow. For this choleretic capacity some bile salts, like for instance

UDCA, are used in clinical situations were decreased bile flow or bile duct obstruction is

hypothesized to be contributing to the biliary disease (89).

The second level at which bile flow is determined are the bile ducts. Cholangiocytes form the

lining of the bile ducts. Cholangiocytes are capable of secreting water and e.g. bicarbonate

into the bile, thereby increasing bile flow and diluting the (canalicular) bile content (90). This

portion of the bile production is called the ductular bile flow. Different ion and anion

membrane transporter proteins are involved in the complex mechanism of water secretion

over the apical membrane by cholangiocytes (91).

The actual driving force behind this process of water secretion of cholangiocytes is the active

Cl- transport into the bile duct lumen (92). The most important Cl

- transporter is CFTR, but

other Cl- channels have also been described (93-95). The active Cl

- transport creates a Cl

-

gradient over the apical membrane. The luminal Cl- is then exchanged for bicarbonate (HCO3

-)

via a protein called Anion exchanger-2. In this way, the Cl- gradient is replaced by a HCO3

--

osmotic gradient. Water leaves the cholangiocytes, via water channels or so called aquaporins

on the basis of this osmotic HCO3- gradient. In this manner, the water secretion by

cholangiocytes is directly related to the Cl- transport capacity and CFTR function (96)

The chloride transport of CFTR is an active ATP consuming, process (97). CFTR belongs to the

extensive family of ATP binding cassette membrane proteins (98). The CFTR chloride channel

only opens after binding and hydrolysis of the intracellular energy source ATP (99). The ATP

binding necessary for CFTR activation is regulated via the protein cycle AMP (c-AMP). In the

cholangiocytes, c-AMP activity is controlled via different pathways. The most important

stimulating pathway of c-AMP is the hormone secretin (100). This gastro-intestinal hormone

is released during feeding and in this manner influences the bile flow rate. c-AMP can also be

induced via intraluminal factors like bile salts or intraluminal ATP (94, 101).

BILE SALT SYNTHESIS

Bile salts are synthesized in the hepatocytes from cholesterol (figure 4). The synthesis

requires several sequential enzymatic steps (78). The synthesis of the bile acids is

quantitatively the predominant pathway of cholesterol catabolism in mammals. The major

pathway for the synthesis of the bile acids is initiated via hydroxylation of cholesterol at the 7

position via the action of cholesterol 7α-hydroxylase (CYP7A1), an ER localized enzyme.

CYP7A1 is a member of the cytochrome P450 family of metabolic enzymes. This pathway

Chapter 1

20

1

initiated by CYP7A1 is referred to as the "classic" pathway of bile acid synthesis. There is an

alternative pathway via the mitochondrial enzyme sterol 27-hydroxylase (CYP27A1).

Although, in rodents the alternative pathway can account for up to 25% of total bile acid

synthesis, in humans it has been suggested to account for no more than 6% (102).

Figure 4. Schematic representation of the bile salt synthesis pathways. (Illustration from Lipid research 2009;50:s120-

s125)(102)

General introduction

21

1 1

THE REGULATION OF BILE SALT SYNTHESIS

Bile salts have an important physiological contribution to the intestinal fat digestion, besides

elimination of cholesterol and other endogenic and exogenic catabolytes and metabolytes

from the body. Bile salt synthesis is under the control of at least two different negative

feedback pathways (figure 5.). In the first pathway, the sensing of bile salt takes place in the

hepatocytes (103). In the other pathway, bile salt sensing takes place at the level of the

enterocytes of the gut. In both pathways binding of bile salt to Farnesoid X receptor (FXR)

plays a pivotal role in the initiation of the negative feedback regulation. FXR is a nuclear

receptor. Nuclear receptors are transcription factors, whose ligands can determine their DNA

binding and transcription modulating activity. The ligands for the nuclear receptors are,

among others, steroids. In response, these nuclear receptors work with other proteins to

regulate expression of specific genes, thereby controlling and regulating homeostasis in an

organism. In the case of the nuclear receptor FXR bile salt can function as ligands.

Figure 5. Schematic representation of the enterohepatic circulation of bile salts in the context of CFTR. Several bile

salt involving feedback loops regulated bile salt synthesis and liver growth. The bile salt induced nuclear receptor FXR

plays a pivotal role in intestine and in the liver.

Chapter 1

22

1

However different bile salt species vary in their efficacy as FXR ligands (104). In general

hydrophobic bile salts are stronger ligands for FXR compared to hydrophilic bile salts.

Bile salt/FXR interaction results in different physiological responses, in the different cell types,

for example, hepatocytes and intestinal cells. In hepatocytes FXR activation down regulates

the expression of a protein called small heterodimer protein or SHP. SHP can directly down

regulates the expression of CYP7A the rate limiting step in the bile salt synthesis from

cholesterol (105).

The intestinal regulation of bile salt synthesis is more indirect: FXR stimulates the expression

of FGF19 (fibroblast growth factor 19, equivalent to Fgf15 in rodents) that is subsequently

released into the portal blood (106). At the basolateral membrane of the hepatocytes FGF19

can bind to the FGFR4 receptor (107). This binding leads to activation of an intracellular

pathway that, via the so called ERK system, down regulates CYP7A expression and

subsequently reduces bile salt synthesis (108). The hepatic and intestinal sensing of bile salts

by FXR and their respectively negative feedback pathways are functionally independent.

3) THE ROLE OF CFTR IN GASTROINTESTINAL DISEASE.

Different from other organs like lungs and sweat gland, the contribution of CFTR in the

intestine to the clinical phenotype is not entirely clear. In the intestine CFTR is expressed on

the apical membrane of the enterocytes (109). Also in the gut, CFTR functions as a Cl- channel.

The Cl- channel activity of the gut can actually be quantified in ex vivo electrophysiological

studies of gut tissue (110). It is hypothesized that CFTR in the intestine plays a direct role in

the lubrication of the gut contents. Reduced water content of intestinal content may lead to

thickened stools (111). This principle of decreased water content probably lies at the basis of

intestinal pathology in CF like meconium ileus and constipation. Based on studies in mice we

know that in the intestine bile salts are capable of stimulating Cftr via the intestinal

membrane receptor protein Asbt (apical sodium binding protein) (112). The latter indicating

that in Cystic fibrosis intestinal disease, besides the intrinsic malfunction of the CFTR protein,

additionally disease related alternation in bile salt metabolism can be responsible for

decreasing lubrication of the gut contents

FAT MALABSORPTION

One of the most striking features of the gastrointestinal phenotypes in CF is the intestinal fat

malabsorption (113). The fatty stools or steatorrhea is a clinical sign of fat malabsorption. Fat

malabsorption in patients causes several severe problems. Due to the high energy content of

General introduction

23

1 1

fat in general, fecal fat malabsorption causes malnutrition and poor growth. A secondary

effect of the intestinal fat malabsorption is the reduced absorption of the fat soluble vitamins

A, D, E and K. Vitamin malabsorption can lead to hypovitaminosis and vitamin K dependent

coagulopathy (114-116). Therefore, CF patients with intestinal fat malabsorption are usually

dependent on oral vitamin ADEK supplementation.

The main reason the intestinal malabsorption is exocrine pancreatic dysfunction (113). As a

result of the fibrotic destruction, already during pregnancy and infancy, the pancreas loses

the capacity to excrete sufficient amounts of digestive enzymes into the intestine. Pancreatic

enzymes form the bulk of the available intestinal digestive enzymes. The pancreas excretes

lipases for digestion of nutritional fat and proteases for the digestion of nutritional proteins

(117).

The majority of dietary fats consist of triglycerides. These triglycerides are hydrolyzed, by

lipases, into smaller and more polar fatty acids. To overcome the intestinal fat malabsorption,

CF patients with pancreatic insufficiency, are prescribed pancreas enzyme replacement

therapy (PERT) (33). This supplementation is administered simultaneously with every dietary

fat containing meal. PERT dosing is primarily related to the amount of fat calculated from the

diet and adjusted based on clinical effects on symptoms of steatorrhea and growth (118).

INTESTINAL BILE SALT MALABSORPTION

CF patients have a persistently elevated fecal BS excretion compared to healthy controls (26).

The origin of the increased bile salt excretion is not known. First it was suggested to be

related to the intestinal fat malabsorption (119). However increased fecal bile alt excretion is

still present during adequate PERT treatment. Furthermore, increased fecal bile salt excretion

is present in patients with still sufficient exocrine pancreas function, i.e. without fat

malabsorption (120). The latter observation was confirmed in experimental CF mice models.

Also in “mild” CF mouse models, with normal fat absorption, fecal bile salt loss is increased

(121).

Based on the observation that increased fecal bile salt excretion is independent of fecal fat

malabsorption it is assumed, that increased fecal bile excretion in CF is related to the

dysfunctional Cftr, located in the enterocytes of the intestine itself (122, 123). The majority of

bile salts are re-absorbed in the distal part of the small intestine, the terminal ileum. Bile salt

re-absorption into the enterocytes is facilitated by a membrane protein called apical sodium

dependent bile salt transporter (ASBT). In CF mouse models, it has been shown that the

absence or Cftr function in the enterocytes down regulates the transport capacity of bile salt

by the co-located membrane protein ASBT (112). This observation is suggesting that the

Chapter 1

24

1

intestinal re-absorption of bile salt maybe hindered by dysfunctional CFTR. This may in turn

contribute to the observed increased fecal bile salt loss.

Finally, there are suggestions that the increased fecal bile salt excretion could be associated

with CFTR dependent differences in the intestinal bacterial flora (124). Small intestinal

bacterial overgrowth (SIBO) is common in CF. SIBO can result in steatorrhea, abdominal pain,

bloating, flatulence, nausea, and anorexia (125). Also based on results in CF mice models, it

has become known that absence of Cftr function causes significant changes in the intestinal

bacterial composition (126). In mice, differences in intestinal bacterial composition are

associated with changes in bile salt metabolism and with increased fecal bile salt loss(127).

Therefore, it could well be that the reported changes in intestinal bacterial flora in CF

condition contribute to the increased fecal bile salt excretion.

INTESTINAL INFLAMMATION

A more recently recognized GI condition in CF is intestinal inflammation. Based on fecal

calprotectin levels (a marker for intestinal inflammation), intestinal inflammation is a frequent

event in CF patients (128). Evaluation via capsule endoscopy has indicated that macroscopic

inflammatory lesions are frequently present in CF patients (129). It is not known whether the

inflammation is a direct effect of dysfunctional CFTR in the gut or otherwise. Another

hypothesis is that Cftr function interferes with inflammatory regulatory mechanism and

proteins like the peroxisome proliferator activated receptors (PPARs)(130, 131). Finally, it is

also possible that the intestinal inflammation is related to the combination complex

interactions of the bacterial flora of the gut, bile salts metabolism and the immunological

inflammatory response in the context of CFTR dysfunction.

4) CURRENT HEPATIC AND GASTROINTESTINAL ISSUES OF CYSTIC

FIBROSIS DISEASE

Recent developments in the pathophysiology of CF have elucidated many aspects of the

hepatic and gastrointestinal consequences and pathology of cystic fibrosis. However, many

questions have remained unanswered in this research field. To optimize the care for and

prognosis of CF patients in the future, we aimed to gain more insights in the role of CFTR in

the pathology of hepatic and gastrointestinal CF. It is to be expected that insights in the

pathophysiology of CF in the GI tract will also provide knowledge concerning the physiological

roles of CFTR in other tissues. Furthermore, we are facing a new era of therapeutic options in

CF, when CFTR correctors and potentiatiors are currently coming to the market (132). These

General introduction

25

1 1

promising new treatments have to be monitored for their efficacy. New insights in the role of

CF and CFTR in the liver and gastrointestinal tract can offer growing opportunities to test and

evaluated these new therapeutic options (133).

CURRENT ISSUES FOR CFLD

To improve prevention and treatment possibilities of CFLD, a better knowledge of the etiology

and pathophysiology of liver disease in CF is essential. Several potential leads for the etiology

of CFLD have been reported in the literature (134-136). One of the suggestions for the

pathogenesis of CFLD has been the involvement of increased viscosity of bile, due to CFTR

dependent changes in bile composition. Bile duct obstruction then, sequentially, leads to the

development of an obstructive biliary cirrhosis (137)

. However, no firm experimental evidence

is available to support this hypothesis.

Another potentially contributing factor is bile salt cytotoxicity. In the liver CFTR is located at

the apical membrane of the cholangiocytes lining the bile ducts. In cholangiocytes, CFTR

functions, as a chloride channel, indirectly in water and bicarbonate secretion into the bile

(91). Therefore, CFTR plays an important role in the magnitude of bile production and bile

composition (96). Bile salt cytotoxicity can theoretically be enhanced in CF conditions. The

overall bile salt concentration can be increased due to reduced water secretion and reduced

dilution of the bile.

Furthermore, bile cytotoxicity can be increased due to changes in bile composition. Normally

the detergent effects bile salts are reduced by the biliary phospholipids (75, 83). CFTR related

changes in bile composition of the biliary lipids and/or bile salts could play a role in enhanced

susceptibility to bile salt cytotoxicity and the development of biliary liver disease in cystic

fibrosis.

Bile salt cytotoxicity could be enhanced by increasing the contribution of hydrophobic bile

salts and/or by decreasing the biliary secretion of “protective” phospholipids. The biliary bile

salt profile is determined by de novo synthesis of bile salts and by the secondary bile salt

conversion by the intestinal bacterial flora. A change towards a more cytotoxic bile salt

composition could be the result of either altered primary bile salt synthesis or secondary bile

salt conversion.

The treatment of CFLD remains a controversy in the care for CF patients. The only medication

currently used is ursodeoxycholate (UDCA) (64). This relatively hydrophilic bile salt is

administered orally. Its working mechanism is supposed to be related to its choleretic and

anti-inflammatory/anti-fibrotic capacities. In several clinical trials, it has been shown that

UDCA is capable to recover elevated liver functions tests like AST, ALT and GGT to normal

Chapter 1

26

1

values (138, 139). However, there are no long-term follow up studies of the therapeutic

effects of UDCA with respect to clinical endpoints, such as mortality, need for liver

transplantation, or fibrosis/cirrhosis complications. It has never been shown that UDCA is able

to prevent cirrhosis, nor its complications. One of the problems for clinically studies is that we

are not able to identify the subgroup of the CF patients (~10%) that are at risk for cirrhotic

disease.

It has been assumed that the therapeutic effect of UDCA functions via its capacity to increase

bile flow (61). However, based on fundamental studies, it has not been demonstrated that

UDCA indeed is actually capable of increasing bile flow in CF conditions and if so, to what

extent. Measuring bile flow and bile production in humans is invasive and difficult. Therefore,

there is a need for experimental support that clarifies this issue.

Recent studies on cirrhosis by other etiologies indicate that, even in progressive fibrotic liver

diseases, the cirrhosis can be stopped or even reverse on removal of the causative agent

and/or on treatment of the underlying disease (140). In particular antiviral treatment has

been shown to be able to reverse the severity of Hepatitis B virus cirrhosis (141, 142). New

developments are evolving concerning the use of anti-fibrotic therapies in liver fibrosis and

cirrhosis. Although theoretically promising, to date there not yet anti-fibrotic therapies

available in humans (143). However, the scope of these positive developments indicates the

rising opportunity and potential profit for preemptive treatment in CCFLD.

These promising scientific advances indicate the need for reliable and relevant markers to

identify patients at risk for CCFLD. Therefore, one of the most urgent issues in CFLD is the

possibility of early detection of patients at risk for, or in an early phase of, the disease.

Progress in this diagnostic field of CF would offer opportunities for evaluation of current en

new interventional therapies to prevent or treat CFLD. As stated above, no tools for early

detection or recognition are available. As a consequence, CFLD is frequently only recognized

clinically in an advanced stage of severe fibrosis or even cirrhosis. However, cirrhosis typically

becomes manifest around the age of 10 years and it is not present at birth. This observation

indicates that cirrhosis in CFLD develops progressively over a period of years during

childhood.

At this time, cirrhosis is usually diagnosed based on physical examination and abdominal

ultrasound studies of the liver and spleen (53, 144). On physical examination, the most

prominent clinical feature of cirrhosis is splenomegaly. This is often accompanied by

hematological signs of hypersplenism, like thrombo- and leucocytopenia. Ultrasound of the

liver, in case of cirrhosis, can show irregular liver edges and an inhomogeneous, nodular,

pattern of the parenchyma (43, 145, 146). The spleen span can be enlarged for age. Since

portal hypertension is often a characteristic symptom in CF related cirrhosis, abdominal

ultrasound may reveal signs vascular collaterals originating from the portal system. Upper GI

General introduction

27

1 1

endoscopy may reveal evidence for esophageal or gastric variceal disease based on the often

severe portal hypertension (147).

Liver function test (LFTs), like AST, ALT and GGT, are often used to diagnose CFLD. However,

their clinical value for diagnosing (development of) liver fibrosis or even cirrhosis has not been

established. CF patients are routinely checked for elevation LFTs. Frequently patients have

(even persistent) elevations of LFTs above the upper limit of normal (148). By some authors,

this elevation of LFTs has been defined as one of the characteristics of CFLD (52). There is no

proven correlation, however, between elevation of LFTs and development of liver fibrosis or

cirrhosis (60). The current way of evaluating LFTs elevation is thus not validated and useable

to identify patient at risk or actually developing CFLD.

The gold standard for validation of the stage of liver fibrosis in general is histology, normally

obtained via percutaneous liver biopsy. Several studies have shown that liver histology can

classify milder forms of fibrosis in CF (149). However, liver biopsy has several practical and

important drawbacks. Based on autopsy studies in CF patients it has been reported that

fibrosis and cirrhosis may not be evenly spread throughout the liver. This observation

highlights the risk of a sampling error in the case of liver biopsy (60, 150). Another problem

with liver biopsy in CF is the patient selection. It is known that (“only”) around 10% of CF

patients with severe mutations will develop cirrhosis. Since liver biopsy is an invasive

procedure, it is not ethical to use it, with a low threshold, in CF patients. Also in the case of a

liver biopsy, the reliable identification of patients seriously at risk for cirrhosis, could justify

the use of this invasive diagnostic procedure.

Several studies have reported on the possibilities to use ultrasound of the liver as a screening

tool for staging liver fibrosis and cirrhosis in CF patients (43, 145, 150). Ultrasound of the liver

seems promising as a screening measure. It is non-invasive, widely available and already

frequently used in standard clinical CF care. Apart from the inhomogeneity and nodular

abnormalities consistent with cirrhosis in general, several other ultrasound particularities can

be found in CF (146). Often on ultrasound the echogenicity of liver parenchyma is increased.

Increased echogenicity of the liver, however, is not specific for the development of cirrhosis:

rather, it may be related to an increased fat content or steatosis of the liver. There are no

indications that steatosis is preceding the development of fibrosis in CFLD. In general, it is

concluded that ultrasound studies of the liver are a reliable tool for the diagnosis of cirrhosis

in CF, i.e. the end-stage, but are not sensitive enough to recognize or classify CFLD in an

earlier phase.

A more recent development in the diagnosis for liver fibrosis in CF is the use of transient

elastography(151, 152). This technology is based on a method to determine tissue stiffness by

measuring sound wave reflection produced by a special probe. The method is non-invasive

and could potentially be used a screening tool in all pediatric CF patients. Problem with this

method is that it has not been satisfactorily validated for the pediatric (CF) population (153,

Chapter 1

28

1

154). Another difficulty with elastography is that, in CF, it remains to be validated against the

gold standard for liver fibrosis; histology from liver biopsy.

CURRENT ISSUES FOR INTESTINAL CF

Better nutritional status is related to a better survival of CF patients (28). Therefore,

improvement and maintenance of the nutritional status is one of the cornerstones in the care

for CF patients (118). To achieve this goal optimizing dietary intake and improving intestinal

absorption of nutrients are priorities in the research agenda and treatment of CF. Since

dietary fats provide the largest caloric share in the normal dietary energy intake, optimizing

dietary fat intake and absorption provides the greatest potential benefit for improvement of

the nutritional status.

Despite adequate and optimal pancreatic enzyme replacement therapy for exocrine

pancreatic insufficiency, in many CF patients, a persistent fat malabsorption remains (35). This

PERT resistant fat malabsorption is probably associated with other, non-pancreatic, CF related

factors involved in the intestinal fat absorption (36). To further optimize the diagnosis and the

treatment possibilities of PERT resistant fat malabsorption we need more insight in the

factors involved.

Basically, intestinal fat absorption proceeds in two sequential phases. First lipolysis is

facilitated by the digestive enzymes mainly produced by the pancreas. Secondly there are the

processes involved in post-lipolytic fat absorption like the solubilization of fats by bile salts. To

date there is no clear answer to the exact cause of the malfunction of the postlipolytic phase

in CF conditions. However, several factors are associated with PERT resistant fat

malabsorption like changes in bile salt metabolism and the enterohepatic circulation and

differences in bacterial flora. It could also possible that the PERT resistant fat malabsorption

in CF is a direct consequence of the dysfunctional CFTR protein in the apical membrane of the

enterocytes.

5) OUTLINE OF METHODOLOGY USED FOR THIS THESIS

For the studies in this thesis, we applied a variety of methods to obtain more

pathophysiological insights in CF disease in the liver and gastrointestinal tract. Broadly, our

methods could be divided into 3 major groups. First, we aimed to carry out a literature review

study concerning the reported and theoretical background of PERT resistant intestinal fat

malabsorption in the context of CF and CFTR malfunction. Secondly, we performed a

retrospective study in CF patients, to determine the potential value of specific LFTs in

General introduction

29

1 1

identifying CF patients at risk for cirrhosis. Thirdly, we performed fundamental studies in a

variety of CF mice models to determine the role of bile salt metabolism in CF conditions and

the role of bile salts in the development of CFLD. In the CF mice models, we performed

experiments to assess the role of bile salts, their metabolism and enterohepatic circulation.

To aid the reader with limited knowledge on CF mice models, the background of these models

and the methodologies used will first be discussed in more detail.

CYSTIC FIBROSIS MOUSE MODELS

Cystic fibrosis was the first monogenic genetic disease in which the disease causing gene

mutation was identified (5, 6). As a result of this breakthrough in 1989, the genetic and

molecular insights into the functions of the CFTR protein increased rapidly. Based on this

knowledge, several genetically engineered experimental mouse models were developed, in

which the genotype of human CF was mimicked (155, 156). These mice models have been

used successfully to understand CF pathophysiology in different organs. Also in the present

thesis, we used specific CF mouse models to delineate the role of Cftr in the pathophysiology

of cystic fibrosis.

Globally the CF mice models can be divided into two categories. In the first category the Cftr

gene is no longer functionally expressed, the so called complete Cftr knockout mice or Cftr-/-

mice. The knockout mice are not able to produce any functional Cftr protein. The second

category of CF mouse models is engineered in a way that they express a mutated form of the

Cftr gene, in accordance with the most frequent CFTR gene mutations in human CF patients.

Theoretically these Cftr mutated mice resemble the genetic and functional situation of CF

patients more closely than the knock-out mouse models. For example, in parallel to human

CF, mouse models have been constructed that carry the homozygous 508del mutations in

their (murine) CF gene. The delta F508 mutation consists of a deletion of the three

nucleotides that comprise the codon for phenylalanine (F) at position 508. Having two copies

of this mutation (one inherited from each parent) is the leading cause of CF (157).

The phenotype of CF mouse models does not resemble the human CF disease in every aspect.

Some CF disease traits are found similarly in CF mice and humans, including increased fecal

bile salt excretion (121). Others, however, like for instance lung disease, are hardly or not

present in CF mice. Furthermore, strong mouse strain background effects are present among

the different CF gene modifications, irrespective of complete or incomplete Cftr inactivation.

The strain effects are probably related to the presence of disease modifying genes in the

genome of the different genetic strain backgrounds.

Chapter 1

30

1

HEPATIC PHENOTYPE OF CF MOUSE MODELS

CF mouse models display varying types and degrees of hepatic histological phenotypes. Most

mouse models show a normal liver histology. However, there are several reports of CF mice

models with spontaneous developed histopathology. The most severe hepatic phenotype is

described in the University of North Carolina (UNC) Cftrunc/unc

knockout mice. Even without a

further challenge, these Cftr-/-

animals develop focal and progressive hepatobiliary disease

(158). By 3 months of age, they have varying degrees of periportal and bridging fibrosis, which

then progresses with age. Freudenberg et Al. describe histopathology in the homozygous

del508 (Cftr508/508

) mice (135). Liver histopathology of CF and controls livers displays variable,

mild, patchy cholangiopathy characterized by reactive changes in the biliary epithelium, bile

ductular proliferation, and mild portal fibrosis. These findings were only rarely present in WT

mice. Notwithstanding the hepatic phenotype, none of these mice manifested any advanced

degree of liver fibrosis or cirrhosis. The mice did not exhibit any bile duct lesions even though

some animals were more than 12 month old.

Besides spontaneous development of liver histopathology, there are also CF mouse models

with increased susceptibility to an exogenous challenge to induce a liver phenotype. There is

an existing clinical and experimental relationship between colitis and another biliary disease

namely primary sclerosing cholangitis. In experimental animals, a chemical colitis can be

induced via oral dextran sulfate sodium (DSS) administration (136). Based on this concept it

was hypothesized that loss of Cftr function in the setting of a DSS induced colitis can lead to

(enhanced) bile duct injury. Blanco et al. demonstrated that mice homozygous for Cftr

mutations developed bile duct injury following the DSS induction of colitis. In an additional

study from the same research group, the bile duct susceptibility of the Cftr-/-

mouse could be

related to decrease Ppar α expression in Cftr-/-

mice (159). These results seem to suggest that

changes in the systemic inflammatory regulation could be involved in the etiology of CFLD.

THE INTESTINAL PHENOTYPE OF CF MICE MODELS

The most marked phenotypical symptomatology of CF mice is intestinal obstruction (160).

This phenotypical trait resembles in some aspects the clinical picture of intestinal obstruction

in meconium ileus or DIOS in CF patients. In mice, it typically presents when mouse pups are

weaned from breast feeding to solid foods (usually at age ~18-19 days). The obstruction can

be severe and even lethal for the mice. Some mice models have to be treated with oral

laxatives to restore normal bowel movements and/or to prevent mortality. Other mice with

an even more severe intestinal phenotype need persistent liquid feeding to prevent

(re)occurrence of intestinal obstruction (161).

General introduction

31

1 1

Histology of the pancreas does show some preliminary signs of pancreatic fibrosis. However,

based on evaluation of pancreatic function tests, CF mice are pancreatic sufficient. Despite

this normal exocrine pancreatic function, CF mice models do present an increased fecal fat

excretion compared to non CF normal control mice. The latter, indicating other, non-lipolytic,

causes for intestinal fat malabsorption in CF conditions (158).

Another intestinal phenotype of the CF mouse models is intestinal inflammation in

combination with small intestine bacterial overgrowth (SIBO). Oxana et al. describe the novel

finding that a specific innate immune response in the CF mouse small intestine has a role in

inflammation and contributes to the failure to thrive in this mouse model of CF (162). When

then treated with broad-spectrum antibiotics for 3 weeks, the Cftr-/-

mice increased in body

weight when compared with controls (163). Wouthuyzen et al. reported that this positive

effect of antibiotics on body weight was not mediated by increasing the absorption of long-

chain triglycerides (127). Antibiotic treatment of homozygous ΔF508 mice without SIBO

neither augmented body weight nor increased fat absorption. Collectively, the data indicate

that the positive effect of antibiotics on body weight in CFTR-knockout mice may be

attributable to the treatment of SIBO and not necessarily to enhanced absorption of long-

chain triglycerides.

Like human CF patients, CF mice also have an increased fecal bile salt excretion compared to

control mice. To date there is no conclusive explanation for this observation. Since CF mouse

models display normal exocrine pancreatic function, the increased bile salt excretion seems

not related to the pancreatic enzyme dependent lipolysis of the dietary fats (164). Another

possibility is a decreased bile salt uptake in the ileum by the apical sodium dependent bile salt

transporter protein (ASBT). Bijvelds et al. demonstrated that ileal TC uptake was reduced by

17% in Cftr-null mice (112). Because the distal ileum is the discrete site of active BS uptake

and has a pivotal role in the near-complete recovery (∼95%) of the intestinal BS load, this

reduction may well explain the high level of fecal BS excretion reported previously for our

Cftr-null mice (165). However, for homozygous F508del mice, we did not find a reduction in

ileal BS uptake, despite the fact that fecal BS loss is increased to a similar extent as in null

mice. Another explanation may involve interaction between intestinal bile salt metabolism

and Cftr related changes in intestinal bacterial flora. Even under physiological conditions

there is an active and complex interaction between bacterial flora and bile salt metabolism.

For instance, mice with a deficiency in microbiota (germ free mice or mice treated with

antibiotics), display decreased fecal bile acid excretion and an increased bile acid pool size

(166, 167). These studies indicate that bacteria of the gut microflora are interrelated with the

modulation of host bile salts. Since bacterial flora is altered in CF conditions, this could lead

changes in fecal bile salt excretion.

Chapter 1

32

1

GENERAL DESCRIPTION SPECIFIC METHODOLOGY USED IN CF MOUSE

MODELS

There are several methods available to study the hepatic and intestinal consequences of Cftr

function in experimental mice models. There are anthropometry methods like measuring

body weight, liver weight and the determining the liver weight to body weight ratio. These

methods provided the possibility to assess nutritional status and possible trophic factors of

liver growth

FECAL SAMPLE COLLECTION

Fecal sample collection can be used for different purposes. In combination with 72 hour

registration of the oral dietary intake, we used 72 hours stools collection to estimate the

percentage of fat absorption. Fecal fat content was further evaluated to determine the

differential absorbability of the various fatty acids species. Fecal samples were also used to

measure the total bile salt excretion and determine the profile of the different bile salts

species.

BILE DUCT CANNULATION

We performed bile duct cannulation to determine bile production and bile composition. In

this method, the mice are operated via microsurgery. The abdomen is explored and the

gallbladder identified. The choledochal duct is identified and ligated, thereby blocking the bile

flow to the intestine. A hole is punctured in the gallbladder with a needle, after which it is

canulated and fixated. Subsequently all the bile produced by the liver flow freely via the tube

and can be collected in a tube for gravimetric analysis to measure bile production. The bile is

used for biochemical analysis like bile salt concentration and composition, and biliary lipids

(phospholipids and cholesterol).

BILE SALT KINETICS

Additionally we applied the determination of bile salt kinetic parameters, i.e., pool size

(amount of bile salts in the body), fractional turnover rate (the portion of the pool that is

newly synthesized per day), and synthesis rate. These parameters reflect hepatocellular

function, the metabolism of bile salts and the efficiency of enterohepatic cycling. To measure

General introduction

33

1 1

these kinetic parameters we used a previously developed and validated stable isotope

dilution technique without the need to interrupt the enterohepatic circulation. This method

allows simultaneous determination of kinetic parameters. We used the stable isotope

technique with [2H4]-cholate as labeled bile salt. Cholate is a major primary bile salt species

and comprises 50 to 80% (rodents) of the total bile salt pool. Therefore, cholate pool size,

fractional turnover rate (FTR) and synthesis rate are kinetic parameters that allow description

of bile salt kinetics of the quantitatively most important bile salt.

EXPERIMENTAL BILE SALT SUPPLEMENTED DIETS

For several of our mouse experiments, we used bile salt supplemented diets. The bile salt

were added and mixed to regular chow mice feeding. We used ursodeoxycholate (UDCA)

supplemented diets to mimic the clinical situation of UDCA treatment in CF patients. In other

experiments, we used supplementation of the hydrophobic bile salt cholate to mimic the

human hydrophobic bile composition (humans have a higher hydrophobicity than mice). The

purpose of the latter approach was to evaluate if a higher hydrophobicity would contribute to

the cytotoxic effect of bile on the development of CFLD like pathology in the CF mice models.

6) SCOPE OF THIS THESIS:

Despite great progress in the care and treatment of CF, the disease still is severe, lifelong and

lifespan limiting. Besides the clinically progressive pulmonary disease, the CF phenotype

includes a variety of gastro-intestinal and hepatic manifestations. Some of these CF

manifestations, like intestinal fat malabsorption and the development of cirrhosis, give rise to

severe and life threatening complications. Better understanding, concerning the

pathogenesis, the mutual interrelationships and the potential treatment options, of the

gastro-intestinal and hepatic manifestation of CF will have profound positive effects on the

prognosis of CF patients.

The enterohepatic circulation of bile salts connects the physiology of the liver with the

physiology of the intestine. Bile salts are crucial functional and regulatory molecules in a

variety of essential physiological processes of the hepatobiliary and gastrointestinal tract. For

example, bile salt secretion is the driving force behind bile formation and bile salts are

indispensable for intestinal fat, and thus energy, absorption. On the other hand, bile salts are

strong detergents, capable of inflicting serious and permanent cell and tissue injury. The

balance between the essential physiologic functions and the destructive forces of bile salts

requires a precise regulation of this delicate system. Disturbances of the equilibrium of the

enterohepatic circulation can potently lead to physiological malfunction and disease.

Chapter 1

34

1

CF is a multi-organ disease including, among others the hepatobiliary and gastrointestinal

tract. Therefore, it is plausible that primary or secondary consequences of CFTR protein

dysfunction lead to alteration or disturbances in the enterohepatic circulation of bile salts.

The specific aim of the research described in this thesis was to determine the role of bile salts

and their enterohepatic circulation in the hepatic and intestinal phenotype in CF.

Many patients with exocrine pancreatic insufficiency suffer from persistent intestinal fat

malabsorption despite adequate PERT therapy. In chapter 2, we describe, in a comprehensive

literature review, all factors potentially involved in PERT resistant fat absorption including the

enterohepatic circulation of bile salts. In the review, we focus in particularly on the role of bile

salts and enterohepatic circulation.

In the following chapters we switch gear towards CF related liver disease. First we turn to the

CF mouse models to answer basic questions concerning the pathogenesis and development of

CFLD. In chapter 3, we test the hypothesis that biliary bile salt cytotoxicity lays at the basis of

the development of CFLD in CF a specific mice model with spontaneous liver disease. This

initial experimental chapter is followed by chapter 4 in which we study if CFLD can be

induced, by feeding a hydrophobic bile salt-containing diet to CF mice without spontaneous

liver disease. Since CFLD is often treated with UDCA, in chapter 5, we tested the effects of

UDCA on bile production and bile composition in a CF mouse model.

To prevent or treat CFLD, it is essential to be able to recognize CF patient at risk or in an early

phase of the disease. To address this issue, in chapter 6, we return to the clinical aspects of

CCFLD. In this chapter, we aim to identify CF patients at risk for cirrhosis in an early clinical

phase. In a retrospective approach, we tested the hypothesis that the development of

cirrhotic CFLD can be predicted on the basis of follow up of biochemical liver function tests. In

chapter 7, we discuss our overall results, place these in a clinical and experimental

framework, and discuss future perspectives.

General introduction

35

1 1

REFERENCE LIST

1. Ratjen F, Döring G. Cystic fibrosis. Lancet. 2003;361(9358):681-9.

2. Vernooij-van Langen AM, Loeber JG, Elvers B, Triepels RH, Gille JJ, Van der Ploeg, Catharina

PB, et al. Novel strategies in newborn screening for cystic fibrosis: A prospective controlled

study. Thorax. 2012;67(4):289-95.

3. Slieker MG, Uiterwaal CS, Sinaasappel M, Heijerman HG, van der Laag J, van der Ent,

Cornelis K. Birth prevalence and survival in cystic FibrosisA national cohort study in the

netherlands. CHEST Journal. 2005;128(4):2309-15.

4. Cystic fibrosis foundation patient registry 2011. 2012.

5. Kerem B, Rommens JM, Buchanan JA, Markiewicz D, Cox TK, Chakravarti A, et al.

Identification of the cystic fibrosis gene: Genetic analysis. Science. 1989

09/08;245(4922):1073-80.

6. Riordan JR, Rommens JM, Kerem B, Alon N, Rozmahel R, Grzelczak Z, et al. Identification of

the cystic fibrosis gene: Cloning and characterization of complementary DNA. Science.

1989;245(4922):1066.

7. Collins FS. Cystic fibrosis: Molecular biology and therapeutic implications. Science. 1992

05/08;256(5058):774-9.

8. Cystic fibrosis mutation database (CFTR1) [Internet]. Available from:

http://www.genet.sickkids.on.ca/cftr/Home.html.

9. The Clinical and Functional Translation of CFTR (CFTR2) website [Internet]. Available from:

http://www.cftr2.org/index.php.

10. Cystic fibrosis manifestion .png [Internet]. Available from:

http://commons.wikimedia.org/wiki/Main_Page.

11. Kerem E. Pharmacological induction of CFTR function in patients with cystic fibrosis:

Mutation-specific therapy. Pediatr Pulmonol. 2005;40(3):183-96.

12. Hoffman LR, Ramsey BW. Cystic fibrosis TherapeuticsCurrent and future cystic fibrosis

TherapiesThe road ahead. CHEST Journal. 2013;143(1):207-13.

13. Dalemans W, Barbry P, Champigny G, Jallat S, Dott K, Dreyer D, et al. Altered chloride ion

channel kinetics associated with the ΔF508 cystic fibrosis mutation. Nature.

1991;354(6354):526-8.

14. Kerem E, Corey M, Kerem B, Rommens J, Markiewicz D, Levison H, et al. The relation

between genotype and phenotype in cystic fibrosis—analysis of the most common mutation

(ΔF508). N Engl J Med. 1990;323(22):1517-22.

Chapter 1

36

1

15. Trezíse AE, Buchwald M. In vivo cell-specific expression of the cystic fibrosis

transmembrane conductance regulator. . 1991.

16. Wilschanski M, Zielenski J, Markiewicz D, Tsui L, Corey M, Levison H, et al. Correlation of

sweat chloride concentration with classes of the cystic fibrosis transmembrane conductance

regulator gene mutations. J Pediatr. 1995;127(5):705-10.

17. Quinton P, Reddy M. Control of CFTR chloride conductance by ATP levels through non-

hydrolytic binding. . 1992.

18. Wine JJ. The genesis of cystic fibrosis lung disease. J Clin Invest. 1999;103(3):309.

19. Quinton PM. Cystic fibrosis: Impaired bicarbonate secretion and mucoviscidosis. The

Lancet. 2008;372(9636):415-7.

20. Littlewood J. Gastrointestinal complications in cystic fibrosis. J R Soc Med. 1992;85(Suppl

19):13.

21. Zielenski J. Genotype and phenotype in cystic fibrosis. Respiration. 2000;67(2):117-33.

22. Shwachman H, Dooley RR, Guilmette F, Patterson PR, Weil C, Leubner H. Cystic fibrosis of

the pancreas with varying degrees of pancreatic insufficiency. AMA journal of diseases of

children. 1956;92(4):347-68.

23. Oppenheimer EH, Esterly JR. Hepatic changes in young infants with cystic fibrosis: Possible

relation to focal biliary cirrhosis. J Pediatr. 1975;86(5):683-9.

24. Shwachman H, Kulczycki LL. Long-term study of one hundred five patients with cystic

fibrosis: Studies made over a five-to fourteen-year period. Arch Pediatr Adolesc Med.

1958;96(1):6.

25. Rosenstein BJ, Langbaum TS. Incidence of distal intestinal obstruction syndrome in cystic

fibrosis. J Pediatr Gastroenterol Nutr. 1983;2(2):299-301.

26. Weber AM, Roy CC, Morin CL, Lasalle R. Malabsorption of bile acids in children with cystic

fibrosis. N Engl J Med. 1973 11/08;289(19):1001-5.

27. Corey M, McLaughlin F, Williams M, Levison H. A comparison of survival, growth, and

pulmonary function in patients with cystic fibrosis in boston and toronto. J Clin Epidemiol.

1988;41(6):583-91.

28. Yen EH, Quinton H, Borowitz D. Better nutritional status in early childhood is associated

with improved clinical outcomes and survival in patients with cystic fibrosis. J Pediatr. 2012.

29. Borowitz D, Robinson KA, Rosenfeld M, Davis SD, Sabadosa KA, Spear SL, et al. Cystic

fibrosis foundation evidence-based guidelines for management of infants with cystic fibrosis. J

Pediatr. 2009;155(6):S73-93.

General introduction

37

1 1

30. Marino CR, Matovcik LM, Gorelick FS, Cohn JA. Localization of the cystic fibrosis

transmembrane conductance regulator in pancreas. J Clin Invest. 1991;88(2):712.

31. Couper R, Corey M, Moore D, Fisher L, Forstner G, Durie P. Decline of exocrine pancreatic

function in cystic fibrosis patients with pancreatic sufficiency. Pediatr Res. 1992;32(2):179-82.

32. Harper A, Raper HS. Pancreozymin, a stimulant of the secretion of pancreatic enzymes in

extracts of the small intestine. J Physiol (Lond ). 1943;102(1):115-25.

33. Fieker A, Philpott J, Armand M. Enzyme replacement therapy for pancreatic insufficiency:

Present and future. Clinical and experimental gastroenterology. 2011;4:55.

34. Borowitz D, Stevens C, Brettman LR, Campion M, Chatfield B, Cipolli M. International phase

III trial of liprotamase efficacy and safety in pancreatic-insufficient cystic fibrosis patients.

Journal of Cystic Fibrosis. 2011.

35. Kalivianakis M, Minich DM, Bijleveld CMA, van Aalderen WMC, Stellaard F, Laseur M, et al.

Fat malabsorption in cystic fibrosis patients receiving enzyme replacement therapy is due to

impaired intestinal uptake of long-chain fatty acids. Am J Clin Nutr. 1999 01;69(1):127-34.

36. Wouthuyzen-Bakker M, Bodewes F, Verkade H. Persistent fat malabsorption in cystic

fibrosis; lessons from patients and mice. Journal of Cystic Fibrosis. 2011.

37. Valman H, France N, Wallis P. Prolonged neonatal jaundice in cystic fibrosis. Arch Dis Child.

1971;46(250):805-9.

38. Lykavieris P, Bernard O, Hadchouel M. Neonatal cholestasis as the presenting feature in

cystic fibrosis. Arch Dis Child. 1996;75(1):67-70.

39. Farrell PM, Kosorok MR, Rock MJ, Laxova A, Zeng L, Lai H, et al. Early diagnosis of cystic

fibrosis through neonatal screening prevents severe malnutrition and improves long-term

growth. Pediatrics. 2001;107(1):1-13.

40. Scottjupp R, Lama M, Tanner MS. Prevalence of liver-disease in cystic-fibrosis. Arch Dis

Child. 1991 06;66(6):698-701.

41. Fishbein MH, Gardner KG, Potter CJ, Schmalbrock P, Smith MA. Introduction of fast MR

imaging in the assessment of hepatic steatosis. Magn Reson Imaging. 1997;15(3):287-93.

42. Lindblad A, Glaumann H, Strandvik B. Natural history of liver disease in cystic fibrosis.

Hepatology. 1999 11;30(5):1151-8.

43. Patriquin H, Lenaerts C, Smith L, Perreault G, Grignon A, Filiatrault D, et al. Liver disease in

children with cystic fibrosis: US-biochemical comparison in 195 Patients1. Radiology.

1999;211(1):229-32.

44. Dodge JA, Turck D. Cystic fibrosis: Nutritional consequences and management. Best

Practice & Research Clinical Gastroenterology. 2006;20(3):531-46.

Chapter 1

38

1

45. Strandvik B, Hultcrantz R. Liver function and morphology during long‐term fatty acid

supplementation in cystic fibrosis. Liver. 1994;14(1):32-6.

46. Stern RC, Rothstein FC, Doershuk CF. Treatment and prognosis of symptomatic gallbladder

disease in patients with cystic fibrosis. J Pediatr Gastroenterol Nutr. 1986;5(1):35-40.

47. Willi UV, Reddish JM, Teele RL. Cystic fibrosis: Its characteristic appearance on abdominal

sonography. AJR Am J Roentgenol. 1980 May;134(5):1005-10.

48. Dobson RL, Johnson MA, Hennig RC, Brown NE. Sonography of the gallbladder, biliary tree,

and pancreas in adults with cystic fibrosis. Can Assoc Radiol J. 1988 Dec;39(4):257-9.

49. Colombo C, Bertolini E, Assaisso ML, Bettinardi N, Giunta A, Podda M. Failure of

ursodeoxycholic acid to dissolve radiolucent gallstones in patients with cystic fibrosis. Acta

Paediatrica. 1993;82(6‐7):562-5.

50. Freudenberg F, Leonard MR, Liu SA, Glickman JN, Carey MC. Pathophysiological

preconditions promoting mixed “black” pigment plus cholesterol gallstones in a ΔF508 mouse

model of cystic fibrosis. Am J Physiol Gastrointest Liver Physiol. 2010;299(1):G205-14.

51. Potter CJ, Fishbein M, Hammond S, McCoy K, Qualman S. Can the histologic changes of

cystic fibrosis-associated hepatobiliary disease be predicted by clinical criteria? J Pediatr

Gastroenterol Nutr. 1997;25(1):32-6.

52. Colombo C, Battezzati PM, Crosignani A, Morabito A, Costantini D, Padoan R, et al. Liver

disease in cystic fibrosis: A prospective study on incidence, risk factors, and outcome.

Hepatology. 2002;36(6):1374-82.

53. Debray D, Kelly D, Houwen R, Strandvik B, Colombo C. Best practice guidance for the

diagnosis and management of cystic fibrosis-associated liver disease. J Cyst Fibros.

2011;10:S29-36.

54. Ooi CY, Nightingale S, Durie PR, Freedman SD. Ursodeoxycholic acid in cystic fibrosis-

associated liver disease. Journal of Cystic Fibrosis. 2012 1;11(1):72-3.

55. Giniès JL, Couetil JP, Houssin D, Guillemain R, Champion G, Bernard O. Hepatopulmonary

syndrome in a child with cystic fibrosis. J Pediatr Gastroenterol Nutr. 1996;23(4):497-500.

56. Nash KL, Allison ME, McKeon D, Lomas DJ, Haworth CS, Bilton D, et al. A single centre

experience of liver disease in adults with cystic fibrosis 1995-2006. J Cyst Fibros. 2008

05;7(3):252-7.

57. Bartlett JR, Friedman KJ, Ling SC, Pace RG, Bell SC, Bourke B, et al. Genetic modifiers of

liver disease in cystic fibrosis. JAMA: the journal of the American Medical Association.

2009;302(10):1076-83.

General introduction

39

1 1

58. Debray D, Lykavieris P, Gauthier F, Dousset B, Sardet A, Munck A, et al. Outcome of cystic

fibrosis-associated liver cirrhosis: Management of portal hypertension. J Hepatol.

1999;31(1):77-83.

59. Mueller-Abt PR, Frawley KJ, Greer RM, Lewindon PJ. Comparison of ultrasound and biopsy

findings in children with cystic fibrosis related liver disease. Journal of Cystic Fibrosis.

2008;7(3):215-21.

60. Lewindon PJ, Shepherd RW, Walsh MJ, Greer RM, Williamson R, Pereira TN, et al.

Importance of hepatic fibrosis in cystic fibrosis and the predictive value of liver biopsy.

Hepatology. 2011;53(1):193-201.

61. Paumgartner G, Beuers U. Ursodeoxycholic acid in cholestatic liver disease: Mechanisms of

action and therapeutic use revisited. Hepatology. 2002 09;36(3):525-31.

62. Colombo C, Battezzati PM, Podda M, Bettinardi N, Giunta A. Ursodeoxycholic acid for liver

disease associated with cystic fibrosis: A double‐blind multicenter trial. Hepatology.

2003;23(6):1484-90.

63. Nousia-Arvanitakis S, Fotoulaki M, Economou H, Xefteri M, Galli-Tsinopoulou A. Long-term

prospective study of the effect of ursodeoxycholic acid on cystic fibrosis-related liver disease. J

Clin Gastroenterol. 2001;32(4):324-8.

64. Cheng K, Ashby D, Smyth RL. Ursodeoxycholic acid for cystic fibrosis-related liver disease.

Cochrane Database Syst Rev. 2012;10.

65. Hultcrantz R, Mengarelli S, Strandvik B. Morphological findings in the liver of children with

cystic fibrosis: A light and electron microscopical study. Hepatology. 1986 09;6(5):881-9.

66. Shier KJ, Horn Jr RC. The pathology of liver cirrhosis in patients with cystic fibrosis of the

pancreas. Can Med Assoc J. 1963;89(13):645.

67. Sinaasappel M. Hepatobiliary pathology in patients with cystic fibrosis. Acta Paediatr

Scand. 1988;363:45,50; discussion 50-1.

68. Sokol RJ, Durie PR. Recommendations for management of liver and biliary tract disease in

cystic fibrosis. J Pediatr Gastroenterol Nutr. 1999;28:S1.

69. Cotting J, Lentze M, Reichen J. Effects of ursodeoxycholic acid treatment on nutrition and

liver function in patients with cystic fibrosis and longstanding cholestasis. Gut.

1990;31(8):918-21.

70. Rowland M, Gallagher CG, Ó'Laoide R, Canny G, Broderick A, Hayes R, et al. Outcome in

cystic fibrosis liver disease. Am J Gastroenterol. 2010;106(1):104-9.

71. Salvatore F, Scudiero O, Castaldo G. Genotype–phenotype correlation in cystic fibrosis: The

role of modifier genes. Am J Med Genet. 2002;111(1):88-95.

Chapter 1

40

1

72. Vawter GF, Shwachman H. Cystic fibrosis in adults: An autopsy study. Pathol Annu.

1979;14 Pt 2:357-82.

73. Gooding I, Dondos V, Gyi KM, Hodson M, Westaby D. Variceal hemorrhage and cystic

fibrosis: Outcomes and implications for liver transplantation. Liver transplantation.

2005;11(12):1522-6.

74. Nash K, Collier J, French J, McKeon D, Gimson A, Jamieson N, et al. Cystic fibrosis liver

disease: To transplant or not to transplant? American Journal of Transplantation.

2008;8(1):162-9.

75. Hofmann AF. Bile acid secretion, bile flow and biliary lipid secretion in humans.

Hepatology. 1990 09;12(3):17S-22S.

76. Hofmann A. Fat digestion: The interaction of lipid digestion products with micellar bile acid

solutions. In: Lipid Absorption: Biochemical and Clinical Aspects. Springer; 1976. p. 3-21.

77. Hofmann AF, Roda A. Physicochemical properties of bile acids and their relationship to

biological properties: An overview of the problem. J Lipid Res. 1984;25(13):1477-89.

78. Chiang JY. Bile acids: Regulation of synthesis. J Lipid Res. 2009;50(10):1955-66.

79. Trauner M, Claudel T, Fickert P, Moustafa T, Wagner M. Bile acids as regulators of hepatic

lipid and glucose metabolism. Digestive Diseases. 2010;28(1):220-4.

80. Hofmann AF. Enterohepatic circulation of bile acids. . 1969.

81. Small DM, Dowling RH, Redinger RN. The enterohepatic circulation of bile salts. Arch Intern

Med. 1972;130(4):552.

82. Hofmann A, Small D. Detergent properties of bile salts: Correlation with physiological

function. Annu Rev Med. 1967;18(1):333-76.

83. Hofmann AF. Bile acids: The good, the bad, and the ugly. Physiology. 1999;14(1):24-9.

84. De Vree, J Marleen L, Jacquemin E, Sturm E, Cresteil D, Bosma PJ, Aten J, et al. Mutations

in the MDR3 gene cause progressive familial intrahepatic cholestasis. Proceedings of the

National Academy of Sciences. 1998;95(1):282-7.

85. Mauad TH, van Nieuwkerk CM, Dingemans KP, Smit JJ, Schinkel AH, Notenboom RG, et al.

Mice with homozygous disruption of the mdr2 P-glycoprotein gene a novel animal model for

studies of nonsuppurative inflammatory cholangitis and hepatocarcinogenesis. The American

journal of pathology. 1994;145(5):1237.

86. van Nieuwerk CM, Groen AK, Ottenhoff R, van Wijland M, Van Den Bergh Weerman MA,

Tytgat GN, et al. The role of bile salt composition in liver pathology of mdr2 (-/-) mice:

Differences between males and females. J Hepatol. 1997 01;26(1):138-45.

General introduction

41

1 1

87. Boyer JL, Klatskin G. Canalicular bile flow and bile secretory pressure. evidence for a non-

bile salt dependent fraction in the isolated perfused rat liver. Gastroenterology. 1970

Dec;59(6):853-9.

88. Drew R, Priestly B. Choleretic and cholestatic effects of infused bile salts in the rat.

Experientia. 1979;35(6):809-11.

89. Hofmann AF. Biliary secretion and excretion in health and disease: Current concepts. Ann

Hepatol. 2007 01;6(1):15-27.

90. Roberts SK, Kuntz SM, Gores GJ, LaRusso NF. Regulation of bicarbonate-dependent

ductular bile secretion assessed by lumenal micropuncture of isolated rodent intrahepatic bile

ducts. Proceedings of the National Academy of Sciences. 1993;90(19):9080-4.

91. Fitz JG. Regulation of cholangiocyte secretion. Semin Liver Dis. 2002 08;22(3):241-9.

92. Fitz JG, Basavappa S, McGill J, Melhus O, Cohn JA. Regulation of membrane chloride

currents in rat bile duct epithelial cells. J Clin Invest. 1993 01;91(1):319-28.

93. Schlenker T, Romac JM, Sharara AI, Roman RM, Kim SJ, Larusso N, et al. Regulation of

biliary secretion through apical purinergic receptors in cultured rat cholangiocytes. American

Journal of Physiology-Gastrointestinal and Liver Physiology. 1997;273(5):G1108-17.

94. Minagawa N, Nagata J, Shibao K, Masyuk AI, Gomes DA, Rodrigues MA, et al. Cyclic AMP

regulates bicarbonate secretion in cholangiocytes through release of ATP into bile.

Gastroenterology. 2007;133(5):1592-602.

95. Chen B, Nicol G, Cho WK. Role of calcium in volume-activated chloride currents in a mouse

cholangiocyte cell line. J Membr Biol. 2007;215(1):1-13.

96. Feranchak AP, Sokol RJ. Cholangiocyte biology and cystic fibrosis liver disease. Seminars in

liver disease; ; 2001.

97. Gunderson KL, Kopito RR. Conformational states of CFTR associated with channel gating:

The role of ATP binding and hydrolysis. Cell. 1995;82(2):231-9.

98. Schinkel AH, Jonker JW. Mammalian drug efflux transporters of the ATP binding cassette

(ABC) family: An overview. Adv Drug Deliv Rev. 2012.

99. Schwiebert EM, Egan ME, Hwang T, Fulmer SB, Allen SS, Cutting GR, et al. CFTR regulates

outwardly rectifying chloride channels through an autocrine mechanism involving ATP. Cell.

1995;81(7):1063-73.

100. Banales JM, Arenas F, Rodríguez‐Ortigosa CM, Sáez E, Uriarte I, Doctor RB, et al.

Bicarbonate‐rich choleresis induced by secretin in normal rat is taurocholate‐dependent and

involves AE2 anion exchanger. Hepatology. 2006;43(2):266-75.

Chapter 1

42

1

101. Fiorotto R, Spirli C, Fabris L, Cadamuro M, Okolicsanyi L, Strazzabosco M. Ursodeoxycholic

acid stimulates cholangiocyte fluid secretion in mice via CFTR-dependent ATP secretion.

Gastroenterology. 2007 11;133(5):1603-13.

102. Russell DW. Fifty years of advances in bile acid synthesis and metabolism. J Lipid Res.

2009;50(Supplement):S120-5.

103. Lu TT, Makishima M, Repa JJ, Schoonjans K, Kerr TA, Auwerx J, et al. Molecular basis for

feedback regulation of bile acid synthesis by nuclear receptors. Mol Cell. 2000;6(3):507-15.

104. Wang H, Chen J, Hollister K, Sowers LC, Forman BM. Endogenous bile acids are ligands for

the nuclear receptor FXR/BAR. Mol Cell. 1999;3(5):543-53.

105. Goodwin B, Jones SA, Price RR, Watson MA, McKee DD, Moore LB, et al. A regulatory

cascade of the nuclear receptors FXR, SHP-1, and LRH-1 represses bile acid biosynthesis. Mol

Cell. 2000;6(3):517-26.

106. Lundåsen T, Gälman C, Angelin B, Rudling M. Circulating intestinal fibroblast growth

factor 19 has a pronounced diurnal variation and modulates hepatic bile acid synthesis in

man. J Intern Med. 2006;260(6):530-6.

107. Xie M, Holcomb I, Deuel B, Dowd P, Huang A, Vagts A, et al. FGF-19, a novel fibroblast

growth factor with unique specificity for FGFR4. Cytokine. 1999;11(10):729-35.

108. Rao Y, Studer EJ, Stravitz RT, Gupta S, Qiao L, Dent P, et al. Activation of the

Raf‐1/MEK/ERK cascade by bile acids occurs via the epidermal growth factor receptor in

primary rat hepatocytes. Hepatology. 2002;35(2):307-14.

109. Trezíse AE, Buchwald M. In vivo cell-specific expression of the cystic fibrosis

transmembrane conductance regulator. . 1991.

110. De Jonge HR, Ballmann M, Veeze H, Bronsveld I, Stanke F, Tümmler B, et al. Ex vivo CF

diagnosis by intestinal current measurements (ICM) in small aperture, circulating ussing

chambers. Journal of Cystic Fibrosis. 2004;3:159-63.

111. Wilschanski M, Durie P. Pathology of pancreatic and intestinal disorders in cystic fibrosis.

J R Soc Med. 1998;91(Suppl 34):40.

112. Bijvelds MJC, Jorna H, Verkade HJ, Bot AGM, Hofmann F, Agellon LB, et al. Activation of

CFTR by ASBT-mediated bile salt absorption. American Journal of Physiology-Gastrointestinal

and Liver Physiology. 2005;289(5):G870.

113. Andersen DH. Cystic fibrosis of the pancreas and its relation to celiac diseasea clinical and

pathologic study. American journal of Diseases of Children. 1938;56(2):344-99.

114. Hahn TJ, Squires AE, Halstead LR, Strominger DB. Reduced serum 25-hydroxyvitamin D

concentration and disordered mineral metabolism in patients with cystic fibrosis. J Pediatr.

1979;94(1):38-42.

General introduction

43

1 1

115. Walters TR, Koch CHF. Hemorrhagic diathesis and cystic fibrosis in infancy. Arch Pediatr

Adolesc Med. 1972;124(5):641.

116. Farrell PM, Bieri JG, Fratantoni JF, Wood RE, di Sant'Agnese PA. The occurrence and

effects of human vitamin E deficiency: A study in patients with cystic fibrosis. J Clin Invest.

1977;60(1):233.

117. Domínguez‐Muñoz JE. Pancreatic exocrine insufficiency: Diagnosis and treatment. J

Gastroenterol Hepatol. 2011;26(s2):12-6.

118. Stallings VA, Stark LJ, Robinson KA, Feranchak AP, Quinton H. Evidence-based practice

recommendations for nutrition-related management of children and adults with cystic fibrosis

and pancreatic insufficiency: Results of a systematic review. J Am Diet Assoc. 2008;108(5):832-

9.

119. Weber AM. Relationship between bile acid malabsorption and pancreatic insufficiency in

cystic fibrosis. Gut. 1976;17(4):295.

120. Watkins J, Tercyak A, Szczepanik P, Klein P. Bile salt kinetics in cystic fibrosis: Influence of

pancreatic enzyme replacement. Gastroenterology. 1977;73(5):1023.

121. Kalivianakis M, Bronsveld I, Jonge de H, Sinaasappel M, Havinga R, Kuipers F, et al.

Increased fecal bile salt excretion is independent of the presence of dietary fat malabsorption

in two mouse models for cystic fibrosis. Proefschrift Mini Kalivianalis. 2003.

122. Fondacaro JD, Heubi JE, Kellogg FW. Intestinal bile acid malabsorption in cystic fibrosis: A

primary mucosal cell defect. Pediatr Res. 1982;16(6):494-8.

123. O'Brien S, Mulcahy H, Fenlon H, O'Broin A, Casey M, Burke A, et al. Intestinal bile acid

malabsorption in cystic fibrosis. Gut. 1993;34(8):1137.

124. Ridlon JM, Kang D, Hylemon PB. Bile salt biotransformations by human intestinal

bacteria. J Lipid Res. 2006;47(2):241-59.

125. Lisowska A, Wójtowicz J, Walkowiak J. Small intestine bacterial overgrowth is frequent in

cystic fibrosis: Combined hydrogen and methane measurements are required for its detection.

Acta Biochim Pol. 2009;56(4):631.

126. De Lisle RC, Roach EA, Norkina O. Eradication of small intestinal bacterial overgrowth in

the cystic fibrosis mouse reduces mucus accumulation. J Pediatr Gastroenterol Nutr.

2006;42(1):46-52.

127. Wouthuyzen-Bakker M, Bijvelds MJC, de Jonge HR, De Lisle RC, Burgerhof JGM, Verkade

HJ. Effect of antibiotic treatment on fat absorption in mice with cystic fibrosis. Pediatr Res.

2011;71(1):4-12.

Chapter 1

44

1

128. Bruzzese E, Raia V, Gaudiello G, Polito G, Buccigrossi V, Formicola V, et al. Intestinal

inflammation is a frequent feature of cystic fibrosis and is reduced by probiotic administration.

Aliment Pharmacol Ther. 2004;20(7):813-9.

129. Werlin SL, Benuri-Silbiger I, Kerem E, Adler SN, Goldin E, Zimmerman J, et al. Evidence of

intestinal inflammation in patients with cystic fibrosis. J Pediatr Gastroenterol Nutr.

2010;51(3):304.

130. Ollero M, Junaidi O, Zaman MM, Tzameli I, Ferrando AA, Andersson C, et al. Decreased

expression of peroxisome proliferator activated receptor γ in CFTR−/− mice. J Cell Physiol.

2004;200(2):235-44.

131. Dekkers JF, van der Ent, Cornelis K, Kalkhoven E, Beekman JM. PPARγ as a therapeutic

target in cystic fibrosis. Trends Mol Med. 2012;18(5):283-91.

132. Accurso FJ, Rowe SM, Clancy J, Boyle MP, Dunitz JM, Durie PR, et al. Effect of VX-770 in

persons with cystic fibrosis and the G551D-CFTR mutation. N Engl J Med. 2010;363(21):1991-

2003.

133. Dekkers JF, Wiegerinck CL, de Jonge HR, Bronsveld I, Janssens HM, de Winter-de Groot,

Karin M, et al. A functional CFTR assay using primary cystic fibrosis intestinal organoids. Nat

Med. 2013.

134. Pereira TN, Lewindon PJ, Greer RM, Hoskins AC, Williamson RM, Shepherd RW, et al.

Transcriptional basis for hepatic fibrosis in cystic fibrosis–associated liver disease. J Pediatr

Gastroenterol Nutr. 2012;54:328-35.

135. Freudenberg F, Broderick AL, Yu BB, Leonard MR, Glickman JN, Carey MC.

Pathophysiological basis of liver disease in cystic fibrosis employing a DeltaF508 mouse model.

Am J Physiol Gastrointest Liver Physiol. 2008 06;294(6):G1411-20.

136. Blanco PG, Zaman MM, Junaidi O, Sheth S, Yantiss RK, Nasser IA, et al. Induction of colitis

in cftr–/–mice results in bile duct injury. American Journal of Physiology-Gastrointestinal and

Liver Physiology. 2004;287(2):G491.

137. Moyer K, Balistreri W. Hepatobiliary disease in patients with cystic fibrosis. Curr Opin

Gastroenterol. 2009;25(3):272-8.

138. Colombo C, Battezzati PM, Podda M, Bettinardi N, Giunta A. Ursodeoxycholic acid for

liver disease associated with cystic fibrosis: A double‐blind multicenter trial. Hepatology.

1996;23(6):1484-90.

139. Lindblad A, Glaumann H, Strandvik B. A two‐year prospective study of the effect of

ursodeoxycholic acid on urinary bile acid excretion and liver morphology in cystic fibrosis–

associated liver disease. Hepatology. 1998;27(1):166-74.

General introduction

45

1 1

140. Ellis EL, Mann DA. Clinical evidence for the regression of liver fibrosis. J Hepatol. 2012

5;56(5):1171-80.

141. Brown A, Goodman Z. Hepatitis B-associated fibrosis and fibrosis/cirrhosis regression

with nucleoside and nucleotide analogs. Expert review of gastroenterology & hepatology.

2012;6(2):187-98.

142. D'Ambrosio R, Aghemo A, Rumi MG, Ronchi G, Donato MF, Paradis V, et al. A

morphometric and immunohistochemical study to assess the benefit of a sustained virological

response in hepatitis C virus patients with cirrhosis. Hepatology. 2012;56(2):532-43.

143. Friedman SL. Evolving challenges in hepatic fibrosis. Nature Reviews Gastroenterology

and Hepatology. 2010;7(8):425-36.

144. Leeuwen L, Fitzgerald DA, Gaskin KJ. Liver disease in cystic fibrosis. Paediatric Respiratory

Reviews. 2013.

145. Lenaerts C, Lapierre C, Patriquin H, Bureau N, Lepage G, Harel F, et al. Surveillance for

cystic fibrosis-associated hepatobiliary disease: Early ultrasound changes and predisposing

factors. J Pediatr. 2003 09;143(3):343-50.

146. Williams SGJ, Evanson JE, Barrett N, Hodson ME, Boultbee JE, Westaby D. An ultrasound

scoring system for the diagnosis of liver disease in cystic fibrosis. J Hepatol. 1995;22(5):513-21.

147. Debray D, Lykavieris P, Gauthier F, Dousset B, Sardet A, Munck A, et al. Outcome of cystic

fibrosis-associated liver cirrhosis: Management of portal hypertension. J Hepatol.

1999;31(1):77-83.

148. Mayer-Hamblett N, Kloster M, Ramsey BW, Narkewicz MR, Saiman L, Goss CH. Incidence

and clinical significance of elevated liver function tests in cystic fibrosis clinical trials.

Contemporary clinical trials. 2012.

149. Potter CJ, Fishbein M, Hammond S, McCoy K, Qualman S. Can the histologic changes of

cystic fibrosis-associated hepatobiliary disease be predicted by clinical criteria? J Pediatr

Gastroenterol Nutr. 1997;25(1):32-6.

150. Mueller-Abt PR, Frawley KJ, Greer RM, Lewindon PJ. Comparison of ultrasound and biopsy

findings in children with cystic fibrosis related liver disease. Journal of Cystic Fibrosis. 2008

5;7(3):215-21.

151. de Lédinghen V, Le Bail B, Rebouissoux L, Fournier C, Foucher J, Miette V, et al. Liver

stiffness measurement in children using FibroScan: Feasibility study and comparison with

fibrotest, aspartate transaminase to platelets ratio index, and liver biopsy. J Pediatr

Gastroenterol Nutr. 2007;45(4):443-50.

152. Foucher J, Chanteloup E, Vergniol J, Castera L, Le Bail B, Adhoute X, et al. Diagnosis of

cirrhosis by transient elastography (FibroScan): A prospective study. Gut. 2006;55(3):403-8.

Chapter 1

46

1

153. Witters P, De Boeck K, Dupont L, Proesmans M, Vermeulen F, Servaes R, et al. Non-

invasive liver elastography (fibroscan) for detection of cystic fibrosis-associated liver disease.

Journal of Cystic Fibrosis. 2009;8(6):392-9.

154. Malbrunot-Wagner A, Bridoux L, Nousbaum J, Riou C, Dirou A, Ginies J, et al. Transient

elastography and portal hypertension in pediatric patients with cystic fibrosis: Transient

elastography and cystic fibrosis. Journal of Cystic Fibrosis. 2011;10(5):338-42.

155. Wilke M, Buijs-Offerman RM, Aarbiou J, Colledge WH, Sheppard DN, Touqui L, et al.

Mouse models of cystic fibrosis: Phenotypic analysis and research applications. J Cyst Fibros.

2011;10:S152-71.

156. Scholte BJ, Davidson DJ, Wilke M, de Jonge HR. Animal models of cystic fibrosis. J Cyst

Fibros. 2004;3:183-90.

157. Bobadilla JL, Macek M, Fine JP, Farrell PM. Cystic fibrosis: A worldwide analysis of CFTR

mutations—correlation with incidence data and application to screening. Hum Mutat.

2002;19(6):575-606.

158. Durie PR, Kent G, Phillips MJ, Ackerley CA. Characteristic multiorgan pathology of cystic

fibrosis in a long-living cystic fibrosis transmembrane regulator knockout murine model. Am J

Pathol. 2004 04;164(4):1481-93.

159. Pall H, Zaman MM, Andersson C, Freedman SD. Decreased peroxisome proliferator

activated receptor [alpha] is associated with bile duct injury in cystic fibrosis transmembrane

conductance regulator-/-mice. J Pediatr Gastroenterol Nutr. 2006;42(3):275.

160. van Doorninck JH, French PJ, Verbeek E, Peters RH, Morreau H, Bijman J, et al. A mouse

model for the cystic fibrosis delta F508 mutation. EMBO J. 1995 09/15;14(18):4403-11.

161. Ratcliff R, Evans MJ, Cuthbert AW, MacVinish LJ, Foster D, Anderson JR, et al. Production

of a severe cystic fibrosis mutation in mice by gene targeting. Nat Genet. 1993 05;4(1):35-41.

162. Norkina O, Burnett TG, De Lisle RC. Bacterial overgrowth in the cystic fibrosis

transmembrane conductance regulator null mouse small intestine. Infect Immun.

2004;72(10):6040-9.

163. De Lisle RC, Roach EA, Norkina O. Eradication of small intestinal bacterial overgrowth in

the cystic fibrosis mouse reduces mucus accumulation. J Pediatr Gastroenterol Nutr.

2006;42(1):46.

164. Bijvelds MJC, Bronsveld I, Havinga R, Sinaasappel M, de Jonge HR, Verkade HJ. Fat

absorption in cystic fibrosis mice is impeded by defective lipolysis and post-lipolytic events. Am

J Physiol Gastrointest Liver Physiol. 2005 04;288(4):G646-53.

General introduction

47

1 1

165. Dawson PA, Haywood J, Craddock AL, Wilson M, Tietjen M, Kluckman K, et al. Targeted

deletion of the ileal bile acid transporter eliminates enterohepatic cycling of bile acids in mice.

J Biol Chem. 2003;278(36):33920-7.

166. Miyata M, Takamatsu Y, Kuribayashi H, Yamazoe Y. Administration of ampicillin elevates

hepatic primary bile acid synthesis through suppression of ileal fibroblast growth factor 15

expression. J Pharmacol Exp Ther. 2009;331(3):1079-85.

167. Islam K, Fukiya S, Hagio M, Fujii N, Ishizuka S, Ooka T, et al. Bile acid is a host factor that

regulates the composition of the cecal microbiota in rats. Gastroenterology.

2011;141(5):1773-81.

Chapter 1

48

1

CHAPTER 2

PERSISTENT FAT MALABSORPTION

IN CYSTIC FIBROSIS; LESSONS

FROM PATIENTS AND MICE

Adapted from: Journal of Cystic Fibrosis 10.3 (2011): 150-158.

Marjan Wouthuyzen-Bakker

Frank A. J. A. Bodewes,

Henkjan J. Verkade

University Medical Center Groningen, the Netherlands

Chapter 3

50

2

ABSTRACT

Fat malabsorption in pancreatic insufficient cystic fibrosis (CF) patients is classically treated

with pancreatic enzyme replacement therapy (PERT). Despite PERT, intestinal fat absorption

remains insufficient in most CF patients. Several factors have been suggested to contribute to

the persistent fat malabsorption in CF (CFPFM). We reviewed the current insights concerning

the proposed causes of CFPFM and the corresponding intervention studies. Most data are

obtained from studies in CF patients and CF mice. Based on the reviewed literature, we

conclude that alterations in intestinal pH and intestinal mucosal abnormalities are most likely

to contribute to CFPFM. The presently available data indicate that acid suppressive drugs and

broad spectrum antibiotics could be helpful in individual CF patients for optimizing fat

absorption and/or nutritional status.

Persistent fat malabsorption in cystic fibrosis; lessons from patients and mice

51

2

INTRODUCTION

Most patients with cystic fibrosis (CF) have a suboptimal nutritional status. In order to

minimize morbidity and mortality, obtaining and maintaining a good nutritional status

remains one of the important challenges in CF (1-3). Next to preventing and controlling

pulmonary infections, a high dietary energy intake diet combined with pancreatic enzyme

replacement therapy (PERT) are classic cornerstones in CF care.

Despite optimizing and maximizing PERT, fat malabsorption in most CF patients persist

(CFPFM). Rather than the physiological absorption above 95%, most CF patients show a

decreased intestinal fat absorption of ± 85-90% of dietary fat intake (4,5). It is reasonable to

assume that at least part of the failure to approach a physiological degree of fat absorption

could still be attributed to inefficiency and inaccurate dosing of PERT. Also, a non-

synchronous entrance of PERT with dietary energy intake into the duodenum may play a role

in CFPFM (6). However, besides PERT related issues, other factors have been implied to be

involved in CFPFM (7,8).

During the last decades, several of these potential contributing factors to CFPFM have been

addressed and investigated in patient studies as well as animal studies. In this review, we will

provide an overview of the different mechanisms that have been implied to contribute to

CFPFM and describe the results of corresponding interventional studies on fat absorption

and/or nutritional status. We will mainly describe studies performed in CF patients and CF

mice. CF mice are especially suitable to exclusively investigate CFPFM, as most CF mouse

models do not suffer from pancreatic insufficiency but do exhibit impaired absorption of fat

(9). In order to fully recognize and understand the CF-related factors that might be involved in

CFPFM, we start to describe the physiological process of fat digestion and absorption.

PHYSIOLOGY OF FAT DIGESTION AND ABSORPTION

Dietary fat intake mainly consists of long-, medium- and short-chain triglycerides. The

mechanism of digestion and absorption of these lipids can be dissected into several

sequential steps. In contrast to long-chain triglycerides, medium- and short-chain triglycerides

are known to circumvent several steps in lipid digestion and absorption. For example, unlike

long-chain triglycerides, medium- and short chain triglycerides are less dependent upon

solubilization, escape the re-esterification into triglycerides and are directly absorbed into the

portal system without being assembled into chylomicrons (10). Since the majority of dietary

fat and energy intake consists of long-chain triglycerides (92-96%), we exclusively focused on

their mechanism of intestinal digestion and absorption. The mechanism of digestion and

absorption of long-chain triglycerides can be dissected into several sequential processes

(figure 1A).

Chapter 3

52

2

Figure 1. Simplified scheme of fat digestion and absorption in health and in CF conditions. A) Simplified

representation of the process of long-chain triglyceride digestion and absorption in health as described

in the text. B) Overview of the various steps in fat digestion and absorption in relation to findings in CF

patients and CF mice. +; disturbed, -; not disturbed, +/-; likely to be disturbed. ND: not determined.

References are described in the text.

.

CF

patients

CFTR knockout

mice

Homozygous delta

F508 mice

1. Emulsification

2. Lipolysis

3. Solubilization

4. Translocation

5. Intracellular processing

6. Chylomicron production

+

+

+

±

±

±

-

+

+

±

ND

ND

-

-

+

±

ND

ND

A

B

Persistent fat malabsorption in cystic fibrosis; lessons from patients and mice

53

2

1) Emulsification. Emulsification of triglycerides is a process in which (water-insoluble) fat

droplets are suspended in an aqueous environment. Emulsification can be achieved by

mechanical as well as chemical means. Mechanical emulsification is attained by chewing and

forcing dietary fat with high pressure through a small opening (e.g. the pylorus) and

dispersing large fat droplets into smaller droplets. Chemical emulsification is attained by the

action of bile and prevents the emulsion from re-coalescing. Emulsification results in a fine,

relatively stable oil-in-water emulsion with an increased surface area. The water soluble

digestive / lipolytic enzymes are active at the site of the water-oil interface.

2) Lipolysis. During lipolysis, the breakdown (hydrolysis) of lipids into smaller particles is

continued. Around 10-30% of the triglycerides are hydrolyzed in the stomach into diglycerides

and free fatty acids, catalyzed by lingual and gastric lipase (a process also known as

biochemical/lipolytic emulsification (10)). Under physiological conditions, pancreatic lipase

completes the process in the proximal part of the small intestine, by hydrolyzing the

remaining triglycerides and diglycerides into monoglycerides and free fatty acid molecules.

Bile salts are amphipathic molecules, i.e. they possess a hydrophilic and a hydrophobic site.

Upon exposure to dietary lipid emulsion at the level of the proximal small intestine, the

hydrophobic site orients itself towards the hydrophobic lipid droplets, whereas their

hydrophilic site exposes itself to the aqueous (water) phase. Lipids droplets whose surface is

thus covered bile salts are not accessible to pancreatic lipase. Pancreatic co-lipase allows the

anchoring of the pancreas lipase to the oil-water interface and is essential for lipolytic activity.

Under physiological conditions, pancreatic lipase is present in vast excess. Thus, in CF, fat

maldigestion only occurs during severe pancreatic malfunction (11). During exocrine

pancreatic insufficiency when pancreatic lipase is severely reduced, lipolysis is more

dependent upon the activity of lingual and gastric lipase. In CF conditions, lingual lipase

remains fully active in the intestine, where it accounts for more than 90% of lipase activity,

even when the intestinal pH drop below the optimal level for bile salt dependent lipolysis

(12).

3) Solubilization. The process in which fat molecules are dissolved in water in the form of

(mixed) micelles is called solubilization. Solubilization enhances the aqueous solubility of fatty

acids by several orders of magnitude (100 to 1000 fold) (10). Solubilization is necessary for

monoglycerides and free fatty acids to efficiently overcome the diffusion barrier of the so

called unstirred water layer of the enterocytes (10). The unstirred water layer separates the

enterocytes from the luminal contents of the intestine. Solubilization is achieved by the

formation of mixed micelles; mainly consisting of phospholipids and bile salts derived from

biliary secretion. The diameter of the lipid droplets ranges from 100 – 1000 nm, whereas that

of mixed micelles ranges 3-5 nm. The hydrophobic part of the lipid molecules, such as the

acyl-chains, will be oriented inwards, whereas the hydrophilic parts, (such as the carboxylic

headgroups) orients towards the aqueous outside of the micelle. Saturated fatty acids are

more dependent on solubilization than unsaturated fatty acids, due to the higher

Chapter 3

54

2

hydrophobicity of the former. The difference in bile salt dependency can be derived from

absorption studies in rats with chronic bile diversion: in such an intestinal bile deficient

condition, absorption of saturated fatty acids is below 30% of the ingested amount, while the

absorption of unsaturated fatty acids is relatively maintained (~80%) (13). Just like the

digestion of fat, the process of micelle formation is also pH sensitive. Low intestinal pH levels

can severely inhibit micelle formation or induce premature release of lipolytic products out of

micelles (10).

4) Translocation. Once the mixed micelles diffuse through the unstirred water layer and arrive

at the proximity of the (apical) enterocyte membrane, the free fatty acids and monoglycerides

dissociate from the micelles. It has been postulated that the acidic microclimate near the

apical membrane of intestinal mucosal cells, induces micelle- disintegration and favours the

translocation of fatty acid molecules across the enterocyte membrane (10). It has remained

unclear whether intestinal membrane transporters are essential for the translocation of fatty

acids and monoglycerides. It is suggested that free fatty acid uptake is concentration-

dependent, in which a high intra-luminal concentration drives passive diffusion. Two putative

intestinal transporters have been proposed to be involved fatty acid uptake; the fatty acid

binding protein (FABP) and the fatty acid translocase / cluster determinant 36 (FAT/CD36)

(14). However, both transporters are not likely to be involved in fatty acid uptake into the

enterocytes. FABP is only expressed in a small area of the crypt-villus of the intestine and

knockout mice for FABP and CD36 do not exhibit impairments in fatty acid uptake (15-17).

5) Intracellular processing. After being absorbed into the enterocytes, the lipolytic products

migrate to the endoplasmic reticulum, possibly mediated via the fatty acid binding proteins

(14,15). At the cytoplasmic surface of the endoplasmic reticulum, the fatty acids and

monoglycerides are re-esterified into triglycerides. Under physiological conditions, re-

esterification mainly occurs via the monoacylglycerol pathway, i.e. the sequential acylation of

monoacylglycerol by acyl-CoA (10).

6) Chylomicron production. Newly synthesized triglycerides are transferred into the smooth

endoplasmic reticulum and are assembled into lipoprotein particles called chylomicrons.

Intestinal phospholipids are required for chylomicron production in order to prevent

accumulation in the enterocyte (18,19). Maturation of chylomicrons (i.e. assembly of fat

particles with a phospholipid-cholesterol-apolipoprotein surface) takes place in the Golgi

apparatus. Chylomicron formation is followed by exocytosis via the secretory pathway at the

basolateral surface of the enterocyte. The chylomicrons are released into the circulation via

the mesenteric lymph system, via the thoracic duct into the venous system, after which their

contents are systemically delivered.

Persistent fat malabsorption in cystic fibrosis; lessons from patients and mice

55

2

FACTORS PROPOSED TO BE ASSOCIATED WITH PERSISTENT FAT

MALABSORPTION IN CF

Several factors have been proposed to contribute to CFPFM. For each of these factors we will

describe; (1) how these factors are affected in CF conditions; (2) what is known about the

relation of these factors with CFPFM; and, (3) what the result of intervention studies are in CF

patients and CF mice when improving or altering these factors (figure 1B).

3.1 INTESTINAL PH

1) CF conditions. CFTR dysfunction reduces pancreatic and duodenal bicarbonate secretion

(20,21). Gastric bicarbonate secretion in CF patients is not affected (22). Hence, the duodenal

pH is (on average) 1-2 units lower in CF patients compared with healthy controls. Also, CF

patients have significantly longer periods (postprandial) in which the duodenal pH drops

below 4 (23). The pH values in jejunal and ileal contents from CF patients vary from lower to

similar pH values compared with healthy controls (23). CFTRtm1Cam

knockout mice have

adequate pancreatic bicarbonate secretion, but probably do have reduced secretions of

duodenal bicarbonate (9). This might explain the observation in CFTRtm1Unc

knockout mice,

where duodenal pH is only minimally reduced (~0.25) (21).

2) Relation with CFPFM. A normal intestinal pH is essential for adequate digestion and

absorption of fat (10). Recently, Quinton et al. showed that adequate bicarbonate secretion is

essential for mucin expansion in the intestine (24,25). When mucin expansion is disturbed by

inadequate release of bicarbonate in CF conditions, viscous and sticky mucus might impair

translocation of lipids to the enterocyte. A strong relation has been described between

postprandial duodenal pH levels and the degree of fat malabsorption in CF patients treated

with pancreatic enzymes (26). These data indicate that interventions to increase intestinal pH

values might improve CFPFM.

3) Intervention studies. Two double-blind crossover studies evaluated fat absorption in

pediatric and adult CF patients after the addition of bicarbonate to enteric coated pancreatic

enzymes. In the first study, the average fat absorption increased from 75% to 82% of the

ingested amount (27). In this study, a beneficial effect on fat absorption was observed in

seventy-five percent of the patients. In the second study, the average fat absorption did not

increase, but fifty percent of patients did show an improvement in fat absorption (>5%) after

bicarbonate supplementation (28) . Due to the overall minimal observed changes in fat

absorption, pancreatic enzymes buffered with bicarbonate are not implemented in CF care.

In a systematic review, Jones A. evaluated the effect of acid suppressant therapy on fat

absorption and/or faecal fat excretion and nutritional status (29). While multiple studies

reported improved fat absorption or reduced faecal fat excretion after acid suppressive

Chapter 3

56

2

therapy, other studies reported no improvements. Improved fat absorption was not

accompanied with an improvement in nutritional status. The performed studies showed high

variability in the used dosage of pancreatic enzymes, choice of acidic suppressive medication

and differed in the inclusion criteria for the degree of fat malabsorption. Due to these large

inter-study differences, it remains difficult to draw an overall definite conclusion about the

effectiveness of acid suppressive drugs. Evaluating individual data in these intervention

studies, a subgroup of patients clearly showed improved fat absorption. The results indicate

that some CF patients do benefit from acid suppressive therapy. Why some patients respond,

but others show no improvement in fat absorption, illustrates that an altered pH is not the

only factor responsible CFPFM. Next to this, it is known that acid suppressive drugs can induce

small intestinal bacterial overgrowth (SIBO) and can alter bile salt metabolism, with a

potentially negative effect on fat absorption (30). Therefore, acid suppressive therapy should

be imposed with caution, especially in CF patients who are relatively more prone to SIBO. A

trial of proton pump inhibitors might be useful to evaluate its effectiveness in the individual

CF patient.

Bijvelds et al. investigated the effect of the acidic suppressive drug omeprazole on lipolysis

and uptake of lipolytic products in CFTRtm1CAM

knockout mice by using radio-isotope labelled

triglycerides and fatty acids. The researchers showed that lipolytic activity and lipid

absorption improved after omeprazole treatment (9). It is likely that increasing the intestinal

pH has a general positive effect on intestinal fat absorption, since lipolytic and post-lipolytic

activity also partially improved in the omeprazole treated wild type mice (9).

3.2 INTRA-LUMINAL BILE SALTS

1) CF conditions. CF patients have an increased loss of faecal bile salts (31,32). It is speculated

that the loss of bile salts is due to impaired bile salt uptake by intestinal mucosal alterations;

like thickening of the mucus barrier and/or SIBO (33). Because the biosynthesis of taurine is

limited in humans, the faecal loss of bile salts induces an increased glycine/taurine ratio of

conjugated bile salts in CF patients (34). In addition, CF patients have an altered bile salt

composition in gallbladder bile. Due to a quantitative increase in biliary cholate, the percent

contribution of cholate is higher, and the percent contribution of chenodeoxycholate and

deoxycholate is lower in the bile of CF patients (35).

2) Relation with CFPFM. It has been proposed that excessive faecal loss of bile salts diminishes

the bile salt pool in CF patients and consequently impairs fat absorption by reducing the

solubilization capacity of bile (31). However, a subsequent study indicated that the amount of

faecal bile salt excretion was not related to the degree of fat malabsorption in CF patients

(32). Strandvik et al. showed that adult CF patients have normal to large bile salt pool sizes

and similar amounts of duodenal bile salts as healthy controls (35). Bile salt synthesis was

normal or even increased, indicating that CF patients adequately compensate for the faecal

Persistent fat malabsorption in cystic fibrosis; lessons from patients and mice

57

2

bile salt loss. We confirmed in CF mouse models that faecal loss of bile salts does not

influence the absorption of fat (9). Homozygous ∆F508 mice and CFTRtm1CAM

knockout mice

both exhibit, to the same extent, an increased faecal loss of bile salts, but only the CFTRtm1CAM

knockout mice had fat malabsorption (9). In conclusion, bile salt malabsorption in itself, at the

levels observed in CF patients and CF mice, does not seem to contribute to CFPFM.

Theoretically, the increased glycine/taurine ratio may impair fat absorption in an acidic

intestinal lumen. Due to the higher pKa of glycine, glycine conjugated bile salts are less able to

remain in micellar solution (36). In addition, part of the glycoconjuates are passively absorbed

in the in the proximal part of the intestine and are, in comparison to tauroconjugates, less

resistant to bacterial degradation (37,38).

Deoxycholate and chenodeoxycholate are relatively hydrophobic in comparison to other bile

salts. Therefore, a reduction in these bile salts may theoretically impair fat absorption by

diminishing the solubilization capacity of bile. In general, the percent contribution, and not

the quantitative amount, of deoxycholate and of chenodeoxycholate is diminished in CF

patients. Thus, the altered bile salt composition is not likely to contribute to CFPFM. This

hypothesis is supported by the fact that both homozygous ∆F508 mice and CFTR tm1CAM

knockout mice show these alterations, while only the CFTR tm1CAM

knockout mice exhibit fat

malabsorption (9).

3) Intervention studies. Multiple studies showed that taurine supplementation reduces faecal

fat excretion and improves the nutritional status of CF patients, particularly in patients with

severe steatorrhoea (39-41). However, the beneficial effect of taurine supplementation on fat

absorption is not unequivocally demonstrated (42-45) . Furthermore, the degree of fat

absorption did not relate to changes in the serum glycine/taurine ratio in CF children (45).

Altogether, the use of taurine supplementation remains controversial and is not implemented

in nutritional CF care.

3.3 INTESTINAL MUCOSAL ABNORMALITIES

1) CF conditions. Several intestinal mucosal abnormalities are described in CF patients. These

abnormalities include accumulation of viscous and sticky mucus, SIBO, ileal hyperthrophy and

villous atrophy, increased intestinal permeability and (chronic) inflammation of the small

intestine (46-53).

2) Relation with CFPFM. All the above mentioned factors may theoretically contribute to fat

malabsorption in CF patients, as they might impair adequate translocation of fatty acids in(to)

the enterocyte. In addition, SIBO might impair micelle formation by the bacterial

deconjugation of bile salts (54). However, no studies evaluated the actual contribution of

intestinal mucosal abnormalities to CFPFM.

Chapter 3

58

2

3) Intervention studies. The group of De Lisle et al. evaluated the effect of antibiotic treatment

on intestinal inflammation, mucus accumulation and SIBO in CFTRtm1Unc

knockout mice (55).

CFTRtm1Unc

knockout mice display a phenotype of SIBO with a 400 fold increase in small

intestinal bacterial content compared to wild type littermates (56). Broad spectrum antibiotic

treatment with ciprofloxacine and metronidazole did not only reduce the bacterial load in the

small intestine, but also decreased intestinal mucus accumulation and inflammation. More

importantly, three weeks of treatment substantially improved the body weight of these mice.

The same effect on the bacterial load, intestinal mucus accumulation, inflammation and body

weight was observed after laxative treatment (57). It remains to be elucidated whether the

growth benefit was due to improved fat absorption, reduced competition for nutrients by

intestinal bacteria or due to reduced intestinal inflammation. As O’Brien et al. showed that a

seven-day treatment with metronidazole reduced faecal fat excretion in four CF patients (32),

it is reasonable to assume that the increased weight is due to improved fat absorption

Although the effect on fat absorption was not assessed in either of these studies, the results

suggest that metronidazole might be a potential therapeutic option for increasing fat

absorption or nutritional status in CF patients.

One study evaluated the effect of probiotics on intestinal inflammation in CF patients (49). It

has been suggested that probiotics improve intestinal barrier function and modify the

immune response (58). Bruzzese et al. reported that a four week treatment with Lactobacillus

rhamnosus GG, reduced faecal calprotectin levels (marker for intestinal inflammation) in 8

out of 10 treated CF patients (49). The long term effect of probiotics on intestinal

inflammation was not evaluated, nor the effect on fat absorption or nutritional status. We

found only one mice study which evaluated fat absorption after probiotics; lactobacillus

supplementation increased intestinal absorption of dietary lipids in germ-free mice colonized

with human baby flora (59). Until now, the clinical value of probiotics and its relation with fat

absorption in CF patients is not clear.

3.4 SMALL INTESTINAL TRANSIT TIME

1) CF conditions. No studies exclusively investigated small intestinal transit in CF patients, but

several groups have evaluated the oro-cecal transit time (OCTT) (23). The lactulose/hydrogen

breath test was most commonly applied to for determination of the OCTT. Eighty five percent

of CF patients had a prolonged OCTT (at least +50%) in the fasted state (60,61). Another

method to evaluate intestinal activity is by measuring the intestinal muscle activity during the

inter-digestive state, also known as the migrating motor complex. The migrating motor

complex indirectly reflects the ability to transit food through the intestine. Hallberg et al.

showed that duodenal motility during fasting was normal in 8 out of 10 CF patients (22). This

contradiction with transit studies underlines the necessity to evaluate small intestinal transit

in CF patients. Especially, since intestinal smooth muscle activity shows an erratic pattern and

Persistent fat malabsorption in cystic fibrosis; lessons from patients and mice

59

2

is unresponsive to cholinergic stimulation in CFTRtm1Unc

knockout mice (62). It has been

repeatedly shown, that these mice have a slower small intestinal transit time in comparison

to wild type littermates on the same diet (57-64).

2) Relation with CFPFM. Fat absorption partly depends on the time fat is in contact with the

absorptive epithelium of the intestine. A classical thought is that during fat malabsorption the

intestine prolongs the intestinal transit time (as a compensatory mechanism), in order to

enhance its ability to absorb fat; a phenomenon also known as the ‘ileal brake’ (65). It is

reasonable to assume that this compensatory mechanism also occurs in CFPFM as intestinal

transit time is prolonged in CF. However, prolonged intestinal transit time is also a risk factor

for the occurrence of SIBO in CF conditions (8). Therefore, one might ask, if prolonged

intestinal transit time might actually induce or worsen CFPFM. The relation between intestinal

transit time and CFPFM is not yet evaluated.

3) Intervention studies. The research group of De Lisle et al. performed several (intervention)

studies on gastro-intestinal transit in CFTRtm1Unc

knockout mice (57,62,64). Oral laxative

treatment improved circular smooth muscle function in the small intestine and normalized

intestinal transit time (62). Moreover, as earlier described, laxative treatment reduced

intestinal mucus accumulation, eradicated overgrowth of bacteria in the small intestine and

improved body weight (57). Unfortunately, fat absorption was not evaluated in these studies.

It is tempting to speculate that laxative treatment may also improve the nutritional status in

CF patients, but before recommendations could be made, more clinical data are needed.

3.5 ESSENTIAL FATTY ACID DEFICIENCY

1) CF conditions. Despite high fat diets and PERT, CF patients can still have a deficiency in

essential fatty acids (EFA) (14). In general, many CF patients have subnormal levels of serum

and tissue linoleic acid (LA) and cervonic acid (DHA), often with increased levels of arachidonic

acid (AA) (14,65,66). The reduction in EFA levels is, apart from fat malabsorption, due to many

other CF-related factors. The exact mechanism behind the alterations in EFA’s still requires

further exploration, but increased lipid turnover in cellular membranes and increased

oxidation of EFA’s are considered to play the most prominent role (14,67-70). The prevalence

of EFA deficiency in CF patients strongly depends on which biochemical marker is applied as

diagnostic criterion. Magbool et al. recently demonstrated that the LA status in serum

phospholipids is more relevant than the triene/tetraene ratio as clinical indicator for EFA

deficiency in CF patients (71). When serum LA values are applied as a diagnostic criterion for

EFA-deficiency, around forty percent of pediatric CF patients are classified as EFA deficient .

The literature is not consistent on the occurrence of EFA deficiency in CF mice. Werner et al.

reported that the LA and DHA content in serum and jejunum was not altered in CFTRtm1Cam

knock-out mice and homozygous delta F508 mice (72), while Freedman et al. reported

decreased levels of DHA in the ileum of CFTRtm1Unc

knock-out mice (73).

Chapter 3

60

2

2) Relation with CFPFM. In relation to fat malabsorption, the EFA content in the intestine is

very important. EFA’s provide fluidity and flexibility to a cell membrane and play an important

role in the maintenance of cell membrane functionality of the enterocyte. EFA deficiency in

itself can cause fat malabsorption by affecting the absorption capacity of the intestine (74).

Apart from reduced LA and DHA levels, increased levels of AA in the CF intestine could also

(indirectly) affect fat absorption, by contributing to intestinal inflammation, mucus secretion

and intestinal smooth muscle relaxation (75) . It has been shown that EFA deficiency affects

both the intra-luminal and intra-cellular process of fat absorption, as exemplified in (non-CF)

animal models (69). While the degree of EFA deficiency in CF patients is generally mild, we

cannot exclude the possibility that it might interact with fat absorption. Especially since serum

EFA levels do not necessarily correlate with EFA levels in the intestine (76,77). Ex vivo

experiments on intestinal biopsies from CF patients indicated that abnormal intracellular lipid

processing occurs in the enterocyte (76). A low incorporation of [14

C]palmitic acid in

triglycerides was found in cultures of duodenal biopsies after 18 hours of incubation. The

secretion of triglyceride-rich chylomicrons was reduced in comparison to intestinal biopsies

from healthy controls (76). Whether this impaired lipid transport is due to EFA deficiency, to

aberrant CFTR itself or to a complex relation between the two, requires further investigation

(78,79). To what extent reduced intestinal EFA levels in CF patients interfere with the actual

percent absorption of dietary fat intake is unknown. An association has been described

between the degree of EFA deficiency and the nutritional status of CF patients (80), but EFA

deficiency can also be present in well-nourished CF patients (81). Thus, the clinical relevance

of the reduced EFA levels in CF patients on fat malabsorption remains to be determined.

3) Intervention studies. Despite the abundance of (intervention) studies (66,82-85), no study

determined the effect of EFA supplementation on CFPFM. Improved DHA levels in duodenal

mucosa of CF patients after 70 mg of DHA/kg body weight for six weeks has been reported,

but whether this improved fat absorption or nutritional status was not described. Reviews on

omega 3 fatty acid supplementation or DHA supplementation alone, conclude that there is

still insufficient evidence for the therapeutic effect of omega 3 fatty acids to recommend the

routine use in CF patients (82,85). Other studies showed that energy supplements rich in LA

improved the body weight in CF patients, but it is hard to differentiate between the effects of

LA from that of a high energy intake (83,84). Accordingly, we do not recommend the routine

use of EFA supplements in CF for reasons to enhance fat absorption, but it could be

considered to improve nutritional status in individual cases.

Persistent fat malabsorption in cystic fibrosis; lessons from patients and mice

61

2

CF patients CF mice

(primarily Cftr-/- mice)

Treatment effect CF patients

(fat absorption+weight)

Treatment effect CF mice

(fat absorption+weight)

Intestinal pH Duodenum: ↓ Duodenum: minor ↓ Bicarbonate enteric coated pancreatic enzymes: fat absorption: ↔ or minor ↑ weight: not known

Acid suppressant drugs:triglyceride uptake: ↑ fatty acid uptake: ↑

weight: not known

Jejunum: ↓ or ↔ Jejunum: not known

Ileum: ↔ Ileum: not known Acid suppressant drugs: fat absorption: ↔ or ↑ weight: ↔

Intraluminal bile salts

Fecal bile salt excretion: ↑ Fecal bile salt excretion: ↑ Taurine supplementation: fat absorption: ↔ or minor ↑ weight: ↔ or minor ↑

No interventional studies

Glycine/taurine ratio: ↑ Altered bile salt composition

Altered bile salt composition

Intestinal wall abnormalities

SIBO (~ 40%) SIBO No interventional studies Antibiotic treatment: fat absorption: not known weight: ↑ Intestinal mucus accumulation Intestinal mucus accumulation

Intestinal inflammation Intestinal inflammation

Intestinal permeability Intestinal permeability

Small intestinal transit time

Not known Prolonged No interventional studies Laxative treatment: fat absorption: not known weight: ↑

Essential fatty acid deficiency

Serum/tissue LA and DHA: ↓ or ↔

Serum/tissue LA and DHA: ↓ or ↔

Energy supplements with LA: fat absorption: not known weight: ↑

No interventional studies

Serum/tissue AA: ↑ or ↔ Serum/tissue AA: ↑ or ↔

Table 1. Factors that are suggested to contribute to fat malabsorption in cystic fibrosis (CF). Table presents an

overview of the described alterations observed in CF patients and CF mice and, interventional studies that evaluated

fat absorption and/or nutritional status. ↔ : not affected, ↑: increased, ↓: decreased . The described data on CF

mice are mainly based on studies in CFTR knockout mice. References are described in the text. SIBO: small intestinal

bacterial overgrowth. LA: linoleic acid. DHA: cervonic acid. AA: arachidonic acid

SUMMARY, CLINICAL IMPLICATIONS AND FUTURE PERSPECTIVES

Several factors have been proposed to contribute to CFPFM (table 1). Based on the available

data we consider it not likely that bile salt alterations do contribute to CFPFM. The

quantitative contribution of intestinal mucosal abnormalities, small intestinal transit time and

EFA deficiency to CFPFM remains to be determined. Based on the findings in CF mouse

models, CF patients with signs of small bacterial overgrowth might benefit from treatment

with broad-spectrum antibiotics. Antibiotic treatment should preferably be evaluated with

changes in nutritional status or fat absorption, using three day fat balances. However, we

should realize that three day fat balances show high intra-individual variability, which

exemplifies the need for novel methodologies to assess fat absorption in CF patients. So far,

the influence of reduced duodenal pH levels seems most convincingly demonstrated to play a

role in CFPFM. Although not necessarily beneficial in all CF patients, a trial of acid suppressive

drugs might be helpful to optimize fat absorption in individual cases.

Chapter 3

62

2

REFERENCE LIST

(1) Corey M, McLaughlin F, Williams M, Levison H. A comparison of survival, growth, and

pulmonary function in patients with cystic fibrosis in Boston and Toronto. J Clin Epidemiol

1988;41(6):583-591.

(2) Beker LT, Russek-Cohen E, Fink RJ. Stature as a prognostic factor in cystic fibrosis survival. J

Am Diet Assoc 2001;101(4):438-442.

(3) Pencharz P, Durie P. Pathogenesis of malnutrition in cystic fibrosis, and its treatment.

Clinical Nutrition 2000;19(6):387-394.

(4) Littlewood JM, Wolfe SP, Conway SP. Diagnosis and treatment of intestinal malabsorption

in cystic fibrosis. Pediatr Pulmonol 2006;41(1):35-49.

(5) FitzSimmons SC, Burkhart GA, Borowitz D, Grand RJ, Hammerstrom T, Durie PR, et al. High-

dose pancreatic-enzyme supplements and fibrosing colonopathy in children with cystic

fibrosis. N Engl J Med 1997;336(18):1283-1289.

(6) Taylor C, Hillel P, Ghosal S, Frier M, Senior S, Tindale W, et al. Gastric emptying and

intestinal transit of pancreatic enzyme supplements in cystic fibrosis. Arch Dis Child

1999;80(2):149-152.

(7) Kalivianakis M, Minich DM, Bijleveld CMA, van Aalderen WMC, Stellaard F, Laseur M, et al.

Fat malabsorption in cystic fibrosis patients receiving enzyme replacement therapy is due to

impaired intestinal uptake of long-chain fatty acids. Am J Clin Nutr 1999 01;69(1):127-134.

(8) Borowitz D, Durie PR, Clarke LL, Werlin SL, Taylor CJ, Semler J, et al. Gastrointestinal

outcomes and confounders in cystic fibrosis. J Pediatr Gastroenterol Nutr 2005;41(3):273-285.

(9) Bijvelds MJC, Bronsveld I, Havinga R, Sinaasappel M, de Jonge HR, Verkade HJ. Fat

absorption in cystic fibrosis mice is impeded by defective lipolysis and post-lipolytic events. Am

J Physiol Gastrointest Liver Physiol 2005 04;288(4):G646-G653.

(10) Verkade HJ, Tso P. Biophysics of intestinal luminal lipids. Intestinal Lipid Metabolism:

Springer; 2001. p. 1-18.

(11) Zoppi G, Andreotti G, Pajno-Ferrara F, Njai D, Gaburro D. Exocrine pancreas function in

premature and full term neonates. Pediatr Res 1972;6(12):880-886.

(12) Abrams CK, Hamosh M, Hubbard VS, Dutta SK, Hamosh P. Lingual lipase in cystic fibrosis.

Quantitation of enzyme activity in the upper small intestine of patients with exocrine

pancreatic insufficiency. J Clin Invest 1984;73(2):374-382.

(13) Minich DM, Voshol PJ, Havinga R, Stellaard F, Kuipers F, Vonk RJ, et al. Biliary

phospholipid secretion is not required for intestinal absorption and plasma status of linoleic

Persistent fat malabsorption in cystic fibrosis; lessons from patients and mice

63

2

acid in mice. Biochimica et Biophysica Acta (BBA)-Molecular and Cell Biology of Lipids

1999;1441(1):14-22.

(14) Peretti N, Marcil V, Drouin E, Levy E. Mechanisms of lipid malabsorption in Cystic Fibrosis:

the impact of essential fatty acids deficiency. Nutr Metab (Lond) 2005;2(1):11-19.

(15) Vassileva G, Huwyler L, Poirier K, Agellon LB, Toth MJ. The intestinal fatty acid binding

protein is not essential for dietary fat absorption in mice. The FASEB Journal

2000;14(13):2040-2046.

(16) Levy E, Ménard D, Delvin E, Montoudis A, Beaulieu J, Mailhot G, et al. Localization,

function and regulation of the two intestinal fatty acid-binding protein types. Histochem Cell

Biol 2009;132(3):351-367.

(17) Goudriaan JR, Dahlmans VE, Febbraio M, Teusink B, Romijn JA, Havekes LM, et al.

Intestinal lipid absorption is not affected in CD36 deficient mice. Cellular Lipid Binding

Proteins: Springer; 2002. p. 199-202.

(18) Lightfoot FG, Lauretta Grau MS S, Satchitanandum M. Lipid accumulation in jejunal and

colonic mucosa following chronic cholestyramine (Questran) feeding. Dig Dis Sci

1985;30(5):468-476.

(19) Werner A, Havinga R, Perton F, Kuipers F, Verkade HJ. Lymphatic chylomicron size is

inversely related to biliary phospholipid secretion in mice. American Journal of Physiology-

Gastrointestinal and Liver Physiology 2006;290(6):G1177-G1185.

(20) Tang L, Fatehi M, Linsdell P. Mechanism of direct bicarbonate transport by the CFTR anion

channel. Journal of Cystic Fibrosis 2009;8(2):115-121.

(21) Kaur S, Norkina O, Ziemer D, Samuelson LC, De Lisle RC. Acidic duodenal pH alters gene

expression in the cystic fibrosis mouse pancreas. American Journal of Physiology-

Gastrointestinal and Liver Physiology 2004;287(2):G480-G490.

(22) Hallberg K, Abrahamsson H, Dalenbäck J, Fändriks L, Strandvik B. Gastric secretion in

cystic fibrosis in relation to the migrating motor complex. Scand J Gastroenterol

2001;36(2):121-127.

(23) Gregory P. Gastrointestinal pH, motility/transit and permeability in cystic fibrosis. J

Pediatr Gastroenterol Nutr 1996;23(5):513-523.

(24) Garcia MAS, Yang N, Quinton PM. Normal mouse intestinal mucus release requires cystic

fibrosis transmembrane regulator–dependent bicarbonate secretion. J Clin Invest

2009;119(9):2613-2622.

(25) Quinton PM. Birth of mucus. American Journal of Physiology-Lung Cellular and Molecular

Physiology 2010;298(1):L13-L14.

Chapter 3

64

2

(26) PJ Robinson MB B. Duodenal pH in cystic fibrosis and its relationship to fat malabsorption.

Dig Dis Sci 1990;35(10):1299-1304.

(27) Brady MS, Garson JL, Krug SK, Kaul A, Rickard KA, Caffrey HH, et al. An enteric-coated

high-buffered pancrelipase reduces steatorrhea in patients with cystic fibrosis: a prospective,

randomized study. J Am Diet Assoc 2006;106(8):1181-1186.

(28) Kalnins D, Ellis L, Corey M, Pencharz PB, Stewart C, Tullis E, et al. Enteric-coated

pancreatic enzyme with bicarbonate is equal to standard enteric-coated enzyme in treating

malabsorption in cystic fibrosis. J Pediatr Gastroenterol Nutr 2006;42(3):256-261.

(29) Jones DE, Palmer JM, Kirby JA, De Cruz DJ, McCaughan GW, Sedgwick JD, et al.

Experimental autoimmune cholangitis: a mouse model of immune-mediated cholangiopathy.

Liver 0226 10;20(5):351-356.

(30) Shindo K, Machida M, Fukumura M, Koide K, Yamazaki R. Omeprazole induces altered bile

acid metabolism. Gut 1998;42(2):266-271.

(31) Weber AM. Relationship between bile acid malabsorption and pancreatic insufficiency in

cystic fibrosis. Gut 1976;17(4):295.

(32) O'Brien S, Mulcahy H, Fenlon H, O'Broin A, Casey M, Burke A, et al. Intestinal bile acid

malabsorption in cystic fibrosis. Gut 1993;34(8):1137.

(33) Fondacaro JD, Heubi JE, Kellogg FW. Intestinal bile acid malabsorption in cystic fibrosis: a

primary mucosal cell defect. Pediatr Res 1982;16(6):494-498.

(34) Roy CC, Weber AM, Morin CL, Combes JC, Nusslé D, Mégevand A, et al. Abnormal biliary

lipid composition in cystic fibrosis. N Engl J Med 1977;297(24):1301-1305.

(35) Strandvik B, Einarsson K, Lindblad A, Angelin B. Bile acid kinetics and biliary lipid

composition in cystic fibrosis. J Hepatol 1996 07;25(1):43-48.

(36) Regan PT, Malagelada J-, Dimagno EP, Go VLW. Reduced intraluminal bile acid

concentrations and fat maldigestion in pancreatic insufficiency: Correction by treatment.

Gastroenterology 1979 08/01;77(2):285-289.

(37) Krag E, Phillips SF. Active and passive bile acid absorption in man. Perfusion studies of the

ileum and jejunum. J Clin Invest 1974;53(6):1686.

(38) Hepner GW, Sturman JA, Hofmann AF, Thomas PJ. Metabolism of Steroid and Amino Acid

Moieties of Conjugated Bile Acids in Man III. CHOLYLTAURINE (TAUROCHOLIC ACID). J Clin

Invest 1973;52(2):433-440.

(39) Belli DC, Levy E, Darling P, Leroy C, Lepage G, Giguère R, et al. Taurine improves the

absorption of a fat meal in patients with cystic fibrosis. Pediatrics 1987;80(4):517-523.

Persistent fat malabsorption in cystic fibrosis; lessons from patients and mice

65

2

(40) Darling PB, Lepage G, Leroy C, Masson P, Roy CC. Effect of taurine supplements on fat

absorption in cystic fibrosis. Pediatr Res 1985;19(6):578-582.

(41) Smith LJ, Lacaille F, Lepage G, Ronco N, Lamarre A, Roy CC. Taurine Decreases Fecal Fatty

Acid and Sterol Excretion in Cystic Fibrosis: A Randomized Double-blind Trial. Arch Pediatr

Adolesc Med 1991;145(12):1401-1404.

(42) Colombo C, Arlati S, Curcio L, Maiavacca R, Garatti M, Ronchi M, et al. Effect of taurine

supplementation on fat and bile acid absorption in patients with cystic fibrosis. Scand J

Gastroenterol 1988;23(S143):151-156.

(43) Merli M, Bertasi S, Servi R, Diamanti S, Martino F, De Santis A, et al. Effect of a medium

dose of ursodeoxycholic acid with or without taurine supplementation on the nutritional status

of patients with cystic fibrosis: a randomized, placebo-controlled, crossover trial. J Pediatr

Gastroenterol Nutr 1994;19(2):198-203.

(44) De Curtis M, Santamaria F, Ercolini P, Vittoria L, De Ritis G, Garofalo V, et al. Effect of

taurine supplementation on fat and energy absorption in cystic fibrosis. Arch Dis Child

1992;67(9):1082-1085.

(45) Thompson GN, Robb TA, Davidson GP. Taurine supplementation, fat absorption, and

growth in cystic fibrosis. J Pediatr 1987;111(4):501-506.

(46) Van Elburg RM, Uil JJ, Van Aalderen WMC, Mulder CJJ, Heymans HSA. Intestinal

permeability in exocrine pancreatic insufficiency due to cystic fibrosis or chronic pancreatitis.

Pediatr Res 1996;39(6):985-991.

(47) Lewindon P, Robb T, Moore DJ, Davidson GP, Martin AJ. Bowel dysfunction in cystic

fibrosis: importance of breath testing. J Paediatr Child Health 1998;34(1):79-82.

(48) Lisowska A, Wójtowicz J, Walkowiak J. Small intestine bacterial overgrowth is frequent in

cystic fibrosis: combined hydrogen and methane measurements are required for its detection.

Acta Biochim Pol 2009;56(4):631.

(49) Bruzzese E, Raia V, Gaudiello G, Polito G, Buccigrossi V, Formicola V, et al. Intestinal

inflammation is a frequent feature of cystic fibrosis and is reduced by probiotic administration.

Aliment Pharmacol Ther 2004;20(7):813-819.

(50) Raia V, Maiuri L, de Ritis G, de Vizia B, Vacca L, Conte R, et al. Evidence of chronic

inflammation in morphologically normal small intestine of cystic fibrosis patients. Pediatr Res

2000;47(3):344-350.

(51) Sbarbati A, Bertini M, Catassi C, Gagliardini R, Osculati F. Ultrastructural lesions in the

small bowel of patients with cystic fibrosis. Pediatr Res 1998;43(2):234-239.

(52) Smyth RL, Croft NM, O'Hea U, Marshall TG, Ferguson A. Intestinal inflammation in cystic

fibrosis. Arch Dis Child 2000;82(5):394-399.

Chapter 3

66

2

(53) Quigley EM, Quera R. Small intestinal bacterial overgrowth: roles of antibiotics,

prebiotics, and probiotics. Gastroenterology 2006;130(2):S78-S90.

(54) Martin FJ, Dumas M, Wang Y, Legido-Quigley C, Yap IK, Tang H, et al. A top-down systems

biology view of microbiome-mammalian metabolic interactions in a mouse model. Molecular

systems biology 2007;3(1):1-16.

(55) De Lisle RC, Roach EA, Norkina O. Eradication of small intestinal bacterial overgrowth in

the cystic fibrosis mouse reduces mucus accumulation. J Pediatr Gastroenterol Nutr

2006;42(1):46-52.

(56) Norkina O, Burnett TG, De Lisle RC. Bacterial overgrowth in the cystic fibrosis

transmembrane conductance regulator null mouse small intestine. Infect Immun

2004;72(10):6040.

(57) De Lisle RC, Roach E, Jansson K. Effects of laxative and N-acetylcysteine on mucus

accumulation, bacterial load, transit, and inflammation in the cystic fibrosis mouse small

intestine. American Journal of Physiology-Gastrointestinal and Liver Physiology

2007;293(3):G577-G584.

(58) Perdigón G, Fuller R, Raya R. Lactic acid bacteria and their effect on the immune system.

Curr Issues Intestinal Microbiol 2001;2(1):27-42.

(59) Martin FJ, Wang Y, Sprenger N, Yap IK, Lundstedt T, Lek P, et al. Probiotic modulation of

symbiotic gut microbial–host metabolic interactions in a humanized microbiome mouse

model. Molecular systems biology 2008;4(1):1-14.

(60) Dalzell A, Freestone N, Billington D, Heaf D. Small intestinal permeability and orocaecal

transit time in cystic fibrosis. Arch Dis Child 1990;65(6):585-588.

(61) Bali A, Stableforth DE, Asquith P. Prolonged small-intestinal transit time in cystic fibrosis.

Br Med J (Clin Res Ed) 1983;287(6398):1011-1013.

(62) De Lisle R, Sewell R, Meldi L. Enteric circular muscle dysfunction in the cystic fibrosis

mouse small intestine. Neurogastroenterology & Motility 2010;22(3):341-e87.

(63) Lin HC. Ileal brake: neuropeptidergic control of intestinal transit. Curr Gastroenterol Rep

2006;8(5):367-373.

(64) De Lisle RC. Altered transit and bacterial overgrowth in the cystic fibrosis mouse small

intestine. Am J Physiol Gastrointest Liver Physiol 2007;293(1):G104-G111.

(65) Freedman SD, Blanco PG, Zaman MM, Shea JC, Ollero M, Hopper IK, et al. Association of

cystic fibrosis with abnormalities in fatty acid metabolism. N Engl J Med 2004;350(6):560-569.

(66) Jumpsen JA, Brown NE, Thomson AB, Paul Man S, Goh YK, Ma D, et al. Fatty acids in blood

and intestine following docosahexaenoic acid supplementation in adults with cystic fibrosis.

Journal of Cystic Fibrosis 2006;5(2):77-84.

Persistent fat malabsorption in cystic fibrosis; lessons from patients and mice

67

2

(67) Andersson C, Al-Turkmani MR, Savaille JE, Alturkmani R, Katrangi W, Cluette-Brown JE, et

al. Cell culture models demonstrate that CFTR dysfunction leads to defective fatty acid

composition and metabolism. J Lipid Res 2008;49(8):1692-1700.

(68) Strandvik B. Fatty acid metabolism in cystic fibrosis. Prostaglandins, Leukotrienes and

Essential Fatty Acids 2010;83(3):121-129.

(69) Levy E, Garofalo C, Thibault L, Dionne S, Daoust L, Lepage G, et al. Intraluminal and

intracellular phases of fat absorption are impaired in essential fatty acid deficiency. American

Journal of Physiology-Gastrointestinal and Liver Physiology 1992;262(2):G319-G326.

(70) Rogiers V, Dab I, Michotte Y, Vercruysse A, Crokaert R, Vis HL. Abnormal Fatty Acid

Turnover in the Phospholipids of the Red Blood Cell Membranes of Cystic Fibrosis Patients

( in Vitro Study&rpar. Pediatr Res 1984;18(8):704-709.

(71) Maqbool A, Schall JI, Garcia-Espana JF, Zemel BS, Strandvik B, Stallings VA. Serum linoleic

acid status as a clinical indicator of essential fatty acid status in children with cystic fibrosis. J

Pediatr Gastroenterol Nutr 2008;47(5):635-644.

(72) Werner A, Bongers ME, Bijvelds MJ, de Jonge HR, Verkade HJ. No indications for altered

essential fatty acid metabolism in two murine models for cystic fibrosis. J Lipid Res

2004;45(12):2277-2286.

(73) Freedman SD, Katz MH, Parker EM, Laposata M, Urman MY, Alvarez JG. A membrane lipid

imbalance plays a role in the phenotypic expression of cystic fibrosis in cftr−/− mice.

Proceedings of the National Academy of Sciences 1999;96(24):13995-14000.

(74) Snipes RL. The effects of essential fatty acid deficiency on the ultrastructure and

functional capacity of the jejunal epithelium. Lab Invest 1968 Feb;18(2):179-189.

(75) De Lisle RC, Meldi L, Flynn M, Jansson K. Altered eicosanoid metabolism in the cystic

fibrosis mouse small intestine. J Pediatr Gastroenterol Nutr 2008;47(4):406-416.

(76) Peretti N, Roy CC, Drouin E, Seidman E, Brochu P, Casimir G, et al. Abnormal intracellular

lipid processing contributes to fat malabsorption in cystic fibrosis patients. American Journal

of Physiology-Gastrointestinal and Liver Physiology 2006;290(4):G609-G615.

(77) Korotkova M, Strandvik B. Essential fatty acid deficiency affects the fatty acid composition

of the rat small intestinal and colonic mucosa differently. Biochimica et Biophysica Acta (BBA)-

Molecular and Cell Biology of Lipids 2000;1487(2):319-325.

(78) Mailhot G, Ravid Z, Barchi S, Moreau A, Rabasa-Lhoret R, Levy E. CFTR knockdown

stimulates lipid synthesis and transport in intestinal Caco-2/15 cells. American Journal of

Physiology-Gastrointestinal and Liver Physiology 2009;297(6):G1239-G1249.

Chapter 3

68

2

(79) Mailhot G, Rabasa-Lhoret R, Moreau A, Berthiaume Y, Levy E. CFTR depletion results in

changes in fatty acid composition and promotes lipogenesis in intestinal Caco 2/15 cells. PloS

one 2010;5(5):e10446.

(80) Shoff SM, Ahn H, Davis L, Lai H. Temporal associations among energy intake, plasma

linoleic acid, and growth improvement in response to treatment initiation after diagnosis of

cystic fibrosis. Pediatrics 2006;117(2):391-400.

(81) Roulet M, Frascarolo P, Rappaz I, Pilet M. Essential fatty acid deficiency in well nourished

young cystic fibrosis patients. Eur J Pediatr 1997;156(12):952-956.

(82) Van Biervliet S, Van Biervliet J, Robberecht E, Christophe A. Docosahexaenoic acid trials in

cystic fibrosis: a review of the rationale behind the clinical trials. Journal of Cystic Fibrosis

2005;4(1):27-34.

(83) Steinkamp G, Demmelmair H, Rühl-Bagheri I, von der Hardt H, Koletzko B. Energy

supplements rich in linoleic acid improve body weight and essential fatty acid status of cystic

fibrosis patients. J Pediatr Gastroenterol Nutr 2000;31(4):418-423.

(84) Van Egmond A, Kosorok MR, Koscik R, Laxova A, Farrell PM. Effect of linoleic acid intake

on growth of infants with cystic fibrosis. Am J Clin Nutr 1996;63(5):746-752.

(85) Coste TC, Armand M, Lebacq J, Lebecque P, Wallemacq P, Leal T. An overview of

monitoring and supplementation of omega 3 fatty acids in cystic fibrosis. Clin Biochem

2007;40(8):511-520.

CHAPTER 3

INCREASE OF SERUM GAMMA

GLUTAMYLTRANSFERASE (GGT)

ASSOCIATED WITH THE

DEVELOPMENT OF CIRRHOTIC

CYSTIC FIBROSIS LIVER DISEASE PREDICTION OF CIRRHOTIC CYSTIC FIBROSIS LIVER DISEASE.

Submitted

Frank A.J.A. Bodewes1

Hubert. P.J. van der Doef2

Roderick H.J. Houwen2

Henkjan J Verkade1

1Department of Pediactric Gastroentrology and Hepatology, Beatrix Children’s Hospital, University

Medical Center Groningen, University of Groningen, the Netherlands

2Department of Pediactric Gastroentrology and Hepatology University Medical Center Utrecht, the

Netherlands

Chapter 3

70

3

ABSTRACT

Background:

Cirrhotic liver disease (CCFLD) develops in 5-10% of CF patients. Identification of patients at

risk on CCFLD is potentially beneficial for preventive treatment. We studied the evolution of

liver enzymes (ALT, AST and GGT) in years preceding the diagnosis of CCFLD.

Methods:

We analyzed medical records of 277 pediatric CF patients. The time point of CCFLD diagnosis

was defined as the date when the ultrasonographic macronodular liver and the clinical

splenomegaly were first recorded. We analyzed liver enzymes in the 2 years preceding the

diagnosis of CCFLD. We compared these results with the annual liver enzymes of no-CCFLD

controls (>15 years of age and no ultrasonographic or physical signs of CCFLD).

Results:

At group level the median GGT, and not AST or ALT, of CCFLD patients, in the 2 years

preceding their diagnosis, significantly higher than the median of all GGT results of no-CCFLD

controls (45 vs. 17 U/l, respectively, P<0.001). The value of a single GGT result, for predicting

future CCFLD, was low. However, for CF patients with a mean GGT>35 U/L, based on repeated

measurements, the Odds ratio to develop CCFLD was 39 (95% CI: [9-175], sensitivity: 95%,

specificity: 64%, positive predictive value: 50%).

Conclusion:

In pediatric CF patients a persistent, high-normal, serum GGT is strongly associated with the

diagnosis of CCFLD within 2 years. The prognostic value of a single GGT measurement remains

limited. Our results indicate that groups of patients at increased risk for CCFLD can be

identified on the basis of repeated GGT measurements.

Increase of serum GGT associated with the development of cirrhotic cystic fibrosis liver disease

71

3

INTRODUCTION

Cirrhotic cystic fibrosis liver disease (CCFLD) develops in 5-10% of cystic fibrosis patients (1). It

is a serious complication of CF and patients with CCFLD tend to have a more severe CF

phenotype than CF patients without liver disease (2, 3).CCFLD is characterized by extensive

and often inhomogeneous cirrhosis. Clinically, CCFLD patients frequently have splenomegaly,

hypersplenism and complications of portal hypertension, including variceal bleeding and

ascites (4, 5). Liver synthesis and detoxification functions are often spared, and the need for

primary liver transplantation for CF liver disease has remained rather rare (6, 7).

CCFLD is an acquired complication of CF and is not yet present in infancy. Most CCFLD

patients develop the disease during childhood and have established disease before puberty

(8). The diagnosis of CCFLD is mostly based on clinical, imaging and biochemical (liver

enzymes) parameters. CCFLD patients develop splenomegaly and thrombocytopenia (9). The

specific findings of hepatic nodularity and splenomegaly on ultrasound (US) are reported as

reliable markers for advanced fibrosis with only limited discrepancy with liver biopsy (10).

Histology of liver biopsies specimens typically shows a severe bridging type of portal fibrosis

and proliferation and destruction of bile ducts (11).

The recognition CCFLD in an early phase and/or the identification of patients at risks may

have relevant clinical and therapeutic consequences (12). To date the only treatment option

used for CCFLD in clinical practice is ursodeoxycholic acid (UDCA)(13). It has been

hypothesized that UDCA improves or recovers the compromised or obstructed bile flow in CF

conditions. Theoretically, it would seem favorable to start treatment of CCFLD either in

patients at risk (prevention) or an early phase of the disease. Not only UDCA, but also other

bile salt analogues and anti-fibrogenic agents are currently under development and evaluated

(14). Identification of patients at risk for CCFLD or in an early, preclinical stage of the disease

would be of value to test the preventive and/or disease course modifying capacity of these

agents.

Diagnosis of CCFLD is presently only possible in the cirrhotic phase of the disease. To date no

reliable diagnostic tool or measurement are established to predict patients at risk for CCFLD

or to recognize the early phase of the disease. Different clinical signs are suggested to be

related to the development of CCFLD. However, no strong relations are established between,

for example, hepatic US findings and the risk for the development of CCFLD (15, 16).

Liver enzymes such as the transaminases AST (aspartate transaminase) and ALT (alanine

transaminases) and gamma-glutamyl transpeptidase (GGT) are frequently evaluated during

routine clinical checkups of CF patients (17). Elevated AST, ALT, and GGT are sometimes

regarded as indicators for the presence or development of CFLD. Colombo et al. reported that

UDCA treatment in CF patients frequently corrects the elevation of transaminases (18). It

Chapter 3

72

3

needs to be realized, however, that elevation of liver enzymes occurs frequently and

transiently in CF (19). These elevations of transaminases could be induced by hepatotoxic

therapies like antibiotics or be a para-infectious phenomenon associated with pulmonary

exacerbation.

The relation between liver enzymes and the presence of CCFLD was first studied by Potter et

al.(17). These scientists related liver biopsy findings of 43 CF patients to their (ALT, AST and

GGT levels at the time of biopsy. The biopsies were performed based on clinical indications

for liver disease including hepatomegaly, abnormal liver function tests, splenomegaly or

esophageal varices. Potter et al. found liver fibrosis (grade≤3) in 37% of their biopsies and that

GGT, at the time of the biopsy, did not correlate with the presence of fibrosis. Unfortunately,

no historical biochemical results were available to include in the analysis. Therefore, the

results of this study could not address the role of liver enzymes during the development of

CCFLD. Lindblad et al. reported on a historical cohort of liver biopsy results in 41 CF patients

(4). Nine out of these 41 patients had histology proven cirrhosis, of which 5 indeed had

clinical signs of cirrhosis. In this cohort the sensitivity of liver enzymes was 100% and the

specificity 41% for the presence of moderate or severe fibrosis and cirrhosis.

In this retrospective, controlled, study we aimed to identify potential biochemical risk factors

for the future development of CCFLD. Therefore we focused evolution of liver enzymes (ALT,

AST and GGT) in years preceding the diagnosis of CCFLD in patients with already established

cirrhosis.

METHODS

Study cohort

We performed a retrospective analysis in a cohort that contained all pediatric CF patients (2 –

18 years) from the Cystic Fibrosis centers of the University Medical Center Utrecht and the

Beatrix Children’s Hospital, University Medical Center Groningen, The Netherlands (reference

date January 1st 2007). According to the clinical protocols patients were seen and

evaluated at least yearly in our CF centers. The annual medical checkup included blood testing

for ALT, AST and GGT and ultrasonography of liver and spleen.

Study method

We defined the existence of cirrhosis as the ultrasonographic appearance of multilobular

macronodularity of the liver and the presence of splenomegaly (10) For this purpose we

reviewed the annual radiology reports for the ultrasonographic description of

macronodularity. Additionally we reviewed medical records for reported an enlarged spleen

Increase of serum GGT associated with the development of cirrhotic cystic fibrosis liver disease

73

3

on physical examination or ultrasonographic enlarged spleen compared to the maximum

references spleen size according to age (20). We reviewed all biochemistry results for ALT,

AST and GGT. We defined the upper limit of normal for AST, ALT and GGT of 50 U/l (21). Since

liver enzymes could be transiently elevated due to neonatal cholestasis in CF patients,

unrelated to the development of CCFLD, we excluded liver enzymes results obtained before

the age of 2 years. Based on these results we categorized patients into 4 groups:

A) Macronodularity and splenomegaly

B) Macronodularity without splenomegaly

C) Splenomegaly without macronodularity

D) No established macronodularity or splenomegaly

CCFLD study group

We considered group A (macronodularity and splenomegaly) as patients with an established

diagnosis of CCFLD. We defined the date of CCFLD diagnosis as the first evaluation date on

which the patient met the CCFLD criteria of ultrasonographic liver macronodularity and

splenomegaly.

No-CCFLD control group

From group D (no macronodularity and no splenomegaly) we selected a no-CCFLD control

group. This group consisted of CF patients that, at the age of 15 years, had never developed

any signs of CCFLD (e.g. both normal liver ultrasounds and no reported splenomegaly). Of this

group we use all the annually collected liver enzymes results after the second year of life.

Statistics

For statistical analysis, we used IBM SPPS version 20. For comparison of the nominal

variables, we use the Chi square testing or Fischer exact, when appropriate. For comparison

of the continuous variables, we used the Mann–Whitney U test. We use receiver operating

characteristic (ROC) to determine AOC and the cutoff value. We used contingency tables to

analyze frequency distribution and risk ratios.

Chapter 3

74

3

Figure 1. Schematic representation of the composition of the studies population. Group A are patients

with cirrhotic cystic fibrosis liver disease (CCFLD) defined as ultrasonographic macronodularity and

splenomegaly, Group B are patients with macronodularity and no splenomegaly, group C are patients

with splenomegaly without macro nodular disturbances of the liver on US and group D with no signs of

CCFLD. Group D is referred to as the reference population. The study group contained all patients of

group A of whom biochemistry results were available in period 2 years prior to the date of diagnosis of

CCFLD. The control group was defined as CF patients out of the reference population that had not

developed any signs of CCFLD (e.g. normal liver ultra sound and no splenomegaly) at the age of 15 years.

Increase of serum GGT associated with the development of cirrhotic cystic fibrosis liver disease

75

3

RESULTS

Study cohort

The total study cohort consisted of 277 pediatric CF patients (Figure 1). Nineteen (7%)

patients met the criteria for established CCFLD (Group A). Additionally we identified patients

with either ultrasonographic signs of macronodularity but no splenomegaly [group B: N=10

(4%)], or presence of splenomegaly but no macronodularity on liver ultra sound [group C:

N=12 (4%)]. Two hundred thirty six (85%) patients had not displayed any signs of

ultrasonographic macronodularity or splenomegaly (group D, reference group). In the current,

retrospective, study we found a relatively high and variable age of diagnosis for CF. This is

explained by the fact that in the Netherlands a nationwide neonatal CF screening program

was introduced only in 2011 (Table 1).

We evaluated different potential risk factors or factors reported being related with the

development of cystic fibrosis related liver disease (Table 1). Almost all patients with signs of

liver involvement had started UDCA treatment. We found a significantly increased prevalence

of DIOS in the two patients groups with macro-nodular abnormalities on liver ultrasound

(group A and B). In patients with only splenomegaly (group C), on the other hand, the

prevalence of DIOS was low.

We analyzed the relationships between liver enzymes and prevalence of liver involvement in

CF (Table 1). We did this by evaluating the proportion of patients per study group, in which

any liver enzymes results had been above the upper limit of normal (AST, ALT and GGT> 50

U/l). We found no significant higher proportion of patients with increased liver transaminases

(AST or ALT) in the study groups with any signs of liver disease (group A, B and C) compared to

group D. However, in the CCFLD patient (group A) we did observe a significantly higher

proportion of patients with a GGT above the upper limit of normal of 50 U/l compared to the

patients without any signs of liver disease (group D). The latter indicative for a potential

relation between GGT elevations and the development or presence of CCFLD.

CCFLD study group

We evaluated the age of presentation of CCFLD (group A) in the study population. We

determined the date of diagnosis as the first day patients met the defined criteria of CCLFD.

The peak incidence of CF liver disease was around the age of 10 years. We found no patients

who developed CCFLD before the age of 5 or after the age of 15 years. The median age at

presentation of CCFLD in our population was 10 years. Almost all (95%) CCFLD patients from

group A were treated with UDCA.

Chapter 3

76

3

Table 1. Description of clinical symptoms and biochemistry data of total studied pediatric CF

population.

A B C D1

macronodularity

and

splenomegaly

macronodularity

and

no splenomegaly

splenomegaly

and

no macronodularity

no macronodularity

and

no splenomegaly

Total

Patients (N) 19 (7%) 10 (4%) 12 (4%) 236 (85%) 277

Median age at

evaluation date

(years)2

16* (8-17) 11 (6-15) 13* (8-17) 10 (2-17)

Median age at

diagnosis CF

(days)2

61 (4-4894) 84 (9-1009) 195 (16-3511) 126(0-3440)

Clinical symptoms N % P value2 Odds ratio (CI) N % N % N % N %

Male3 13 (68%) 0.092 7 (70%) 8 (67%) 114 (48%) 142 (51%)

UDCA use3 18 (95%) >0.001 72 (9-556) 10 (100%) 9 (75%) 47 (20%) 84 (30%)

DIOS3 4 (21%) 0.009 5 (1-16) 4* (40%) 1 (8%) 13 (6%) 22 (8%)

Meconium ileus3 5 (26%) 0.112 3 (30%) 3 (25%) 31 (13%) 42 (15%)

Pancreatic

insufficient (PERT)3 19 (100%) 0.621 10 (100%) 12 (100%) 233 (99%) 274 (99%)

Severe genotype3# 19 (100%) 0.062 8 (80%) 11 (92%) 199 (84%) 237 (86%)

DF508/DF5083 11 (58%) 0.622 6 (60%) 9 (75%) 150 (64%) 176 (64%)

Biochemistry

ASAT>50 U/l3 2 (14%) 0.285 3 (33%) 3 (30%) 33 (17%) 41 (18%)

ALAT>50U/l3 2 (14%) 0.543 0 (0%) 2 (20%) 24 (12%) 28 (12%)

GGT>50U/l3 6 (43%) <0.001 16 (5-52) 2 (22%) 2 (20%) 6 (3%) 16 (7%)

1 Reference population

2Mann–Whitney U (either group A, B or C individually compared to group D)

3Chi square test, group (either group A, B or C individually compared to group D)

*P<0.05

#Severe genotype defined as CFTR class1-3 mutations

Increase of serum GGT associated with the development of cirrhotic cystic fibrosis liver disease

77

3

Table 2. Comparisons of study groups for GGT sub analysis.

CCFLD

Study group

N=14, 33 data sets

No-CCFLD

Control group

N=36, 205 data sets

N % N %

Male 9 64 22 61

Severe genotype# 14 100 26 72

DF508/DF508 7 50 23 63.9

Meconium ileus 3 21 3 8.3

Pancreatic insufficient 14 100 36 100

DIOS 2 14 1 3

UDCA use 13 93 8 22

#Severe genotype defined as CFTR class1-3 mutations

No-CCFLD control group

For an additional analysis, we established a no- CCFLD control group out of the patient group

D, of which we assumed that they had and would not develop CCFLD (Figure 1). By definition,

they were all older than 15 years of age at the time of study evaluation. The no-CCFLD control

group consisted of 36 patients (Table 2). The control group contained a slightly higher

proportion of males. Meconium ileus and DIOS had also occurred in the control group.

Remarkably 22% of no-CCFLD control group patients had reported use of UDCA. Since in this

group no splenomegaly or macronodularity has been found at any time, the indication for

UDCA most likely was increased liver enzymes, then considered as an indication for UDCA in

our centers.

Chapter 3

78

3

Controls CCFLD0

10

20

30

40

50

A

GG

T (

U/L

)

0 100 200 3000

50

100

150Controls

CCFLD

B

ALT (U/L)

GG

T (

U/l

)

0.0 0.2 0.4 0.6 0.8 1.00.0

0.2

0.4

0.6

0.8

1.0

C

1- Specificity

Sen

sit

ivit

y

Figure 2. GGT in CCFLD compared to no-CCFLD CF patients. The mean gamma-glutamyl transpeptidase

(GGT) results of the period 2 years prior to the diagnosis cirrhotic cystic fibrosis liver disease (CCFLD) and

GGT results of no-CCFLD CF (A). Relation between alanine aminotransferase (ALT) results and

corresponding GGT results of CCFLD patient (closed black dots) 2 years prior to the diagnosis CCFLD

compared and all corresponding GTT results(open gray dots, obtained between age 2-15 years, from CF

patients without any signs of liver disease at the age of 15 years) (B). Receiver operating curve (ROC)

analysis for the relation between presence of CCFLD and GGT results within 2 years prior to the diagnosis

CCFLD or no-CCFLD controls shows a AOC of 0.9±0.1; P<0.001, at a cutoff value for GGT of 21 U/l (C)

Increase of serum GGT associated with the development of cirrhotic cystic fibrosis liver disease

79

3

Table 3. Overview of effects, of different cut off values for GGT, on diagnostic testing for risk of

development of CCFLD

Cutoff value Specificity Sensitivity Odds ratio (CI) PPV NPV NNT

GGT > 20U/l 82 93 57 (7-1206) 26 99 4

GGT > 25U/l 87 71 16 (4-68) 27 98 4

GGT > 30U/l 92 71 28 (7-119) 37 98 3

GGT > 35U/l 95 64 39 (9-175) 50 98 2

GGT > 40U/l 96 50 28 (7-127) 50 96 2

CCFLD patients vs. no-CCFLD control patients

We compared GGT results of the CCFLD patients in the 2 years preceding the diagnosis CCFLD

(N=28 GGT measurements) with those of the no-CCFLD control group (N=205 GGT

measurements). We found that the mean GGT in the CCFLD patient group was significantly

higher than in the controls (45 vs. 17 U/l, respectively, P<0.001) (Figure 2A). However the

mean GGT of the CCFLD group was still within the normal-high region. Elevation of GGT in

CCFLD patients appeared, in most cases, an isolated event, i.e. increase in GGT was almost

never accompanied by simultaneous elevation in ALT (Figure 2B). In the CCFLD group 82% of

the patients had a repeated GGT level over 30 U/l in consecutive repeated measurements

versus 18% in the no-CCFLD controls. Only 14% of the patients in the no-CCFLD control group

ever showed a GGT over 35 U/l, however, in the CFFLD group this was 79%. In 2 (9%) of no-

CCFLD controls the GGT was never below 30 U/l (N=5/5 measurements, range 30-141U/l)

despite the fact that there were no ultrasonographic or physical signs of CCFLD in these

patients. In the CCFLD group 58% of patients never had a GGT below 30 U/l.

To establish a cutoff value of the GGT above which the risk for developing CCFLD would be

significantly and relevantly increased, we applied a ROC analysis. We used all the GGT results

of the CCFLD study group and the no-CCFLD control group. If in the CCFLD group, more than

one GGT results per patients was available in the period 2 years before the diagnosis CCFLD,

we used the mean value of all available results. In this manner each CCFLD patient only

contributed one GGT value to the analysis. We compared these mean GGT results of CCFLD

patients to all the available GGT results of the control group. We found an AOC of 0.9±0.1

(P<0.001), and determined a cutoff value for GGT of 21 U/l (Figure 3C). To establish an

indication for potential clinical relevance of cutoff values, we performed cross tabulations

analysis (Table 3). This analysis shows the different consequences for Odds ratio, specificity,

Chapter 3

80

3

sensitivity, positive predictive value and negative predictive value for different, arbitrarily,

GGT cutoff values. In a potential interventional trail for prevention of CCFLD the numbers

need to treat based on the applied cut of values would be highly relevant. Based on the

different chosen cutoff values for GGT, we calculated the theoretical numbers needed to

treat, and found them to vary from 2 to 4 (Table 3).

DISCUSSION

Cirrhotic liver disease is a severe and chronic complication of cystic fibrosis. CF patients in

early stage of disease could benefit from preventive treatment strategies. However, to date,

we do not have the possibility to test and identify patients at risk for CCFLD. In this study we

found that, at group level, in pediatric CF patients in the period 2 years preceding their

diagnosis CCFLD the serum GGT levels are significantly higher than in CF patients who never

develop CCFLD. It is notable that although generally higher than in no-CCFLD controls, also in

the CCFLD group GGT results often remain within the normal and high-normal range.

However, for CF patients with a mean GGT>35 U/L, based on repeated measurements, the

Odds ratio to develop CCFLD was 39. These results indicate that in pediatric CF patients a

persistent, high-normal, serum GGT is strongly associated with the diagnosis of CCFLD within

2 years. However the prognostic value of a single GGT measurement remains limited. Our

results do indicate that groups of patients at increased risk for CCFLD can be identified on the

basis of repeated GGT measurements.

Recent studies in cirrhosis research indicate that progression of fibrotic liver diseases can be

stopped or even reversed by removal of the causative agent or treatment of the underlying

disease (22). In particular in the field of Hepatitis B and C disease, antiviral treatment is shown

to be able to reverse the severity of fibrosis (23, 24). In addition, new developments are

evolving concerning the use of anti-fibrotic therapies in forms of liver disease. Although

theoretically promising, to date there have not yet anti-fibrotic therapies become available

for humans (25). However the scope of these positive developments indicates the rising

opportunity and potential profit for preemptive treatment in CCFLD. These scientific advances

also indicate the need for reliable and relevant markers to identify patients at risk for CCFLD.

For future treatment strategies, aimed at preventing CCFLD, the number of patients that need

to be treated, based on the used risk factors, would be relevant. To date, CF patients often

start with UDCA treatment based on persistent elevation of liver enzyme above the upper

limit of normal. Liver enzymes elevations, in particular AST and ALT, are rather frequent

events in CF patient. Additionally AST and ALT elevation seem not to be strongly associated

with the development of the cirrhotic form of CF liver disease. Therefore many patients are

probably treated with UDCA even though they do not actually carry an increased risk for the

Increase of serum GGT associated with the development of cirrhotic cystic fibrosis liver disease

81

3

development of CCFLD. Our results indicate that a cutoff value for GGT of 35 U/l could be

clinically relevant to identify patient at risks for the diagnosis CCFLD. When this cutoff value is

used in patient selection for preemptive treatment strategies the numbers needed to treat

would be two with high sensitivity, positive and negative predictive value and a relative

acceptable specificity.

For most of our CCFLD patients, GGT results of repeated measurements were available, in the

2 years preceding the diagnosis of CCFLD. In our current analysis, we used the mean of

available GGT measurements as an indicator for the course of GGT in this period. However, in

the ROC analysis for the no-CCFLD group we used all available GGT results per patients from

the age 2-15 years including outliers. We did not correct the no-CCFLD group for any isolated

single GGT elevations. Still, only two no-CCFLD patients had a mean GGT, of all their GGT

results, over 30 U/l. Therefore we feel that this methodology, that involved all available GGT

results of the control group, strengthens our conclusions. Based on these findings we

hypothesize that in a prospective study design that includes only persistent GGT elevation in

the analysis; an even more distinct difference between CCFLD and no-CCFLD can be obtained.

The current study provides additional information on the presence of (isolated) elevated liver

transaminases in CF patients. In the no-CCFLD control group we found that respectively 18%

and 12% of patients had recorded episodes of AST and ALT elevation above the upper limited

of normal. This observation indicates that increased transaminases are rather frequent events

in pediatric CF patients, not related to future development of CCFLD. Based on this

observation we conclude that elevation of transaminases are not a sensitive predictive

biochemical markers for development of CCFLD. It is possible that elevation of transaminases

is associated with other forms of CF related liver disease like for example, non-alcoholic

steatohepatitis. Elevation of transaminases can also result from secondary hepatotoxic effects

of medication or clinical infectious episodes..

We are aware that our retrospective study design has potential inherent weaknesses. First a

retrospective analysis of the clinical follow up may not be as accurate and complete as

desired. Secondly the date of the CCFLD diagnosis cannot be pinpointed to a specific day.

However the date of diagnosis used in this study is based, on the rather random date, of the

annual clinical checkup. Thirdly, for the no-CCFLD control group, we used a cut-off age of 15

years. This cut-off is based on the reasonable assumption that CCFLD does not develop

anymore beyond this age. However, we do realize that it cannot be ruled out that new cases

of CCFLD could still develop after the age of 15 years. Based on the above-listed limitations of

the current study we realize that the reported data, ideally, need confirmation in a

prospective study design.

We defined CCFLD as ultrasonographic macronodularity of the liver in combination with

splenomegaly. However in our analysis, we also encounter patients with either

macronodularity or splenomegaly and not the combination of both. We cannot exclude that

Chapter 3

82

3

these observations indicate an earlier phase of CCFLD development. For ultrasonographic

macronodularity, without splenomegaly, the only likely explanation is an advanced stage of

fibrosis, however still in the absence of relevant portal hypertension. For isolated

splenomegaly the situation is different and less clear. For example, splenomegaly has been

described in cystic fibrosis patients, in the absence of any signs of liver disease (26). On the

other hand Mueller-Abt et al. described a case of one patient with splenomegaly as the only

pathologic ultrasonographic finding in a patient with liver histology that matched cirrhosis

(10).

In summary, we found evidence that in pediatric CF patients, a persistent, high-normal,

serum GGT is strongly associated with the diagnosis of CCFLD within 2 years. The prognostic

value of a single GGT measurement however, remains limited. Our results indicate that

groups of patients at increased risk for CCFLD can be identified on the basis of repeated GGT

measurements.

Increase of serum GGT associated with the development of cirrhotic cystic fibrosis liver disease

83

3

REFERENCE LIST

1. Scottjupp R, Lama M, Tanner MS. Prevalence of liver-disease in cystic-fibrosis. Arch Dis

Child. 1991 06;66(6):698-701.

2. Rowland M, Gallagher CG, Ó'Laoide R, Canny G, Broderick A, Hayes R, et al. Outcome in

cystic fibrosis liver disease. Am J Gastroenterol. 2010;106(1):104-9.

3. Corbett K, Kelleher S, Rowland M, Daly L, Drumm B, Canny G, et al. Cystic fibrosis–

associated liver disease: A population-based study. J Pediatr. 2004 9;145(3):327-32.

4. Lindblad A, Glaumann H, Strandvik B. Natural history of liver disease in cystic fibrosis.

Hepatology. 1999 11;30(5):1151-8.

5. Debray D, Kelly D, Houwen R, Strandvik B, Colombo C. Best practice guidance for the

diagnosis and management of cystic fibrosis-associated liver disease. J Cyst Fibros.

2011;10:S29-36.

6. Cystic fibrosis foundation patient registry 2011. 2012.

7. Nash KL, Allison ME, McKeon D, Lomas DJ, Haworth CS, Bilton D, et al. A single centre

experience of liver disease in adults with cystic fibrosis 1995-2006. J Cyst Fibros. 2008

05;7(3):252-7.

8. Bartlett JR, Friedman KJ, Ling SC, Pace RG, Bell SC, Bourke B, et al. Genetic modifiers of liver

disease in cystic fibrosis. JAMA: the journal of the American Medical Association.

2009;302(10):1076-83.

9. Debray D, Lykavieris P, Gauthier F, Dousset B, Sardet A, Munck A, et al. Outcome of cystic

fibrosis-associated liver cirrhosis: Management of portal hypertension. J Hepatol.

1999;31(1):77-83.

10. Mueller-Abt PR, Frawley KJ, Greer RM, Lewindon PJ. Comparison of ultrasound and biopsy

findings in children with cystic fibrosis related liver disease. Journal of Cystic Fibrosis.

2008;7(3):215-21.

11. Lindblad A, Hultcrantz R, Strandvik B. Bile-duct destruction and collagen deposition: A

prominent ultrastructural feature of the liver in cystic fibrosis. Hepatology. 1992 08;16(2):372-

81.

12. Moyer K, Balistreri W. Hepatobiliary disease in patients with cystic fibrosis. Curr Opin

Gastroenterol. 2009;25(3):272-8.

13. Cheng K, Ashby D, Smyth RL. Ursodeoxycholic acid for cystic fibrosis-related liver disease.

Cochrane Database Syst Rev. 2012;10.

14. Fallowfield JA. Therapeutic targets in liver fibrosis. American Journal of Physiology-

Gastrointestinal and Liver Physiology. 2011;300(5):G709-15.

Chapter 3

84

3

15. Lenaerts C, Lapierre C, Patriquin H, Bureau N, Lepage G, Harel F, et al. Surveillance for

cystic fibrosis-associated hepatobiliary disease: Early ultrasound changes and predisposing

factors. J Pediatr. 2003 09;143(3):343-50.

16. Colombo C, Apostolo MG, Ferrari M, Seia M, Genoni S, Giunta A, et al. Analysis of risk

factors for the development of liver disease associated with cystic fibrosis. J Pediatr. 1994

3;124(3):393-9.

17. Potter CJ, Fishbein M, Hammond S, McCoy K, Qualman S. Can the histologic changes of

cystic fibrosis-associated hepatobiliary disease be predicted by clinical criteria? J Pediatr

Gastroenterol Nutr. 1997;25(1):32-6.

18. Colombo C, Battezzati PM, Podda M, Bettinardi N, Giunta A. Ursodeoxycholic acid for liver

disease associated with cystic fibrosis: A double‐blind multicenter trial. Hepatology.

2003;23(6):1484-90.

19. Mayer-Hamblett N, Kloster M, Ramsey BW, Narkewicz MR, Saiman L, Goss CH. Incidence

and clinical significance of elevated liver function tests in cystic fibrosis clinical trials.

Contemporary Clinical Trials;34(2):232-8.

20. Loftus WK, Metreweli C. Ultrasound assessment of mild splenomegaly: Spleen/kidney

ratio. Pediatr Radiol. 1998;28(2):98-100.

21. Schwimmer JB, Dunn W, Norman GJ, Pardee PE, Middleton MS, Kerkar N, et al. SAFETY

study: Alanine aminotransferase cutoff values are set too high for reliable detection of

pediatric chronic liver disease. Gastroenterology. 2010 4;138(4):1357,1364.e2.

22. Ellis EL, Mann DA. Clinical evidence for the regression of liver fibrosis. J Hepatol. 2012

5;56(5):1171-80.

23. Brown A, Goodman Z. Hepatitis B-associated fibrosis and fibrosis/cirrhosis regression with

nucleoside and nucleotide analogs. Expert review of gastroenterology & hepatology.

2012;6(2):187-98.

24. D'Ambrosio R, Aghemo A, Rumi MG, Ronchi G, Donato MF, Paradis V, et al. A

morphometric and immunohistochemical study to assess the benefit of a sustained virological

response in hepatitis C virus patients with cirrhosis. Hepatology. 2012;56(2):532-43.

25. Friedman SL. Evolving challenges in hepatic fibrosis. Nature Reviews Gastroenterology and

Hepatology. 2010;7(8):425-36.

26. Penketh A, Wise A, Mearns M, Hodson M, Batten J. Cystic fibrosis in adolescents and

adults. Thorax. 1987;42(7):526-32.

CHAPTER 4

CHOLIC ACID INDUCES A CFTR

DEPENDENT BILIARY SECRETION

AND LIVER GROWTH RESPONSE IN

MICE Submitted

Frank A. J. A. Bodewes1

Marcel J. Bijvelds2

Willemien de Vries1

Julius F.W. Baller1

Annet S.H. Gouw3

Hugo R. de Jonge2

Henkjan J. Verkade1

1 Department of Pediatrics, University of Groningen, Beatrix Children’s Hospital - University

Medical Center, Groningen, The Netherlands

2 Department of Gastroenterology & Hepatology, Erasmus University Medical Center,

Rotterdam, The Netherlands

3 Department of Pathology, University Medical Center, Groningen, The Netherlands

Chapter 4

86

4 4

ABSTRACT

Introduction

Cystic fibrosis (CF) mouse models do not display cirrhotic CF liver disease. We hypothesized

that a cirrhotic CF liver disease equivalent may be provoked in CF mice by hydrophobic bile

salt exposure.

Methods

We studied the effect of acute (intravenous) and chronic (dietary) cholate administration on

bile formation, bile composition and histology in a CF mouse model.

Results

Basal bile production, i.e. without bile salt administration, was similar in Cftr-/-

mice and

controls. Intravenous taurocholate stimulated bile salt secretion and bile flow similarly in both

genotypes. Cholate administration for 3 weeks also increased bile salt secretion similarly in

Cftr-/-

mice and controls. However, cholate administration for 3 weeks increased bile flow in

Cftr-/-

mice to a lower extend, resulting in a significantly higher biliary bile salt concentration,

compared with controls. In controls, but not in Cftr-/-

mice, chronic cholate increased both the

biliary phospholipid-to-bile salt (molar) ratio and bile hydrophobicity. Cholate administration

for 3 weeks did not induce liver histology analogous to cirrhotic cystic fibrosis liver disease in

Cftr-/-

mice. However, chronic cholate administration induced a liver growth (+52%, p<0.001)

and parenchymal proliferation response (Ki67) in controls that was lacking in CF mice.

Conclusion

We conclude that chronic CA administration induces a Cftr dependent biliary secretion and

liver growth and proliferation response in mice. Chronic CA exposure did not induce cirrhotic

cystic fibrosis liver disease in CF mice. This response includes Cftr dependent cytoprotective

changes in the bile composition. These findings point to a, potentially protective, Cftr

dependent, hepatic response to prolonged hydrophobic bile salt exposure.

Cholic acid induces a Cftr dependent biliary secretion and liver growth response in mice

87

4 4 4

INTRODUCTION

Cystic fibrosis (CF) is caused by mutations in the CFTR gene (1,2). Cystic fibrosis liver disease

(CFLD) develops in 5-10% of cystic fibrosis patients (3). It is a serious complication of CF (4,5).

CFLD is characterized by cirrhosis and patients often present with splenomegaly,

hypersplenism and complications of portal hypertension, including variceal bleeding and

ascites (6,7). Although the synthesis and excretory liver function are usually spared, in some

CFLD patients liver transplantation can be indicated (8,9). Despite major advances in basic

knowledge about cystic fibrosis, the pathogenesis of CFLD is still unexplained.

In the liver, CFTR is expressed exclusively at the apical membrane of the cholangiocytes lining

the bile ducts (10,11). CFTR in cholangiocytes mediates water secretion, and controls the pH

of bile (12,13). Consequently, loss of CFTR function may increase the concentration of

potentially cytotoxic bile salts in bile and increase its viscosity, leading to occlusion of small

bile ducts and, ultimately, obstructive biliary cirrhosis (14-16).

CF mice may show minor gallbladder and liver abnormalities that do not lead to loss of liver

function (17-20). In general, mice produce bile that is relatively hydrophilic (21). Hydrophilic

bile salts are considered less cytotoxic than hydrophobic ones (22,23). The hydrophilic profile

of murine bile could prevent or mitigate liver disease in CF mice.

To test this hypothesis we sought to change the composition of murine bile to one containing

more hydrophobic bile salts. To this end, we either acutely or chronically challenged CF mice

with cholic acid (CA), a bile salt with strong detergent action or monitored effects on bile

production, composition, serum alanine transaminase (ALT), liver morphology, and histology

(23)

MATERIALS AND METHODS

Animals.We used C57Bl/6;129 Cftr-/-tm1CAM

mice and Cftr +/+tm1CAM

littermate controls. All mice

were bred and accommodated at the Animal Experimental Center of the Erasmus Medical

Center in Rotterdam, The Netherlands (17). Mice were housed in a light-controlled (lights on

6 AM to 6 PM) and temperature-controlled (21oC) facility, and had free access to tap water

and a semi-synthetic diet (SRM-A; Hope Farms BV Woerden, The Netherlands) from the time

of weaning. All experiments were performed with mice of 10-20 weeks of age. Group size

varied per experiment from 5-7 mice per genotype. Experimental protocols were approved by

the Ethical Committee for Animal Experiments of Erasmus MC. All surgery was performed

under sodium pentobarbital anesthesia, and all efforts were made to minimize suffering.

Experimental procedures. To test our hypothesis we used a mouse gall bladder cannulation

model (24,25). We first applied an acute bile salt infusion, as previously described (25). We

infused taurine-conjugated cholate (TCA) (43 mM dissolved in phosphate-buffered saline, pH

Chapter 4

88

4 4

7.4) via the jugular vein using an infusion pump. The bile salt dosage was increased every 30

minutes in a stepwise manner (dosage steps 0, 50, 300, 450 and 600 nmol.min-1

). During bile

salt infusion, bile was collected in 15 minutes fractions.

Different groups of Cftr-/-

and control mice were either fed a control diet consisting of

standard chow or the same diet enriched with cholate (CA 0.5% wt/wt) for 3 weeks.

Additional experimental groups of Cftr-/-

mice and controls were fed the diet for 3 months.

After gallbladder cannulation as described above, genuine bile production was determined by

bile collection for 30 minutes. The bile was collected for analysis. Mice were then sacrificed by

cardiac puncture, blood was drawn by cardiac puncture and livers samples were immersed in

neutral buffered formalin.

Analytical techniques. Biliary bile salt concentrations were determined by an enzymatic

fluorometric assay (26). Lipids were extracted from the bile (27). The phospholipid

concentrations were determined using a spectrophotometric assay (28). Biliary bile salt

composition was determined by capillary gas chromatography (29). The hydrophobicity of bile

salts in bile was calculated according to the Heuman index based on the fractional

contribution of the mayor bile salt species (23). ALT was determined in plasma samples.

For histological analysis of the liver samples, hematoxylin/eosin staining was applied on

paraffin embedded sections. To quantify mitotic active cells we used Ki-67 immunostaining

using a rabbit anti Ki-67p monoclonal antibody (Novocastra Laboratories Ltd., Newcastle upon

Tyne, UK) (30).

An experienced hepato-pathologist (A.S.H.G.) assessed the liver samples in a blinded fashion.

Histology of the liver parenchyma and portal tracts were evaluated separately for

inflammation, fibrosis and Ki-67 mitotic activity. The score was performed in a semi-

quantitative approach on a scale from 0 to 3 (0: absent, 1: sporadic, 2: regular, 3: frequent)

(31).

Statistical analysis. Analyses were performed using SPSS version 18.0 for Windows (SPSS Inc.,

Chicago, IL). All nominal results are reported as means ± SEM. The ordinal histology results

are reported as median and range. Differences between study groups were evaluated using

the Mann-Whitney U test. The level of significance was set at a P value of less than 0.05.

RESULTS

Bile production and bile salt secretion after interruption of the enterohepatic circulation. Bile

production, assessed over 30 minutes immediately after interruption of the enterohepatic

circulation, did not differ significantly between Cftr-/-

mice and controls (mean of 2 time points

in first 30 minutes: 6.3±0.7 vs. 5.8±0.5 μl.min-1

.100 g -1, respectively; Figure 1A). Bile salt

Cholic acid induces a Cftr dependent biliary secretion and liver growth response in mice

89

4 4 4

concentrations and bile salt secretion rates were also similar between Cftr-/-

mice and controls

(Figure 1B and C).

Bile production and bile salt secretion during acute IV taurocholic acid (TCA) administration.

Administration of TCA (iv) dose-dependently increased bile production (Figure 1A). The dose-

response relationship was similar for Cftr-/-

mice and controls, indicating that bile production

was not impaired in CF mice. A concomitant dose-dependent increase in bile salt

concentrations and bile salt secretion rates was observed upon TCA infusion, in both Cftr-/-

mice and controls (Figure 1B and C). A linear correlation was observed between the bile salt

secretion rate and bile flow (Figure 1D). The slopes derived after linear regression analysis,

were similar for Cftr-/-

mice and controls, indicating similar levels of bile salt dependent bile

flow. These observations strongly suggest that the bile flow in mice, both under basal

conditions and after acute IV TCA administration, is (predominantly) generated via CFTR

independent mechanisms.

0

5

10

15

20

Cftr+/+

ACftr

-/-

150300

450600 TCA infusion rate

(nmol.min -1)

30

Time (minutes)

60 90 120 1650

Bile f

low

( l.m

in-1

.100 g

ram

s B

W-1

)

0

50

100

150

Cftr+/+

BCftr

-/-

150300

450600 TCA infusion rate

(nmol.min -1)

30

Time (minutes)

60 90 120 1650

Bil

e s

alt

co

ncen

trati

on

(mM

)

0

500

1000

1500

2000

Cftr+/+

CCftr

-/-

150300

450600 TCA infusion rate

(nmol.min -1)

30

Time (minutes)

60 90 120 1650

Bil

e s

alt

secre

ton

rate

(nm

ol.m

in-1

.100 g

ram

s B

W-1

)

0 500 1000 1500 20000

5

10

15

Cftr+/+

DCftr

-/-

Bile salt secreton rate

(nmol.min-1

.100 grams BW-1

)

Bile f

low

( l.m

in-1

.100 g

ram

s B

W-1

)

Figure 1. Biliary parameters during intravenous taurocholic acid (TCA) administration. Biliary bile flow (A), bile salt

concentration (B), bile salt secretion rate (C) and relationship between bile salt secretion rate and bile flow (D) in Cftr

knockout mice (Cftr-/) and control littermates (Cftr+/+) during intravenous infusion with TCA in stepwise increasing

dosage of. The grey symbols in (D) represent baseline values before the start of TCA infusion. Data are presented as

means ± SEM of N=5-7 mice per group. There was no significant difference between Cftr-/- and Cftr+/+ mice, at any of

the individual time points, for bile flow, bile salt concentration and bile salt secretion rate.

Chapter 4

90

4 4

Bile production and bile salt composition during chronic cholic acid feeding. Feeding a CA

containing diet for 3 weeks increased bile flow, both in Cftr-/-

mice and controls. However, the

increase was larger in controls than in Cftr-/-

mice (Figure 2A). The CA diet increased the biliary

bile salt concentration in Cftr-/-

animals but, not in controls (Figure 2B). Chronic CA feeding

increased the bile salt secretion rate in both Cftr-/-

mice and controls (Figure 2C).

To assess cytotoxicity, we measured the biliary phospholipid concentration and calculated the

phospholipid -to-bile salt ratio. During regular chow feeding, biliary phospholipid

concentrations were similar in Cftr-/-

mice and controls (Figure 2D), as were the phospholipid -

to- bile salt ratios (Figure 2E). CA feeding appeared to increase phospholipid concentrations to

a similar magnitude in Cftr-/-

and controls although the apparent differences between means

did not reach statistical significance (Figure 2D). CA administration increased the phospholipid

-to- bile salt ratio in controls. Because CA feeding increased the biliary bile salt concentration

in Cftr-/-

mice (but not in controls; see above) the phospholipid -to- bile salt ratio, under these

conditions, was significantly lower in Cftr-/-

mice than in controls (Figure 2E). We found no

significant difference in serum ALT in Cftr-/-

mice after CA diet compared to normal diet

(Figure 2F). The CA diet, however, significantly increased ALT, compared to normal in controls

and not in Cftr-/-

mice.

CA feeding markedly increased the fractional contribution of CA to the total bile salt pool of

the bile. In control mice, it rose to 90%, whereas in Cftr-/-

mice it reached near 100%. In the

control animals, the remaining 10% was attributed to deoxycholic acid (DCA), which is formed

from CA by intestinal bacteria (Figure 2G). Based on this result, after CA diet, the Heuman

hydrophobicity index of the bile was higher in controls than in Cftr-/-

mice (0.061 vs. 0.003,

respectively, P<0.01, Figure 2H).

On regular chow, serum ALT levels were similar in Cftr-/-

mice and controls (Figure 2F). CA

feeding increased serum ALT had in controls, suggesting that, despite the concomitant

increase in the phospholipid-to-bile salt ratio, the bile produced was more cytotoxic (Figure

2F). In Cftr-/-

mice, the CA diet induced a similar mean increase in serum ALT levels, despite a

markedly lower hydrophobicity index. However, the variation between individual animals was

larger than for controls, suggesting that a subset of Cftr-/-

animals may be more prone to

develop lesions of the biliary tract.

Cholic acid induces a Cftr dependent biliary secretion and liver growth response in mice

91

4 4 4

Normal diet CA diet0

5

10

15

20

25

Cftr+/+

Cftr-/-

A**

*

Bile f

low

(l.m

in-1

.100 g

ram

s B

W-1

)

Normal diet CA diet0

50

100

150

Cftr+/+

Cftr-/-

B*

*

Bile s

alt

co

ncen

trati

on

(mM

)Normal diet CA diet

0

500

1000

1500

2000

Cftr+/+

Cftr-/-

C*

*

Bile s

alt

secre

ton

rate

(nm

ol.m

in-1

.100 g

ram

s B

W-1

)

Normal diet CA diet0

5

10

15

Cftr+/+

Cftr-/-

D

Ph

osp

ho

lip

id c

on

cen

trati

on

(mM

)

Normal diet CA diet0.00

0.05

0.10

0.15

0.20

0.25

Cftr+/+

Cftr-/-

E

*

Ph

osp

ho

lip

id

to b

ile s

alt

rati

o

Normal diet CA diet0

50

100

150

Cftr+/+

Cftr-/-

F

*

AL

T (

U/l)

Cftr+/+

Cftr-/-

Cftr+/+

Cftr-/-

0

25

50

75

100

CholateDCA

Normal diet CA diet

G

Others

%

-0.10

-0.05

0.00

0.05

0.10

Cftr+/+

Cftr-/-*

H

Normal diet CA diet

*

Hyd

rop

ho

bic

ity i

nd

ex

Figure 2. Biliary parameters and bile composition before and after chronic dietary cholic acid (CA) administration.

Biliary bile flow (A), bile salt concentration (B), bile salt secretion rate (C), phospholipid concentration (D),

phospholipid-to- bile salt ratio (E) Serum ALT levels (F), percent contribution of the bile salts cholate, deoxycholate and

others (chenodeoxycholate, deoxycholate, ursodeoxcholate, α-muri cholate and β-muricholate) in bile (G) and

Heuman index of total bile salts in bile representing the hydrophobicity of bile salts (H) in Cftr knockout mice (Cftr-/-)

and control littermates (Cftr+/+) after a regular or 0.5%-CA (wt/wt) chow diet for 3 weeks. Data are presented as

means ± SEM or percentage if appropriate of N=6-7 mice per group. *P-value<0.05.

Chapter 4

92

4 4

Cftr dependent effects of chronic CA exposure on liver weight. Similar to previous observations

in other CF mouse models, our Cftr-/-

mice tended to be smaller than littermate controls

(Figure 3A) (32). CA feeding did not markedly affect body weight. On regular chow, Cftr-/-

mice

and controls had similar absolute and relative liver weights (Figure 3C and 3E). Chronic CA

administration increased the absolute and relative liver weight in controls, but not in Cftr-/-

mice (+52%, p<0.001), indicating that chronic CA administration induces a hepatic, Cftr-

dependent adaptive mechanism to stimulate liver enlargement (Figure 3C and 3E). The CA-

induced liver growth response (+33%, p<0.01) was also observed in control mice with a

genetic background (FVB;129Sv) different from the one of the Cftr-/-

littermates (C57Bl6;129).

The liver growth response to CA-diet was lost in mice expressing mutant Cftr (homozygous

F508del Cftr mice; Cftrtm1EUR

), indicating that it is largely independent of subtle differences

between mouse strains and requires functional CFTR, Figure 3 B, 3D and 3F) (33).

Cftr dependent effects of chronic CA exposure on liver histology. On a regular chow diet,

parenchymal cell proliferation was enhanced in Cftr-/-

mice, relative to controls (Figure 4C),

but no signs of parenchymal inflammation were evident. Chronic CA administration increased

parenchymal mitotic activity in controls to a level similar to that found in Cftr-/-

mice on either

regular chow or after chronic CA exposure. Concomitantly, 3 week CA feeding significantly

increased the parenchymal inflammation score in controls (Figure 4D). Prolonged (3 month)

CA exposure exceeds the 3 weeks findings, showing increased parenchymal mitotic activity

and parenchymal inflammation, particularly observed in controls. CA feeding did not generate

pathological changes in the portal tract areas of the Cftr-/-

mice, or controls (Figure 4 A and B).

DISCUSSION

In mice, both acute and chronic hydrophilic bile salt administration markedly increased biliary

bile salt production and flow rates. Cftr-/-

mice showed a reduced capacity to adapt to chronic

bile salt administration. Although biliary bile salt production was similar the increase in bile

flow was less pronounced in Cftr-/-

mice compared controls. This reduced capacity, of Cftr-/-

mice, to increase bile flow resulted in significantly higher biliary bile salt concentrations. No

such defect was observed upon acute intravenous bile salt loading, suggesting that only long-

term adaptations to bile salt loading, such as hepatic cell proliferation (discussed below), is

perturbed in CF. These results imply that in mice, in contrast to humans, either Cftr does not

significantly contribute to canalicular or ductular fluid secretion or that, in the absence of Cftr

is substituted by other anion channels.

Cholic acid induces a Cftr dependent biliary secretion and liver growth response in mice

93

4 4 4

Normal diet CA diet0

10

20

30

40

Cftr+/+

Cftr-/-

A*

Bo

dy w

eig

ht

(gra

ms)

Normal diet CA diet0

10

20

30

Cftrwt/wt

Cftrdf508/df508

B

Bo

dy w

eig

ht

(gra

ms)

Normal diet CA diet0

1

2

3

Cftr+/+

Cftr-/-

C**

Liv

er

weig

ht

(gra

ms)

Normal diet CA diet0

1

2

Cftrwt/wt

Cftrdf508/df508

D**

Liv

er

weig

ht

(gra

ms)

Normal diet CA diet0.00

0.02

0.04

0.06

0.08

Cftr+/+

Cftr-/-

E**

Liv

er

to b

od

y w

eig

ht

rati

o

Normal diet CA diet0.00

0.02

0.04

0.06

0.08

Cftrwt/wt

Cftrdf508/df508

F**

Liv

er

to b

od

y w

eig

ht

rati

o

Figure 3. Body and liver weight before and after chronic dietary cholic acid (CA) or UDCA administration. Body weight

(A/B), absolute liver weight (C/D) and relative liver weight reported as a percentage of the body weight (E/F) in Cftr

knockout mice (Cftr-/-) or ΔF508 Cftr mice (Cftrtm1EUR) and their respective control littermates (Cftr+/+) after a

regular or 0.5%-CA (wt/wt) chow diet for 3 weeks. Data are presented as means ± SEM of N=5-7 mice per group. *P-

value<0.05.

Chapter 4

94

4 4

0

1

2

3 Cftr-/-

Normaldiet

CA diet3 weeks

CA diet3 months

Cftr+/+

AP

ort

al

Ki6

7

sco

re

0

1

2

3Cftr

-/-

Normaldiet

CA diet3 weeks

CA diet3 months

Cftr+/+

B

Po

rtal

infl

am

tio

n

sco

re

0

1

2

3

Normaldiet

CA diet3 weeks

CA diet3 months

*

*

C

Cftr-/-

Cftr+/+

Pare

nch

ym

al

Ki6

7

sco

re

0

1

2

3Cftr

-/-

Normaldiet

CA diet3 weeks

CA diet3 months

Cftr+/+

D *

*

*

Pare

nch

ym

al

infl

am

tio

n

sco

re

Figure 4. Liver histology before and after chronic cholic acid (CA) administration. Histological evaluation of liver

parenchymal inflammation (A) parenchymal Ki67 activity (B), portal inflammation (C) and portal Ki67 activity (D) in

Cftr knockout mice (Cftr-/-) and control littermates (Cftr+/+) after a regular, a 0.5%-CA (wt/wt) chow diet for 3 weeks or

0.5%-CA (wt/wt) chow diet for 3 months. Histology was scored in a blinded and semiquantitative approach on an

ordinal scale: 0: absent, 1: sporadic, 2: regular, 3 frequent. All individual results are presented; the horizontal black

line represents the median of N=4-6 mice per group

We had expected that in CF conditions prolonged CA feeding would the increase in biliary bile

salt concentration together with increased level of secondary hydrophobic bile salts. We

anticipated that this combination of hydrophilic bile salts would result in more cytotoxic bile.

However, we found that under these conditions, bile of Cftr-/-

mice contained little of the

secondary bile salt deoxycholate (DCA), whereas in controls its level in bile was markedly

enhanced by CA feeding. Since DCA is produced exclusively by intestinal bacteria, these

results suggest that alterations in the composition of this micro flora, which are typical of CF,

reduce DCA production (34-36). Alternatively, absorption of DCA may be reduced in CF. This

low availability of DCA may reduce the cytotoxicity of bile. Therefore, this hitherto unknown

mechanism may protect CF mice against bile salt-induced biliary disease.

A previous study reported the development of spontaneous liver disease in CF mice (18). The

mice used in these studies had a genetic background (C57BL/6J) different from the animals

used by us. This difference between strains indicates that genetic modifiers might affect the

Cholic acid induces a Cftr dependent biliary secretion and liver growth response in mice

95

4 4 4

hepatobiliary phenotype. Considering the potential effect of the intestinal micro flora on bile

composition, environmental factors may also play a role (36). Peptamen, a liquid diet

formulation used to prevent intestinal obstruction (used in ref. 14, but not by us), has been

shown to promote small intestinal bacterial overgrowth in CF mice (37). This may lead to

enhanced production of secondary bile salts, in turn leading to the production of a more

cytotoxic bile and, consequently, liver disease.

We found that CA feeding significantly increased liver mass in control mice. Similar findings

have been reported previously (38). This earlier study showed that CA feeding also enhanced

liver regeneration after partial hepatectomy. It was proposed that this response is mediated

through the Farnesoid X receptor (FXR) nuclear receptor that is directly activated by specific

bile salt, including CA and DCA. In accordance, we have found earlier that UDCA, which is a

limited FXR ligand, does not resemble the effect of CA on liver growth (39). In this study, oral

feeding of UDCA (0.5% wt/wt) to Cftr-WT mice for 3 weeks, did not result in a significant

increase in relative liver weight (0.048 ±0.003 vs. 0.044±0.001, respectively; unpublished

data).

Presently, we found no increase in liver mass in CA-fed Cftr-/-

and CftrΔF508/ΔF508

mice. This

indicates that FXR signaling may be perturbed in CF mice, a notion, which is reinforced by a

previous study showing that FXR-controlled genes are down-regulated in the ileum of CF

mice, possibly, because of reduced bile salt uptake in ileocytes (40).

In summary, we show that, in mice, CFTR plays a role in the hepatic adaptive response to high

bile salt exposure. The CF condition is characterized by a reduction in bile salt-induced hepatic

cell proliferation, increase in biliary bile salt levels and difference in secondary bile salt

composition. These changes suggest that the CF liver may be more prone to develop hepato-

biliary disease upon (chronic) exposure to hepatotoxic substances.

ACKNOWLEDGMENTS:

Rick Havinga for the technical assistance and the mice micro-surgery procedures

REFERENCE LIST

(1) Collins FS. Cystic fibrosis: molecular biology and therapeutic implications. Science 1992

05/08;256(5058):774-779.

(2) Kerem B, Rommens JM, Buchanan JA, Markiewicz D, Cox TK, Chakravarti A, et al.

Identification of the cystic fibrosis gene: genetic analysis. Science 1989 09/08;245(4922):1073-

1080.

Chapter 4

96

4 4

(3) Scottjupp R, Lama M, Tanner MS. Prevalence of Liver-Disease in Cystic-Fibrosis. Arch Dis

Child 1991 06;66(6):698-701.

(4) Rowland M, Gallagher CG, Ó'Laoide R, Canny G, Broderick A, Hayes R, et al. Outcome in

Cystic Fibrosis Liver Disease. Am J Gastroenterol 2010;106(1):104-109.

(5) Corbett K, Kelleher S, Rowland M, Daly L, Drumm B, Canny G, et al. Cystic fibrosis–

associated liver disease: a population-based study. J Pediatr 2004 9;145(3):327-332.

(6) Lindblad A, Glaumann H, Strandvik B. Natural history of liver disease in cystic fibrosis.

Hepatology 1999 11;30(5):1151-1158.

(7) Debray D, Kelly D, Houwen R, Strandvik B, Colombo C. Best practice guidance for the

diagnosis and management of cystic fibrosis-associated liver disease. J Cyst Fibros

2011;10:S29-S36.

(8) Cystic Fibrosis Foundation Patient Registry 2011. Annual Data Report Cystic Fibrosis

Foundation Bethesda, Maryland 2012.

(9) Nash KL, Allison ME, McKeon D, Lomas DJ, Haworth CS, Bilton D, et al. A single centre

experience of liver disease in adults with cystic fibrosis 1995-2006. J Cyst Fibros 2008

05;7(3):252-257.

(10) Cohn JA, Strong TV, Picciotto MR, Nairn AC, Collins FS, Fitz JG. Localization of the cystic

fibrosis transmembrane conductance regulator in human bile duct epithelial cells.

Gastroenterology 1993 12;105(6):1857-1864.

(11) Fitz JG, Basavappa S, McGill J, Melhus O, Cohn JA. Regulation of membrane chloride

currents in rat bile duct epithelial cells. J Clin Invest 1993 01;91(1):319-328.

(12) Beuers U, Hohenester S, de Buy Wenniger LJM, Kremer AE, Jansen PLM, Elferink RPJO. The

biliary HCO3- umbrella: A unifying hypothesis on pathogenetic and therapeutic aspects of

fibrosing cholangiopathies. Hepatology 2010;52(4):1489-1496.

(13) Fitz JG. Regulation of cholangiocyte secretion. Semin Liver Dis 2002 08;22(3):241-249.

(14) Sinaasappel M. Hepatobiliary pathology in patients with cystic fibrosis. Acta Paediatr

Scand 1988;363:45-50; discussion 50-1.

(15) Moyer K, Balistreri W. Hepatobiliary disease in patients with cystic fibrosis. Curr Opin

Gastroenterol 2009;25(3):272-278.

(16) Feranchak AP, Sokol RJ. Cholangiocyte biology and cystic fibrosis liver disease. Semin Liver

Dis 2001 11;21(4):471-488.

(17) Ratcliff R, Evans MJ, Cuthbert AW, MacVinish LJ, Foster D, Anderson JR, et al. Production

of a severe cystic fibrosis mutation in mice by gene targeting. Nat Genet 1993 05;4(1):35-41.

Cholic acid induces a Cftr dependent biliary secretion and liver growth response in mice

97

4 4 4

(18) Durie PR, Kent G, Phillips MJ, Ackerley CA. Characteristic multiorgan pathology of cystic

fibrosis in a long-living cystic fibrosis transmembrane regulator knockout murine model. Am J

Pathol 2004 04;164(4):1481-1493.

(19) Freudenberg F, Leonard MR, Liu SA, Glickman JN, Carey MC. Pathophysiological

preconditions promoting mixed “black” pigment plus cholesterol gallstones in a ΔF508 mouse

model of cystic fibrosis. Am J Physiol Gastrointest Liver Physiol 2010;299(1):G205-G214.

(20) Fiorotto R, Scirpo R, Trauner M, Fabris L, Hoque R, Spirli C, et al. Loss of CFTR Affects

Biliary Epithelium Innate Immunity and Causes TLR4-NF-kB-Mediated Inflammatory Response

in Mice. Gastroenterology 2011;141(4):1498-1508.e5.

(21) van Nieuwerk CM, Groen AK, Ottenhoff R, van Wijland M, Van Den Bergh Weerman MA,

Tytgat GN, et al. The role of bile salt composition in liver pathology of mdr2 (-/-) mice:

differences between males and females. J Hepatol 1997 01;26(1):138-145.

(22) Benedetti A, Alvaro D, Bassotti C, Gigliozzi A, Ferretti G, La Rosa T, et al. Cytotoxicity of

bile salts against biliary epithelium: a study in isolated bile ductule fragments and isolated

perfused rat liver. Hepatology 1997;26(1):9-21.

(23) Heuman D, Hylemon P, Vlahcevic Z. Regulation of bile acid synthesis. III. Correlation

between biliary bile salt hydrophobicity index and the activities of enzymes regulating

cholesterol and bile acid synthesis in the rat. J Lipid Res 1989;30(8):1161-1171.

(24) Bijvelds MJC, Bronsveld I, Havinga R, Sinaasappel M, de Jonge HR, Verkade HJ. Fat

absorption in cystic fibrosis mice is impeded by defective lipolysis and post-lipolytic events. Am

J Physiol Gastrointest Liver Physiol 2005 04;288(4):G646-G653.

(25) Verkade HJ, Havinga R, Shields DJ, Wolters H, Bloks VW, Kuipers F, et al. The

phosphatidylethanolamine N-methyltransferase pathway is quantitatively not essential for

biliary phosphatidylcholine secretion. J Lipid Res 2007;48(9):2058-2064.

(26) Mashige F, Imai K, Osuga T. A simple and sensitive assay of total serum bile acids. Clin

Chim Acta 1976 07/01;70(1):79-86.

(27) Bligh EG, Dyer WJ. A rapid method of total lipid extraction and purification. Am J Physiol

Gastrointest Liver Physiol 1959;37(8):911-917.

(28) Böttcher C, Van Gent C. Changes in the composition of phospholipids and of phospholipid

fatty acids associated with atherosclerosis in the human aortic wall. J Atheroscler Res

1961;1(1):36-46.

(29) Kok T, Hulzebos CV, Wolters H, Havinga R, Agellon LB, Stellaard F, et al. Enterohepatic

circulation of bile salts in FXR-deficient mice: Efficient intestinal bile salt absorption in the

absence of ileal bile acid-binding protein (Ibabp). J Biol Chem 2003;278(43):41930-41937.

Chapter 4

98

4 4

(30) Mauad TH, Van Nieuwkerk CM, Dingemans KP, Smit JJ, Schinkel AH, Notenboom RG, et al.

Mice with homozygous disruption of the mdr2 P-glycoprotein gene. A novel animal model for

studies of nonsuppurative inflammatory cholangitis and hepatocarcinogenesis. Am J Pathol

1994 11;145(5):1237-1245.

(31) Van Nieuwkerk CM, Elferink RP, Groen AK, Ottenhoff R, Tytgat GN, Dingemans KP, et al.

Effects of Ursodeoxycholate and cholate feeding on liver disease in FVB mice with a disrupted

mdr2 P-glycoprotein gene. Gastroenterology 1996 07;111(1):165-171.

(32) Scholte BJ, Davidson DJ, Wilke M, de Jonge HR. Animal models of cystic fibrosis. J Cyst

Fibros 2004;3:183-190.

(33) van Doorninck JH, French PJ, Verbeek E, Peters RH, Morreau H, Bijman J, et al. A mouse

model for the cystic fibrosis delta F508 mutation. EMBO J 1995 09/15;14(18):4403-4411.

(34) De Lisle RC, Roach EA, Norkina O. Eradication of small intestinal bacterial overgrowth in

the cystic fibrosis mouse reduces mucus accumulation. J Pediatr Gastroenterol Nutr

2006;42(1):46-52.

(35) De Lisle RC, Mueller R, Boyd M. Impaired Mucosal Barrier Function in the Small Intestine

of the Cystic Fibrosis Mouse. J Pediatr Gastroenterol Nutr 2011;53(4):371-379.

(36) Wouthuyzen-Bakker M, Bijvelds MJC, de Jonge HR, De Lisle RC, Burgerhof JGM, Verkade

HJ. Effect of antibiotic treatment on fat absorption in mice with cystic fibrosis. Pediatr Res

2011;71(1):4-12.

(37) De Lisle RC. Altered transit and bacterial overgrowth in the cystic fibrosis mouse small

intestine. Am J Physiol Gastrointest Liver Physiol 2007;293(1):G104-G111.

(38) Huang W, Ma K, Zhang J, Qatanani M, Cuvillier J, Liu J, et al. Nuclear receptor-dependent

bile acid signaling is required for normal liver regeneration. Science 2006

04/14;312(5771):233-236.

(39) Bodewes FAJA, Wouthuyzen-Bakker M, Bijvelds MJC, Havinga R, de Jonge HR, Verkade HJ.

Ursodeoxycholate modulates bile flow and bile salt pool independently from the cystic fibrosis

transmembrane regulator (Cftr) in mice. Am J Physiol Gastrointest Liver Physiol

2012;302(9):G1035-G1042.

(40) Debray D, Rainteau D, Barbu V, Rouahi M, El Mourabit H, Lerondel S, et al. Defects in

gallbladder emptying and bile acid homeostasis in mice with cystic fibrosis transmembrane

conductance regulator deficiencies. Gastroenterology 2012;142(7):1581-1591. e6.

CHAPTER 5

BILE SALT METABOLISM IN A CF

MICE MODEL WITH

SPONTANEOUS LIVER DISEASE

Submitted

Frank A.J.A. Bodewes1

Mariëtte Y.M. van der Wulp 1

Satti Beharry2

Rick Havinga1

Renze Boverhof1

M. James Phillips2

Peter R. Durie2

Henkjan J. Verkade1

Affiliations

1Department of Pediatric Gastroenterology and Hepatology, University Medical Center

Groningen, The Netherlands,

2 Department of Gastroenterology, Hepatology and Nutrition, The Hospital for Sick Children,

Toronto, Canada

Chapter 5

100

5

ABSTRACT

Introduction:

Long-living congenic C57BL/6J cystic fibrosis transmembrane regulator (Cftr)-/-

mice are

reported to spontaneously develop liver pathology in an age-dependent fashion. Alterations

in bile production and composition have been postulated to be involved in the development

of CFLD. To address this hypothesis, we determined bile production and composition in this

CF mouse model and controls.

Methods:

C57BL/6J Cftr-/-

mice and control littermates (N=10per genotype), mean age 74±25 days, were

fed a liquid diet (Peptamen Nestle) ad libitum from the time of weaning. We determined bile

production and composition after collection via gall bladder cannulation and sampled stools

for bile salt secretion and bile salt composition.

Results:

Cftr-/-

and control mice did not differ in bile production rate, biliary bile salt and phospholipid

concentration and secretion rate, and phospoholipid to bile salt ratio. Bile salt secretion– bile

flow graphs indicated that Cftr-/-

mice had a significantly higher bile salt dependent bile flow

compared with controls (Cftr+/+

, +130%, p=0.01). The biliary bile salt profile had a similar bile

hydrophobicity index in Cftr-/-

and control mice. However, the proportion of secondary bile

salts was significantly higher in Cftr-/-

mice than in controls (17±3 vs. 29±10%, resp.; p< 0.001).

In particular, the secondary, hydrophilic bile salt ursocholate was ~7 fold higher in Cftr-/-

mice

than in controls. (25% vs. 3%of total bile salts resp.; p=0.004).

Conclusion:

The reported development of liver disease in the C57BL/6J Cftr-/-

mice is not related to

alterations in bile production or biliary lipid composition. Cftr-/-

mice, on a liquid diet, have

increased amounts of secondary bile salts, which may be due to specific changes in bile salt

metabolism in the intestinal lumen of this CF mouse model.

Bile salt metabolism in a CF mice model with spontaneous liver disease

101

5

INTRODUCTION

Cystic fibrosis can be accompanied by liver disease (CFLD), which in ~10 % of all patients can

results in cirrhosis (1). Liver histology of these patients shows a severe biliary type of fibrosis

and cirrhosis (2).

To this day the pathogenesis and developmental pathways of CFLD remains obscure. It has

long been suggested that, in parallel to the suggested development of the obstructive CF lung

disease, local biliary obstruction by thickened, sticky bile formed the pathological basis for the

liver disease (3). However, this hypothesis is not well supported by experimental evidence.

Most information concerning the pathology of CFLD in humans has been derived from post

mortem evaluations (4). Prospective histological studies in humans are scarce, mainly because

of the need for invasive diagnostic procedures in asymptomatic patients (5).

Different CF mice models, with a wide variation in genetic background, are available (6). Most

CF mice models do not show liver pathology. Durie et al., however, described, age related,

histological abnormalities in the liver of C578L/6J Cftr−/−

mice (7). Additionally Freudenberg et

al. reported that older ΔF508 CF mice (100-200 days) had more liver fibrosis than their

controls (8).

Bile salts play an important role in intestinal lipid absorption and in biliary lipid excretion. Bile

salts can act as detergents and have cytotoxic properties. Bile salt composition and in

particular biliary bile hydrophobicity is related to development of liver disease in general (9).

The primary bile salts are synthesized in the liver and secreted into the intestine via the bile

ducts. In the intestine they serve mainly as detergents to solubilise water insoluble

components such as cholesterol and fats. In the intestine primary bile salts are subject to

biotransformation by the intestinal flora. In this way a variety of different secondary bile salts

are formed. All bile acids are absorbed in the terminal ileum and return to the liver, thus

undergoing enterohepatic circulation (10).

We studies the potential role of bile salt metabolism and biliary bile salt cytotoxicity in the

pathogenesis of the liver pathology the C578L/6J Cftr−/−

mice model. To address this question

we measured bile production and determined biliary and fecal bile salt excretion and

composition.

Chapter 5

102

5

METHODS

Animals. All experimental protocols were conducted in the Hospital for Sick children, the

Research Institute, Toronto, Canada, after approval by the institutional Animal Care

Committee. Mice were bred to wild-type C57BL/ 6J mice, obtained from the Jackson

Laboratories (Bar Harbor, ME). Mice were genotyped at 14 days of age using polymerase

chain reaction analysis of tail clip DNA. To minimize bowel obstruction and optimize long-

term viability, 20- to 23-day-old congenic C57BL/ 6J Cftr-/-tm1Unc

mice and their Cftr+/+

littermates were weaned to a liquid diet (Peptamen, Nestlé Nutrition, Canada) using glass

liquid mouse feeders, prepared in sterile water according to the manufacturer’s instructions.

Fresh diet and feeders, sterilized by autoclave, were replaced daily. Mice were housed in a

non-sterile conventional housing unit in micro-isolators cages, with corncob bedding changed

daily, and provided with sterile water in addition to the liquid diet. The colony was

maintained at a pathogen-free status by serological screening at a commercial laboratory.

Mice were kept in a 12-hour light-dark cycle. The animals used for the experiments were

approximately 1.5-4 months old

Experimental procedures. The total feces of 3 consecutive days was collected, dried and

homogenized. Bile was collected after surgical ligation of the common bile duct and gall

bladder cannulation using silicone tubing (size 0.020" x 0.037", Degania Silicone Ltd.)

under anesthesia. The anesthetic mixture consisted of: 0.75 ml ketamine (100 mg/ ml),

0.25 ml xylazine (20 mg/ ml) and 4 ml saline. The mixture (0.1 ml per 10 grams of body

weight) was intraperitoneal administered. Body temperature was maintained by placement of

the animal in a temperature and humidity controlled incubator. Bile secretions were collected

in 30 min. fractions, after a 10 equilibration period. Bile flow rate was assessed

gravimetrically, assuming that 1 g of secretion corresponds to 1 ml.

Analytical techniques. Biliary BS concentrations in bile and feces were determined by an

enzymatic fluorimetric assay (11). Lipids were extracted from the bile (12).The phospholipids

and cholesterol concentrations were determined using a spectrophotometric assay. Biliary BS

composition in bile and feces was determined by capillary gas chromatography (13). The

hydrophobicity of BS in bile was calculated according to the Heuman index based on the

fractional contribution of the mayor BS species (14).

Liver histology. Hemotoxylin and eosin (H&E) sections of the liver were assessed blindly by an

expert hepato-pathologist (JP) at The Hospital for Sick Children (Toronto, Canada) for the

degree of biliary duct obstruction and inflammation. We applied a previously published

numerical scoring system for each parameter, using an arbitrary 0 to 4 scale, in which zero

represents normal histology and four represents the most severe pathology for each

parameter (7). A minimum of five portal tracts was assessed for each mouse.

Bile salt metabolism in a CF mice model with spontaneous liver disease

103

5

RNA isolation from whole livers. Quantitative real time polymerase chain reactions (qRT-PCR,

combined for in vivo and in vitro experiments PCR was performed on a 7900HT Fast Real-Time

PCR system (Applied Biosystems).

Statistical analysis. Analyses were performed using SPSS version 18.0 for Windows (SPSS Inc.,

Chicago, IL). All nominal results are reported as means ± SEM. Differences between study

groups were evaluated using the Mann-Whitney U test. Bile flow was analyzed by correlation

and regression analysis. The categorized histology results were analyzed using cross tabs and

Fischer’s exact test. The level of significance was set at a P value of less than 0.05.

RESULTS

Body and liver weights. Figure 1 shows body weight and liver weight of CF and control mice.

C57BL/ 6J Cftr-/-

mice bodyweight was significantly lower compared to control littermates. The

liver weight of the Cftr-/-

was significantly less compared to controls, but the difference was

less pronounced than for the body weight. Accordingly, the relative liver weight compared to

the total bodyweight was significantly higher in the Cftr-/-

mice.

0

10

20

30

Cftr+/+

Cftr-/-

A*

Body w

eig

ht

(gra

ms)

0.0

0.5

1.0

1.5

Cftr+/+

Cftr-/-

B*

Liv

er

weig

ht

(gra

ms)

0.0

2.5

5.0

7.5

Cftr+/+

Cftr-/-

C*

Liv

er

to b

ody w

eig

ht ra

tio

Figure 1. Body and liver weight of Cftr+/+ and Cftr-/- mice. Body weight (A), absolute liver weight (B) and relative liver

weight reported as a percentage of the BW (C) in Cftr knockout mice (Cftr-/-) and control littermates (Cftr+/+). Data are

presented as means ± SEM of N=9-9 mice per group. *P-value<0.05.

Bile production and bile salt secretion. To determine if Cftr plays a role in the magnitude of

the bile production in C57BL/ 6J CF mice in vivo we measured bile production after

interruption of the enterohepatic circulation via gall bladder cannulation (Figure 2).

Interestingly, we found that the total bile production, was ~20-30% higher in Cftr-/-

mice

compared to control animals (P< 0.05 in the last time period, Figure 2A)

Chapter 5

104

5

Bile production is largely determined by the bile salt concentration and bile salt secretion

rate. The total bile biliary salt concentration was not different between Cftr-/-

versus control

mice (Figure 2B). The total biliary bile salt secretion rate was not significantly different

between Cftr-/-

and control mice (Figure 2C), but a consistent tendency towards higher values

was observed in the former. To evaluate the magnitude by which bile salts determine bile

production in CF conditions we related the bile flow to the bile salt secretion rate. (Figure 2D)

In this figure the BA-dependent flow (slope) and BA independent flow (intercept y-axis) can

be distinguished. Regression analysis showed a similar BA-independent bile flow between

both groups (P= 0.193). In contrast, the BA-dependent flow on the contrary, was higher in

Cftr-/-

vs. Cftr+/+

mice (P= 0.014).

30 min 60 min 90 min0

2

4

6

8

10

Cftr+/+

Cftr-/-

*

A

Bile

flo

w

(l.m

in-1

.100 g

ram

s B

W-1

)

30 min 60 min 90 min0

20

40

60

B

Cftr+/+

Cftr-/-

Bile

salt

concentr

atio

n

(mM

)

30 min 60 min 90 min0

100

200

300

Cftr+/+

Cftr-/-

C

Bile

salt s

ecr

etion r

ate

(nm

ol.m

in-1

.100 g

ram

BW

-1)

0 100 200 300 4000

2

4

6

8

10

Cftr+/+

Cftr-/-

D

Bile salt secreton rate

(nmol.min-1

.100 grams BW-1

)

Bile

flo

w

(l.m

in-1

.100 g

ram

s B

W-1

)

Figure 2. Biliary parameters of Cftr+/+ and Cftr-/- mice. Biliary bile flow (A), bile salt flow (B) bile salt secretion rate (C)

and bile secretion vs. bile flow (D) in Cftr knockout mice (Cftr-/-) and control littermates (Cftr+/+). In panel D there is a

significant correlation between BASR and bile flow in Cftr+/+ (Spearman’s Rho= 0.394, P=0.042), as well as in Cftr-/-

mice (Spearman’s Rho=0.762, P=0.000).Data are presented as means ± SEM of N=9-9 mice per group. *P-value<0.05.

Bile salt metabolism in a CF mice model with spontaneous liver disease

105

5

Biliary bile salt composition. Bile salt composition and in particular biliary bile hydrophobicity

is related to development of liver disease in general (9). To address the hypothesis that bile

salt hydrophobicity plays a role in CFLD we evaluated biliary and fecal bile salt composition.

Significant differences in bile salt composition were observed between the genotypes (Figure

3A). In Cftr-/-

compared to controls there were significant lowered biliary proportion of CA

(40.3±8.5 vs. 53.0±7.5% respectively; p<0.05), DCA (1.6±2.0 vs. 5.3±2.7% respectively; p<0.05)

and ω-MCA (0.9±0.5 vs. 6.1± 2.6% respectively; p<0.05) content in bile. Primary BA (produced

in the liver: CA, CDCA, α- and β-MCA) content was decreased in Cftr-/-

bile compared to Cftr+/+

(70% vs. 84% respectively; p<0.05) with a replacement by secondary BA (HDCA, UDCA, DCA,

ω-MCA). (Figure 3C) Gas chromatographic analysis of bile showed an unexpectedly, ~7-fold

higher, UCA content in Cftr-/-

mice compared to controls (24.6 vs. 3.3% respectively; p<0.01).

Mass spectrometry confirmed the identification and abundance of UCA in CF mice. As a result

of increased proportion of the hydrophilic UCA the biliary bile salt hydrophobicity index of

Cftr-/-

mice compared to Cftr+/+

mice was significantly lower vs. (-0.2.5±0.09 vs. -0.13±0.05

respectively; P<0.05; Figure 3C).

-MCA

DCA

CA

CDCA

HDCA

UCA

-MCA

-MCA

0

20

40

60

80

100

Cftr+/+

Cftr-/-

*

*

*

A

% o

f to

tal b

ile a

cid

s

Cftr+/+ Cftr-/-0

25

50

75

100

Primary bile salts

Secondary bile salts

B

%

-0.4

-0.2

-0.0

0.2

0.4

Cftr+/+

Cftr-/-

C

*H

ydro

phobic

ity in

dex

Figure 3. Biliary bile composition in bile of Cftr+/+ and Cftr-/- mice. BA profiles (A), data are presented as percentage of

different BA species present in bile. Abbreviations: CA = cholic acid, DCA = deoxycholic acid, CDCA = chenodeoxycholic

acid, HDCA = hyodeoxycholic acid, UDCA= ursodeoxycholic acid, UCA= ursocholic acid, MCA= muricholic acid. Relative

distribution between primary BA content and secondary BA (B). Primary BA content is decreased in Cftr-/- mice,

whereas secondary BA content is increased (P= 0.004). Heuman index (C) of total BS in bile representing the

hydrophobicity of bile salts. Data are presented as means ± SEM or percentage if appropriate of N=6-7 mice per

group. *P-value<0.05.

Fecal bile salt composition. Increased total fecal bile salt loss has been frequently observed in

CF conditions, including by ourselves. In the C57BL/ 6J CF mice model we found that fecal bile

salt loss was ~80% higher compared to control animals. (Figure 4A) Corresponding with the

biliary bile salt composition results, Cftr-/-

mice had a profoundly higher UCA content (27%) in

feces than the Cftr+/+

mice (1%, mass spectrometry analysis; Figure 4B) Cftr-/-

mice had a

Chapter 5

106

5

significantly lower DCA (11.4 ± 10.3 vs 42.0 ± 10.7% respectively; P<0.01), HDCA (0.3 ± 0.2 vs.

2.2 ± 0.9% respectively; P<0.01) and ω-MCA (1.2 ± 0.9 vs. 18.1 ± 8.3 respectively; P<0.01)

content in their feces. Cftr-/-

mice had significantly more CA in their feces, compared to

controls (26.5 ± 9.1% vs. 8.4 ± 7.4% respectively; P<0.01).

To address whether the observes differences were genotype or (microbial) environment

related, we performed fecal bile salt analysis in mice of the same CF and control genotypes

(N=5/5) from a different colony at another institution (courtesy of Dr R.C. De Lisle University

of Kansas, Kansas City, Missouri). The pattern and of the bile salt composition, in particular

the presence of UCA, was virtually identical, supporting a genotype related effect. (Data not

shown)

0

10

20

30

40

A

Cftr+/+

Cftr-/-

*

Fecal b

ile s

alt

excre

tion

(um

ol/1

00gB

W/d

ay)

-MCA

DCA

CA

CDCA

HDCA

UCA

-MCA

-MCA

0

20

40

60

80

100

Cftr+/+

Cftr-/-

**

*

B%

of to

tal b

ile a

cid

s

Figure 4. Fecal bile salt excretion and fecal bile salt composition of Cftr+/+ and Cftr-/- mice. Total bile salt excretion (A),

presented as means ± SEM. BA profiles (B), data are presented as percentage of different BA species present in bile.

Abbreviations: CA = cholic acid, DCA = deoxycholic acid, CDCA = chenodeoxycholic acid, HDCA = hyodeoxycholic acid,

UDCA= ursodeoxycholic acid, UCA= ursocholic acid, MCA= muricholic acid. N=9-9 mice per group. *P-value<0.05.

Bile salt metabolism in a CF mice model with spontaneous liver disease

107

5

30 min 60 min 90 min0

2

4

6

8

10

Cftr+/+

Cftr-/-

AP

ho

sp

ho

lip

s (

mM

)

0 100 200 300 4000

20

40

60Cftr+/+

Cftr-/-

B

Bile salt secretion rate

(nmol.min-1

.100 gram BW-1

)

PL s

ecre

tion r

ate

30 min 60 min 90 min0.00

0.05

0.10

0.15

0.20

Cftr+/+

Cftr-/-

C

Ph

os

po

ho

lip

id t

o b

ile

sa

lt r

ati

o

Figure 5. Phospholipid secretion in Cftr+/+ and Cftr-/- mice. PL secretion rate (a) and the PL secretion vs. BS secretion

rate (B) and PL to BS ratio in Cftr knockout mice (Cftr-/-) and control littermates (Cftr+/+). Data are presented as means

± SEM of N=9-9 mice per group.

Biliary lipids composition. Apart from bile salt hydrophobicity the biliary cytotoxicity and

detergent activity is determined by the biliary lipid composition (15). We found that the

phospholipid secretion was not significantly different between Cftr-/-

mice and controls

(Figure 5A). In both Cftr-/-

and controls the phospholipid secretion rate was linearly related to

the bile salt secretion rate, despite the significant differences in biliary bile salt

hydrophobicity. (Figure 5B).We found no difference in the biliary phospholipid to bile salt

ratio between Cftr-/-

and control mice (Figure 5C). Additionally we found no difference in

biliary cholesterol concentration and cholesterol secretion rate between Cftr-/-

and Cftr+/+

mice (data not shown).

Bile salt synthesis genes. Different bile salts have distinct influence in the induction of bile salt

synthesis. In particular DCA and CDCA are regarded as strong ligands for Fxr. We found a

significantly difference in CA and DCA proportions (Figure 6). This could imply difference in

bile synthesis activation between genotypes. To address the issue of potential difference we

determined the expression of different bile salt synthesis genes. (Figure 6) We found no

difference in expression of Fxr between Cftr-/-

and control mice, but the expression of the

orphan receptor Shp was significantly lower in the former. We did not observe significant

differences in the hepatic expression of Cyp 7a, encoding the rate limiting protein in the bile

salt synthesis process.

Chapter 5

108

5

*

FxrShp

Cyp

7a1

Cyp

8b1

BSEP

0

1

2

3

Cftr+/+

Cftr-/-

Rela

tive m

RN

A le

vel

norm

aliz

ed to b

eta

-actin

Figure 6. Gene expression in isolated livers in Cftr+/+ and Cftr-/- mice. mRNA levels were normalized to a housekeeping

gene (β-actin) in Cftr knockout mice (Cftr-/-) and control littermates (Cftr+/+). Data are presented as means ± SEM of

N=9-9 mice per group. *P-value<0.05.

Histology. Durie et al described in their initial report concerning liver pathology in Unc mice

that all Cftr-/-

animals showed focal and progressive hepatobiliary disease (7). The mean age

of the mice in the current experiment was 75±25 days in both the Cftr-/-

and control group.

Although, in general the current histo-pathological findings are consistent with the previous

report they are less pronounced and the global duct proliferation score, in this cohort,

revealed no significant difference between Cftr-/-

and control mice. (Figure 7)

0

1

2

3

4Cftr-/-

Cftr+/+

Port

al i

nfla

matio

n s

core

0

1

2

3

4Cftr-/-

Cftr+/+

Ducta

l change s

core

0

1

2

3

4Cftr-/-

Cftr+/+

NS (P=0.09)

Bridgin

g fib

rosis

score

Figure 7. Histological bile duct proliferation score. Bile duct proliferation score base on the observation of 10 portal

tracts per mouse graded: normal: 1 point, mild proliferation: 2 points, severe proliferation: 3 points. The bile duct

proliferation represents the mean of al scored points per mouse. Data are presented as means ± SEM of N=10-10 mice

per group.

DISCUSSION

We evaluated bile salt metabolism and bile salt composition and in the C57BL/ 6J CF mice

model that has been reported to spontaneously develop CFLD like liver disease (7). We

Bile salt metabolism in a CF mice model with spontaneous liver disease

109

5

hypothesized that this Cftr-/-

mice develop liver disease in the context of a cytotoxic biliary bile

salt profile. However in contrast of what could be expected we found that in Cftr-/-

mice on a

liquid diet, the biliary bile salt profile is substantially more hydrophilic and the bile flow is

increased compared to control mice. This increase in bile production is explained by the

increased presence of the secondary hydrophilic and highly choleretic bile salt ursocholate in

this mice model. Since UCA metabolism is converted form cholate by intestinal microbiota our

results emphasize the critical role of intestinal microflora in modifying the enterohepatic

circulation of bile salt in CF conditions.

Based on hydrophobicity and Pl content we did not find an increased cytotoxic profile of the

biliary bile of Cftr-/-

mice that could explain the reported spontaneous development of biliary

liver disease in this model. Others have suggested different potential mechanism for the

development of liver disease in the CF mouse models. Blanco et al. reported that chemical

induction of colitis, by dextran, in Cftr-/-

Unc-mice results in a increase in bile duct injury (16).

In an additional report from the same group they describe that decreased peroxisome

proliferator activated receptor alpha is associated with bile duct Injury in Cftr-/-

mice report

(17). These findings point in the direction of a Cftr related difference in immune response to

development of liver disease in CF mice. Beharry et al. reported a highly significant protective

effect of long-term docosahexaenoic acid therapy on the progression of CF-like hepatobiliary

disease in C57BL/ 6J CF Cftr-/-

mice (18). They suggested that the inhibition of inflammation by

DHA may account for the anti-inflammatory effects in the liver. The latter report again

endorsing a potentially immunologic/inflammatory genesis of CF related liver disease.

UCA, the 7α-hydroxy epimer of cholic acid, is a natural bile acid characterized by sixfold and

threefold greater hydrophilicity than chenodeoxycholic and ursodeoxycholic acid, respectively

(19). The contribution of UCA to the bile salt pool is usually rather low, due to rapid

biotransformation of UCA during enterohepatic circulation into DCA by intestinal micoflora. In

our Cftr-/-

mice, however, DCA contribution was significantly lower than in control mice,

suggesting a decreased biotransformation of UCA into DCA in the CF intestine. Due to it high

choleretic potency the enrichment of UCA in our model leads to a significant increased bile

salt dependent flow and low hydrophobicity index in Cftr-/-

mice.

It cannot be excluded that the exclusively liquid feeding could have influenced biliary bile salt

composition in our model. The C57BL/ 6J CF Cftr-/-

mice are dependent on liquid feeding after

weaning to prevent lethal intestinal obstruction (20). Therefore Peptamen liquid feeding is

used. Peptamen is a commercial infant formula use for instance for patients with short bowel

syndrome, intractable malabsorption, patients with proven inflammatory bowel disease and

bowel fistulae. Peptamen is a semi-elementary formula with a high proportion of protein and

medium chain triglycerides. Different from regular chow, Peptamen does not contain dietary

fibre or grain products. Since Cftr-/-

mice and controls were both fed the liquid Peptamen, diet

effects cannot explain the differences in bile salt composition between the genotypes.

Chapter 5

110

5

In our model we found a distinct increased fecal bile salt excretion. Increased fecal bile salt

loss has reported before in CF patients and animals models (21, 22). It has been suggested

that the underlying mechanism involves increased fecal fat excretion in CF conditions (23),

but previously we reported that increased fecal bile salt loss was also observed in CF mice

without fat malabsorption. Increased hepatic bile salt synthesis compensates for the higher

bile salt loss in Cftr-/-

mice (24). However, in the current study, the proportion of primary bile

salt in the biliary bile was actually lower in Cftr-/-

mice, acutely indicating a decreased

synthesis of bile salt in these CF mice.

Several publications in recent years have revealed the complex regulatory interaction of bile

salt metabolism between intestine and liver (25, 26). Bile salt synthesis in rodents is regulated

in the intestine via the Fxr-Fgf15 axes as well as more directly in the hepatocyte, via the Fxr-

Shp axis. Both routes eventually target the expression of Cyp7a, the rate limiting enzyme in

the bile salt synthesis. UCA is reported to have no or limited influence on bile salt synthesis, in

contrast to the hydrophobic DCA, which is one of the most potent FXR ligands to suppress bile

salt synthesis (27). The reduction of intestinal transformation of UCA to DCA may have lead to

a reduced bile salt synthesis as represented by the reduction in the proportion of primary

biliary bile salt. However this finding did not correspond to the Cyp7a expression levels that

were not different in Cftr-/-

mice compared to controls. On the other hand the reported

reduction in Shp expression would imply a partial down regulation of bile salt synthesis could

be induced via the alternative intestinal Fxr-Fgf15 pathway.

We found that the reported biliary disease in the Unc Cftr-/-

mice is not related to increased

biliary cytotoxicity. However Cftr-/-

mice shows distinctive changes in the metabolism and

enterohepatic circulation of the bile salts. Recent discoveries in the intestinal regulation of

bile salt synthesis shed a new light on the importance and role of the intestinal bile salt

composition. The reported difference in bile salt metabolism could be origin of the intestinal

bile salt loss and in CF condition. However additional research is needed to further determine

the role of the enterohepatic circulation in CF.

REFERENCE LIST

1. Scottjupp R, Lama M, Tanner MS. Prevalence of liver-disease in cystic-fibrosis. Arch Dis

Child. 1991 06;66(6):698-701.

2. Hultcrantz R, Mengarelli S, Strandvik B. Morphological findings in the liver of children with

cystic fibrosis: A light and electron microscopical study. Hepatology. 1986 09;6(5):881-9.

3. Sinaasappel M. Hepatobiliary pathology in patients with cystic fibrosis. Acta Paediatr Scand.

1988;363:45,50; discussion 50-1.

Bile salt metabolism in a CF mice model with spontaneous liver disease

111

5

4. Shier KJ, Horn Jr RC. The pathology of liver cirrhosis in patients with cystic fibrosis of the

pancreas. Can Med Assoc J. 1963;89(13):645.

5. Lewindon PJ, Shepherd RW, Walsh MJ, Greer RM, Williamson R, Pereira TN, et al.

Importance of hepatic fibrosis in cystic fibrosis and the predictive value of liver biopsy.

Hepatology. 2011;53(1):193-201.

6. Wilke M, Buijs-Offerman RM, Aarbiou J, Colledge WH, Sheppard DN, Touqui L, et al. Mouse

models of cystic fibrosis: Phenotypic analysis and research applications. J Cyst Fibros.

2011;10:S152-71.

7. Durie PR, Kent G, Phillips MJ, Ackerley CA. Characteristic multiorgan pathology of cystic

fibrosis in a long-living cystic fibrosis transmembrane regulator knockout murine model. Am J

Pathol. 2004 04;164(4):1481-93.

8. Freudenberg F, Broderick AL, Yu BB, Leonard MR, Glickman JN, Carey MC.

Pathophysiological basis of liver disease in cystic fibrosis employing a DeltaF508 mouse model.

Am J Physiol Gastrointest Liver Physiol. 2008 06;294(6):G1411-20.

9. Attili AF, Angelico M, Cantafora A, Alvaro D, Capocaccia L. Bile acid-induced liver toxicity:

Relation to the hydrophobic-hydrophilic balance of bile acids. Med Hypotheses. 1986

1;19(1):57-69.

10. Hofmann AF. Enterohepatic circulation of bile acids. In: Comprehensive Physiology. John

Wiley & Sons, Inc.; 2010.

11. Mashige F, Imai K, Osuga T. A simple and sensitive assay of total serum bile acids. Clin

Chim Acta. 1976 07/01;70(1):79-86.

12. Bligh EG, Dyer WJ. A rapid method of total lipid extraction and purification. Am J Physiol

Gastrointest Liver Physiol. 1959;37(8):911-7.

13. Kok T, Hulzebos CV, Wolters H, Havinga R, Agellon LB, Stellaard F, et al. Enterohepatic

circulation of bile salts in FXR-deficient mice: Efficient intestinal bile salt absorption in the

absence of ileal bile acid-binding protein (ibabp). J Biol Chem. 2003;278(43):41930-7.

14. Heuman D, Hylemon P, Vlahcevic Z. Regulation of bile acid synthesis. III. correlation

between biliary bile salt hydrophobicity index and the activities of enzymes regulating

cholesterol and bile acid synthesis in the rat. J Lipid Res. 1989;30(8):1161-71.

15. Loria P, Carulli N, Medici G, Menozzi D, Salvioli G, Bertolotti M, et al. Effect of ursocholic

acid on bile lipid secretion and composition. Gastroenterology. 1986 Apr;90(4):865-74.

16. Blanco PG, Zaman MM, Junaidi O, Sheth S, Yantiss RK, Nasser IA, et al. Induction of colitis

in cftr–/–mice results in bile duct injury. American Journal of Physiology-Gastrointestinal and

Liver Physiology. 2004;287(2):G491.

Chapter 5

112

5

17. Pall H, Zaman MM, Andersson C, Freedman SD. Decreased peroxisome proliferator

activated receptor [alpha] is associated with bile duct injury in cystic fibrosis transmembrane

conductance regulator-/-mice. J Pediatr Gastroenterol Nutr. 2006;42(3):275.

18. Beharry S, Ackerley C, Corey M, Kent G, Heng YM, Christensen H, et al. Long-term

docosahexaenoic acid therapy in a congenic murine model of cystic fibrosis. American Journal

of Physiology-Gastrointestinal and Liver Physiology. 2007;292(3):G839-48.

19. Roda A, Hofmann AF, Mysels KJ. The influence of bile salt structure on self-association in

aqueous solutions. Journal of Biological Chemistry. 1983 May 25;258(10):6362-70.

20. Snouwaert JN. An animal model for cystic fibrosis made by gene targeting. Science.

1992;257(5073):1083.

21. O'Brien S, Mulcahy H, Fenlon H, O'Broin A, Casey M, Burke A, et al. Intestinal bile acid

malabsorption in cystic fibrosis. Gut. 1993;34(8):1137.

22. Bijvelds MJC, Jorna H, Verkade HJ, Bot AGM, Hofmann F, Agellon LB, et al. Activation of

CFTR by ASBT-mediated bile salt absorption. American Journal of Physiology-Gastrointestinal

and Liver Physiology. 2005;289(5):G870.

23. Weber AM. Relationship between bile acid malabsorption and pancreatic insufficiency in

cystic fibrosis. Gut. 1976;17(4):295.

24. Bijvelds MJC, Bronsveld I, Havinga R, Sinaasappel M, de Jonge HR, Verkade HJ. Fat

absorption in cystic fibrosis mice is impeded by defective lipolysis and post-lipolytic events. Am

J Physiol Gastrointest Liver Physiol. 2005 04;288(4):G646-53.

25. Inagaki T, Moschetta A, Lee YK, Peng L, Zhao G, Downes M, et al. Regulation of

antibacterial defense in the small intestine by the nuclear bile acid receptor. Proc Natl Acad Sci

U S A. 2006;103(10):3920.

26. Lundåsen T, Gälman C, Angelin B, Rudling M. Circulating intestinal fibroblast growth factor

19 has a pronounced diurnal variation and modulates hepatic bile acid synthesis in man. J

Intern Med. 2006;260(6):530-6.

27. Heuman DM, Hylemon PB, Vlahcevic ZR. Regulation of bile acid synthesis. III. correlation

between biliary bile salt hydrophobicity index and the activities of enzymes regulating

cholesterol and bile acid synthesis in the rat. Journal of Lipid Research. 1989 August

01;30(8):1161-71.

CHAPTER 6

URSODEOXYCHOLATE MODULATES

BILE FLOW AND BILE SALT POOL

INDEPENDENTLY FROM CFTR IN

MICE Based on: American Journal of Physiology-Gastrointestinal and Liver Physiology302.9 (2012):

G1035-G1042.

Frank A.J.A. Bodewes1

Marjan Wouthuyzen-Bakker1

Marcel J. Bijvelds2

Rick Havinga1

Hugo R. de Jonge2

Henkjan J. Verkade1

1 Department of Pediatrics, University of Groningen, Beatrix Children’s Hospital - University

Medical Center, Groningen, The Netherlands

2 Department of Gastroenterology & Hepatology, Erasmus University Medical Center,

Rotterdam, The Netherlands

Chapter 6

114

6

ABSTRACT

Introduction

Cystic fibrosis liver disease (CFLD) is treated with ursodeoxycholate (UDCA). Our aim was to

evaluate, in Cftr-/- mice and wild type controls, if the supposed therapeutic action of UDCA is

mediated via choleretic activity or effects on bile salt metabolism.

Methods

Cftr-/- mice and controls, under general anesthesia, were IV infused with TUDCA in increasing

dosage or were fed either standard or UDCA enriched chow (0.5%wt/wt) for 3 weeks. Bile

flow and bile composition were characterized. In chow fed mice, we analyzed bile salt

synthesis and pool size of cholate (CA).

Results

In both Cftr-/- and controls IV TUDCA stimulated bile flow by ~250% and dietary UDCA by

~500%, compared with untreated animals (p<0.05). In non-UDCA treated Cftr-/- mice, the

proportion of CA in bile was higher compared to controls (61±4% vs. 46±4%, resp.; p<0.05),

accompanied by an increased CA synthesis (16±1 vs. 10±2 μmol/hour/100gramBW resp.;

p<0.05) and CA pool size (28±3 vs. 19±1 μmol/100gramBW, resp.; p<0.05). In both Cftr-/- and

controls UDCA treatment drastically reduced the proportion of CA in bile below 5% and

diminished CA synthesis (2.3±0.3 vs. 2.2±0.4 μmol/day/100gramBW, resp.; NS) and CA pool

size (3.6±0.6 vs. 1.5±0.3 µmol/100gramsBW, resp.; p<0.05).

Conclusion

Acute TUDCA infusion and chronic UDCA treatment both stimulate bile flow in CF conditions

independently from Cftr function. Chronic UDCA treatment reduces the hydrophobicity of the

bile salt pool in Cftr-/- mice. These results support a potential beneficial effect of UDCA on

bile flow and bile salt metabolism in CF conditions.

Ursodeoxycholate modulates bile flow and bile salt pool independently from Cftr in mice

115

6

INTRODUCTION

Cystic Fibrosis (CF) is caused by mutations in the Cystic Fibrosis Transmembrane conductance

Regulator (Cftr) gene which result in dysfunction of the CFTR protein (1, 2). CFTR is a cAMP

activated chloride channel and is present in the apical membrane of various epithelial and

non-epithelial tissues (3-5). Cirrhotic Cystic Fibrosis Liver Disease (CFLD) develops in ~5-10% of

CF patients and is the second leading cause of mortality in CF patients (6-8). Absence of

functional CFTR in biliary epithelium is thought to initiate abnormal secretin/cAMP-stimulated

chloride and bicarbonate secretion, leading to decreased bile flow and bile duct plugging by

thickened secretions. Secondary cholangiocyte and hepatocyte injury can ultimately lead to

the development of cirrhosis (9).

Treatment with ursodeoxycholate (UDCA) is applied for different cholangiopathies, including

CFLD. Although routinely applied in CF patients with increased levels of liver function

parameters in serum, the therapeutic action of UDCA in CF conditions remains unclear. It has

been hypothesized that UDCA is beneficial by its choleretic activity and/or its capacity to

correct aberrant bile salt metabolism (10). Indeed, it has been shown that UDCA stimulates

biliary secretion of bile acids in patients with primary biliary cirrhosis and primary sclerosing

cholangitis (11, 12). However, there are only a few trials assessing the efficacy of UDCA for

the treatment of CFLD and currently, there is insufficient evidence to justify the routine use of

UDCA in CF (13).

Shimokura et al. demonstrated in biliary cells that UDCA, in pharmacological concentrations,

increased intracellular calcium and induced chloride efflux. These scientists speculated that

UDCA increased bile flow by direct stimulation of ductular secretion, what could be of

therapeutic benefit for patients with CF who have impaired cyclic AMP-dependent biliary

secretion (14). However, Fiorotto et al. suggested that UDCA induced bile flow in a Cftr

dependent manner (15). They showed that UDCA stimulated cholangiocytic fluid secretion in

vitro in bile duct units and in isolated perfused livers from wild type, but not from Cftr

knockout mice. UDCA induced a Cftr-dependent ATP release, which, by activating the

purinergic signaling pathway, induced cholangiocyte secretion by stimulation of calcium-

activated chloride channels. These in vitro and ex vivo observations clearly supported the

concept that UDCA-induced fluid secretion is Cftr dependent and that, in CF conditions, UDCA

induced cholangiocytic choleresis can’t be achieved.

An alternative beneficial effect of UDCA in CF conditions could be related to changes in bile

salt metabolism. It is known, that the size and composition of the bile salt pool is critical for

adequate bile formation(16). In addition, a more hydrophobic bile salt composition is

suggested to be toxic to the hepatobiliary tract and might contribute to the development of

CFLD (17). Freudenberg et. al reported an increased biliary hydrophobicity index in Cftr508/508

mice compared to controls. Different effects are published on the influence of UDCA on bile

salt metabolism. Frenkiel et al. reported that UDCA reduced the cholate pool size and cholate

Chapter 6

116

6

synthesis rate in patients with gallstones and decreased the hydrophobicity of bile (18). In

contrast, Beuers et al. described that UDCA did not decrease the bile salt pool size in patients

with cholestatic liver disease (19). In UDCA treated patients with CFLD, duodenal bile became

enriched with the conjugated species of UDCA (accounting for 12% to 3% of the total biliary

bile salts), indicating an increased hydrophilic bile composition (20). Thus far, the effect of

UDCA on bile salt metabolism under CF conditions has not been conclusive.

We have reasoned that insights in the effects of UDCA on bile production, cholate

biosynthesis rate and cholate pool size under controlled in vivo CF conditions have been

lacking. In the present study we aimed to overcome this knowledge gap by exploiting recently

developed techniques allowing the study of these parameters in small experimental animals

(21, 22). We determined the effect of UDCA on bile flow and bile salt metabolism in Cftr

knockout mice and wild type littermates during acute intravenous tauroursodeoxycholate

(TUDCA) infusion and after chronic dietary administration of UDCA. Although biliary

phenotypes have been described in different CF mouse models the C57Bl/6;129 Cftr-/- tm1CAM

mouse model we use does not exhibit CF related gallbladder or liver abnormalities (23-25).

The lack of hepatobiliary abnormalities allows us to exclusively investigate the action UDCA

under complete Cftr null conditions, without the secondary interference of cholestasis on

biliary parameters.

MATERIAL AND METHODS

Animals and diets. C57Bl/6;129 Cftr-/- tm1CAM

mice and Cftr+/+tm1CAM

littermates were bred and

accommodated at the Animal Experimental Center of the Erasmus Medical Center in

Rotterdam, The Netherlands. Southern blotting of tail-clip DNA was performed to verify the

genotype of individual animals (25). In accordance with previous studies, the Cftr-/-

mice did

not exhibit CF liver or gallbladder abnormalities (data not shown). Mice were housed in a

light-controlled (lights on 6 AM to 6 PM) and temperature-controlled (21oC) facility, and were

allowed access to tap water and a semi-synthetic chow diet (SRM-A; Hope Farms BV

Woerden, The Netherlands) from the time of weaning. All experiments were performed on

female and male animals of 10-20 weeks of age. Group size varied per experiment from 5-9

animals per genotype. Experimental protocols were approved by the Ethical Committee for

Animal Experiments of the Erasmus Medical Center.

Experimental procedures. To evaluate the effect of UDCA on bile production and bile

composition, we used a mouse bile duct cannulation experimental model. This model

provides the option to measure bile production over an extended period of time. The animals

were placed in a temperature and humidity controlled incubator. Bile was collected after

surgical ligation of the common bile duct and cannulation of the gallbladder using

polyethylene tubing under intraperitoneal anesthesia with hypnorm (fentanyl/fluanisone 1

µl/gBW) and diazepam (10 µg/gram BW), as previously described (26). Bile secretions were

Ursodeoxycholate modulates bile flow and bile salt pool independently from Cftr in mice

117

6

collected in 15 minute fractions during the stepwise dose increase phase and in 10 minutes

fraction at the highest dose (600 nmol.min-1

) for 60 minutes. Bile samples were used for bile

salt composition analysis. Bile flow rate was assessed gravimetrically, assuming a density of

1g/ml.

Acute intravenous TUDCA administration. We performed an acute bile salt infusion

experiment to evaluate the dose-response effect of acute UDCA supplementation on bile flow

and bile salt composition in Cftr-/-

and control mice. Acute TUDCA administration provides the

possibility to measure the direct choleretic effects of bile salts without the possible interfering

effects of adaptations to long-term bile salt administration. We infused taurine-conjugated

UDCA (TUDCA) for our acute infusion experiment, to closely mimic the physiological condition

in which bile salts are secreted into bile almost exclusively in conjugated form. An intravenous

line was placed in the jugular vein and the gallbladder was cannulated. After equilibration of

the bile flow for 5-10 minutes, spontaneous bile production was assessed for 30 minutes, i.e.

without TUDCA infusion. Subsequently, TUDCA solution (43 mM dissolved in phosphate-

buffered saline, pH 7.4) was administered using an IV pump(27). The TUDCA dosage was

increased every 30 minutes in a stepwise manner (dosage steps 150, 300, 450 and 600

nmol/min). The maximal dosage was given for 60 minutes. During TUDCA infusion, bile was

collected in 15 minute- fractions.

Chronic dietary UDCA administration. In CF patients, UDCA supplementation is given

chronically via the enteral route (28). Analogous to the human situation, we evaluated the

effect of chronic enteral (dietary) UDCA treatment in Cftr-/-

and control mice compared to

untreated animals. Mice were fed either a control diet consisting of standard chow or the

same diet enriched with UDCA (0.5% wt/wt) for 3 weeks. Body weight was measured after the

diet period. After gallbladder cannulation, spontaneous bile production was determined by

bile collection for 30 minutes. We additionally determined cholate synthesis and pool size,

using a previously developed and validated stable isotope dilution technique (22). In short,

3.0 mg of [2H4]-cholate in a solution of 0.5% NaHCO3 in phosphate-buffered saline was slowly

injected via the penile vein during isoflurane anesthesia. Blood samples were taken before

injection and at 12, 24, 36, 48, 60, and 72 h after injection. Blood samples (100 µl) were

collected by tail bleeding. Blood was collected in EDTA tubes and centrifuged to obtain

plasma. After centrifuging (3,000 rpm for 10 min at 4°C), plasma was stored at –20°C until

analysis. At the last day of the experiment (72 h), mice were anesthetized and equipped with

a catheter in the bile duct as described above. Subsequently, bile was collected for 30 min,

after an initial equilibration period of 5-10 min. Animals were euthanized by heart puncture.

Analytic procedures. Biliary bile salt concentrations were determined by an enzymatic

fluorimetric assay (29). Biliary bile salt composition was determined by capillary gas

chromatography (30). The hydrophobicity of bile salts in bile was calculated according to the

Heuman index based on the fractional contribution of the major murine bile salt species

cholate, chenodeoxycholate, deoxycholate, ursodeoxcholate, α-muricholate and β-

Chapter 6

118

6

muricholate (31). Alanine transaminase (ALT) was determined in plasma samples. Plasma and

bile samples were prepared for GLC-MS analysis on a Finnigan SSQ7000 Quadrupole GC-MS

machine as described previously by Stellaard et al. (32). The isotope dilution technique is

based on the dilution of a labeled tracer into the pool of the metabolite of interest. It has

been demonstrated to result in virtual identical cholate synthesis rates as obtained with a 14

C-

cholesterol bile salt synthesis measuring methods (33). The tracer is administered as a bolus,

which mixes into the pool. Shortly after mixing, the isotopic enrichment is highest. Thereafter,

the enrichment decreases due to dilution with unlabelled molecules introduced by de novo

synthesis. Enrichment of 2H4-cholate in plasma was determined as the increase of the M4-/M0-

cholate, relative to baseline measurements and was expressed as the natural logarithm of

atom percent excess (ln APE). From the decay curve of ln APE (calculated by linear regression

analysis), daily fractional turnover rate (FTR; equals the slope of the regression line) and pool

size ([administered amount of label x isotopic purity x 100] / intercept of the y-axis of the ln

APE curve) of cholate were calculated. Multiplying the pool size with the FTR results in a value

for the absolute turnover rate. In the steady-state situation, the absolute turnover rate equals

the synthesis rate. (22, 33, 34).

Statistical analysis. Statistical analysis was performed using SPSS version 16.0 for Windows

(SPSS Inc., Chicago, IL). All results are reported as means±SEM. Differences between

genotypes or diet groups were evaluated using the Mann-Whitney U test. The level of

significance was set at a P value of less than 0.05.

Ursodeoxycholate modulates bile flow and bile salt pool independently from Cftr in mice

119

6

RESULTS

The physiological state of the enterohepatic circulation in Cftr-/-

mice. We evaluated bile

production during the first 30 min after acute interruption of the enterohepatic circulation,

i.e. closely reflecting the physiological state (represented by time points 15 and 30 minutes in

Figure 1A-C). We found that bile flow (4.5±0.6 vs. 4.0±0.4 μl/min/100gram BW, NS), biliary

bile salt concentration (67±12 vs. 54±5 mM, NS) and biliary bile salt output (313±72 vs.

208±25 nmol/min/100gram BW, NS) were similar in Cftr-/-

and control mice respectively,

indicating no quantitative differences in te choleretic capacity between Cftr-/-

mice and their

controls at baseline.

0

5

10

15

Cftr+/+

ACftr-/-

150300

450

30

Time (minutes)

60 90 120 180

0

600 TUDCA infusion rate

(nmol/min)

Bile f

low

( l/

min

/100 g

ram

BW

)

0

500

1000

1500

2000

2500

Cftr+/+

CCftr

-/-

150300

450

30

Time (minutes)

60 90 120 180

600 TUDCA infusion rate

(nmol/min)

Bil

e s

alt

secre

tio

n r

ate

(nm

ol/

min

/100 g

ram

BW

)

0 500 1000 1500 2000 25000

5

10

15

D Cftr+/+

Cftr-/-

Bile salt secretion rate(nmol/min/100 gram BW)

Bil

e f

low

( l/

min

/100 g

ram

BW

)

0

50

100

150

200

Cftr+/+

BCftr-/-

150300

450

30

Time (minutes)

60 90 120 180

600 TUDCA infusion rate

(nmol/min)

Bil

e s

alt

co

ncen

trati

on

(mM

)

Figure 1 Biliary parameters during acute intravenous TUDCA administration. Biliary bile flow (A), bile salt

concentration (B), bile salt secretion rate (C) and relationship between bile salt secretion rate and bile flow (D) in Cftr

knockout mice (Cftr-/-) and control littermates (Cftr+/+) during acute intravenous infusion with TUDCA. The dosage of

TUDCA was increased at indicated intervals. The grey symbols in Fig. D represent baseline values without TUDCA

infusion. Data are presented as means ± SEM of N=5 mice per group. There was no significant difference between Cftr-

/- and Cftr+/+ mice, at any of the individual time points, for bile flow, bile salt concentration and bile salt secretion rate.

TUDCA administration increased bile flow equally in Cftr-/-

and control mice in a dose-

dependent manner. Infusions with TUDCA increased bile flow to a similar extend in Cftr-/-

mice

Chapter 6

120

6

and controls (+ ~250%; Figure 1A). The results indicate that TUDCA is capable of generating a

bile salt induced bile flow independent of the expression of CFTR. The biliary bile salt

concentration increased in parallel to the infused bile salt dosage (Fig. 1B). The bile salt

secretion rates, i.e. the product of flow (Figure 1A) and concentration (Figure 1B), increased in

equal pace with an increased TUDCA IV dosage and was not different between Cftr-/-

and

control mice (Figure 1C). We performed Mann–Whitney U tests for all individual time point

during the TUDCA infusion experiments. There was no significant difference between Cftr-/-

and Cftr+/+

mice at any of the individual time points for bile flow, bile salt concentration and

bile salt secretion rate. To determine the choleretic capacity of TUDCA in both Cftr-/-

and

control mice we related the bile flow to the bile salt secretion rate (Figure 1D) The choleretic

capacity of TUDCA was similar in Cftr-/-

and control mice, as was the estimated bile salt

dependent bile flow, based on the Y-intercepts (4.6 vs. 5.3 μl/min/100gram BW, respectively,

NS). The increase in bile flow was linearly related to the bile salt output (control: R2 0.9, slope

0.003 µmol/nmol; Cftr-/-

R2 0.9, slope 0.003 µmol/nmol).

Normal diet UCDA diet0

10

20

30

40

50

Cftr+/+

Cftr-/-

A

Bil

e f

low

( l/

min

/100 g

ram

BW

)

Normal diet UCDA diet0

25

50

75

100

Cftr+/+

Cftr-/-

*

BB

A c

on

cen

trati

on

(mM

)

Normal diet UCDA diet0

500

1000

1500

2000

Cftr+/+

Cftr-/-

*

C

BS

SR

(nM

/min

/100 g

ram

s B

W)

Normal diet UCDA diet0

25

50

Cftr+/+

Cftr-/-

*

D

Bo

dy w

eig

ht

(gra

m)

Figure 2 Biliary parameters after chronic dietary UDCA administration. Biliary bile flow (A), bile salt secretion (B) bile

salt secretion rate (C) body weight (D) in Cftr knockout mice (Cftr-/-) and control littermates (Cftr+/+) after a normal or

0.5%-UDCA chow diet for 3 weeks. Data are presented as means ± SEM of N=5-6 mice per group. *P-value<0.05.

UDCA treatment increased bile flow equally in Cftr-/-

and control mice. Chronic UDCA

treatment for 3 weeks increased bile flow with ~500% when compared to untreated mice

(p<0.05). This increase was comparable for Cftr-/-

and control mice (29.0±2.6 vs. 31.0±1.9

Ursodeoxycholate modulates bile flow and bile salt pool independently from Cftr in mice

121

6

μl/min/100gram BW, resp.; NS; Figure 2A). Chronic UDCA treatment decreased the total

biliary bile salt concentration in both genotypes, but to a lower extent in Cftr-/-

mice (Cftr+/+

-

51% vs. Cftr-/-

mice -37%), resulting in significantly higher BS concentration in Cftr-/-

mice

compared to controls (42±3 vs. 26±3 mM, resp., p<0.05, Figure 2B). The bile salt output was

significantly higher in Cftr-/-

mice compared with controls during chronic UDCA treatment

(Cftr-/-

, 1232±147 vs. control, 827±113 μmol/min/100gram BW, resp.; p<0.05, Figure 2C). A

significant reduced growth of the Cambridge Cftr-/-

mice compared to wild type animals has

been reported (25). Since the findings of the differences in the bile composition on treatment

with UDCA may be affected by the nutritional status we measured body weight of the treated

and untreated mice (Figure 2D). In our current experiment the phenotype of the Cftr-/-

mice

includes a decreased body weight compared to controls, however this is not affected by

UDCA treatment.

Increased hydrophobic bile salt composition of Cftr-/-

mice compared to control mice. In non-

UDCA treated Cftr-/-

mice, the fractional biliary cholate content was higher compared with

control mice (61.4±1.5% vs. 46.5±3.8%, resp.; p<0.05; Table 1.) The natural biliary UDCA

enrichment was ~50% lower in Cftr-/-

mice compared with controls (2.7±0.4 vs. 6.0±0.5% resp.;

p<0.05; Fig. 2A). In non UDCA treated mice β-muricholate is the major hydrophilic biliary bile

salt in both Cftr-/-

and control mice (42±3,2 vs. 33±1,6% resp.; NS). Based on the fractional

contribution of the bile salt we calculated the biliary hydrophobicity index, according to

Heuman et al (14). The bile salt composition of non-UDCA treated Cftr-/-

mice was significantly

more hydrophobic than the control mice (0.05±0.002 vs. -0.07±0.005 resp.; p<0.01; Figure

3B).

Chapter 6

122

6

Biliary bile salt profile: proportional contribution of the major murine

bile salt species

Non-UDCA treatment UDCA treatment

Cftr+/+

Cftr-/-

Cftr+/+

Cftr-/-

CA 46.5 ± 3.8 61.5 ± 1.5* 2.7 ± 0.9† 4.0 ± 1.3

†

CDC 3.4 ± 0.4 2.1 ± 0.3 2.8 ± 0.5 1.3 ± 0.2*

DC 1.7 ± 0.4 0.7 ± 0.4 ND ND

UDCA 6.0 ± 0.5 2.7 ± 0.4* 83.0 ± 2.7† 82.0 ± 1.9

†

α-M ND ND 3.3 ± 0.3† 2.5 ± 0.3

†

β-M 42.0 ± 3.2 33.0 ± 1.6 6.5 ± 1.8† 8.7 ± 1.6

†

Table 1. Values are means ±SE in percent; N 4–6 mice per group. Biliary bile salt composition is shown of the mayor

bile salt species [cholate (CA), chenodeoxycholate (CDC), deoxycholate (DC), ursodeoxycholate (UDCA),α-muricholate

(α-M), and β-muricholate (β-M)] in cystic fibrosis transmembrane regulator (Cftr) knockout mice (Cftr_/_) and control

littermates (Cftr_/_) after a normal or 0.5% UDCA chow diet for 3 wk. ND, not detectable. Mann-Whitney U-test: *P<

0.05, difference between genotype (Cftr_/_ vs. Cftr_/_); †P <0.05, difference between treatment group (non-UDCA vs.

UDCA).

UDCA treatment decreased the biliary hydrophobicity of Cftr-/-

mice. After UDCA treatment,

the biliary bile salt composition changes extensively (Table 1.) After treatment the bile salt

composition consisted for more than ~80% of UDCA in both Cftr-/-

mice and controls (Figure

3A). UDCA treatment drastically reduced the fraction of cholate in the bile to below 5% in

both Cftr-/-

mice and controls. However, the fraction of β-muricholate was also reduced in Cftr-

/- and controls mice compared to the untreated animals (6.5±1.8 vs. 8.7±1.6% resp.; NS).

Taken together, UDCA treatment decreased the bile salt hydrophobicity index of Cftr-/-

mice to

wild type levels (-0.08±0.003 vs. -0.08±0.002 resp.; NS; Figure 3B). To evaluate the potential

hepatotoxic effects of UDCA and differences in biliary bile hydrophobicity we measured

plasma ALT levels in during normal diet and in UDCA treated animals (Figure 3C). Under

control diet conditions, although not significant, ALT was higher in CF than in control mice

(40±8 vs. 60±13 U/l resp.; NS). UDCA decreased ALT both in CF and control mice (31±9 vs.

Ursodeoxycholate modulates bile flow and bile salt pool independently from Cftr in mice

123

6

15±1 U/l resp.; NS). UDCA treatment significantly lowered plasma ALT levels in Cftr-/-

mice

compared to Cftr-/-

on control diet (15±1 vs. 60±13 U/l resp.; p<0.05).

Cftr+/+

Cftr-/-

Cftr+/+

Cftr-/-

0

25

50

75

100

Cholate

OthersUDCA

Normal diet UDCA Diet

A

%

-0.10

-0.05

0.00

0.05

0.10

Cftr+/+

Cftr-/-

*

B

Normal diet UDCA diet

Hyd

rop

ho

bic

ity i

nd

ex

Normal diet UCDA diet0

25

50

75

100

Cftr+/+

Cftr-/-

*

C

AL

T (

U/l

)

Figure 3 Bile salt composition after chronic dietary UDCA administration. (A) Percent contribution of the bile salts

cholate, ursodeoxcholate and others (chenodeoxycholate, deoxycholate, α-muri cholate and β-muricholate) in bile, (B)

Heuman index of total bile salts in bile representing the hydrophobicity of bile salts, (C) Alanine aminotransferase

(ALT) in plasma of Cftr-/- and control littermates (Cftr+/+) after a normal or 0.5%-UDCA chow diet for 3 weeks. Data are

presented as means ± SEM of 4-9 mice per group. *P-value<0.05.

Chapter 6

124

6

*

Normal diet UCDA diet0

10

20

30

40

Cftr+/+

Cftr-/-

*

A

Ch

ola

te p

oo

l siz

e

( m

ol/

100 g

ram

s B

W)

Normal diet UCDA diet0

5

10

15

20

Cftr+/+

Cftr-/-

*

B

Ch

ola

te s

yn

thesis

rate

( m

ol/

day/1

00 g

ram

BW

)

Figure 4 Cholate pool size and cholate synthesis after chronic dietary UDCA administration. Cftr knockout mice (Cftr-/-)

and control littermates (Cftr+/+) were intravenously injected with 2H4-cholate after a normal or 0.5%-UDCA chow diet

for 3 weeks Enrichment of the administered 2H4-cholate was determined in plasma until 72 hours after the

administered label. From the plasma decay curve, cholate pool size (A) and cholate synthesis rate (B) were calculated,

as detailed in the Materials and Methods. Data are presented as means ± SEM of 4-6 mice per group. *P-value<0.05.

UDCA treatment decreased cholate synthesis and pool size in Cftr-/-

and control mice. Non-

UDCA treated Cftr-/-

mice had a higher cholate synthesis rate (16±1 vs. 10±2 μmol/day/100

gram BW resp.; p<0.05) and larger cholate pool size (28±3 vs. 19±1 μmol/100 gram BW, resp.;

p<0.05) compared to controls (Figure 4).

Chronic UDCA treatment reduced the cholate pool size by ~90% in both Cftr-/-

and control

mice (p<0.05). Nevertheless, the pool size of Cftr-/-

mice remained higher compared with

controls (3.6±0.6 vs. 1.5±0.3 µmol/100 grams BW, resp.; p<0.05). Under the assumption that

steady state conditions of the bile salt kinetics were obtained after 3 weeks of treatment we

found that UDCA decreased the cholate synthesis rate by ~85% in both Cftr-/-

mice and

controls compared to untreated mice (p<0.05; Fig. 4B). Interestingly, UDCA treatment

straightened out the difference in synthesis rate between Cftr-/-

and control mice (2.3±0.3 vs.

2.2±0.4 μmol/hour/100 gram BW, resp.; NS).

DISCUSSION

The major physiologic effect of UDCA is its capacity to increase bile flow. This property

supports the therapeutic use of UDCA in a variety of cholangiopathies, including CFLD (35). In

vitro and ex vivo studies indicated that the stimulatory of UDCA on cholangiocyte secretion

depends on the presence of CFTR (14, 15). In the present study, we demonstrated that UDCA,

in vivo, either during acute or chronic administration, induced a significant Cftr independent

increase of bile flow in mice. Therefore, our results indicate that a positive choleretic effect of

UDCA can also to be expected in CF conditions. UDCA treatment reduced the relative

Ursodeoxycholate modulates bile flow and bile salt pool independently from Cftr in mice

125

6

hydrophobic biliary bile salt composition in Cftr-/-

mice by replacing the high percent

contribution of the bile salt cholate by UDCA and by the quantitative reduction of the cholate

pool size. These properties could contribute to the assumed beneficial effects of UDCA in

CFLD.

Our present in vivo results are in apparent contrast with in vitro and ex vivo studies, which

report on the interaction between Cftr and bile salt stimulated biliary secretion (14, 15). In

these studies, an important role is ascribed to the function of calcium induced chloride

channels in UDCA stimulated bile flow, through the induction of purinergic signaling via Cftr

dependent ATP release. There can be several possibilities underlying the divergence of our in

vivo results from the in vitro and ex vivo reported studies. First, the choleretic effect found in

our in vivo studies probably predominantly reflects an osmotic, bile salt induced canalicular

bile flow, rather than a major ductular bile flow. The canalicular bile flow may thereby

predominate the Cftr dependent secretion effect of UDCA at the level of the cholangiocytes.

Second, the activation of Cftr independent routes may differ between the in vitro and in vivo

conditions. In vivo, UDCA may stimulate hepatocytes to secrete ATP into bile in a CFTR-

independent manner and subsequently induce bicarbonate secretion via paracrine purinergic

pathways linked to calcium-activated chloride channels (CaCC) in the cholangiocytes (36).

CaCCs have been suggested to play a more prominent role in epithelial fluid secretion in mice

than in other species, including pigs (37). This might therefore explain the relatively mild

phenotype in CF mouse models (38). Recently, Beuers et al. postulated the “bicarbonate

umbrella hypothesis”, by suggesting that biliary bicarbonate secretion in humans serves to

maintain an alkaline pH near the apical surface of hepatocytes and cholangiocytes in order to

prevent the uncontrolled membrane permeation of protonated glycine-conjugated bile(39).

In this concept, the bile acid receptor TGR5 (GPBAR-1), localized on the tip of the cilia of

apical cholangiocyte membranes in mice and humans could trigger a Cftr independent

signaling cascade resulting in cholangiocyte secretion.

During chronic UDCA administration, a CF biliary phenotype became apparent: an increased

bile salt secretion rate (Fig. 2C, p<0.05). The difference in secretion rate is based on the

product of two measured parameters. Since the majority of biliary bile salts are derived from

enterohepatic circulation, the most plausible explanation is an increased total BS pool size,

which during treatment is predominantly accounted for by UDCA (Fig 3A). It seems therefore

logical to assume that the total amount of bile salts in the enterohepatic circulation is

increased in CF mice. Indeed, this explanation seems to be supported by the ~50% higher

cholate pool size in CF mice fed a normal diet (Figure 5).

During bile salt treatment, however, CF and control mice were administered the same,

supraphysiological dosages of TUDCA and UDCA in the acute and chronic experiment,

respectively. We previously reported an increased fecal loss of bile salts in the Cftr-/-

mice

(40). Although UDCA could potentially influence intestinal fat malabsorption we could not find

an effect of UDCA treatment on the body weight in Cftr-/-

or control mice. Furthermore we

Chapter 6

126

6

recently reported in rats that UDCA does not influence fecal fat excretion(41). Therefore, our

results suggest a different bile salt kinetic steady state in CF mice, in which overcompensation

of bile salt synthesis results in an enlargement of the bile salt pool. Since the higher bile salt

secretion rate in CF mice was seen most prominently during chronic treatment, we speculate

that CFTR is involved in adaptation of the enterohepatic circulation to chronic bile salt

treatment, either at the level of the intestine, of the liver, or both.

Although differences exist in bile salt metabolism between mice and man (42), we found clear

similarities between CF patients and Cftr-/-

mice in bile composition. Untreated Cftr-/-

mice (i.e.

without UDCA treatment) have a more hydrophobic bile salt pool composition and an

increased cholate synthesis rate and pool size compared to control mice. This is in line with

the report by Freudenberg et al of the increased biliary hydrophobicity in their Cftr508/508

mice

model. Similar results have been found in CF patients: Strandvik et al. reported an increased

proportion of primary bile salts in serum and bile of CF, which has been attributed to an

enhanced bile salt biosynthesis in response to increased fecal bile salt disposal (43). The Cftr-/-

mice used in the present study are therefore comparable to human CF patients in this

respect.

The relatively hydrophobic bile salt composition in CF has been implied in the development of

liver disease in CF conditions, but definitive proof for this concept is (still) lacking (44).

Nevertheless, chronic UDCA treatment is apparently capable to partially correct the

hydrophobic bile salt profile in CF mice. Additionally UDCA treatment normalized the initially

increased liver function tests in CF mice, in agreement with our hypothesis.

We investigated the effect of UDCA under CF conditions in the absence of CF-related liver

disease (CLFD). The advantage of this approach is the ability to exclusively examine UDCA

effects on several bile salt parameters during CFTR deficient conditions, without possible

interference of secondary changes due to liver disease. The major induction of bile flow and

alterations in bile salt composition could contribute to a preventive role of UDCA on the

development of CLFD. It is unclear whether effects on bile flow and bile salt composition can

be found under conditions of CFLD. Nakagawa et. al. evaluated duodenal bile salt composition

after a two month UDCA-treatment in nine CF patients with CFLD (45). Similar to our results,

the percent contribution of UDCA increased, resulting in a more hydrophilic bile salt pool. The

contribution of UDCA was not as high as in our study (25%, compared with ~80% in the

present study), possibly due to the relatively low dosage of UDCA that was used (10-15

mg/kg/BW/day). In patients with primary sclerosing cholangitis UDCA treatment dosage of

25-30 mg/kg/day resulted in UDCA comprising 74% or the bile salt pool, with a corresponding

cholate reduction from 29% to 6% (46). In addition, a reduction in cholate pool size and

cholate synthesis is also described in gallstone patients treated with 750 mg UDCA (47).

Therefore, regarding the effect of UDCA on bile composition, we have no indication that the

observed effects will be absent during CF-related cholestasis or other signs of CFLD.

Ursodeoxycholate modulates bile flow and bile salt pool independently from Cftr in mice

127

6

Figure 5 Schematic representation of Cftr dependent effects of UDCA treatment on bile salt kinetics. Schematic

representation of the enterohepatic circulation of the primary bile salt cholate in Cftr knockout mice (Cftr-/-) and

control littermates, either untreated (top) or treated for 3 weeks with a 0.5%-UDCA chow diet for 3 weeks. Cholate

undergoes enterohepatic cycling (dotted lines). In the liver bile salt are excreted via the bile into the intestine, and

then almost completely reabsorbed at the level of the terminal ileum. Under steady state conditions the fecal cholate

loss is compensated for by de novo cholate synthesis in the liver as represented by the width of the black arrows

coming from the liver (µmol/day/100 gram BW), maintaining the total cholate pool size in equilibrium. The diameters

of the circles represent the magnitude of the pool size (μmol/100 gram BW). Cholate synthesis rate and pool size are

determined as detailed in the Materials and Methods. Data are presented as means ± SEM of 4-6 mice per group.

In conclusion, our results in mice in vivo indicate that UDCA exerts a choleretic effect and

influences the bile salt profile and synthesis, independent of the presence of functional CFTR.

When extrapolated to the human situation, this might imply that UDCA treatment results

both in CF and in non-CF individuals in increased choleresis, reduced bile salt synthesis and a

more hydrophilic bile salt pool. Interpretation of the present results for the human (CF)

condition needs to take into account the possibility of species specificity of the observed

effects. However the outcome of this study does provides a firm experimental basis to explain

the beneficial effects of UDCA observed in CF patients

Chapter 6

128

6

REFERENCE LIST

1. Collins FS. Cystic fibrosis: Molecular biology and therapeutic implications. Science. 1992

05/08;256(5058):774-9.

2. Kerem B, Rommens JM, Buchanan JA, Markiewicz D, Cox TK, Chakravarti A, et al.

Identification of the cystic fibrosis gene: Genetic analysis. Science. 1989

09/08;245(4922):1073-80.

3. Stutts MJ, Canessa CM, Olsen JC, Hamrick M, Cohn JA, Rossier BC, et al. CFTR as a cAMP-

dependent regulator of sodium channels. Science. 1995 08/11;269(5225):847-50.

4. Lamhonwah AM, Bear CE, Huan LJ, Chiaw PK, Ackerley CA, Tein I. Cystic fibrosis

transmembrane conductance regulator in human muscle: Dysfunction causes abnormal

metabolic recovery in exercise. Ann Neurol. 2010;67(6):802-8.

5. Li H, Ganta S, Fong P. Altered ion transport by thyroid epithelia from CFTR−/− pigs suggests

mechanisms for hypothyroidism in cystic fibrosis. Exp Physiol. 2010;95(12):1132-44.

6. Lindblad A, Hultcrantz R, Strandvik B. Bile-duct destruction and collagen deposition: A

prominent ultrastructural feature of the liver in cystic fibrosis. Hepatology. 1992 08;16(2):372-

81.

7. Colombo C, Battezzati PM, Crosignani A, Morabito A, Costantini D, Padoan R, et al. Liver

disease in cystic fibrosis: A prospective study on incidence, risk factors, and outcome.

Hepatology. 2002 12;36(6):1374-82.

8. Nash KL, Allison ME, McKeon D, Lomas DJ, Haworth CS, Bilton D, et al. A single centre

experience of liver disease in adults with cystic fibrosis 1995-2006. J Cyst Fibros. 2008

05;7(3):252-7.

9. Fitz JG, Basavappa S, McGill J, Melhus O, Cohn JA. Regulation of membrane chloride

currents in rat bile duct epithelial cells. J Clin Invest. 1993 01;91(1):319-28.

10. Hofmann AF. Biliary secretion and excretion in health and disease: Current concepts. Ann

Hepatol. 2007 01;6(1):15-27.

11. Lindor K, Therneau T, Jorgensen R, Malinchoc M, Dickson E. Effects of ursodeoxycholic acid

on survival in patients with primary biliary cirrhosis. Gastroenterology. 1996;110(5):1515-8.

12. Lindor K, Jorgensen R, Anderson M, Gores G, Hofmann A, LaRusso N. Ursodeoxycholic acid

and methotrexate for primary sclerosing cholangitis: A pilot study. Am J Gastroenterol.

1996;91(3):511-5.

13. Cheng K, Ashby D, Smyth R. Ursodeoxycholic acid for cystic fibrosis-related liver disease.

Cochrane Database Syst Rev. 2000(2):CD000222.

Ursodeoxycholate modulates bile flow and bile salt pool independently from Cftr in mice

129

6

14. Shimokura GH, McGill JM, Schlenker T, Fitz JG. Ursodeoxycholate increases cytosolic

calcium concentration and activates cl- currents in a biliary cell line. Gastroenterology. 1995

09;109(3):965-72.

15. Fiorotto R, Spirli C, Fabris L, Cadamuro M, Okolicsanyi L, Strazzabosco M. Ursodeoxycholic

acid stimulates cholangiocyte fluid secretion in mice via CFTR-dependent ATP secretion.

Gastroenterology. 2007 11;133(5):1603-13.

16. Hofmann AF. Chemistry and enterohepatic circulation of bile acids. Hepatology.

1984;4(S2):4S-14S.

17. Benedetti A, Alvaro D, Bassotti C, Gigliozzi A, Ferretti G, La Rosa T, et al. Cytotoxicity of bile

salts against biliary epithelium: A study in isolated bile ductule fragments and isolated

perfused rat liver. Hepatology. 1997;26(1):9-21.

18. Frenkiel PG, Lee D, Cohen H, Gilmore C, Resser K, Bonorris G, et al. The effect of diet on bile

acid kinetics and biliary lipid secretion in gallstone patients treated with ursodeoxycholic acid.

Am J Clin Nutr. 1986;43(2):239.

19. Beuers U, Spengler U, Zwiebel FM, Pauletzki J, Fischer S, Paumgartner G. Effect of

ursodeoxycholic acid on the kinetics of the major hydrophobic bile acids in health and in

chronic cholestatic liver disease. Hepatology. 1992;15(4):603-8.

20. Colombo C, Setchell KDR, Podda M, Crosignani A, Roda A. Effects of ursodeoxycholic acid

therapy for liver disease associated with cystic fibrosis*. J Pediatr. 1990;117(3):482-9.

21. Scholte BJ, Davidson DJ, Wilke M, de Jonge HR. Animal models of cystic fibrosis. J Cyst

Fibros. 2004 08;3 Suppl 2:183-90.

22. Hulzebos CV, Renfurm L, Bandsma RH, Verkade HJ, Boer T, Boverhof R, et al. Measurement

of parameters of cholic acid kinetics in plasma using a microscale stable isotope dilution

technique: Application to rodents and humans. J Lipid Res. 2001;42(11):1923.

23. Durie PR, Kent G, Phillips MJ, Ackerley CA. Characteristic multiorgan pathology of cystic

fibrosis in a long-living cystic fibrosis transmembrane regulator knockout murine model. Am J

Pathol. 2004 04;164(4):1481-93.

24. Freudenberg F, Broderick AL, Yu BB, Leonard MR, Glickman JN, Carey MC.

Pathophysiological basis of liver disease in cystic fibrosis employing a DeltaF508 mouse model.

Am J Physiol Gastrointest Liver Physiol. 2008 06;294(6):G1411-20.

25. Ratcliff R, Evans MJ, Cuthbert AW, MacVinish LJ, Foster D, Anderson JR, et al. Production of

a severe cystic fibrosis mutation in mice by gene targeting. Nat Genet. 1993 05;4(1):35-41.

26. Kuipers F, Van Ree J, Hofker MH, Wolters H, Veld G, Havinga R, et al. Altered lipid

metabolism in apolipoprotein e–deficient mice does not affect cholesterol balance across the

liver. Hepatology. 1996;24(1):241-7.

Chapter 6

130

6

27. Verkade HJ, Havinga R, Shields DJ, Wolters H, Bloks VW, Kuipers F, et al. The

phosphatidylethanolamine N-methyltransferase pathway is quantitatively not essential for

biliary phosphatidylcholine secretion. J Lipid Res. 2007 09;48(9):2058-64.

28. Colombo C, Setchell KDR, Podda M, Crosignani A, Roda A. Effects of ursodeoxycholic acid

therapy for liver disease associated with cystic fibrosis*. J Pediatr. 1990;117(3):482-9.

29. Mashige F, Imai K, Osuga T. A simple and sensitive assay of total serum bile acids. Clin

Chim Acta. 1976 07/01;70(1):79-86.

30. Kok T, Hulzebos CV, Wolters H, Havinga R, Agellon LB, Stellaard F, et al. Enterohepatic

circulation of bile salts in FXR-deficient mice: Efficient intestinal bile salt absorption in the

absence of ileal bile acid-binding protein (ibabp). J Biol Chem. 2003;278(43):41930-7.

31. Heuman D, Hylemon P, Vlahcevic Z. Regulation of bile acid synthesis. III. correlation

between biliary bile salt hydrophobicity index and the activities of enzymes regulating

cholesterol and bile acid synthesis in the rat. J Lipid Res. 1989;30(8):1161-71.

32. Stellaard F, Langelaar S, Kok R, Jakobs C. Determination of plasma bile acids by capillary

gas-liquid chromatography-electron capture negative chemical ionization mass

fragmentography. J Lipid Res. 1989;30(10):1647.

33. Mitchell JC, Stone BG, Duane WC. Measurement of bile acid synthesis in man by release of

14 CO 2 from [26-14 C] cholesterol: Comparison to isotope dilution and assessment of

optimum reference cholesterol specific activity. Lipids. 1992;27(1):68-71.

34. Stellaard F, Brufau G, Boverhof R, Jonkers EZ, Boer T, Kuipers F. Developments in bile acid

kinetic measurements using 13C and 2H: 105 times improved sensitivity during the last 40

years†. Isotopes Environ Health Stud. 2009;45(4):275-88.

35. Paumgartner G, Beuers U. Ursodeoxycholic acid in cholestatic liver disease: Mechanisms of

action and therapeutic use revisited. Hepatology. 2002 09;36(3):525-31.

36. Minagawa N, Nagata J, Shibao K, Masyuk AI, Gomes DA, Rodrigues MA, et al. Cyclic AMP

regulates bicarbonate secretion in cholangiocytes through release of ATP into bile.

Gastroenterology. 2007;133(5):1592-602.

37. Clarke LL, Grubb BR, Yankaskas JR, Cotton CU, McKenzie A, Boucher RC. Relationship of a

non-cystic fibrosis transmembrane conductance regulator-mediated chloride conductance to

organ-level disease in cftr(-/-) mice. Proc Natl Acad Sci U S A. 1994 01/18;91(2):479-83.

38. Liu X, Luo M, Zhang L, Ding W, Yan Z, Engelhardt JF. Bioelectric properties of chloride

channels in human, pig, ferret, and mouse airway epithelia. American journal of respiratory

cell and molecular biology. 2007;36(3):313.

Ursodeoxycholate modulates bile flow and bile salt pool independently from Cftr in mice

131

6

39. Beuers U, Hohenester S, de Buy Wenniger LJM, Kremer AE, Jansen PLM, Elferink RPJO. The

biliary HCO3- umbrella: A unifying hypothesis on pathogenetic and therapeutic aspects of

fibrosing cholangiopathies. Hepatology. 2010;52(4):1489-96.

40. Bijvelds MJC, Bronsveld I, Havinga R, Sinaasappel M, de Jonge HR, Verkade HJ. Fat

absorption in cystic fibrosis mice is impeded by defective lipolysis and post-lipolytic events. Am

J Physiol Gastrointest Liver Physiol. 2005 04;288(4):G646-53.

41. Cuperus FJC, Hafkamp AM, Havinga R, Vitek L, Zelenka J, Tiribelli C, et al. Effective

treatment of unconjugated hyperbilirubinemia with oral bile salts in gunn rats.

Gastroenterology. 2009;136(2):673,682. e1.

42. Hofmann AF, Hagey LR, Krasowski MD. Bile salts of vertebrates: Structural variation and

possible evolutionary significance. J Lipid Res. 2010;51(2):226.

43. Strandvik B, Einarsson K, Lindblad A, Angelin B. Bile acid kinetics and biliary lipid

composition in cystic fibrosis. J Hepatol. 1996 07;25(1):43-8.

44. Smith JL, Lewindon PJ, Hoskins AC, Pereira TN, Setchell KDR, O'Connell NC, et al.

Endogenous ursodeoxycholic acid and cholic acid in liver disease due to cystic fibrosis.

Hepatology. 2004;39(6):1673-82.

45. Nakagawa M, Colombo C, Setchell KDR. Comprehensive study of the biliary bile acid

composition of patients with cystic fibrosis and associated liver disease before and after UDCA

administration. Hepatology. 1990;12(2):322-34.

46. Sinakos E, Marschall HU, Kowdley KV, Befeler A, Keach J, Lindor K. Bile acid changes after

high‐dose ursodeoxycholic acid treatment in primary sclerosing cholangitis: Relation to

disease progression. Hepatology. 2010;52(1):197-203.

47. Hofmann AF. Medical treatment of cholesterol gallstones by bile desaturating agents.

Hepatology. 1984;4(S2):199S-208S.

Chapter 6

132

6

CHAPTER 7

SUMMARY AND GENERAL

DISCUSSION

Chapter 7

134

7

GENERAL SUMMARY AND DISCUSSION

In this thesis, we studied clinical and basic aspects of the hepatic and gastrointestinal

involvement in cystic fibrosis. We focused on the physiological and pathological role of bile

salts and the enterohepatic circulation in Cystic fibrosis conditions. From our clinical study,

we learned that, based on the involvement of gamma-GT (GGT), cirrhotic Cystic fibrosis liver

disease (CCFLD) is primarily a biliary disease since GGT elevation is a predictor for the

development of CCFLD. From our experiments in CF mice models we conclude that cystic

fibrosis related liver disease is not related to biliary bile salt cytotoxicity. Nevertheless, in

different CF mice models we could establish a CFTR dependent and bile salt specific, biliary

phenotype. We provided proof that this biliary phenotype can, partially, be corrected by the

hydrophilic bile salt UDCA. Finally, we demonstrated that the CF phenotype in mice includes

alterations in the metabolism and enterohepatic circulation of bile salts. These changes are

due to interaction between the CFTR protein and the bacterial flora of the gut. Our findings

provide new experimental evidence for the significant role of CFTR in the equilibrium of the

enterohepatic circulation. Impaired CFTR function, like in Cystic fibrosis disease conditions,

can therefore give rise to negative alterations in the enterohepatic circulation and bile salt

metabolism. Our results provide new understanding in the gastro-intestinal and hepatic role

of CFTR that provides new insights for development of future treatment options in Cystic

fibrosis.

CCFLD is a life-threatening, hepatic complication of CF (1, 2). The clinical presentation of

CCFLD is often characterized by severe portal hypertension and gastrointestinal variceal

bleeding (3). In patients with end stage CCFLD liver transplantation might be indicated (4, 5).

Therefore patients at risk for or still in early stages of the disease could benefit from

preventive treatment for CCFLD. However, to date no predictors to identify patients at risk for

CCFLD were available. In chapter 2, we report that, in the time period 2 years prior to the

diagnosis of CCFLD, an increase of GGT, even if the level remains within the normal range,

indicates a significantly increased risk for the development CCFLD. This finding provides new

opportunities for identifying high risk patient for CCFLD and more targeted preventive or

preemptive therapy strategies like UDCA.

Within a few years after the CFTR gene was identified in humans, the first genetically

engineered CF mouse models were developed (6-9). These models have been of great

importance for increasing the understanding Cftr physiology and pathology in various organ

systems. However, the phenotype of the developed CF mice models differs in several aspects

from the human CF disease equivalent. For example lung disease, the most prominent CF

disease feature in CF patients, is barely found in mice models (10). On the other hand, CF

mouse models exhibit a wide range of gastrointestinal phenotypes that closely resemble the

human CF condition (11). In chapter 3, we reviewed current knowledge on the hepatic and

Summary and general discussion

135

7

gastrointestinal CF phenotypes in mouse models and in human patients. We focused on

chronic intestinal fat malabsorption, present in many CF patients. We indentified useful

similarities in the phenotype between mice and man. For example, the comparable

phenotype of increased fecal bile salt loss and alterations in bile salt composition in both CF

mice and CF patients (12). Also, the intestinal mucosal abnormalities like small intestinal

bacterial overgrowth (SIBO), intestinal mucus accumulation, intestinal inflammation and

increased intestinal permeability have corresponding phenotypes in CF humans and mice(13,

14). From these findings we concluded that the CF mouse, despite differences in phenotypic

expression among different genetic mice backgrounds, can well serve as useful experimental

model for gastrointestinal CF phenomena. Based on this analysis, our studies in CF mice

models could provide an educated basis for future translational research.

We investigated to what extent CF mouse models can provide pathophysiological and

mechanistic understanding in the development of CF related hepatic disease (chapter 4-6).

Our main focus was to determine whether biliary bile salt cytotoxicity is involved in the

development of CFLD like disease in CF mice. To answer this question we analyzed the liver

involvement in various CF mouse models. In the Cftr-/-tm1Unc

CF mice, a model characterized by

spontaneous development of CFLD like disease, we could not establish a contribution of

biliary bile salt cytotoxicity to its hepatic histo-pathologic phenotype. On the other hand, in

other CF mouse models we did find a significant increase in the hydrophobic profile of the

biliary bile. However, even in these mice models, this increased biliary cytotoxicity could not

be related to CFLD like disease. Finally, we could not prove that CF mice, when challenged

with and prolonged hydrophobic bile salt exposure, were more susceptible to develop CFLD

like disease. Based on these combined results we conclude that the development of CFLD like

disease, at least in mice, is not related to Cftr related changes in biliary cytotoxicity. We

postulate that this conclusion can be extrapolated to the development of liver involvement in

humans with CF.

Although we could not establish a specific histo-pathological phenotype in mice that was

corresponding with CCFLD in humans, we did find a unique hepatic and biliary phenotype.

(chapter 4). Cftr-/-

mice, when exposed to the hydrophobic bile salt cholate, over an extended

period of time, display an increased hydrophobicity of bile and a decreased capacity to

increase bile flow. Cftr-/-

mice are apparently less able to dilute the bile during chronic cholate

administration. This is clearly illustrated by a significantly increased biliary BS concentration

but unaffected BS secretion rates. We observed a distinct Cftr dependent liver growth and

proliferation response, on prolonged hydrophobic BS exposure: in wild type mice the liver

weight increased by ~50% upon a chronic BS administration, whereas liver weight was stable

in different CF mouse models. These observations point to a, hitherto unrecognized, role of

CFTR in the hepatic adaptive response to prolonged hydrophobic BS exposure. We speculate

that, this impaired regenerative capability of the CF phenotype, could be related to an

increased hepatic vulnerability, for example in cholestatic conditions. This deficiency in

regenerative hepatic capacity may play a role in liver disease development.

Chapter 7

136

7

Our results in Cftr-/-tm1Unc

mice (chapter 5) demonstrated Cftr dependent differences in

intestinal bile salt metabolism. We found, Cftr specific, alterations in the bacterial formation

of the secondary bile salts and accordingly in the fecal and biliary bile salt composition. As a

result we found that in Cftr-/-tm1Unc

mice, exclusively fed a liquid diet, the biliary bile salt profile

was considerably more hydrophilic and the bile flow was increased compared to wild type

mice. The increase in bile production, in the Cftr-/-tm1Unc

, could be explained by increased

concentrations of ursocholate (UCA). The contribution of UCA to the bile salt pool is usually

relatively small, due to its rapid biotransformation into DCA by the intestinal bacterial flora.

Therefore, the reported increased UCA enrichment of the Cftr-/-tm1Unc

mice can only be

explained by, specific Cftr related, changes in the functioning of intestinal bacterial flora. UCA

is a very hydrophilic and highly choleretic bile salt. Therefore, UCA enrichment in the CftrUnc-/-

mice, results in a significantly increased bile salt dependent bile flow and in a significantly

decreased biliary bile hydrophobicity. Our results provide a clear example of the way in which

alterations in the interaction between intestinal flora and bile salt formation can significantly

affect the enterohepatic circulation of bile salt. In CF conditions these changes, in bile salt

metabolism, could be related to its phonotypical characteristics or even result in induction or

prevention of liver disease development. Several intestinal mucosal abnormalities are

described in CF patients and CF mice models that could influence the composition of the

bacterial micro flora. These abnormalities include accumulation of viscous and sticky mucus,

small intestinal bacterial over growth, increased intestinal permeability and inflammation of

the small intestine (15-24).

The importance of the intestinal bacterial flora for the CF gastrointestinal phenotype was also

illustrated by other experiments. We treated control and CF mice with either the hydrophilic

bile salt UDCA or the hydrophobic bile salt cholate (CA). We found that control mice, on CA

diet, displayed a slightly more hydrophobic bile composition compared to the Cftr-/-

mice. This

effect, in wild type mice, can, in its entirety, be explained by the presence of the secondary

bile salt deoxycholate (DCA), not found in bile of Cftr-/-

mice. DCA is produced exclusively, via

biotransformation, by the intestinal micro flora. Recently, we and others have reported that

intestinal micro flora and the intestinal mucosal permeability are changed in CF (25-28). Our

results emphasize the critical role of the intestinal micro flora, for CF conditions, in modifying

the enterohepatic circulation of bile salts. We believe that, Cftr specific, genotypic changes in

intestinal flora are primary responsible for a majority of the, CF related, alterations in bile salt

composition and production. These observations open the perspective for targeted

manipulation of the intestinal micro flora in CF mice and patients.

As stated, the Cftr dependent changes in bile salt composition and metabolism can have

profound effects on the enterohepatic circulation in CF conditions. Yet, to some extent, the

observed changes in bile salt formation can also result from, Cftr related, alterations in the

enterohepatic circulation itself. The reported difference in bacterial BS biotransformation, in

particular the conversion CA to DCA, can indirectly affect hepatic bile salt synthesis. For

example DCA is a strong activating ligand of the farnesoid X receptor (FXR) (29). This nuclear

Summary and general discussion

137

7

receptor is highly expressed in the liver and ileum (30). In the liver, an important function of

FXR activation is the suppression of bile acid synthesis from cholesterol (31). Activation of

intestinal FXR induces the expression of FGF19 (fibroblast growth factor 19, equivalent to

Fgf15 in rodents) by the enterocytes (32). FGF19 is excreted and transported via the portal

circulation to the liver. Here it can induce the cascade of bile salt synthesis from cholesterol.

Additionally, recent publications describe the central role of FGF15/19 in hepatocyte

proliferation (33, 34). Normally FGF15/19 is only expressed in the terminal ileum and not in

normal liver. However, hepatic expression of FGF15/19 can occur in cholestatic conditions

and is associated with hepatic adaptation aimed at protecting against BS injury (35).

In our chronic hydrophobic bile salt (cholate) exposure model, expression of Fgf15 could

potentially be induced both in the intestine and in the liver. The increased biliary fraction of

DCA, found in the controls after chronic CA diet, can induce Fgf15 expression via FXR in the

enterocytes or directly by inducing Fgf15 expression in the liver. In either situation, induction

of Fgf15 could induce liver proliferation and liver growth (36). There could also be a primary

role for the intestinal flora. A recent study in germ free mice showed that gut microbiota do

regulate bile salt metabolism by influencing naturally occurring FXR antagonists like tauro-

beta-muricholic acid (37). On the other hand cholate induced liver growth in mice, similar to

our current findings in controls, was described previously. Huang et al. reported that CA

feeding (0.2%) for 5 days stimulated normal liver growth and increased the relative liver

weight by approximately 30% in wild type mice (38). These investigators also observed that

CA supplementation enhanced liver regeneration after partial hepatectomy. They concluded

that normal liver regeneration depends on BS activation via nuclear receptor-dependent

signaling pathways, in particular FXR.

The potential role of FXR is underlined by our finding that liver growth is not observed during

chronic UDCA exposure. UDCA does not lead to increased presence of strong FXR ligands like

DCA. Current findings offer opportunities for future research. To differentiate the flora effect

from the bile salt effect in regulating FXR it would be of value to develop germ free CF mice to

repeat the experiments. Furthermore it would be of great interest to perform a cholate diet

experiment in FXR intestinal and liver tissue specific knockout mice models. These models

provide the opportunity to separate the intestinal FXR stimulation from the hepatic FXR

effects on liver growth.

UDCA as a preventive treatment option for cirrhotic CFLD remains controversial (39). An

important factor in this argument is the lack of support by basic experimental evidence.

Additionally there or no well-designed clinical study showing unequivocally a long term

clinical benefit on relevant clinical outcome (40). A major physiological effect of UDCA is its

capacity to increase bile flow. This property has supported the therapeutic use of UDCA in a

variety of cholangiopathies, including CFLD (41). In vitro and ex vivo studies indicated,

however that the stimulatory role of UDCA on cholangiocyte secretion depends on the

presence of CFTR (42, 43). In the present thesis (chapter 6), we studied the effects of UDCA in

Chapter 7

138

7

vivo in CF mice. We demonstrated that UDCA, in vivo, either during acute or chronic

administration, induced a significant, Cftr independent, increase of bile flow in mice.

Therefore, our results suggest that a positive choleretic effect of UDCA can also be expected

in human CF conditions. Furthermore, UDCA treatment reduced the relative hydrophobic

biliary bile salt composition in Cftr-/-

mice. In bile UDCA replaced cholate and reduced the

quantitative total pool size of cholate. These properties could contribute to the assumed

beneficial effects of UDCA in CFLD. However, we would like to underline that our results do

not provide evidence for preventive effects of UDCA on the development of liver fibrosis.

We feel that the results of our studies lead to an improved understanding of the development

of Cystic fibrosis liver disease. Our results provide new opportunities for future clinical and

basic research. This is especially relevant given the current groundbreaking therapeutic

developments in CF. These new therapies are target to directly correct the basic genetic CFTR

defect (44). To explore these new therapeutic options further, early intervention in young

children, who have not yet developed severe pulmonary disease, would be most favorable

(45). Classic lung function measurements, the traditional clinical endpoint in CF clinical

research, may not be the most indicative in young patients still without a measurable

decrease pulmonary function. To perform future therapeutic studies new clinical endpoints

and intermediate endpoints for Cystic fibrosis disease need to be developed (46).

The results of his thesis offer additional options for the development of new gasto-intestinal

and hepatic outcome parameters for future clinical trials (47). For example measuring

differences in fecal bile secretion or changes in bile salt metabolism might have high potential

as clinical outcome variables. Total bile salt secretions, as well as differences in fecal bile salt

formation, are relatively easily obtainable results and do not require an invasive approach.

These measurements are feasible to achieve even in young children. Another example that

open new research opportunities could be the ever growing possible study methods to

identify and quantify the intestinal bacterial micro flora. The CFTR specific changes in the

interaction of the intestinal bacterial flora with intestinal bile salt metabolism and absorption

as reported in this thesis can serve as the basis for further research in this direction.

Moreover, our results proved new opportunities for treatment options for CCFLD. Based on

our results, it appears justifiable to evaluate the prognostic value of GGT to identify patients

at risk for severe cirrhotic CFLD in a large prospective cohort study. This study can than

provide the necessary validation and support for clinical use and interventional therapeutic

strategies. Simultaneously, therapeutic interventional and observational (registry) studies

could provide additional candidate parameters on gastrointestinal function. These could

serve as new parameters to evaluate either the severity of the gastrointestinal phenotype in

general or for the development of CCFLD in particular.

In addition to clinical, patient-oriented studies, we feel that continuation of the basic studies

in relevant model systems is essential. Concerning the search for the pathogenesis of CCFLD

we believe the focus should switch from bile salt cytotoxicity to the liver-intestine-micro flora

Summary and general discussion

139

7

interaction. The results presented in this thesis provide strong support for the concept that

CFTR dependent changes in the intestinal bacterial flora composition are strongly related to

intestinal inflammation and intestinal bile salt metabolism. Focus on pathophysiology of small

intestinal bacterial overgrowth (SIBO) and of the intestinal flora in general in CF seems

indicated. The role of intestinal paneth cells, secreting anti-bacterial proteins into the

intestinal lumen, may also be relevant to delineate in CF conditions.

Another line of logical continuation of our current research involves the hepatic-intestinal axis

and the enterohepatic circulation of bile salt in CF conditions. Our results show a distinct CFTR

dependent phenotype of bile salt synthesis, intestinal bile salt biotransformation and fecal

bile salt loss. This interaction of CFTR function with the enterohepatic circulation of bile salts

including the interaction with the intestinal bacterial flora needs further exploration.

Finally there are new developments in possibilities to study different CF animal models.

Recently CF ferrets and CF pigs were generated and characterized in experimental studies (48,

49). The firsts result of these studies show that, in particular the CF pigs, demonstrate a

hepatic CFLD like phenotype (50). These new models provide opportunities to further study

CFLD development and treatment.

In conclusion we believe that our research provides key additional expertise in the field of

gastro-intestinal and hepatic consequence of Cystic fibrosis disease. We feel that our present

results, in combination with the current promising developments in CF drug development, the

expansion of global patient registries and the development of new experimental animal

models, add to improve the understanding and the treatment for CF disease in the near

future.

Chapter 7

140

7

REFERENCE LIST

1. Leeuwen L, Fitzgerald DA, Gaskin KJ. Liver disease in cystic fibrosis. Paediatric Respiratory

Reviews. 2013.

2. Lindblad A, Glaumann H, Strandvik B. Natural history of liver disease in cystic fibrosis.

Hepatology. 1999 11;30(5):1151-8.

3. Debray D, Lykavieris P, Gauthier F, Dousset B, Sardet A, Munck A, et al. Outcome of cystic

fibrosis-associated liver cirrhosis: Management of portal hypertension. J Hepatol.

1999;31(1):77-83.

4. Chryssostalis A, Hubert D, Coste J, Kanaan R, Burgel P, Desmazes-Dufeu N, et al. Liver

disease in adult patients with cystic fibrosis: A frequent and independent prognostic factor

associated with death or lung transplantation. J Hepatol. 2011;55(6):1377-82.

5. Rowland M, Gallagher CG, Ó'Laoide R, Canny G, Broderick A, Hayes R, et al. Outcome in

cystic fibrosis liver disease. Am J Gastroenterol. 2010;106(1):104-9.

6. Kerem B, Rommens JM, Buchanan JA, Markiewicz D, Cox TK, Chakravarti A, et al.

Identification of the cystic fibrosis gene: Genetic analysis. Science. 1989

09/08;245(4922):1073-80.

7. Riordan JR, Rommens JM, Kerem B, Alon N, Rozmahel R, Grzelczak Z, et al. Identification of

the cystic fibrosis gene: Cloning and characterization of complementary DNA. Science.

1989;245(4922):1066.

8. Ratcliff R, Evans MJ, Cuthbert AW, MacVinish LJ, Foster D, Anderson JR, et al. Production of

a severe cystic fibrosis mutation in mice by gene targeting. Nat Genet. 1993 05;4(1):35-41.

9. van Doorninck JH, French PJ, Verbeek E, Peters R, Morreau H, Bijman J, et al. A mouse model

for the cystic fibrosis delta F508 mutation. EMBO J. 1995;14(18):4403.

10. Wilke M, Buijs-Offerman RM, Aarbiou J, Colledge WH, Sheppard DN, Touqui L, et al. Mouse

models of cystic fibrosis: Phenotypic analysis and research applications. J Cyst Fibros.

2011;10:S152-71.

11. Durie PR, Kent G, Phillips MJ, Ackerley CA. Characteristic multiorgan pathology of cystic

fibrosis in a long-living cystic fibrosis transmembrane regulator knockout murine model. Am J

Pathol. 2004 04;164(4):1481-93.

12. Kalivianakis M, Bronsveld I, Jonge de H, Sinaasappel M, Havinga R, Kuipers F, et al.

Increased fecal bile salt excretion is independent of the presence of dietary fat malabsorption

in two mouse models for cystic fibrosis. Proefschrift Mini Kalivianalis. 2003.

13. Fridge JL, Conrad C, Gerson L, Castillo RO, Cox K. Risk factors for small bowel bacterial

overgrowth in cystic fibrosis. J Pediatr Gastroenterol Nutr. 2007;44(2):212-8.

Summary and general discussion

141

7

14. Norkina O, Burnett TG, De Lisle RC. Bacterial overgrowth in the cystic fibrosis

transmembrane conductance regulator null mouse small intestine. Infect Immun.

2004;72(10):6040-9.

15. Van Elburg RM, Uil JJ, Van Aalderen WMC, Mulder CJJ, Heymans HSA. Intestinal

permeability in exocrine pancreatic insufficiency due to cystic fibrosis or chronic pancreatitis.

Pediatr Res. 1996;39(6):985-91.

16. Lewindon P, Robb T, Moore DJ, Davidson GP, Martin AJ. Bowel dysfunction in cystic

fibrosis: Importance of breath testing. J Paediatr Child Health. 1998;34(1):79-82.

17. Lisowska A, Wójtowicz J, Walkowiak J. Small intestine bacterial overgrowth is frequent in

cystic fibrosis: Combined hydrogen and methane measurements are required for its detection.

Acta Biochim Pol. 2009;56(4):631.

18. Bruzzese E, Raia V, Gaudiello G, Polito G, Buccigrossi V, Formicola V, et al. Intestinal

inflammation is a frequent feature of cystic fibrosis and is reduced by probiotic administration.

Aliment Pharmacol Ther. 2004;20(7):813-9.

19. Raia V, Maiuri L, de Ritis G, de Vizia B, Vacca L, Conte R, et al. Evidence of chronic

inflammation in morphologically normal small intestine of cystic fibrosis patients. Pediatr Res.

2000;47(3):344-50.

20. Sbarbati A, Bertini M, Catassi C, Gagliardini R, Osculati F. Ultrastructural lesions in the

small bowel of patients with cystic fibrosis. Pediatr Res. 1998;43(2):234-9.

21. Smyth RL, Croft NM, O'Hea U, Marshall TG, Ferguson A. Intestinal inflammation in cystic

fibrosis. Arch Dis Child. 2000;82(5):394-9.

22. Quigley EM, Quera R. Small intestinal bacterial overgrowth: Roles of antibiotics, prebiotics,

and probiotics. Gastroenterology. 2006;130(2):S78-90.

23. Norkina O, Burnett TG, De Lisle RC. Bacterial overgrowth in the cystic fibrosis

transmembrane conductance regulator null mouse small intestine. Infect Immun.

2004;72(10):6040.

24. De Lisle RC, Roach E, Jansson K. Effects of laxative and N-acetylcysteine on mucus

accumulation, bacterial load, transit, and inflammation in the cystic fibrosis mouse small

intestine. American Journal of Physiology-Gastrointestinal and Liver Physiology.

2007;293(3):G577-84.

25. De Lisle RC, Roach EA, Norkina O. Eradication of small intestinal bacterial overgrowth in

the cystic fibrosis mouse reduces mucus accumulation. J Pediatr Gastroenterol Nutr.

2006;42(1):46-52.

26. De Lisle RC, Mueller R, Boyd M. Impaired mucosal barrier function in the small intestine of

the cystic fibrosis mouse. J Pediatr Gastroenterol Nutr. 2011;53(4):371-9.

Chapter 7

142

7

27. Wouthuyzen-Bakker M, Bijvelds MJC, de Jonge HR, De Lisle RC, Burgerhof JGM, Verkade

HJ. Effect of antibiotic treatment on fat absorption in mice with cystic fibrosis. Pediatr Res.

2011;71(1):4-12.

28. Martin CR, Zaman MM, Ketwaroo GA, Bhutta AQ, Coronel E, Popov Y, et al. CFTR

dysfunction predisposes to fibrotic liver disease in a murine model. American Journal of

Physiology-Gastrointestinal and Liver Physiology. 2012;303(4):G474-81.

29. Wang H, Chen J, Hollister K, Sowers LC, Forman BM. Endogenous bile acids are ligands for

the nuclear receptor FXR/BAR. Mol Cell. 1999;3(5):543-53.

30. Vaquero J, Monte MJ, Dominguez M, Muntané J, Marin JJG. Differential activation of the

human farnesoid X receptor depends on the pattern of expressed isoforms and the bile acid

pool composition. Biochem Pharmacol. 2013 10/1;86(7):926-39.

31. Goodwin B, Jones SA, Price RR, Watson MA, McKee DD, Moore LB, et al. A regulatory

cascade of the nuclear receptors FXR, SHP-1, and LRH-1 represses bile acid biosynthesis. Mol

Cell. 2000;6(3):517-26.

32. Kim I, Ahn SH, Inagaki T, Choi M, Ito S, Guo GL, et al. Differential regulation of bile acid

homeostasis by the farnesoid X receptor in liver and intestine. J Lipid Res. 2007

12;48(12):2664-72.

33. Wu X, Ge H, Lemon B, Vonderfecht S, Weiszmann J, Hecht R, et al. FGF19-induced

hepatocyte proliferation is mediated through FGFR4 activation. J Biol Chem.

2010;285(8):5165-70.

34. Wu AL, Coulter S, Liddle C, Wong A, Eastham-Anderson J, French DM, et al. FGF19

regulates cell proliferation, glucose and bile acid metabolism via FGFR4-dependent and

independent pathways. PloS one. 2011;6(3):e17868.

35. Schaap FG, van der Gaag NA, Gouma DJ, Jansen PLM. High expression of the bile

salt‐homeostatic hormone fibroblast growth factor 19 in the liver of patients with extrahepatic

cholestasis. Hepatology. 2009;49(4):1228-35.

36. Uriarte I, Fernandez-Barrena MG, Monte MJ, Latasa MU, Chang HC, Carotti S, et al.

Identification of fibroblast growth factor 15 as a novel mediator of liver regeneration and its

application in the prevention of post-resection liver failure in mice. Gut. 2013;62(6):899-910.

37. Sayin S, Wahlström A, Felin J, Jäntti S, Marschall H, Bamberg K, et al. Gut microbiota

regulates bile acid metabolism by reducing the levels of tauro-beta-muricholic acid, a naturally

occurring FXR antagonist. Cell Metabolism. 2013 2/5;17(2):225-35.

38. Huang W, Ma K, Zhang J, Qatanani M, Cuvillier J, Liu J, et al. Nuclear receptor-dependent

bile acid signaling is required for normal liver regeneration. Science. 2006

04/14;312(5771):233-6.

Summary and general discussion

143

7

39. Ooi CY, Nightingale S, Durie PR, Freedman SD. Ursodeoxycholic acid in cystic fibrosis-

associated liver disease. Journal of Cystic Fibrosis. 2012 1;11(1):72-3.

40. Cheng K, Ashby D, Smyth RL. Ursodeoxycholic acid for cystic fibrosis-related liver disease.

Cochrane Database Syst Rev. 2012;10.

41. Paumgartner G, Beuers U. Ursodeoxycholic acid in cholestatic liver disease: Mechanisms of

action and therapeutic use revisited. Hepatology. 2002 09;36(3):525-31.

42. Fiorotto R, Spirli C, Fabris L, Cadamuro M, Okolicsanyi L, Strazzabosco M. Ursodeoxycholic

acid stimulates cholangiocyte fluid secretion in mice via CFTR-dependent ATP secretion.

Gastroenterology. 2007 11;133(5):1603-13.

43. Shimokura GH, McGill JM, Schlenker T, Fitz JG. Ursodeoxycholate increases cytosolic

calcium concentration and activates cl- currents in a biliary cell line. Gastroenterology. 1995

09;109(3):965-72.

44. Ramsey BW, Davies J, McElvaney NG, Tullis E, Bell SC, Dřevínek P, et al. A CFTR potentiator

in patients with cystic fibrosis and the G551D mutation. N Engl J Med. 2011 11/03;

2012/09;365(18):1663-72.

45. Ashlock MA, Olson ER. Therapeutics development for cystic fibrosis: A successful model for

a multisystem genetic disease. Annu Rev Med. 2011;62:107-25.

46. De Boeck K, Bulteel V, Tiddens H, Wagner T, Fajac I, Conway S, et al. Guideline on the

design and conduct of cystic fibrosis clinical trials: The european cystic fibrosis Society–Clinical

trials network (ECFS-CTN). Journal of Cystic Fibrosis. 2011;10:S67-74.

47. Rowe SM, Borowitz DS, Burns JL, Clancy JP, Donaldson SH, Retsch-Bogart G, et al. Progress

in cystic fibrosis and the CF therapeutics development network. Thorax. 2012;67(10):882-90.

48. Rogers CS, Stoltz DA, Meyerholz DK, Ostedgaard LS, Rokhlina T, Taft PJ, et al. Disruption of

the CFTR gene produces a model of cystic fibrosis in newborn pigs. Science.

2008;321(5897):1837-41.

49. Sun X, Sui H, Fisher JT, Yan Z, Liu X, Cho H, et al. Disease phenotype of a ferret CFTR-

knockout model of cystic fibrosis. J Clin Invest. 2010;120(9):3149.

50. Stoltz DA, Rokhlina T, Ernst SE, Pezzulo AA, Ostedgaard LS, Karp PH, et al. Intestinal CFTR

expression alleviates meconium ileus in cystic fibrosis pigs. J Clin Invest. 2013;123(6):2685.

Chapter 7

144

7

CHAPTER 8

NEDERLANDSE DISCUSSIE EN

SAMENVATTING

Chapter 8

146

8

NEDERLANDSE SAMENVATTING

Dit proefschrift beschrijft onze klinische en basale experimentele onderzoeken naar de

hepatologische- en gastro-intestinale aspecten van de ziekte Cystic fibrosis (CF). We hebben

ons hierbij in het bijzonder gericht op de fysiologische en pathologische rol van galzouten en

de enterohepatische circulatie. In onze klinische studie beschrijven we de vroege verhoging

van serum gamma-GT (GGT) bij de ontwikkeling van de ernstige cirrotische vorm van Cystic

fibrosis leverziekte (CCFLD). Deze resultaten ondersteunen een rol voor het meten en

vervolgen van GGT verhoging als een voorspeller voor de ontwikkeling van CCFLD. Uit onze

experimenten in verschillende CF muismodellen kunnen we concluderen dat Cystic fibrosis

gerelateerde leverziekte niet is geassocieerd met een veranderde cytotoxiciteit van biliaire

galzoutsamenstelling. Toch konden we, op basis van experimenten in CF muismodellen, wel

een Cftr specifiek, biliair fenotype vaststellen. We tonen aan dat dit biliaire fenotype,

gedeeltelijk, kan worden gecorrigeerd door het toedienen van het hydrofiele galzouten

UDCA. Tot slot hebben we aangetoond dat het CF fenotype bij muizen wordt gekenmerkt

door specifiek veranderingen in het metabolisme en de enterohepatische circulatie van

galzouten. Deze variaties worden veroorzaakt door de interactie van het Cystic fibrosis

transmembrane conductance regulator (CFTR) eiwit en de intestinale bacteriële flora. Onze

bevindingen bieden nieuwe experimenteel bewijs voor de belangrijke rol van CFTR in het

equilibrium van de enterohepatische kringloop van galzouten. Een verminderde of afwezig

CFTR functie, zoals bij de ziekte Cystic fibrosis, kan dan daardoor aanleiding geven

tot pathologische veranderingen in de enterohepatische circulatie en het galzouten

metabolisme. Onze resultaten geven nieuwe inzichten in belangrijke de gastro-intestinale en

hepatologisch rol van CFTR. Deze nieuwe inzichten bieden een perspectief voor het verdere

begrip en behandeling van de ziekte Cystic fibrosis.

CCFLD is een levensbedreigende complicatie van CF (1, 2). De kliniek van CCFLD wordt vaak

gekenmerkt door ernstige portale hypertensie en hoge kans op gastro-intesintale

varicesbloedingen (3). In sommige patiënten met CCFLD is levertransplantatie als therapie

noodzakelijk (4, 5). De piekleeftijd voor de presentatie van CCFLD ligt rond de leeftijd van 10

jaar. CCFLD kent een “stille” preklinische fase waarin de litteken vorming (fibrose) in de lever

zich geleidelijk, in een periode van jaren, ontwikkelt tot cirrose. Patiënten met een verhoogd

risico voor CCFLD of patiënten in een vroege fase van de ziekte zouden daarom theoretische

kunnen profiteren van preventieve behandeling. Echter op dit moment zijn er geen testen

beschikbaar om patiënten met een risico voor CCFLD te identificeren. In hoofdstuk 2,

beschrijven wij dat een toename in het serum GGT, vaak nog binnen de grenzen van normaal,

een indicator kan zijn voor de presentatie van CCFLD binnen 2 jaar. Deze bevingen biedt de

Nederlandse discussie en samenvatting

147

8

mogelijkheid om risico patiënten voor de ontwikkeling van CCFLD te identificeren en bij hen

preventieve behandeling, bijvoorbeeld met ursodeoxy cholzuur (UDCA) te overwegen.

Binnen een paar jaar nadat de het CFTR gen bij de mens was geïdentificeerd werden de eerste

genetische gemodificeerde muis modellen ontwikkeld. (6-9). Deze experimentele

diermodellen zijn van groot belang geweest bij de totstandkoming van het groeiende inzicht

in de fysiologie en pathologie van het functioneren van CFTR in verschillende orgaan

systemen. Echter het CF fenotype van de muis modellen verschillen op een aantal kenmerken

met de humane uitingsvormen van Cystic fibrosis. Zo wordt bijvoorbeeld longziekte of

chronische longschade, de belangrijkste ziekte uiting van CF bij de mens, bij muizen

nauwelijks gezien (10). Echter CF muis modellen presenteren wel verschillende gastro-

intestinale fenotypes die goed overeenkomen met het humane gastro-intestinale

ziektebeelden van CF (11). In hoofdstuk 3, geven wij een overzicht van de kennis in de

literatuur op het gebied van hepatologische en gastro-intestinale fenotypes van CF muis

modellen in vergelijking met het humane ziekte beeld van CF. We hebben ons hierbij

geconcentreerd op klinisch beeld van de persisterende intestinale vetmalabsorptie dat bij de

veel CF patiënten aanwezig is.

Galzouten spelen als emulgatoren een zeer belangrijke bij de intestinale vetabsorptie.

Opvallend hierbij is dat bij zowel CF patiënten als CF muizen op vergelijkbare manier sprake is

van een toegenomen fecaal verlies van galzouten en veranderingen in het

galzoutmetabolisme (12). Deze CF gerelateerde veranderingen op het gebied van

galzoutmetabolisme en galzouttransport zouden goed gerelateerd kunnen zijn aan de

beschreven persisterende vetmalabsorptie in CF. Andere gastro-intestinale fenotypische

overeenkomsten tussen CF patiënten en muismodellen zijn bijvoorbeeld het voorkomen van

intestinale bacteriële overgroei (“small intestinal bacterial overgrowth” of SIBO), intestinale

inflammatie en een toegenomen intestinale mucosale permeabiliteit (13, 14). Uit onze studie

concluderen we dan ook dat CF muizen een geschikt basaal experimenteel model kunnen

vormen voor studies naar gastro-intestinale CFTR gerelateerde, fenotypes.

Wij hebben onderzocht in welke mate CF muismodellen kunnen bijdragen aan het begrip over

de pathofysiologische oorzaak van CF gerelateerde leverziekte. (hoofdstuk 4-6). Ons

hoofddoel hierbij was om vast te stellen in welke mate de cytotoxiciteit van de biliaire

galzouten oorzakelijk betrokken is bij het ontstaan van CFLD. Om deze vraag te beantwoorden

onderzochten wij de pathologie en functioneren van de lever in verschillende type CF

muismodellen. Bij één van deze muis modellen (Cftr-/-tm1Unc)

, dat zich kenmerkt door de

spontane ontwikkeling van een op humaan CFLD lijkende leverziekte, konden wij geen relatie

vaststellen tussen de biliaire galzouttoxiciteit en het beschreven patho-histologisch beeld van

de lever. In andere CF muismodellen konden wij een wel significante toename vaststellen van

Chapter 8

148

8

het hydrofobe profiel van de biliaire gal samenstelling. Echter ook in deze muismodellen

konden wij deze toegenomen cytotoxiciteit niet relateren aan patho-histolgische

veranderingen in de lever. Ten slotte konden wij niet vaststellen dat, CF muismodellen, ook

niet na een langdurige blootstelling aan hydrofobe galzouten, meer gevoelig waren om een

vorm van CF gerelateerde leverziekte te ontwikkelen. Gebaseerd op deze resultaten

concluderen wij dat de ontwikkeling van CFLD, in ieder geval in muizen, niet gerelateerd is

aan CFTR gerelateerde veranderingen in de biliaire galzoutcytotoxiciteit. We veronderstellen

dat deze conclusie ook doorgetrokken kan worden naar de humane situatie en de

ontwikkeling van CFLD bij patiënten.

Hoewel wij niet de een direct overeenkomstig lever fenotype tussen mens en muis hebben

kunnen vaststellen, hebben wij wel een uniek CFTR afhankelijk hepatisch en biliair fenotype in

CF muizen kunnen aantonen (hoofdstuk 4). Wij vonden dat als Cftr-/-

muizen gedurende een

langere periode worden blootgesteld aan het hydrofobe galzout cholaat, zij een duidelijk

toegenomen hydrofobiciteit van de gal en een verminderd vermogen om de galproductie te

verhogen ontwikkelen. Hier uit blijkt dat Cftr-/-

muizen, in tegenstelling tot wild type muizen,

veel minder goed in staat zijn om de gal te verdunnen tijdens de blootstelling aan cholaat. Dit

blijkt ook duidelijk uit te toegenomen totale biliaire galzoutconcentratie bij basaal gelijk

blijvende galzoutsecretiesnelheden. Verder vonden wij een specifiek, Cftr gerelateerd, groei

effect van cholaat toediening op de levergroei en leverproliferatie. Hierbij zagen wij dat bij

wild type muizen de het levergewicht globaal 50% toenam tijdens de langdurige toediening

van cholaat. Dit in tegenstelling tot de verschillende type CF muizen waarbij, na langdurige

cholaat toediening het levergewicht gelijk bleef. Deze bevinding wijzen op, een tot nu toe

onbekende, rol van CFTR in the normale adaptieve leverrespons op langdurige blootstelling

aan hydrofobe galzouten. Wij veronderstellen dat de verminderde regeneratieve capaciteit

van het CF fenotype en rol kan spelen bij de ontwikkeling van CFLD als gevolg van een

toegenomen kwetsbaarheid van de lever in cholestatische condities.

Onze resultaten van de studies in de Cftr-/-tm1Unc

muizen (hoofdstuk 5) laten duidelijke, Cftr

afhankelijke verschillen zien in het intestinale galzoutmetabolisme. Wij vonden, onder

andere, Cftr specifieke, veranderingen in bacteriële vorming van secundaire galzouten en de

daarbij passende veranderingen in de biliaire en fecale galzoutsamenstelling. Bijvoorbeeld

zagen wij dat in Cftr-/-tm1Unc

muizen dat de biliaire galzoutsamenstelling veel hydrofieler was en

de galproductie veel hoger was in vergelijking met wild type muizen. Deze toename in

galproductie bij de CF muizen kon worden verklaard uit de aanwezigheid van verhoogde

concentraties van het zeer hydrofiele galzout ursocholaat (UCA). Normaal gesproken is de

proportie van UCA in de totale galzoutpool erg klein als gevolg van de zeer snelle omzetting

naar deoxycholaat (DCA) door de intestinale bacteriële flora. De, door ons gevonden, hoge

concentratie van UCA kan dan ook alleen ontstaan als gevolg van Cftr gerelateerde

Nederlandse discussie en samenvatting

149

8

veranderingen in het functioneren van de intestinale bacteriële flora. De hoge concentratie

en hoge hydrofobiciteit van UCA zijn de aanleiding voor de gevonden sterke toename in

galproductie en de lage biliaire hydrofobiciteit bij de Cftr-/-tm1Unc

muizen. Onze resultaten

vormen een duidelijk voorbeeld van de manier waarop, door de interactie tussen intestinale

bacteriële flora en het galzoutmetabolisme, de gehele enterohepatische kringloop van

galzouten kan worden beïnvloed. In de situatie van CF kunnen de veranderingen in het

galzoutmetabolisme gekoppeld zijn aan het induceren van leverziekte of misschien wel het

voorkomen hiervan. Bij CF patiënten en CF muismodellen worden diverse intestinale

mucosale veranderingen beschreven die van invloed zijn op de samenstelling van de

intestinale bacteriële flora. Hierbij kan bijvoorbeeld gedacht worden aan de beschreven

verandering in de accumulatie van viskeus en taai mucus secreet, SIBO, toegenomen

intestinale permeabiliteit en de aanwezigheid van intestinale inflammatie in CF condities(15-

24).

Het grote belang van de intestinale flora op het CF fenotype werd verder nog aangetoond in

één van onze andere experimenten. In deze studie behandelden wij CF muizen en wild type

muizen met of het hydrofiele galzout UDCA of het hydrofobe galzout cholaat. We vonden in

de wild type muizen met de cholaat behandeling er sprake was van een iets toegenomen

hydrofobiciteit in vergelijking met de Cftr-/-

muizen. Deze toename kon in zijn geheel worden

verklaard door de toegenomen aanwezigheid van het hydrofobe secundaire galzout

deoxycholaat (DCA), wat niet werd aangetoond in Cftr-/-

muizen. DCA wordt uitsluitend

geproduceerd door de intestinale bacteriële flora. Recent hebben wij en andere

gerapporteerd over de veranderde intestinale bacteriële flora en intestinale permeabiliteit in

CF condities (25-28). Onze resultaten onderstrepen de grote invloed van intestinale bacteriële

flora op de totale enteroheaptische kringloop in het bijzonder in situatie van CF. Wij zijn van

mening dat CF genotype specifieke veranderingen in de intestinale bacteriële flora primair

verantwoordelijk zijn voor meerderheid van de waargenomen variatie in de samenstelling van

de galzoutpool en de galproductie bij CF muizen. Onze resultaten bieden de mogelijkheden

om de functionele relevantie van de waarnemingen verder te beoordelen met behulp van

interventies gericht op de intestinale bacteriële flora in CF muismodellen en CF patiënten.

Zoals hiervoor aangegeven kunnen, Cftr afhankelijke, veranderingen in de galzoutpool

samenstelling gevolgen hebben op de enterohepatische circulatie in CF condities. Echter, tot

op zeker hoogte, kunnen de veranderingen in galzoutformatie ook het gevolg zijn van CFTR

afhankelijke veranderingen in de enterohepatische circulatie zelf. De gevonden

veranderingen in bacteriële biotransformatie van primaire naar secundaire galzouten, in het

bijzonder van cholaat naar deoxycholaat, kunnen indirect de galzoutsynthese beïnvloeden.

DCA is bijvoorbeeld een zeer krachtige activerende ligand van de farnesoid X receptor (FXR)

(29). Deze nucleaire receptor wordt in een hoge mate tot expressie gebracht in de lever en

Chapter 8

150

8

het ileum (30). In de lever speelt FXR een belangrijke rol bij de activatie van de

galzoutsynthese uit cholesterol(31). Activatie van intestinaal FXR leid tot expressie van

fibroblast growth factor 19 (FGF19) (32). FGF15/19 kan vervolgens vanuit de darm, via het

portale bloed worden getransporteerd naar de lever. Recente publicatie beschrijven de

centrale rol van Fgf15 (Fgf15 is het muis equivalent van het humane FGF19) bij hepatocyten

proliferatie en de regulatie van de systemische galzoutcompositie (33, 34). FGF15/19 worden

alleen tot expressie gebracht in de darm en normale gesproken niet in de lever. Echter

expressie van FGF15/19 in de lever kan voorkomen in cholestatische condities en is dan

geassocieerd met de hepatische adaptatie gericht op de bescherming tegen galzouttoxiciteit

(35).

In ons chronisch hydrofoob galzoutbelastingsmodel zou expressie van Fgf15 in zowel darm als

lever een rol kunnen spelen bij het toegenomen van lever gewicht in wild type muizen. De

verhoogde biliaire fractie van DCA, zoals gevonden in de wild type muizen, na langdurige

cholaat blootstelling, kan zo expressie van Fgf15, via intestinaal FXR kunnen induceren. Het

zou ook kunnen dat DCA direct de expressie van Fgf15 in de lever induceert. In beide situatie

kan Fgf15 expressie aanleiding geven tot hepatocyten proliferatie en levergroei (36). Er zou

hierbij ook nog een essentiële rol kunnen zijn voor de intestinale bacteriële flora. Een recente

studie in bacterie vrije (“germ free”) muismodel toonde aan darm bacteriën mede het

galzoutmetabolisme reguleren door beïnvloeding het natuurlijk voorkomende FXR

antagonistsen zoals tauro beta-muricholaat (37).

Cholaat gestimuleerde levergroei in muizen, vergelijkbaar met onze resultaten bij wild type

muizen, is eerder gerapporteerd. Huang et al. beschreven dat voeren van cholaat (0,2%)

gedurende 5 dagen, bij wild type muizen, de normale levergroei stimuleert en het relatieve

levergewicht verhoogd met gemiddeld 30% (38). Deze onderzoeker zagen verder dat cholaat

supplementatie leverregeneratie versnelde naar een partiële hepatectomie. Zij concluderen

hieruit dat de normale lever regeneratie afhankelijk is van galzoutstimulatie via nucleaire

receptoren zoals een aan FXR gekoppeld feedback mechanisme. De cruciale rol van FXR wordt

nog eens bevestigd door het feit dat langdurige blootstelling aan UDCA niet leidt tot

levergroei in wild type muizen. Dit is zeer waarschijnlijk gerelateerd aan het feit dat UDCA, in

tegenstelling tot DCA, niet functioneert als een ligand voor FXR. Onze huidige bevindingen

openen voor ons mogelijkheden voor toekomstig onderzoek. Om onderscheid te kunnen

maken tussen het rol van galzouten versus intestinale bacteriële flora zou het zeer waarde vol

kunnen zijn om een “germ free” CF muismodel te ontwikkelen en hierbij onze huidige

experimenten te herhalen. Tevens zou het zeer interessant kunnen zijn cholaat

blootstellingsexperimenten te herhalen in lever en darm specifiek FXR knockout muis

modellen. Deze modellen zouden de mogelijkheid bieden om de onderscheid te maken

tussen de rol van intestinaal en lever FXR bij de stimulatie van levergroei.

Nederlandse discussie en samenvatting

151

8

De behandeling van CFLD patiënten door middel van UDCA is een controversieel punt (39).

Een belangrijk aspect hierbij is de het ontbreken van basaal experimenteel bewijs over de

mogelijke werking van UDCA bij CFLD. Tevens ontbreken er goed ontworpen prospectieve

studies die onomstotelijk een gunstige lange termijn klinisch effect bewijzen en onderbouwen

(40). Een belangrijkste fysiologische eigenschappen van UDCA is dat de galproductie kan

verhogen. Deze eigenschap vormt de basis voor het therapeutisch gebruik van UDCA bij

diverse cholangiopathien, zo ook CFLD (41). In vitro en ex vivo studies hebben echter laten

zien dat het choleretisch effect van UDCA afhankelijk is van de aanwezigheid en de functie

van CFTR (42, 43). In onze huidige studie (hoofdstuk 6) onderzochten wij het effect van UDCA,

in vivo, in een CF (Cftr-/-

) muismodel. Wij toonden hierbij aan dat, in vivo, zowel bij acute als

bij langdurige toediening van UDCA er sprake is van een significante, Cftr onafhankelijke,

toename van de galproductie. Hiermee laten onze resultaten zien dat, ook in CF condities,

bijvoorbeeld bij patiënten, er ten minste een positieve choleretisch effect van UCA kan

worden verwacht. Verder zagen wij dat langdurige UDCA ook de hydrofobiciteit van de

galsamenstelling in CF muizen verlaagt. UDCA neemt hierbij de plaats in van cholaat en tevens

verlaagt UDCA de kwantitatieve grootte van de cholaat galzoutpool. Deze additionele

eigenschappen zouden nog verder kunnen bijdragen aan het vermeende positieve effect van

UDCA in de behandeling van CFLD. Wij willen echter onderstrepen dat onze resultaten geen

onderbouwing leveren voor eventuele gunstige effecten op de voorkoming van lever fibrose

of andere lange termijn gunstige effecten van UDCA in CF condities.

Wij zijn van mening dat de resultaten van onze studies leiden tot een verbeterd begrip van de

ontwikkeling van CF gerelateerde leverziekte. Onze studies bieden nieuwe mogelijkheden

voor toekomstig klinische en basaal experimenteel onderzoek. Dit positieve vooruitzicht met

betrekking tot mogelijkheden voor onderzoek is juist nu essentieel, in het licht van de huidige

baanbrekende ontwikkelingen op het gebied van de behandeling van CF (44). Om het effect

van deze nieuwe therapieën te optimaliseren is het theoretisch te verkiezen om de

behandeling al te starten bij jonge kinderen, nog zonder tekenen van chronische longschade

(45). Echter klassieke outcome parameters voor effect meting bij CF, zoals longfunctie

metingen, zullen echter weinig richtinggevend zijn bij patiënten nog zonder significante

longziekte. Om in de toekomst toch relevante klinische effectiviteitsstudie te kunnen

uitvoeren moeten er nieuwe klinische resultaten metingen ontwikkeld worden (46). De

gevonden en beschreven resultaten in dit proefschrift bieden aanvullende mogelijkheden om

nieuwe outcome gastro-intestinale en hepatologische outcome parameters te ontwikkelen

welke al toepasbaar zijn vanaf jonge leeftijd (47).

Zo hebben bijvoorbeeld het meten van de fecale galzoutexcretie en het meten van

veranderingen in het galzoutmetabolisme duidelijke potentie om te kunnen functioneren als

potentiële resultaat variabelen. De totale fecale galzoutsecretie en verandering in de

Chapter 8

152

8

galzoutcompositie zijn relatief eenvoudig te meten. Het zijn weinig invasieve onderzoeken,

die potentieel goed uitvoerbaar en haalbaar zijn, ook bij jonge kinderen. Verder bieden ook

de steeds maar in aantal toenemende onderzoeksmethodes voor het kwantificeren en

kwalificeren van de intestinale bacteriële flora grote potentie als toekomstige

onderzoeksmethodes. De door ons beschreven CFTR specifieke verandering met betrekking

tot de veranderde interactie tussen van de intestinale bacteriële flora met het

galzoutmetabolisme kunnen goed dienen als basis voor verder onderzoek in deze richting.

De resultaten zoals beschreven in dit proefschrift bieden nieuwe mogelijkheden voor de

behandeling van de cirrotische vorm van CFLD (CCFLD). Gebaseerd op onze resultaten is goed

te verantwoorden om, de prognostische waarde van het serum GGT als voorspeller van de

ontwikkeling van CCFLD, te onderzoeken in een groot prospectief cohort. Een dergelijke

studie bied de mogelijkheid om preventieve therapeutische interventies te evalueren op

effectiviteit. Gelijktijdig kunnen observationele (registratie) studie de mogelijkheden bieden

om nieuwe potentiele gasto-intestinale en hepatologische outcome resultaten te evalueren

en te valideren.

In aanvulling op klinische georiënteerde studies is het van het grootste belang is om ook

basaal experimenteel onderzoek in relevante experimentele modellen te continueren. Met

betrekking tot de ontwikkeling en de pathogenesis van CFLD zijn wij van mening dat de focus

van onderzoek moet worden verlegt van galzoutcytotoxiciteit naar de lever/intestinale

bacteriële flora interactie. De resultaten beschreven in dit proefschrift bieden een duidelijke

onderbouwing voor de relatie tussen de CFTR afhankelijke verandering in de intestinale

bacteriele flora, galzout metabolisme en daarmee ook potentieel met intestinale inflammatie.

Hiermee lijkt het goed te rechtvaardigen om de focus van onderzoek te richten op intestinale

bacteriële flora en bacteriële overgroei in CF condities. Hierbij zouden bijvoorbeeld de

intestinale panethcellen, betrokken van de secretie van anti-bacteriel eiwitten in het

intestinale lumen, ook bij CF een duidelijke rol kunnen spelen.

Gebaseerd op onze bevindingen zou een andere logische onderzoekslijn de intestinale-lever

as, in de regulatie van de enterohepatische kringloop en synthese van galzouten, in CF

condities zijn. Onze resultaten tonen een duidelijk CFTR afhankelijk fenotype van de

galzoutsynthese, de intestinale galzouttransformatie en het fecale galzoutverlies zien.

Tenslotte zijn er de mogelijkheden met de recent nieuw ontwikkelde CF diermodellen. Recent

werden het CF varken en de CF fretten beschreven als experimentele diermodellen voor

Cystic fibrosis. (48, 49). De eerste resultaten van deze studies, in het bijzonder bij het CF

varken, laten een duidelijk CFLD achtige fenotype zien(50). Deze bevindingen bieden

duidelijke mogelijkheden om nieuwe studies te doen naar de ontwikkeling van CFLD.

Nederlandse discussie en samenvatting

153

8

Concluderen kunnen wij stellen dat, onderzoeksresultaten in dit proefschrift, aanvullende

kennis bieden op het gebied van de gastro-intestinale en hepatologische aspecten van Cystic

fibrosis. Wij zijn van mening dat deze resultaten, in combinatie met de huidige ontwikkeling

van nieuwe curatieve therapieën voor CF, het gebruik van CF patiënten registratie data en de

ontwikkeling van de nieuwe diermodellen zullen leiden tot een toegenomen begrip van de

pathofysiologische mechanismen van CF en de behandeling van CF patiënten.

Chapter 8

154

8

REFERENTIELIJST

1. Leeuwen L, Fitzgerald DA, Gaskin KJ. Liver disease in cystic fibrosis. Paediatric Respiratory

Reviews. 2013.

2. Lindblad A, Glaumann H, Strandvik B. Natural history of liver disease in cystic fibrosis.

Hepatology. 1999 11;30(5):1151-8.

3. Debray D, Lykavieris P, Gauthier F, Dousset B, Sardet A, Munck A, et al. Outcome of cystic

fibrosis-associated liver cirrhosis: Management of portal hypertension. J Hepatol.

1999;31(1):77-83.

4. Chryssostalis A, Hubert D, Coste J, Kanaan R, Burgel P, Desmazes-Dufeu N, et al. Liver

disease in adult patients with cystic fibrosis: A frequent and independent prognostic factor

associated with death or lung transplantation. J Hepatol. 2011;55(6):1377-82.

5. Rowland M, Gallagher CG, Ó'Laoide R, Canny G, Broderick A, Hayes R, et al. Outcome in

cystic fibrosis liver disease. Am J Gastroenterol. 2010;106(1):104-9.

6. Kerem B, Rommens JM, Buchanan JA, Markiewicz D, Cox TK, Chakravarti A, et al.

Identification of the cystic fibrosis gene: Genetic analysis. Science. 1989

09/08;245(4922):1073-80.

7. Riordan JR, Rommens JM, Kerem B, Alon N, Rozmahel R, Grzelczak Z, et al. Identification of

the cystic fibrosis gene: Cloning and characterization of complementary DNA. Science.

1989;245(4922):1066.

8. Ratcliff R, Evans MJ, Cuthbert AW, MacVinish LJ, Foster D, Anderson JR, et al. Production of

a severe cystic fibrosis mutation in mice by gene targeting. Nat Genet. 1993 05;4(1):35-41.

9. van Doorninck JH, French PJ, Verbeek E, Peters R, Morreau H, Bijman J, et al. A mouse model

for the cystic fibrosis delta F508 mutation. EMBO J. 1995;14(18):4403.

10. Wilke M, Buijs-Offerman RM, Aarbiou J, Colledge WH, Sheppard DN, Touqui L, et al. Mouse

models of cystic fibrosis: Phenotypic analysis and research applications. J Cyst Fibros.

2011;10:S152-71.

11. Durie PR, Kent G, Phillips MJ, Ackerley CA. Characteristic multiorgan pathology of cystic

fibrosis in a long-living cystic fibrosis transmembrane regulator knockout murine model. Am J

Pathol. 2004 04;164(4):1481-93.

12. Kalivianakis M, Bronsveld I, Jonge de H, Sinaasappel M, Havinga R, Kuipers F, et al.

Increased fecal bile salt excretion is independent of the presence of dietary fat malabsorption

in two mouse models for cystic fibrosis. Proefschrift Mini Kalivianalis. 2003.

Nederlandse discussie en samenvatting

155

8

13. Fridge JL, Conrad C, Gerson L, Castillo RO, Cox K. Risk factors for small bowel bacterial

overgrowth in cystic fibrosis. J Pediatr Gastroenterol Nutr. 2007;44(2):212-8.

14. Norkina O, Burnett TG, De Lisle RC. Bacterial overgrowth in the cystic fibrosis

transmembrane conductance regulator null mouse small intestine. Infect Immun.

2004;72(10):6040-9.

15. Van Elburg RM, Uil JJ, Van Aalderen WMC, Mulder CJJ, Heymans HSA. Intestinal

permeability in exocrine pancreatic insufficiency due to cystic fibrosis or chronic pancreatitis.

Pediatr Res. 1996;39(6):985-91.

16. Lewindon P, Robb T, Moore DJ, Davidson GP, Martin AJ. Bowel dysfunction in cystic

fibrosis: Importance of breath testing. J Paediatr Child Health. 1998;34(1):79-82.

17. Lisowska A, Wójtowicz J, Walkowiak J. Small intestine bacterial overgrowth is frequent in

cystic fibrosis: Combined hydrogen and methane measurements are required for its detection.

Acta Biochim Pol. 2009;56(4):631.

18. Bruzzese E, Raia V, Gaudiello G, Polito G, Buccigrossi V, Formicola V, et al. Intestinal

inflammation is a frequent feature of cystic fibrosis and is reduced by probiotic administration.

Aliment Pharmacol Ther. 2004;20(7):813-9.

19. Raia V, Maiuri L, de Ritis G, de Vizia B, Vacca L, Conte R, et al. Evidence of chronic

inflammation in morphologically normal small intestine of cystic fibrosis patients. Pediatr Res.

2000;47(3):344-50.

20. Sbarbati A, Bertini M, Catassi C, Gagliardini R, Osculati F. Ultrastructural lesions in the

small bowel of patients with cystic fibrosis. Pediatr Res. 1998;43(2):234-9.

21. Smyth RL, Croft NM, O'Hea U, Marshall TG, Ferguson A. Intestinal inflammation in cystic

fibrosis. Arch Dis Child. 2000;82(5):394-9.

22. Quigley EM, Quera R. Small intestinal bacterial overgrowth: Roles of antibiotics, prebiotics,

and probiotics. Gastroenterology. 2006;130(2):S78-90.

23. Norkina O, Burnett TG, De Lisle RC. Bacterial overgrowth in the cystic fibrosis

transmembrane conductance regulator null mouse small intestine. Infect Immun.

2004;72(10):6040.

24. De Lisle RC, Roach E, Jansson K. Effects of laxative and N-acetylcysteine on mucus

accumulation, bacterial load, transit, and inflammation in the cystic fibrosis mouse small

intestine. American Journal of Physiology-Gastrointestinal and Liver Physiology.

2007;293(3):G577-84.

Chapter 8

156

8

25. De Lisle RC, Roach EA, Norkina O. Eradication of small intestinal bacterial overgrowth in

the cystic fibrosis mouse reduces mucus accumulation. J Pediatr Gastroenterol Nutr.

2006;42(1):46-52.

26. De Lisle RC, Mueller R, Boyd M. Impaired mucosal barrier function in the small intestine of

the cystic fibrosis mouse. J Pediatr Gastroenterol Nutr. 2011;53(4):371-9.

27. Wouthuyzen-Bakker M, Bijvelds MJC, de Jonge HR, De Lisle RC, Burgerhof JGM, Verkade

HJ. Effect of antibiotic treatment on fat absorption in mice with cystic fibrosis. Pediatr Res.

2011;71(1):4-12.

28. Martin CR, Zaman MM, Ketwaroo GA, Bhutta AQ, Coronel E, Popov Y, et al. CFTR

dysfunction predisposes to fibrotic liver disease in a murine model. American Journal of

Physiology-Gastrointestinal and Liver Physiology. 2012;303(4):G474-81.

29. Wang H, Chen J, Hollister K, Sowers LC, Forman BM. Endogenous bile acids are ligands for

the nuclear receptor FXR/BAR. Mol Cell. 1999;3(5):543-53.

30. Vaquero J, Monte MJ, Dominguez M, Muntané J, Marin JJG. Differential activation of the

human farnesoid X receptor depends on the pattern of expressed isoforms and the bile acid

pool composition. Biochem Pharmacol. 2013 10/1;86(7):926-39.

31. Goodwin B, Jones SA, Price RR, Watson MA, McKee DD, Moore LB, et al. A regulatory

cascade of the nuclear receptors FXR, SHP-1, and LRH-1 represses bile acid biosynthesis. Mol

Cell. 2000;6(3):517-26.

32. Kim I, Ahn SH, Inagaki T, Choi M, Ito S, Guo GL, et al. Differential regulation of bile acid

homeostasis by the farnesoid X receptor in liver and intestine. J Lipid Res. 2007

12;48(12):2664-72.

33. Wu X, Ge H, Lemon B, Vonderfecht S, Weiszmann J, Hecht R, et al. FGF19-induced

hepatocyte proliferation is mediated through FGFR4 activation. J Biol Chem.

2010;285(8):5165-70.

34. Wu AL, Coulter S, Liddle C, Wong A, Eastham-Anderson J, French DM, et al. FGF19

regulates cell proliferation, glucose and bile acid metabolism via FGFR4-dependent and

independent pathways. PloS one. 2011;6(3):e17868.

35. Schaap FG, van der Gaag NA, Gouma DJ, Jansen PLM. High expression of the bile

salt‐homeostatic hormone fibroblast growth factor 19 in the liver of patients with extrahepatic

cholestasis. Hepatology. 2009;49(4):1228-35.

Nederlandse discussie en samenvatting

157

8

36. Uriarte I, Fernandez-Barrena MG, Monte MJ, Latasa MU, Chang HC, Carotti S, et al.

Identification of fibroblast growth factor 15 as a novel mediator of liver regeneration and its

application in the prevention of post-resection liver failure in mice. Gut. 2013;62(6):899-910.

37. Sayin S, Wahlström A, Felin J, Jäntti S, Marschall H, Bamberg K, et al. Gut microbiota

regulates bile acid metabolism by reducing the levels of tauro-beta-muricholic acid, a naturally

occurring FXR antagonist. Cell Metabolism. 2013 2/5;17(2):225-35.

38. Huang W, Ma K, Zhang J, Qatanani M, Cuvillier J, Liu J, et al. Nuclear receptor-dependent

bile acid signaling is required for normal liver regeneration. Science. 2006

04/14;312(5771):233-6.

39. Ooi CY, Nightingale S, Durie PR, Freedman SD. Ursodeoxycholic acid in cystic fibrosis-

associated liver disease. Journal of Cystic Fibrosis. 2012 1;11(1):72-3.

40. Cheng K, Ashby D, Smyth RL. Ursodeoxycholic acid for cystic fibrosis-related liver disease.

Cochrane Database Syst Rev. 2012;10.

41. Paumgartner G, Beuers U. Ursodeoxycholic acid in cholestatic liver disease: Mechanisms of

action and therapeutic use revisited. Hepatology. 2002 09;36(3):525-31.

42. Fiorotto R, Spirli C, Fabris L, Cadamuro M, Okolicsanyi L, Strazzabosco M. Ursodeoxycholic

acid stimulates cholangiocyte fluid secretion in mice via CFTR-dependent ATP secretion.

Gastroenterology. 2007 11;133(5):1603-13.

43. Shimokura GH, McGill JM, Schlenker T, Fitz JG. Ursodeoxycholate increases cytosolic

calcium concentration and activates cl- currents in a biliary cell line. Gastroenterology. 1995

09;109(3):965-72.

44. Ramsey BW, Davies J, McElvaney NG, Tullis E, Bell SC, Dřevínek P, et al. A CFTR potentiator

in patients with cystic fibrosis and the G551D mutation. N Engl J Med. 2011 11/03;

2012/09;365(18):1663-72.

45. Ashlock MA, Olson ER. Therapeutics development for cystic fibrosis: A successful model for

a multisystem genetic disease. Annu Rev Med. 2011;62:107-25.

46. De Boeck K, Bulteel V, Tiddens H, Wagner T, Fajac I, Conway S, et al. Guideline on the

design and conduct of cystic fibrosis clinical trials: The european cystic fibrosis Society–Clinical

trials network (ECFS-CTN). Journal of Cystic Fibrosis. 2011;10:S67-74.

47. Rowe SM, Borowitz DS, Burns JL, Clancy JP, Donaldson SH, Retsch-Bogart G, et al. Progress

in cystic fibrosis and the CF therapeutics development network. Thorax. 2012;67(10):882-90.

Chapter 8

158

8

48. Rogers CS, Stoltz DA, Meyerholz DK, Ostedgaard LS, Rokhlina T, Taft PJ, et al. Disruption of

the CFTR gene produces a model of cystic fibrosis in newborn pigs. Science.

2008;321(5897):1837-41.

49. Sun X, Sui H, Fisher JT, Yan Z, Liu X, Cho H, et al. Disease phenotype of a ferret CFTR-

knockout model of cystic fibrosis. J Clin Invest. 2010;120(9):3149.

50. Stoltz DA, Rokhlina T, Ernst SE, Pezzulo AA, Ostedgaard LS, Karp PH, et al. Intestinal CFTR

expression alleviates meconium ileus in cystic fibrosis pigs. J Clin Invest. 2013;123(6):2685.

Curriculum vitae

159

CURRICULUM VITAE

Franciscus Albertus Johannes Andreas Bodewes was born on February 14th, 1966 in Eelde in

the Netherlands. He was third in a family of 3 children, with an older brother and sister. He

grew up and went to the Maria primary school in Eelde. After graduating from the

Gemeentelijke MAVO in Eelde, he finished his secondary school period at the Sint

Maartenscollege in Haren.

He went on to study Medicine at the University of Groningen. In his second year he was

offered the opportunity to join the Groningen liver transplant program student team headed

by pioneer transplant surgeon Professor Maarten Slooff.

After he had finished his Medicine study he started to work as a pediatric resident in the

Medical center in Leeuwarden and later on in the Sophia hospital in Zwolle. In 1996 he started

the Pediatrics residency program in University Hospital Groningen lead by, respectively,

Professor Jan Kimpen and Pieter Sauer. As part of this training he worked for two years in the

St Elisabeth Hospital, on the island of Curaçao in the Dutch Antilles. During his residency years

he worked for 3 years as a medical volunteer in holyday camps for Cystic fibrosis patients.

Here he learned about the severity and impacted of Cystic fibrosis diseaseon patients and

their families. After he had become a pediatrician he started his Pediatric gastroenterology,

hepatology and nutrition training in the Beatrix Children’s hospital of the University of

Groningen in de group of Charles Bijleveld.

Since he finished his pediatric GI training program in 2005 he is working as a senior staff

member/consultant in the department of Pediatric gastroenterology and hepatology of the

University medical center Groningen. His clinical focus is gastroenterology, hepatology, organ

transplantation and Cystic fibrosis. His main research topics concern the role of the CFTR

protein in the physiology and patho-physiology of bile formation and the development of

Cystic fibrosis related liver disease as well as the role of the CFTR protein in the physiology

and pathophysiology of intestinal fat absorption in CF patients and experimental animals. He

started this research project, concerning the role of bile salt in the development of Cystic

fibrosis liver disease, in the research group of Professor Henkjan Verkade.

Frank Bodewes married to Els van den Berg in 1997. She is program director for the 400th

anniversary of the University of Groningen. Together they have 3 children: Jelle, Silke and

Douwe.

Dankwoord/acknowledgments

161

DANKWOORD/ACKNOWLEDGMENTS

De totstandkoming van mijn promotieonderzoek heeft zich uitgestrekt over meerdere jaren.

In deze periode hebben vele mensen een bijdrage geleverd aan de studies, aan het schrijven

van de manuscripten en aan alle andere taken die komen kijken bij een promotie. Ik wil

hierbij alle mensen, die mij hebben gesteund en geholpen, van harte bedanken. Een aantal

mensen wil ik persoonlijk noemen vanwege de bijzondere bijdrage die zij hebben geleverd.

Henkjan, Prof. H.J. Verkade. Wat een weg hebben we samen afgelegd! De eerste keer dat ik

je ontmoette was, denk ik, 20 jaar geleden. Ik was student en deed een onderzoeksproject bij

Klary Niezen op Bloemsingel 1 en jij werkte daar op het lab als promovendus. De echte basis

van onze samenwerking werd gelegd in onze opleidingstijd in Groningen. Jij hoorde, toen ik in

1997 begon, al bij de meer senior arts assistenten. Je leerde mij dat je je naast het gewone

werk ook breder moest ontwikkelen en profileren. Als onderzoeker was ik je in Zwolle, waar ik

werkte als agnio, ook al tegengekomen. Je deed daar, samen met Edmond, onderzoek naar de

vetuitscheiding van premature neonaten. Je maakte grote indruk op mij met je verhaal dat je

complete luiers met babypoep oploste in chloroform/methanol om zo de fecale lipiden in

oplossing te brengen. Toen ik, in 2001, begon als staflid bij de kindergeneeskunde wilde ik mij

graag verder gaan specialiseren tot “Kinderarts-gastroenteroloog”. Daarvoor was het doen

van onderzoek een pre. Ik had interesse in Cystic fibrosis en jij wilde graag je lopende Cystic

fibrosis onderzoekslijn van Mini Kalivianakis continueren. Onze onderzoekssamenwerking was

geboren. Ik denk niet dat je toen wist wat je je op de hals haalde. Ik had geen

onderzoekservaring, zeker niet met basaal wetenschappelijk onderzoek en proefdieren.

Daarnaast ben ik een klein beetje dyslectisch. Eerst was het gewoon “onderzoek doen”, later

toen we wat langer bezig waren werd het toch steeds meer een onderzoekslijn en uiteindelijk

een promotietraject. Ik weet wel zeker dat ik jouw langstlopende promotietraject ben

geweest. Henkjan, ik ben je echt dankbaar. Schijnbaar ben je nooit het vertrouwen in mij en

mijn onderzoeksactiviteiten kwijt geraakt. Je behield altijd je positieve houding en bleef mij

voorhouden dat het eindpunt haalbaar was. In de loop van de jaren is onze band als collega’s

steeds hechter geworden. We kennen elkaars sterke punten en elkaars gebruiksaanwijzingen.

Maar het meest belangrijke is dat we respect voor elkaar hebben en eerlijk tegen elkaar

durven zijn. Ik denk met plezier terug aan de vele gezellige en leuke momenten tijdens de

buitenlandse congressen en andere leuke reisjes die we samen over de wereld hebben

gemaakt. De vorm en inhoud van onze samenwerking is in de loop der jaren steeds mee

veranderd met onze veranderende omstandigheden en verantwoordelijkheden. Het

vertrouwen is gebleven.

Hugo en Marcel. Het is in grote mate aan de samenwerking met jullie te danken dat we het

merendeel van de onderzoeken dit van dit proefschrift hebben kunnen uitvoeren. Dankzij

162

jullie wereldwijd befaamde goedlopende onderzoeksprogramma met de CF muismodellen en

jullie bijzondere kennis op dit vakgebied werden onze plannen mogelijk. Hugo volgens mij was

ik 4 jaar geleden al op jouw afscheidssymposium maar “you’re still going strong” met volgens

mij 10 keer meer energie als voor je “pensioen”. Dank voor je zeer deskundige, zeer

inhoudelijke en gedetailleerde commentaren, en dank dat je mijn copromotor wilt zijn.

Marcel bedankt voor de samenwerking, je adviezen en het meedenken. Ik waardeer je

nauwkeurigheid en je doorzettingsvermogen. Ik hoop dat we ook in de komende jaren nog

meer gezamenlijk projecten kunnen uitvoeren.

Willemien, Mariëtte en Marjan. Alle drie kwamen jullie als nieuwe jonge student-

onderzoekers een tijd meelopen op het CF project. Soms was het een beetje onduidelijk

situatie voor jullie als je “bij-begeleider” zelf nog met zijn promotieonderzoek bezig is. Alle

drie hebben jullie mij zeer geholpen met mijn onderzoek. Ieder voor zich hebben jullie een

groot stuk van de puzzel in dit CF onderzoeksproject gelegd. Het waren juist deze essentiële

stukken die maakten dat voor mij het eindpunt van promotie zichtbaar en haalbaar bleef.

Jullie zijn alle drie doorgegaan met je jullie eigen projecten. Allemaal zijn jullie al bij Henkjan

gepromoveerd met prachtige proefschriften. Jullie waren voor mij voorbeelden en

motivatoren om door te gaan.

Peter, Prof. Durie., You’re an Inspirator. I admire the way you make difficult thing sound easy

and the way you provide feedback without a single unkind word. I love your humor and your

generous smile. You’re the global pioneer and expert in Cystic fibrosis gastro-intestinal and

hepatic disease. For me it was an honor to have collaborated with you in our research project.

Thank you and get well soon.

Satti. Actually we met only once, at the NACFC meeting in Anaheim. I appreciate all the work

that you did for our collaborative “Toronto” research project. Thank you for your help and

effort.

Juul. Dank voor de zeer goede samenwerking en het vele werk dat je voor mij hebt gedaan.

Altijd was je bereid mijn vragen te beantwoorden of om mij te helpen. Ik bewonder je

positieve levenshouding en de manier waarop je andere mensen altijd met respect

behandeld.

Annette, Prof. Gouw. Onze samenwerking in dit project was kort maar essentieel. Ik

bewonder de manier waarop jij op en zeer ontspannende manier zeer nauwkeurige en hoog

kwalitatieve beoordeling gaf van de histologie van de muizenlevers. Ik herinner mij nog goed

je uitspraak: “wat heb je met deze muizen gedaan, die lever zien er echt niet normaal uit

hoor!”

Dankwoord/acknowledgments

163

Hans en Marcella. Het is echt geweldig dat jullie nu met Henkjan en mij samenwerken om dit

CF project te continueren en verder te brengen. Met zoveel extra intellect en talent moet dit

wel goed komen.

Alle collega’s van de Kinder-MDL. Renée, Patrick, Nicolette, Hubert, Henkjan en Edmond,

Robert, Hester, Els, Ekkehard en Charles. Bedankt voor de tijd en ruimte die ik gekregen heb

om aan mijn onderzoek en promotie te kunnen werken. In het bijzonder wil ik Hester noemen

met wie ik, als kamer- en lotgenoot, altijd met veel lol en humor, alle verzuchtingen van het

onderzoek en promotie heb mogen delen. Dank voor de steun, het delen van jullie

persoonlijke ervaringen op het vlak van promoveren en de bemoedigende woorden. Ik ben er

trots op om in Groningen bij de KMDL te werken. Ik ben echt een gelukkig mens als ik naar

jullie illustere rijtje namen kijk en ik oprecht kan zeggen dat ik, vooral dankzij jullie, elke dag

met plezier naar mijn werk ging en ga.

Medewerkers van het laboratorium Kindergeneeskunde. In het bijzonder de

“oudgedienden” Renze, Vincent, Henk en Rik. Ik voelde mij vaak als een vis op het droge in

het lab. Het was echt niet mijn natuurlijke habitat. Hoezeer ik veel zaken inhoudelijk zeer

interessant vond, is het werken op het lab en het doen van experimenten andere koek. Jullie

zijn allemaal bijzondere getalenteerde professionals op jullie eigen terrein. Ik wil jullie van

harte bedanken voor jullie tijd en de belangrijke bijdrages die jullie hebben geleverd aan mijn

experimenten.

Onderzoekers van het laboratorium Kindergeneeskunde. Vele van jullie heb ik in de loop der

jaren ontmoet en velen van jullie hebben mij geholpen. Vaak met praktische dingen en zeer

goede tips. De sfeer van collegialiteit en behulpzaamheid van de onderzoekers op het lab is

echt bijzonder.

Leiding en stafleden Beatrix kinderziekenhuis en het UMCG. Velen van jullie hebben

interesse getoond in mijn onderzoeks- en promotieactiviteiten. Harma en Leonie, mijn oud

kamergenoten die mochten meegenieten van de meelwormen die uit muizenvoer op mijn

bureau vielen. Bart en Marc voor het delen van hun promotie ervaringen. De “club van

promoverende stafleden” die, onder leiding van Pieter Sauer, samen kwam in tapas

restaurant “Cervantes”. De meesten van deze club zijn nu klaar of volgen op korte termijn.

Toch een geweldig resultaat.

Bertjan en Bas. Bedankt dat jullie, zonder een moment van twijfel en zonder ook maar een

idee te hebben van de taak en inhoud van jullie functie, mijn paranimfen willen zijn.

Familie vrienden en kennissen. Dank voor jullie ondersteuning en in het bijzonder jullie

oprechte interesse. Het delen van mijn activiteiten en plannen met andere gaf altijd weer

motivatie (en een stok achter de deur) om verder te gaan.

164

Els en Jelle, Silke, Douwe. Zonder jullie geen proefschrift. Zo leuk als ik mijn werk vind zoveel

te leuker vind ik het om naar huis te gaan en bij jullie te zijn. Hoe dynamisch ons gezinsleven

en huishouden vaak ook zijn, er is voor mij geen fijnere plek als bij jullie. Lieve Els dank voor je

onvoorwaardelijke steun, motivatie en begrip voor mijn promotie project. Ik ben er trots op

dat ik mijn promotie nog net binnen de eerste 400 jaar van het bestaan van de

Rijksuniversiteit Groningen heb kunnen afronden. Verder ben ik blij dat jouw en niet mijn

project de naam “For infinity” heeft gekregen. Heel veel succes met het lustrum. Jongens,

papa heeft nu weer lekker veel tijd voor “huiswerkbegeleiding” ;-) en andere leuke dingen

natuurlijk.

Frank Bodewes